HARVARD UNIVERSITY Library of the Museum of Comparative Zoology ThcWlsonBulktm PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY VOLUME 89 1977 QUARTERLY editor: JEROME A. JACKSON REVIEW editor: ROBERT RAIKOW COLOR PLATE EDITOR: WILLIAM A. LUNK EDITORIAL ASSISTANTS: BONNIE J. TURNER PATRICIA RAMEY WAYNE C. WEBER BETTE J. SCHARDIEN MARTHA HAYS President — Douglas A. James, Department of Zoology, University of Arkansas, Fayetteville, Arkansas 72703. First Vice-President — George A. Hall, Department of Chemistry, West Virginia Univer- sity, Morgantown, W. Va. 26506. Second Vice-President — Abbot S. Gaunt, Department of Zoology, Ohio State University, Columbus, Ohio 43210. Editor — Jerome A. Jackson, Department of Zoology, P.O. Drawer Z, Mississippi State Uni- versity, Mississippi State, Mississippi 39762. Secretary' — James Tate, Jr., P.O. Box 2043, Denver, Colorado 80201. Treasurer — Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. Elected Council Members — Sidney A. Gauthreaux, Jr. (term expires 1978) ; James R. Karr (term expires 1979) ; Clait E. Braun (term expires 1980). DATES OF ISSUE OF VOLUME 89 OF THE WILSON BULLETIN AO. 1 — 7 April 1977 NO. 2 — 27 June 1977 NO. 3 — 30 September 1977 NO. 4 — 27 December 1977 CONTENTS OF VOLUME 89 NUMBER 1 NOTES ON THE BIOLOGY AND POPULATION STATUS OF THE MONKEY-EATING EAGLE OF THE PHILIPPINES Robert S. Kennedy COWBIRD PARASITISM AND EGG RECOGNITION OF THE NORTHERN ORIOLE . Stephen I. Rothstein OBSERVATIONS ON THE RED-NECKED GREBE NESTING IN MICHIGAN Michael L. Chamberlin GROWTH AND DEVELOPMENT OF THE PLAIN CHACHALACA IN SOUTH TEXAS Wayne R. Marion SOCIAL DOMINANCE IN WINTER FLOCKS OF CASSIN’s FINCH Fred B. Samson INTER-BROOD MOVEMENTS OF JUVENILE SPRUCE GROUSE Daniel M. Keppie BREEDING BIOLOGY OF YEAR-OLD AND OLDER FEMALE RED-WINGED AND YELLOW-HEADED BLACKBIRDS Richard D. Crawford FOOD HABITS OF OLDSQUAWS WINTERING ON LAKE MICHIGAN Steven R. Peterson and Robert S. Ellarson SEXUAL DIMORPHISM IN THE WHITE IBIS James A. Kushlan EGGSHELL THICKNESS VARIABILITY IN THE WHITE-FACED IBIS David E. Capen CHARACTERISTICS OF A WINTERING POPULATION OF WHITE-TAILED PTARMIGAN IN COLORADO Richard W. Hoffman and Clait E. Braun BREEDING DENSITIES AND MIGRATION PERIODS OF COMMON SNIPE IN COLORADO Bruce R. Johnson and Ronald A. Ryder PRINCIPAL COMPONENT ANALYSIS OF WOODPECKER NESTING HABITAT Richard N. Conner and Curtis S. Adkisson PHENETIC ANALYSIS OF THE SUBFAMILY CARDINALINAE USING EXTERNAL AND SKELETAL CHARACTERS Jenna J. Hellack and Gary D. Schnell GENERAL NOTES WHY OSPREYS HOVER Thonias C. Grubb, Jr. STORAGE OF PINON NUTS BY THE ACORN WOODPECKER IN NEW MEXICO Peter B. Stacey and Roxana Jansma FLOCKING AND FORAGING IN THE SCARLET-RUMPED TANAGER David J. Moriarty YELLOW WARBLER NEST USED BY A LEAST FLYCATCHER J. Paul GoOSSen PINTAIL REPRODUCTION HAMPERED BY SNOWFALL AND AGRICULTURE Gary L. KrapU TICKS AS A FACTOR IN THE 1975 NESTING FAILURE OF TEXAS BROWN PELICANS Kirke A. King, David R. Blankinship, Richard T. Paul, and Robin C. A. Rice PRAIRIE WARBLER FEEDS FROM SPIDER WEB John F. DoUglaSS NOTES ON THE HUMMINGBIRDS OF MONTEVERDE, CORDILLERA DE TILARAN, COSTA RICA Peter Feinsinger NEST-SITE DIFFERENCES BETWEEN RED-HEADED AND RED-BELLIED WOODPECKERS IN SOUTH CAROLINA Lawrence Kilham GROUND FORAGING AND RAPID MOLT IN THE CHUCK-WILL’s WIDOW Sievert Rohwer and James Butler FEEDING RESPONSES OF FALL MIGRANTS TO PROLONGED INCLEMENT WEATHER Elliot J. Tramer and Elora E. Tramer S ’THBOUND MIGRATION OF SHOREBIRDS FROM THE GULF OF ST. LAWRENCE Raymond McNeil and Jean Burton 1 21 33 47 57 67 73 81 92 99 107 116 122 130 149 150 151 153 154 157 158 159 164 165 166 167 FLOCKING AND FORAGING BEHAVIOR OF BROW^N JAYS IN NORTHEASTERN MEXICO Michael L. Morrison and R. Douglas Slack 171 DO MORE BIRDS PRODUCE FEWER YOUNG? A COMMENT ON MAYFIELD’s MEASURE OF NEST SUCCESS Richard F. Green 173 MORTALITY OF NESTLING MISSISSIPPI KITES BY ANTS James W . Parker 176 ORNITHOLOGICAL NEWS 177 ORNITHOLOGICAL LITERATURE 179 NUMBER 2 THE BREEDING BIOLOGY OF THE OAHU ‘elepaio Sheila Conant 193 LIFE HISTORY FACTORS AFFECTING THE INTRINSIC RATE OF NATURAL INCREASE OF BIRDS OF THE DECIDUOUS FOREST BIOME Richard Brewer and Lynda Swander 211 NESTING HABITAT OF COMMON RAVENS IN VIRGINIA Robert G. Hooper 233 GROWTH AND DEVELOPMENT OF KNOWN-AGE RING-BILLED GULL EMBRYOS John P. Ryder and Lynn Somppi 243 HABITAT PARTITIONING IN A COMMUNITY OF PASSERINE BIRDS — Robert C. Whitmore 253 BREEDING DISPLAYS OF THE LOUISIANA HERON James A. Rodgers, Jr. 266 BREEDING BIOLOGY OF CLIFF SWALLOWS IN VIRGINIA Gilbert S. Grant and Thomas L. Quay 286 WEATHER INFLUENCES ON NOCTURNAL BIRD MORTALITY AT A NORTH DAKOTA TOWER Michael Avery, Paul F. Springer, and J. Frank Cassel 291 BREEDING BIOLOGY OF HOUSE SPARROWS IN NORTH MISSISSIPPI James N. Sappington 300 NESTING BEHAMOR OF YELLOW-BELLIED sApsucKERs Lawrence KUham 310 GENERAL NOTES SEX DIFFERENCES IN ALARM RESPONSES OF WINTERING EVENING GROSBEAKS Martha Hatch Ralph 325 NESTING OF TURKEY AND BLACK VULTURES IN PANAMA Laurie A. McHargUC 328 FULVOUS WHISTLING DUCK POPULATIONS IN TEXAS AND LOUISIANA Edward L. Flickinger, David S. Lobpries, and Hugh A. Bateman 329 SLIPPER SHELLS, A MAJOR FOOD ITEM FOR WHITE-WINGED SCOTERS James G. Hoff 331 EGG MOVEMENT BY A FEMALE GADWALL BETWEEN NEST BOWLS Robert F. Johnson, Jr. and Leo M. Kirsch 331 FOODS OF WESTERN CLAPPER RAILS R. D. Ohmart and R. E. Tomlinson 332 AGGRESSION IN FORAGING MIGRANT SEMIPALMATED SANDPIPERS Brian A. Harrington and Sarah Groves 336 HERRING GULL EATING BAYBERRY Roger F. Pasquier 338 THE LESSER ANTILLEAN BULLFINCH IN THE VIRGIN ISLANDS Herbert A. Raffaele and Daniel Roby 338 FORAGING BEHAVIOR OF THE WHITE IBIS James A. Kushlan 342 BIRDS OF FIVE FAMILIES FEEDING FROM SPIDER WEBS Robert B. Waide and Jack P. Hailman 345 WINTER NEST MICROCLIMATE OF MONK PARAKEETS Donald F. Caccamise and W'esley W. W eathers 346 SNAKE PREDATION ON bell’s vireo NESTLINGS Calvin L. Gink 349 CROW PREDATION ON BLACK-CROWNED NIGHT HERON EGGS Joanna Burger and D. Caldwell Hahn 350 ORNITHOLOGICAL LITERATURE CONSERVATION COMMITTEE REPORT (Falconry: Effects on Raptor Populations and Management in North America) ORNITHOLOGICAL NEWS NUMBER 3 NESTING HABITAT OF BACHMAN’s WARBLER — A REVIEW Robert G. Hooper and Paul B. Hamel MORPHOLOGICAL VARIATION IN NORTH AMERICAN PINE GROSBEAKS Curtis S. AdkisSOn BREEDING SUCCESS AND NEST SITE CHARACTERISTICS OF THE RED-WINGED BLACKBIRD Donald F. Caccamise FORAGING BEHAVIOR OF THE EASTERN BLUEBIRD Benedict C. Pinkowski COMPARATIVE FEEDING BEHAVIOR OF IMMATURE AND ADULT HERRING GULLS Nicolaas A. M. Verbeek COMPARATIVE MORTALITY OF BIRDS AT TELEVISION TOWERS IN CENTRAL ILLINOIS James W. Seets and H. David Bohlen EFFECT OF FLOCK SIZE ON FORAGING ACTIVITY IN WINTERING SANDERLINGS James Silliman, G. Scott Mills, and Stephen Alden ACTIVITY PATTERNS OF FEMALE RUFFED GROUSE DURING THE BREEDING SEASON Stephen J. Maxson WEIGHTS AND FAT CONDITION OF SOME MIGRANT WARBLERS IN JAMAICA A. W. Diamond, P. Lack, and R. W. Smith GENERAL NOTES WING MOLT OF THE KiTTLiTz’s MURRELET Spencer G. Sealy INCIDENCE OF RUNT EGGS IN THE CANADA GOOSE AND SEMIPALMATED SANDPIPER T. H. Manning and Brenda Carter LATE FLEDGING DATE FOR HARRIS’ HAWK Eleanor L. Radke and John Klimosewski THE SPATIAL DISTRIBUTION OF WINTERING BLACK-BELLIED PLOVERS Christopher H. Stinson PREDATION AND DISPERSION OF HERRING GULL NESTS Gary W . Shugart and William C. Scharf EGG QUALITY IN RELATION TO NEST LOCATION IN RING-BILLED GULLS John P. Ryder, Donald E. Orr, and Ghomi H. Saedi ROOF-NESTING BY COMMON TERNS Anne E. MacFarlane RAPID CHICK SEPARATION IN WHIP-POOR-WILLS Eric L. Dyer AN INTRASPECIFIC MORTAL ATTACK Vera Lee Grubbs RUFOUS-SIDED TOWHEES MIMICKING CAROLINA WREN AND FIELD SPARROW Donald J. Borror HEAT LOSS FROM THE NEST OF THE HAWAIIAN HONEYCREEPER, “aMAKIHi” G. C. Whittow and A. J. Berger SPREAD OF THE GREAT-TAILED GRACKLES IN SOUTHWESTERN LOUISIANA H. Douglas Pratt, Brent Ortego, and Harland D. Guillory POPLAR LEAF-STEM GALL INSECTS AS FOOD FOR WARBLERS AND WOODPECKERS Sally Hoyt Spofford PERCH HEIGHT PREFERENCES OF GRASSLAND BIRDS Keith G. Harrison SPECIAL REVIEW — JOHN OSTROM's STUDIES ON ARCHAEOPTERYX, THE ORIGIN OF BIRDS, AND THE EVOLUTION OF AVIAN FLIGHT Joel CracraH ORNITHOLOGICAL LITERATURE ORNITHOLOGICAL NEWS 352 360 370 373 380 396 404 415 422 434 439 456 467 469 469 470 472 473 475 476 477 477 480 483 485 485 488 492 503 505 THE president’s PAGE PROCEEDINGS OF THE FIFTY-EIGHTH ANNUAL MEETING 506 SUGGESTIONS TO AUTHORS 520 NUMBER 4 THE LESSER PRAIRIE CHICKEN’s INFLATABLE NECK SACS George Miksch SuttOtl 521 NESTING HABITAT OF CANADA GEESE IN SOUTHEASTERN MICHIGAN Richard M. Kaminski and Harold H. Prince 523 RESIDUES OF ENVIRONMENTAL POLLUTANTS AND SHELL THINNING IN MERGANSER EGGS Donald H. White and Eugene Cromartie 532 BREEDING BIRD SURVEY COUNTS AS RELATED TO HABITAT AND DATE Wayne C. Weber and John B. Theberge 543 ANALYSIS OF MATERIALS IN CLIFF AND BARN SWALLOW NESTS: RELATIONSHIP BETWEEN MUD SELECTION AND NEST ARCHITECTURE Delbert L. Kilgore, Jr. and Kathy L. Knudsen 562 PRODUCTION AND SURVIVAL OF THE VERDIN George T. Austin 572 MALE BEHAVIOR AND FEMALE RECRUITMENT IN THE RED-WINGED BLACKBIRD Patrick J. W eatherhead and Raleigh J. Robertson 583 REDUCTION OF COURTSHIP BEHAVIOR INDUCED BY DDE IN MALE RINGED TURTLE DOVES M. A. Haegele and Rick H. Hudson 593 MOVEMENTS OF THE GREAT-TAILED GRACKLE IN TEXAS .. Keith A. Arnold and Lcon J. Folse, Jr. 602 GENERAL NOTES TROPICAL SCREECH OWL NEST DEFENSE BEHAVIOR AND NESTLING GROWTH RATE Betsy Trent Thomas 609 THREE MORE NEW SPECIMEN RECORDS FOR GUATEMALA Robert W. Dickermau 612 FEEDING BEHAVIOR OF TWO HUMMINGBIRDS IN A COSTA RICAN MONTANE FOREST Barbara K. Snow 613 BLACK-LEGGED KITTIWAKES NESTING ON SNOWBANK George L. Hunt, Jr. and Max C. Thompson 616 EVIDENCE OF DOUBLE BROODING BY AMERICAN KESTRELS IN THE COLORADO HIGH PLAINS Dale W. Stahlecker and Herman J. Griese 618 FURTHER COMMENTS ON SEXUAL SIZE DIMORPHISM IN BIRDS D. Amadon 619 RESPONSE OF INCUBATING BLACK-BELLIED WHISTLING-DUCKS TO LOSS OF MATES Richard E. McCamant and Eric G. Bolen 621 LATE PLEISTOCENE WILLIAMSON’s SAPSUCKER FROM WYOMING Michael Wilson and Amadeo M. Rea 622 AMERICAN KESTREL REJECTS CAPTURED SPADEFOOT TOAD G. ScOtt Mills 623 WINTER DISTRIBUTION OF RED-TAILED HAWKS IN CENTRAL NEW YORK STATE Jonathan Bart 623 OSPREY CATCHES VOLE Noble S. Proctor 625 PATTERNS OF FEEDING FIELD SPARROW YOUNG Louis B. Best 625 AVIAN BONE PATHOLOGIES FROM ARIKARA SITES IN SOUTH DAKOTA Paul W . Parmalee 628 NEST RECIPROCITY IN EASTERN PHOEBES AND BARN SWALLOWS Hamion P. Weeks, Jr. 632 ORNITHOLOGICAL LITERATURE 636 ORNITHOLOGICAL NEWS 655 REQUESTS FOR ASSISTANCE 571, 635 INDEX TO VOLUME 89 657 The Wilson Bulletin PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY VOL. 89, NO. 1 MARCH 1977 ARY APR 2 01977 t“-i A R V A K U I V i£RS i X Y, The Wilson Ornithological Society Founded December 3, 1888 Named after ALEXANDER WILSON, the first American Ornithologist. President — Andrew J. Berger, Department of Zoology, University of Hawaii, Honolulu, Hawaii 96822. First Vice-President — Douglas A. James, Department of Zoology, University of Arkansas, Fayetteville, Arkansas 72703. Second Vice-President — George A. Hall, Department of Chemistry, West Virginia Uni- versity, Morgantown, W. Va. 26506. Editor — Jerome A. Jackson, Department of Zoology, P.O. Drawer Z, Mississippi State Uni- versity, Mississippi State, Mississippi 39762 Secretary — James Tate, Jr., P.O. Box 2043, Denver, Colorado 80201. Treasurer — Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. Elected Council Members — Helmut C. Mueller (term expires 1977) ; Abbott S. Gaunt (term expires 1978) ; James R. Karr (term expires 1979). Membership dues per calendar year are: Active, $10.00; Sustaining, $15.00; Life memberships, $200 (payable in four installments). The Wilson Bulletin is sent to all members not in arrears for dues. The Josselyn Van Tyne Memorial Library The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of Michigan Museum of Zoology, was established in concurrence with the University of Michigan in 1930. Until 1947 tlie Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines from members and friends of the Society. Now two members have generously established a fund for the purchase of new books; members and friends are invited to maintain the fund by regular contribution, thus making available to all Society members the more important new books on ornithology and related subjects. The fund will be administered by the Library Committee, which will be happy to receive suggestions on the choice of new books to be added to the Library. William A. Lunk. University Museums, University of Michi- gan, is Chairman of the Committee. Tlie Library currently receives 104 periodicals as gifts and in exchange for The Wilson Bulletin. With the usual exception of rare books, any item in the Library may be borrowed by members of the Society and will be sent prepaid (by the University of Michigan) to any address in the United States, its possessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers, as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to “The Josselyn Van Tyne Memorial Library, University of Michigan Museum of Zoology, Ann Arbor, Michigan.” Contributions to the New Book Fund should be sent to the Treasurer (small sums in stamps are acceptable). A complete index of the Library’s holdings was printed in the September 1952 issue of The Wilson Bulletin and newly acquired books are listed periodically. The Wilson Bulletin The official organ of the Wilson Ornithological Society, published quarterly, in March, June, September, and December. The subscription price, both in the United States and elsewhere, is S15.00 per year. Single C(>i)ies, S4.00. Subscriptions, changes of address and claims for undelivered copies should be sent to the Treasurer. Most back issues of tlie Bulletin are available and may be ordered from the Treasurer. Special prices will be quoted for <|uantity orders. All articles and communications for publications, books and publications for reviews should be addressed to the Editor. Exchanges should be addressed to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, Michigan. Known office of publication: Department of Zoology, Mississippi State University, Mississippi State, Mississippi 37962. Second class postage paid at Mississippi State, Mississippi and at additional mailing office. Allen Press, Inc., Lawrence, Kansas 66044 THE WILSON BULLETIN A QUARTERLY MAGAZINE OF ORNITHOLOGY Published by the Wilson Ornithological Society VoL. 89, No. 1 March 1977 Pages 1-192 NOTES ON THE BIOLOGY AND POPULATION STATUS OF THE MONKEY-EATING EAGLE OF THE PHILIPPINES Robert S. Kennedy The Monkey-eating Eagle \ Pithecophaga jejferyi^ see Frontispiece) is a huge forest raptor endemic to the Philippines. After its discovery in 1896 on the island of Samar ( Ogilvie Grant 1897b ) , it was considered so rare that nearly every specimen obtained before 1940 prompted a published account. Gonzales ( 1968) provided the first life history data in his 10-month study of a nesting pair in the province of Davao del Sur, on the island of Mindanao. Field censuses of this endangered species in 1969 on Mindanao, where the species is most abundant, produced estimates of 40 ( Alvarez 1970 ) and 36 (Gonzales 1971). Rahor (1971) estimated the numbers in 1970 at 50 to 60 birds. During a study of this species on Mindanao from August 1972 to April 1973, I assisted the Philippine Research, Parks, Range, and Wildlife Division of the Bureau of Forest Development in efforts to conserve this eagle. I col- lected data on the behavior of wild eagles and on their numbers and distribu- tion. Here I report on my findings and include a summary of the former and present status of the species on other islands where it has been recorded. STUDY AREA AND METHODS I studied activities of a pair of Monkey-eating Eagles at Tudaya Falls in Mt. Apo Na- tional Park on Mindanao. The area is dissected hy a series of deep ravines carved by swift mountain rivers. Elevation ranges from 700 to 1200 m. Because of the close prox- imity to the equator, the time for sunrise and sunset varied little over the year. A canyon below Tudaya Falls was the primary study area. It was ca. 400 m wide, 100 to 200 m deep, and 2 km long (Fig. 1). On the ridges surrounding the canyon, Bagoho natives have cleared some of the virgin forest (see Fig. 2). I spent 8 full days and 14 one-half days from September to March (153 hours of observation) at 3 lookouts overlooking the canyon, the choice of which depended on the location of the eagles at the time. With personnel from the Philippine Parks stationed in Davao City and Zamboanga City, I traveled to 10 provinces of Mindanao: Lanao del Norte, Missamis Occidental, Zamboanga del Norte, Zamboanga del Sur, Davao City, Davao del Norte, Davao Oriental, 1 2 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Fig. L View of canyon in tlie Tudaya Falls study area. IMioto taken from Lookout ^I facing tin* southeast. Davao del Sur, North Cotahato, and South Cotahato. Coverage was greatest in the last 6 provinces. Keeords of eagles killed, sighted, or eaj)tured since 1970 were collected from local residents and wer<‘ ae(“»*pted or rejected on the basis of the information provided. Fersonnel from logging companies and natives provided the most reports. Aerial surveys over 12 of the 17 provinces of Mindanao facilitated i)lotting the dis- tribution of Monkey-eating Eagle habitat on .joint Operation (Graphic (AIK) Maps, Series 1501 AIK, 1:250,000, current through 1909. Observations from the ground contributed additional data for habitat j)lotting. Frovinces surveyt'd by air in their entirety were Davao City, Davao del .'*'ur, Lanao del l^ur, and Missamis Occidental. Partially surveyed were Bukidnon, Davao del Norte. Davao Oriental. Lanao del Norte. North Cotahato. South Cotahato. Zamboanga del Norte, and Zamboanga del Sur. For other areas, the extent of the habitat was estimated from the topography and the density of human habitations, as j)rinted on the air maps. Areas of (piestionahle human density were plotted with the 30(K) foot contour line as the {)erimeter of the eagle’s habitat. By use of squared graph paper, the area of potential (‘agle habitat was determined from the air maps. The method does not account for the inereast' in surface area due to variation in altitude. IJEHAVIOR OF WILD EAGLES Hunting, techniques. — Monkey-eating Eagles hunt lioth singly and in pairs. I (lid not see a pair of eagles hunt together, hut several natives and loggers Kennedy • BIOLOGY OF MONKEY-EATING EAGLE 3 DAVAO CITY i DAVAO DEL SUR Fig. 2. Map of Tudaya Falls showing the distribution of cleared land (shaded area), forests (unshaded area), sightings of eagles (dots), and lookouts (triangles). reported that pairs of eagles course through the forest looking for groups of monkeys. An engineer with the Misamis Lumber Co., stated that one eagle would distract the monkey, which would then he captured from behind by the other bird. He reported that, after the kill, the eagle covered the prey with its wings and then gutted and skinned the animal. Gonzales ( 1968 j suggests that eagles are more successful when hunting in pairs than when hunting alone, because of the wariness of monkeys and the defense of the family unit by a lead male. Thirty of 38 observations at Tudaya Falls, and several elsewhere on Min- danao, were of eagles on the hunt. Though I never observed a complete hunting sequence, because of the bird’s sudden appearance and disappearance in the forest, a general 3 part hunting pattern can he reconstructed as follows (Fig. 3) : Part 1 — Preparatory Period : On 28 September I watched an eagle perched on the lower branch of a tree above the near-vertical cliffs across the canyon from Lookout #3 (Fig. 2). There it called from 13:00 to 13:30. It then be- came increasingly alert to the sounds and movements in the canyon below. The usual position of a perched Monkey-eating Eagle is vertical, and, from a distance, the white breast and belly of a bird in this stance closely resemble the light-colored bark of the trunks and main branches of many Philippine 4 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Fig. 3. Schematic representation of a liunting sequence, consisting of three parts; Part 1 — a preparatory jicriod; Part 2 — the act of hunting within the forest canopy; Part 3 — return to a starting point. trees. This may lie a form of cryptic coloration. Because I observed the preparatory period only once, I do not know if the calling activity is charac- teristic of all hunting sequences. Part 2 — dlie Act of Hunting: At 13:35, the eagle dropped from its perch and without flapping, glided about 75 in at a downward diagonal, along the canyon wall, to a resting iilace in a small tree growing out of the canyon wall. Be- cause of the minimal vegetation growth, I could see the bird clearly. It re- mained there for 5 min surveying the surrounding trees. When the eagle seemingly perceived movements, it would shift its head to the right and left of the body axis, fretjuently twisting its head upside down. It made 6 more short flights liefore it disappeared at 14:35. In flights ranging from 75 to 125 m, the bird moved toward the floor of the canyon. The short direct flights from one perch to another usually within the forest canopy (Fig. 3) are the most common hunting techniiiue used by the eagles when they work down a mountain slope or along a heavily forested ridge. I noted one variation from the pattern: on steep slopes, instead of gliding directly to a perch, the eagle would frequently drift out of the forest canopy aw ay from the slope, circle one or more times, and then return to another perch. During these flights, the birds searched the forest around them, apparently for prey or for a suitable Kennedy • BIOLOGY OF MONKEY-EATING EAC;LE 5 perch. J. Hamlet ( pers. comm. ) described the eagle in pursuit of prey as having direct, flapping flight. Part 3 — Return to a Starting Point: If the eagle failed to capture prey, it would return to a higher elevation to initiate a new hunt. From Lookout #2, on 13 February 1973, I saw a Monkey-eating Eagle perched on the lower branch of a large tree at the forest edge, 400 m west of my position. At 09:45, it left the perch and glided parallel to the canyon, heading south for approxi- mately 500 m, with minimal loss of altitude. Then it began to circle slowly, gaining altitude. When it attained a height of 300 to 450 m above its initial ele- vation, the eagle glided directly north for 2 to 3 km, disappearing into the for- est higher up the mountain at 10:05. This same eagle (which had 2 left pri- maries missing) reappeared at the forest edge at 13:45 and repeated a similar sequence. 1 saw 3 eagles elsewhere on Mindanao performing this part of the hunting technique with little deviation from the pattern. Elevations attained varied from 150 to 700 m above the start of the spiral, and the distance glided varied from 400 m to 3 km. Part 1 lasted 35 min. Part 2, 60 min, and Part 3, 20 min. If one assumes that the eagle used to exemplify Part 3 did not engage in any other activities besides hunting from the time it was first seen at 09:45 to the time it reappeared at 13:45, then this hunting cycle lasted 4 h. In another case, the cycle lasted 2 h 55 min. The times for the 3 hunts average 2 h 56 min. Figure 4 indicates 2 peak periods in the day when eagles are likely to be seen. During these periods, the birds often emerge from the forest and fly to another location on the mountain, as described earlier. For the morning peak, 8 of 11 and 7 of 8 sightings made at 09:00 and 10:00, respectively, were of eagles ostensibly hunting. In the afternoon, the sightings at 13:00 and 14:00 were of eagles hunting. The peaks for calling (Fig. 4) suggest that vocalizations occur in Part 1 of the hunting cycle. I observed the result of a successful hunt at Tudaya on the morning of 16 February 1973. While at Lookout #1, I heard an eagle calling at 08:20 from the side of the canyon. It was apparently perched in a tree, and I did not see it until 15 min later, when it stopped calling, flew over the canyon, circled once, then glided 150 m west up Parak Creek and landed in a small tree. I noticed a monkey in the eagle’s talons when the bird flew past me. At the perch, the eagle resumed calling but did not mantle the prey or attempt to eat it. At 08:45, the eagle left its perch, glided west about 500 m, and disappeared into the forest. The manner in which it traveled from one resting place to another was similar to Part 2 of the hunting cycle. Flight. — Brown and Amadon (1968) state that Monkey-eating Eagles “sometimes, but rather rarely, soar over the forest . . . ,” and Grossman 6 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 TIME OF DAY Fig. 4. Plot of time of day when eagles were sighted or were heard calling at the Tudaya Falls study area. and Hamlet (1964) claim that these eagles have “flapping flight with little soaring.” With hut 2 exceptions, I never saw eagles flap their wings, and I fretjuently saw Itirds soaring, though not for extended periods. In traversing an area, they would either glide in a straight line from a higher to a lower elevation or drift over a mountain slope or canyon, riding thermals and mountain updrafts. d4ie most typical glide occurred when the eagles were hunting, and during it the wings were usually fully extended. Sometimes, when greater speed was refiuired, the wings were folded partially or com- l)letely. A gliding or soaring Monkey-eating Eagle holds the wings horizon- tally. When riding updrafts, the eagles usually circle, hut they occasionally tack hack and forth just above a ridge or glide in a straight line at high altitudes. In these situations, the birds usually either maintain or gain altitude. Circling flight exhibits at least 3 forms. During Part 2 of the hunting cycle, when the bird leaves the forest canopy, it may circle one or more times. During these circles, it maintains its altitude hut drifts horizontally as it turns. It reenters the forest at about the same elevation as the exit point. Circling also occurs during Part 3 of the hunting cycle ( Fig. 3 I when the bird gains altitude. Here the eagle leaves the forest and, finding an appropriate updraft, circles slowly without flapping, gaining height all the time. The diameter of the circles I Kennedy • BI()IX)(;Y OF MONKEY-EATINC; EA(;LE 7 observed varied from 25 to 40 m. At the highest point, the birds either glide directly toward a mountain slope, usually without loss in altitude and always without flapping, or begin a series of short glides and large sweeping circles. This third form is frecfuently initiated in instances where a bird has gained altitude. Sometimes, a bird drifts from the forest at a higher elevation and glides over the lower slopes, where it begins soaring. The duration of this third type of circling averaged 5 min. Gonzales ( 1968 ) once noted 3 eagles soaring together. I saw only 1 or 2 birds at a time, though 1 received a report ( Engineer Rizardo, pers. comm.) that as many as 7 eagles were seen soaring together. The Monkey-eating Eagle is relatively short-winged and long-tailed. This structural pattern, which has been termed the Goshawk silhouette by Brown and Amadon ( 1968 ) , is an adaptation of forest-hunting species which reciuire maneuverability and (juick bursts of speed to overtake their prey. Birds with this silhouette usually have flapping flight and seldom soar without flapping. Though the eagle is a forest-hunting bird and is capable of (luick flapping flight in pursuit of prey, it is also a bird that freciuently soars. This soaring ability is, energetically, clearly an adaptive advantage for this species which has a rather large territory. Vocalizations. — Gonzales (1968), Seth-Smith (1910), and Whitehead (1899), have described the calls of the Monkey-eating Eagle. Whitehead (1899) rendered the call phonetically as “zc-af/ zcafz,” and Gonzales noted it as a long, mellow whistle ending sometimes with a downward inflection hut usually with an upward inflection. He stated that the latter call was the one more frequently given by the breeding pair he was studying. When I heard eagles call, the downward inflection was most frequent. A series of 3 to 9 whistles was repeated at intervals ranging from 45 sec to 5 min. The indi- vidual whistles lasted 0.5 to 1.5 sec and were uttered at 1 to 2 sec intervals. The number of series varied from 1 to 15. A possible juvenile bird at Tudaya gave a different call. Calling began in the morning with the typical downward inflection; hut as the day progressed, the call changed to a more plaintive whine-whistle, as if the bird was dis- tressed. Each whine-whistle lasted about 2 sec, and a series of these calls was repeated every 45 sec for up to 0.5 h. This type of vocalization resembled that of an eaglet calling after long periods without food as described by Gonzales (1968). At Tudaya, adult eagles called (1) during Part 1 of the hunting cycle, (2) just after the capture of a prey, and (3 ) during and immediately after being pursued by Rufous Hornbills ( Buceros hydrocorax) . In the first 2 cases, the call had the typical downward inflection. In the third case, the call consisted of a single whine-whistle repeated every 10 sec. This was unlike the call 8 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Gonzales (1968) described as “short, intense, high-pitched notes” during attacks on eagles by avian enemies. Interspecific encounters. — The presence of a IMonkey-eating Eagle in an area is often revealed by the noisy mobbing of the bird by groups of up to 20 Rufous Hornbills whose raucous call can be heard as much as 2 km away. The eagle tries to avoid the hornbills by flying from perch to perch within the forest. If unsuccessful, it leaves the canopy and begins circling slowly, gaining altitude as in Part 3 of the hunting cycle. The hornbills follow for awhile, but eventually are outdistanced and retire to the forest. I saw such mobbing on 5 occasions. On the average, the incidents lasted 2 to 3 min, but tbe duration varied from less than 30 sec to just over 15 min. Though I never observed contact, Gonzales ( 1968) reported that Rufous Hornbills actu- ally strike the eagle’s head. He also noted that Writhed-billed Hornbills (Aceros leucocephalus) and Large-billed Crows (Corvus macrorhynchos ) mob eagles. Both these species were numerous at Tudaya, but I saw no inter- actions between them and eagles. I watched an Oriental Hobby ( Falco severus) attack a Monkey-eating Eagle as the larger bird flew slowly at about 60 m above a ridge. Three times the falcon stooped, nearly grazing the eagle’s head. During the harassment the eagle continued flying normally, but it landed shortly afterwards. HAI5ITAT AM) HOME RANGE Habitat. — 3 he original habitat of the xMonkey-eating Eagle on Mindanao was undoubtedly dipterocarp forest, which comprised 75% of the virgin forest in the Philippines ( W hitford 1911). Dipterocarp forests are characteristic of moist plains and extend up mountain slopes to 800 m ( Brown and Mathews 1911). Today eagles are mainly confined to the larger mountain masses ( Rabor 1971), but at one time they occupied lowland forest down to sea level. A specimen taken in 1954 in Cotabato City at an elevation of ca. 15 m pro- vided evidence for eagles using lowland forests. The highest elevation at which eagles occur is about 2000 rn, where their preferred prey becomes scarce. At l udaya, I saw eagles enter the forest to begin a hunt at about 1700 m. As land has been cleared for agriculture and for luml)er, the lower edges of the forests inhabited by the eagles have been retreating up the sides of moun- tains. Tlie birds have partially adapted to this change by hunting over cleared land and living in second growth forest. This adaptive ability was first indicated by Whitehead (1899), who stated, “He [the eagle] is well known to the natives as a robber of their poultry and small pigs . . . ,” thereby implying that the birds forage near clearings. Gonzales ( 1968) described the habitat of the pair he studied as follows: “Some of the hills are still clothed Kennedy • HIOLOGY OF MONKEY-EATINC; EAGLE 9 with original clipterocarp forest, but others are either naked . . . or covered with coarse cogan grass mixed with shrubs and small trees. The forested hills, however, have not remained virgin for they too have been invaded by the natives as well as logging concessionaires.” At Tudaya, the eagles’ territory included cleared farmland, various stages of secondary growth, and virgin forest. The birds mainly confined their activity to virgin forest or advanced secondary growth (Fig. 2). Of the 11 eagles I sighted on Mindanao, 10 were in areas of virgin forest or in mixed virgin and advanced secondary forest. Occasionally eagles were reported in areas where no typical habitat existed. Most of the reports probably resulted from misidentifications, but one con- fiscated eagle ( LSUMZ 73747 ) was shot in a cornfield about 10 km from the closest forest. The abnormal occurrence of the bird at this location is pos- sibly attributable to destruction of habitat in its former territory. Apparently suitable Monkey-eating Eagle habitat on Mindanao in 1973 (Fig. 5a) comprised 29,000 km- (without allowing for increased area result- ing from elevational differences) or about 30% of the 95,587 km^ of land area of the island. The alarming rate of forest destruction was reported by Gonzales (1971), quoting the Philippine Free Press for 7 June 1969, which stated that the rate of deforestation in the Philippines at that time was 1 ha every 3 min. This problem is not new, however, for Whitehead (1899) stated: “The forests that are left in Samar are still very vast, especially on the Pacific Coast, but for miles inland those of the western coast have been destroyed, leaving ranges of low undulating clay hills chiefly covered with lalang grass. When this country has been passed, the traveler finds himself at an elevation of nearly 1,000 feet and meets with the true virgin forest of Samar. This forest is becoming annually smaller owing to the cultivation of hemp . . . .” Land clearing has confined suitable habitat on Mindanao to the mountain ranges, but even there the trees have been removed up to at least 500 m in most cases, and sometimes to as high as 1586 m ( Gonzales 1971). Nine eagles 1 sighted on Mindanao were associated with steep mountain slopes that formed the sides of deep ravines, canyons, or valleys. Data col- lected on the hunting and soaring behavior of the eagle indicate that it is well adapted to such topography, thus I believe that steep mountains are important in the eagle’s habitat. Home range. — Rabor ( 1968) believes that a pair of Monkey-eating Eagles have a home range comprising from 40 to 50 km^. Gonzales (1968, 1971) says that the range can be as large as 100 km-. Grossman and Hamlet report a smaller range, 31 to 34 km^. To determine the area used by the pair at Tudaya, I have drawn upon my own observations and those of Parks personnel as well as verbal reports by 10 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Fig, 5. a. Distribution of tlie remaining habitat for the Monkey-eating Eagle on Min- danao. 1). (Insert) The Philippine Islands, showing the islands where the species has been positively recorded (colored black) and where it was believed present (shaded area). llie natives living there. Bagolio natives said that the eagles mainly confined their activities to the Fudaya canyon, Parak Creek, and the adjacent ridges and that they were rarely seen elsewhere in the area. Parks personnel first sighted the eagles at Fudaya on 8 April 1972. They found the birds to be restricted to the Tudaya Falls area also. My own observations (Fig. 2) con- firm these reports. I saw eagles on 7 of 8 full days and 10 of 14 half days of surveillance. Clearly, the vicinity around Tudaya Falls has been a center of the eagles’ activity for several years. I calculated the area of the home range of this pair by 2 methods. Con- necting the outermost points where eagles were sighted (shown in Fig. 2) forms a polygon w ith an area of 6.34 km-, the minimum home range. A sec- Kennedy • BIOLOGY OF MONKEY-EATINC; EA(;LE 11 ond method was to determine the range hy measuring the greatest distance between sightings. If this length is considered the diameter of a circular home range, the area would be 12.5 km-. Since the range was measured as hori- zontal surface area and does not account for the increased surface resulting from elevational differences, the total area would be greater, perhaps twice as much. The adjusted home range would then be 12.68 km- (polygon method) to 25.0 km- (circular method). These figures suggest that the area necessary to support a pair of eagles may not be as great as formerly believed. DISTRIBUTION AND STATUS The Monkey-eating Eagle is known from accounts in the literature to occur on 4 of the larger islands in the Philippines: Luzon, Samar, Leyte, and Min- danao (Fig. 5b). John Hamlet, while visiting the island of Negros in 1945- 1946, received a photograph of a Monkey-eating Eagle allegedly captured on that island. He personally sighted 2 birds soaring over a small island in the Surigao Straits, just north of the province of Surigao del Norte on Mindanao. These reports of eagles from additional islands suggest that the species was probaby more widely distributed in the recent past. An account of the present and former status of the eagle on the islands from which specimens have been taken follows. Luzon. — Earlier records of Monkey-eating Eagles from Luzon were fre- quently reported in the literature and chronologically follow: 1 killed near the Agus River in Rizal Province on 11 May 1907, the first specimen posi- tively known from Luzon (McGregor 1907) ; 2 sighted near Montalban, Rizal Province on 13 August 1907 by W. P. Lowe ( Seth-Smith 1910) ; 1 captured on Mt. Ballong, south of Imugan, Nueva Vizcaya Province in January 1917 ( McGregor 1918) ; 1 captured near Pagbilao, Tayabas Province in July 1926 (McGregor 1927); 1 taken from Albay Province (Davidson 1934). More recent data on the status of the eagle on Luzon were gathered by Rabor (1971) during collecting trips in 1959, 1960, and 1961. In the Cordi- llera and Ilocos mountain ranges in northwest Luzon, he received reports that eagles were last seen in the late 1930’s. However, in the Sierra Madre Moun- tain Range in northeastern Luzon, including the provinces of Cagayan, Isabela, Nueva Vizcaya, and Quezon, the eagles were still being sighted by natives as late as 1960. A specimen captured by personnel from the Philippine Parks in the Isabela-Nueva Vizcaya territory in 1963 is the last known record from this island (Gonzales 1968). Samar. — The type of the Monkey-eating Eagle was taken on this island by one of Whitehead’s collectors in 1896 ( Ogilvie Grant 1897a). Davidson (1934) cites one other record of the species on Samar; and Rabor (1971), on the basis of the absence of verbal reports in the 20 years prior to his 12 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 collecting trip in the north central mountains in 1957, considers the bird extinct on Samar. There are no recent data to confirm or refute this sup- position. Leyte. — For a long time, the only indication that the Monkey-eating Eagle occurred on Leyte was provided by Ogilvie Grant (1897a), who mentioned that J. Whitehead had heard the call of the eagle on this island. Recently, personnel from the Philippine Parks learned of the following records in the province of Southern Leyte: eagles reported sighted in 1951, 1963, and 1968; nest site with young observed in 1969; and 2 eagles killed in 1965 ( S. E. Macanas, Regional Director of the Parks, pers. comm.). Macanas and others sighted the bird in Southern Leyte on 21 November 1970, and he estimated the number remaining there to be 8 to 10. Mindanao, early population data. — ^There are many reports in the literature (Clemens 1907; Davidson 1934; McGregor 1907, 1921; Mearns 1905; Seth- Smith 1910) of specimens of the Monkey-eating Eagle accjuired throughout Mindanao in the early part of this century, and apparently the first record came from 1 taken near Davao City on 3 September 1904 (Mearns 1905). Published i)opulation reports are lacking for the first 6 decades of its known existence on the island. However, Hamlet, who worked on Mindanao from 1945 to 1946, has informed me that the eagle was not uncommon there, since he located several active nests and knew of many other pairs. In a recent report, Gonzales ( 1971 ) attempted to estimate the population of Monkey-eating Eagles on Mindanao in 1910, when the forests still covered 65% of the island. Assuming a home range of 100 km‘“ and the use of all the available habitat, he calculated that at least 600 pairs existed on Mindanao in 1910. Mindanao, recent population data. — According to the 3 surveys mentioned earlier, the population on Mindanao in 1969 and 1970 was between 36 and 60 birds. Careful examination of each report reveals discrepancies among the estimates. In 9 of 17 provinces, the authors agreed on the presence or absence of the eagles; and in 5 of the 9. the estimates were the same. However, Rahor (1971) reported the eagle in 4 provinces where neither of the other investi- gators did. Also, in 4 provinces, 2 of the authors were in agreement as to the eagle’s presence, hut the third considered it to he absent. A striking area of such disagreement is the i)iovince of North Cotahato. There, Alvarez (1970) recorded 11 birds, Rahor (1971) 8, and Gonzales (1971) none. In Table 1, I have combined the population data of the 3 workers, and without duplicating records collected from specific locations, derived a popu- lation size of 70 birds. This number is 16.6% greater than the highest total and thus clearly shows their estimates to he low. Reasons for their low figures are that they counted only birds that they saw or that were reported to them. Kennedy • BIOLOGY OF MONKEY-EATING EAGLE 13 I cT fO I I CO I Laiiao del Sur - - 6 6 1800 Misamis 2 2 8 8 400 Occidental Misamis Oriental - - 2 2 900 14 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 • S C: iin :£U III =fl z'. I Ik Ik X 71 I 1 I I i. ? = ^ X a I .11^ VO On Si s ?g bL !- C = n X 5 «Z1^ n X f I S X II ’§ g = 3 X X ^1*"''^ numbers are data I collected. Numbers in parentheses are data collected by personnel from the Philippine Research, Parks, Range, and Wi e Division of the Bureau of Forest Development. These unpublished data are subject to revision. Underscored numbers eciual the combined total. Gonzales (1971) considered the population size to he twice the total records he collected; thus 2 X 18 = 36. Kennedy • lUOLOGY OF M0NKB:Y-EATTNG EAGLE 15 they did not survey all the habitat available to the eagles, and they did not extrapolate their findings to include all the habitat. Mindanao, present population data. — To determine a reasonably accurate population estimate of the Monkey-eating Eagle on Mindanao during the period January 1970 to April 1973, I used 3 censusing methods. Though each method differed slightly, all used the following formula to derive the total population size: t = (N/n) t where T = total population size; t = sample total; N = area of total habitat remaining on Mindanao; and n = area of habitat censused. Similar to the one Gonzales ( 1971 ) used to estimate population size for 1910, the first method involved determination of the extent of the eagle’s habitat remaining on Mindanao. The amount remaining (Ni) was found to be approximately 29,000 km- (see Fig. 5a and Table 1). Using the maximum home range size of 100 km- (rii ) for one pair (ti = 2), the total population estimate ( Ti ) equals 580 birds. A second method involved sampling an area of known size. The Mt. Apo range, west of Davao City proper, was chosen as the site for this study. The amount of suitable habitat in this area was found to be approximately 640 km- (02). Nine eagles were sighted in this area (t2), but as it was physically im- possible to cover the whole mountain range, there were probably more. Since the area of total habitat remains the same, Ni = N2 = 29,000 km-, the total A population estimate (T2 ) is 408 birds. A third method entailed the visitation of as many areas as possible, col- lecting reports of eagles sighted, captured, or killed, and confirming as many reports as possible. The results of this method are shown in Table 1. The combined totals were: number actually sighted by official investigators, 29; number known or reported captured, 16; number known or reported shot, 19; and number of additional eagles reported to the investigators, 74. These data were obtained from 12 of the 17 provinces of Mindanao, hut only 6 provinces ( Davao City, Davao del Norte, Davao Oriental, Davao del Sur, North Cota- bato, and South Cotahato ) were visited regularly, but none was completely surveyed. A rough estimate of the area covered by this method would he 1/3 of the total habitat on the island; thus n^ = 1/3 Ni. Excluding the numbers shot or captured, the total known population in the wild on Mindanao (t;0 was 103 birds ( number actually sighted plus number reported sighted ) during the period of investigation. This gives a 'U of 309 birds. Implicit in these censusing methods are the following assumptions: ( 1) all the habitat is used by eagles; (2) all the birds in areas sampled were known; 16 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 (3) each pair of Birds occupies a fixed range; (4) results found in one area are applicable to other areas not sampled; (5) population is stable; (6) ranges did not overlap. The degree of variance among the different methods results from not meet- ing these assumptions entirely, because of insufficient data and the difficulties of censusing. But these techniques do provide a reasonable range (309 to 5o0j in which the number of surviving eagles probably lies. Mortality should he considered in connection with these estimates since evidence indicates that the population is decreasing. No data are available regarding natural causes of mortality, such as disease and predation, but something is known about 2 unnatural factors. These are: (1 ) loss of habitat (discussed earlier in this i)aper) ; and (2 ) the shooting and capturing of wild birds. During the course of this study, 3S eagles were known to have been taken from the wild either by being captured or shot (see Table 1). Since the dura- tion of the entire study was 40 months, the average number of birds known to have been lost from the population was ().o7 birds jjer month or 10.4 birds per year. As previously mentioned, these data were collected from about 1/3 of tbe available habitat. Thus, the estimated number removed from the pop- ulation is an average of 31.2 per year. This calculation introduces still an- other assumption, that each eagle, regardless of the degree of its isolation, has an e(iual chance of being captured or shot. 4 he importance of this unnatural mortality depends greatly upon the size and annual recruitment rate of the population. If the i)opulation does lie within the limits suggested by the census methods above, i.e., 309 to 580 birds, then the mortality rate would be from 10.1 to 5.4%. With high reproductive success and a great percentage of young surviving to adulthood, the 5.1% and possibly tbe 10.1% mortality could be absorbed on an annual basis. However, few data are available concerning the repro- ductive biology of these birds, (irossman and Hamlet ( lOCvl I summarized what was known in 1961. “Although as a rule only one eaglet seems to sur- vive in each nest, theie may be two eggs, and occasionally (in at least one known instance) two young birds. The adults at several nest sites have pro- duced young every year.” Gonzales (1968) found that the pair he studied produced 1 egg and that 1 eaglet per nesting reached fledging age for 2 con- secutive attempts. Kabor ( 1968) also feels that the eagles breed every year. We do not know the age at which these eagles attain sexual maturity nor do we know the proportion of fledglings that survive to that age. Ihe age of the individuals captured or killed also influences the impor- tance of the unnatural mortality. If all were birds successfully breeding for the first time, the loss would be extremely damaging, as at the 10.1% level it Kennedy • BIOLOGY OF MONKEY-EATING EAGLE 17 would nearly eliminate this, the most valuable age class. Since the ages of the birds that were captured or shot are not known in all cases, the effect of their loss from the population cannot he determined. Population ecologists generally agree that a pair of a species reciuires and usually defends a certain semi-fixed area or territory. From this premise, it follows that the number of individuals a system can support is directly pro- portional to the available habitat. On Mindanao, the habitat is being de- stroyed by logging and other land clearing practices to such a degree that many birds are forced to leave their former range and search out new suitable habitat. I believe that a good percentage of the individuals that are captured or shot are birds whose habitat has been destroyed and that have become “surplus.” Thus, presumably, even if they had not been captured or shot, they would not have contributed significantly again to the continuation of the species, unless they were able to establish a new territory in an unoccupied area. The population surveys conducted have, in the main, shown that the eagles are rather evenly distributed over Mindanao. In some cases, especially parts of Davao del Norte, Davao Oriental, and North Cotahato, the eagles were as common in 1973 in the remaining habitat as they probably had ever been. CONCLUSIONS AND RECOMMENDATIONS The total range of the Monkey-eating Eagle has been greatly reduced during the time in which the species has been known. The reduction has been caused by the loss of the eagle’s habitat, and, since the population varies directly with the amount of habitat, it has suffered also. In this paper, I have brought up- to-date most of what is known concerning this species, knowledge that is still extremely patchy. Though the population is larger than formerly believed, the census techni(iues used are based on a modest amount of data, and thus the resulting figure should he considered only a rough estimate. Since the third census included data from about 1/3 of the remaining habitat and was the most extensive survey, I am inclined to regard it as the most accurate. However, no census techni(iue is entirely dependable, especially one based on extrapolation. Taking everything into account, I feel that the population on Mindanao during the period of investigation was about 300 ± 100. The total number of individuals of the species is unknown, as little population work has been conducted on the other islands where it exists or could exist. Alvarez (1970), Gonzales (1971), and Rahor (1968) have listed the fol- lowing as the principal reasons for the eagles’ decline: (1) the loss of the eagle’s habitat by logging and agricultural practices; (2) shooting the eagles for trophies; and (3) capturing the eagles for private and public display. In addition, Gonzales (1971) and Rahor (1968, 1971) have presented excellent THE WILSON BULLETIN • VoL 89, No. 1, March 1977 1 io recommendations for the conservation of the eagle. Though all their recom- mendations are sound, the 2 that I feel most important are: 1. Educational programs on the conservation of the natural resources in the Philippines, including wildlife. 2. The establishment of Wildlife Sanctuaries and the protection of lands from illegal logging and agricultural practices. Though shooting and capturing of the eagle certainly contribute to the population decrease, I feel that the primary reason for its decline is the loss of habitat, and therefore, base my recommendations for the conservation of the species on maintenance of its natural environment, as follows: 1. The establishment of preserves in mountain ranges where logging and agricultural practices are not economically feasible. The size of these preserves should he at least 200 km-, in order to encompass enough land for several pairs of birds. 2. For areas used as commercial forests, I suggest that the minimum interval between selective logging be 30 years, to allow regeneration of the native forest. 3. When areas are reforested, I recommend that a diversity of native., Philippine species he planted, thereby recreating as closely as pos- sible, the natural state of the forest. The destruction of forests in the Philippines is the result of broad social- economic i)rol)lems that need not he described here. However, if the above minimal recommendations are heeded, they should eventually i)ievent any further decline in this rare endemic. SUMMAKY A study of the Monkey-eating Eagle was eonducted on the island of Mindanao, Republic of Philipi^ines, from August 1972 to April 1973, in conjunction with the Philippine Re- search, Parks, Range, and Wildlife Division of the Bureau of Forest Development. Infor- mation on the hunting teehnicpies, flight, calls, interspecific encounters, territory size, habitat, and population status are presented. Though the eagles at one time occui)ied mature forests from sea level to 2000 m, the forests have been destroyed at lower elevations and thus suitable habitat and the eagles are primarily confined to the mountains. The size of the home range of a pair may range from 12 to 100 km^ The eagles are known to have occurred on Luzon, Samar, possibly on Negros, on a small island in the Surigao Straits north of Mindanao, on Leyte, and on Mindanao. Except for L(‘yte and Mindanao, no recent })opulation data are available. An eagle was sighted on Leyte in 1970 and an estimated 8 to 10 birds were thought present at that time on the southern half of that island. On Mindanao, the combined data collected by the author and by jn'isonnel from the Philippine Parks from .lune 1970 to April 1973 suggest that Kennedy • HIOLOGY OF MONKEY-EATINC; EAGLE 19 earlier population estimates of 36 to 60 for 1969 and 1970 were low, and that the popu- lation size for the period 1970 to 1973 was about 300 ± 100. The species is declining annually because of destruction of its habitat. Recommenda- tions to prevent further decline are jiresented. ACKNOWLEDGMENTS Data for this paper were collected while I was a United States Peace Corps Volunteer assigned to the Philippine Research, Parks, Range, and Wildlife Division of the Bureau of Forest Development. To these agencies I extend my sincere gratitude for allowing me the privilege to study the Monkey-eating Eagle. Special thanks are due Jesus B. Alvarez, Jr. and Armando M. Racelis who supervised the project. Their patience and understand- ing is greatly appreciated. To the participants of the Monkey-eating Eagle Conservation Program, from both the public and private sector, 1 offer my cordial thanks. For his companionship and assistance during most of my work on Mindanao, I am indebted to Antonio 0. Chavez. Robert B. Hamilton, George H. Lowery, Jr., Robert J. Newman, and H. Douglas Pratt kindly read the manuscript and gave many helpful suggestions. Hugh M. Turner helj)ed with photographic processing. Finally, I wish to thank Linda Anne Kennedy for her help during all phases of the study. LITERATURE CITED Alvarez, J. B., Jr. 1970. A report on the 1969 status of the Monkey-eating Eagle of the Philippines. Pap. and Proc. Int. Union Conserv. Nat. Nat. Resour. 11th Technical Meeting, New Delhi, India, 25-28 November 1969, Vol. 11:68-73. Brown, L. and 1). Amadon. 1968. Eagles, hawks, and falcons of the world. McGraw- Hill Book Co., New York. Brown, W. H. and I). M. Mathews. 1914. Philippine dipterocarp forests. Philipp. J. Sci. 9:413-568. Clemens, J. 1907. Notes from the Pliilij)j)ines. Condor 9:92-93. Davidson, M. E. M. 1934. Specimens of the Philippine Monkey-eating Eagle {Pithe- cophaga jefferyi) . Auk 51:338-342. Gonzales, R. B. 1968. A study of the breeding biology and ecology of the Monkey- eating Eagle. Silliman J. 15:461-491. . 1971. Report on the 1969 status of the Monkey-eating Eagle on Mindanao Is- land, Philippines. Bull. Int. Counc. Bird Preserv. 11:154-168. Grossman, M. L. and J. Hamlet. 1964. Birds of prey of the world. New York, Bonanza Books. McGregor, R. C. 1907. Notes on s])ecimens of the Monkey-eating Eagle ( Pithecophaga jefferyi Grant) from Mindanao and Luzon. Philipp. J. Sci. 2:297. . 1918. New or noteworthy Philippine birds, II. Philipp. J. Sci. 13:1-20. . 1921. New or noteworthy Philippine birds, IV. Philipp. J. Sci. 19:691-703. 1927. New or noteworthy Philippine birds, V. Philipp. J. Sci. 32:513-528. Mearns, E. a. 1905. Note on a specitmm of Pithecophaga jefferyi Ogilvie-Grant. Proc. Biol. Soc. Wash. 18:76-77. Ogilvie Grant, W. R. 1897a. On the birds of the Philippine Islands.— Part IX. The islands of Samar and Leite. Ibis 3 (7th Series) : 209 250. . 1897h. I New specimens of birds from the island of Samar.l Bull. Br. Or- nithol. Club No. XL. Ibis 3 (7th Series) :253-254. Raror, I). S. 1968. The present status of the Monkey-eating Eagle, 1*ithecophaga jef- 20 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 feryi Ogilvie-Grant, of the Philippines. Int. Union Conserv. Nat. Nat. Resour. Publ. New Ser. No. 10:312-314. . 1971. The present status of conservation of the Monkey-eating Eagle of the Philippines. Philipp. Geogr. J. 15:90-103. Seth-Smith, D. 1910. On the Monkey-eating Eagle of the Philippines ( Pithecophaga jef feryi) . Ibis 4 (9th Series) :285-290. Whitehead, J. 1899. Field notes on birds collected in the Philippine Islands in 1893-6. Ibis 5 (7th Series) :81-111. Whitford, H. N. 1911. The forest of the Philippines. Part 1. Forest types and prod- ucts. Uept. of the Int., Bur. of Forestry Bull. No. 10. Bureau of Printing, Manila. MUSEUM OF ZOOLOGY, LOUISIANA STATE UNTV., BATON ROUGE 70893. ACCEPTED 28 SEPT. 1976. NEW LIFE MEMBER Robert S. Kennedy is a new life member of the Wilson Ornithological Society. He is presently completing his doctoral work at Louisiana State University. His research int('iests include studies of several species of raptors, including the Monkey-eating Eagle of the Philippines. Other studies deal with avian population dynamics. Mr. Kennedy has had extensive field experience in the Philippines and in North and South America. In addition to his research interests, Mr. Kennedy enjoys traveling, and collecting and marketing wildlife art. COWBIRD PARASITISM AND EGG RECOGNITION OF THE NORTHERN ORIOLE Stephen I. Rothstein Little information exists on host-parasite interactions between the Northern Oriole {Icterus galbula) and the parasitic Brown-headed Cowhird {Molothrus ater) (Friedmann 1963) even though both species are abundant and broadly sympatric. The small number of nests known to have been parasitized is not due to a scarcity of observations on oriole nests; e.g., parasitism was observed at only 8 (2.5%) of 318 oriole nests in Ontario (Peck 1974). Alternative ex- planations can account for the scarcity of observed parasitism: ( 1) Northern Orioles typically accept cowhird eggs but are rarely parasitized; or (2) North- ern Orioles typically eject cowhird eggs causing a large proportion of cow- bird eggs to disappear before observers see them. Under the first explanation, the frequency of observed parasitism would equal the frequency of actual parasitism. But under the second, incidences of observed parasitism would always be less than incidences of actual parasitism and orioles might be fre- quently parasitized even though parasitism is rarely seen. If orioles typically eject cowhird eggs, the cases of natural parasitism most likely to be seen would he those rare ones in which parasitic eggs are accepted. Thus observations of natural parasitism do not give reliable data on the fre- quency with which birds eject cowhird eggs. Reliable data on ejection can be derived by experimentally placing cowhird eggs into nests because the experi- menter can determine the fate of all the cowhird eggs within a sample. In 1968 I placed an artificial cowhird egg in a Northern Oriole nest. The egg was ejected within 24 h. This was one of a series of experiments on many species. These experiments demonstrated little intraspecific variation in re- sponse to experimental cowhird parasitism (Rothstein 1975a, 1975b). Thus species were easily divided into 2 discrete groups — accepters and rejecters. Based on the 1 experiment and on several reports of ejections of naturally deposited cowhird eggs (Friedmann 1963, Smith 1972), I tentatively desig- nated the Northern Oriole as a rejecter species. New experiments on 27 addi- tional nests reported here demonstrate that this designation was correct. Ex- periments on 2 nests also deal with behavioral mechanisms responsible for the oriole’s egg recognition. MATERIALS AND METHODS Artificial eggs. — Artificial cowhird eggs (Fig. 1) made of plaster of Paris were used in most experiments. These eggs are identical to ones described elsewhere (Rothstein 1975a, 1975c) except that eggs used in nests whose number begins with “74-” or “75-” 21 22 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Fig. I. Egg Types Mentioned in Text. Top row: 2 cowbird eggs; an artificial cow- bird egg; 2 artificial cowl)ird eggs damaged by orioles (The surface of the artificial egg on tlu‘. left lias been blackened so as to better reveal the peek marks. The most heavily damaged ai(*a of the egg on the right is within the black lines.). Bottom row: House Sparrow egg after being ejected from nest 75-81; 2 Northern Oriole eggs; Red-winged Blackbird egg; Loggerhead Shrike egg from same clutch as eggs placed in nest 74-86. were coated with a clear aerylie polymer gloss medium, not with shellac. Controls per- formed on other sjiecies show birds do not reject these eggs because they are artificial (Rothstein 1975a, 1975c). 1 performed controls on orioles by experimentally parasitizing nests with a real cowbird egg or with real House Sparrow (Passer domesticus) eggs. The latter simulate cowbird eggs in color and size (Fig. 1, see data in Bent 1958). Experimental procedures. — During a single visit to each nest, one “parasite” egg was added and one “host” (oriole) egg was removed. I experimentally parasitized most nests between 12:00 and 18:30. Elsewhere (Rothstein 1975a) 1 discussed differences between my procedures and those usually employed by cowhirds; hut these differences have no detectable influence on the incidence of rejection. Experimentally parasitized nests were usually checked wdthin 24 ±; 2 h. If an experimental egg was not ejected I left it in the nest for at least 7 days except at nest 73-01 where nest checks ceased after 3 days. Nest stage. — Most naturally parasitized nests receive cowbird eggs during the host’s laying period (Friedmann 1963). The question of whether there is a correlation be- tween host response and nest stage can he answered by parasitizing nests throughout the cycle. I divided nests into 3 stages. Those known to he parasitized on or before the day Northern Oriole Nests Subjected to Experiments Simulating Cowbird Parasitism Rothstein • COWIilHI) PARASITISM OF ORIOLES MS w ^ W) be ^ Q rt .2’^ S S « 0-:Q o o 2 2 OJ t3 bo -H o VO r-l I r-l CM CM cd s ^ s s s ’2’'» ^ cd cd cc ^ o.a Qco \OLOLO'^cOeO'OLOrti'^cOOLOLOTj ■ ■ ■ I II ■ ■ I > ■ . ■ I I < t • • • I 25 30 5 10 15 20 25 30 5 10 15 30 June July August Sept. Fig. 2. Duration of breeding activities of the Red-necked Grebe, 23 June to 29 August 1975. NEST BUILDING On 25 occasions (23 June-12 July) I observed nest construction. Obser- vations ranged from 1-90 min duration (Fig. 2). The very brief periods of nest building ( 1-4 min ) appeared to have more significance as post-copula- tory behavior than actual nest construction. The mean duration of nest building bouts, excluding those occurring immediately after copulation, was 21 min. Nest building was observed at 9 locations which were from 2-70 m apart (Fig. 1). On several occasions the 2 closest nests were worked on simul- taneously. The number of days each nest site was attended by the pair is depicted in Fig. 3. The construction of numerous nests apparently is not un- common. Speirs et al. ( 1964) recorded 7 nests built by one pair of Red- necks on Lake Ontario. The nest site appeared to be chosen by the male either by poking at the future site with his bill, by starting to carry nest materials to a particular spot, or by Invitation. On one occasion the male left the female on a nest site they had been working on for 4 days and had copulated on, swam 6 m to an- other clump of cattails and assumed the Inviting posture ( i.e. lying flat with neck outstretched and low and the bill pointed forward and almost touching the water). The female called several times but the male did not move. After I min the female joined the male and both began building at the new site. All nest sites were among the bulrushes and on floating clumps ( less than I m in diameter) of cattail roots and stems. Bulrushes, water milfoil, and lily pads were incorporated into the nest. These materials were collected within approximately a 5-m radius of the nest. Bulrush stems were picked up 36 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 9 8 7 6 CO ^ 5 ^ A (p 4 Z 3- 2- 20 25 June 30 5 10 July Fig. 3. Duration of nest building aetiviti(‘s at 9 nest sites, 23 June to 12 July 1975. singly and carried crosswise in the bill. Water milfoil was obtained during a brief dive. Lily pads were half-carried, half-dragged through the water and lifted onto the nest. Nest construction appeared to involve 2 steps. J'irst, bulrushes, lily pads, and some water milfoil were placed on a clumi) of cattails. This provided a floating i)latform callable of supporting eggs and an incubating bird. Second, a simple depression was formed by one l>ird on the nest receiving materials (almost exclusively suhmergent vegetation brought up from the bottom I from its mate and pulling these around itself into a low rim. Most of the rim was constructed during the first 2 days of incubation. Nest building was performed by both birds, although the male was the principal builder during the early stages and was observed vigorously piling vegetation on the nest while the female swam hack and forth, rested, or preened a couple meters away. Later, the male frequently carried materials to the nest where the female, on the nest, arranged them around herself. As the day of the laying of the first egg approached both l)irds were often simultaneously involved in the nest building and on one occasion they worked together con- tinuously for 74 min. On several occasions I saw the female building alone. During the nest building period the birds rarely approached the nest site alone, although one freiiuently departed before the other. When one finished foraging before the other, it called, preened, and waited until the other joined it. Only when the birds were together did they cautiously return to the nest site several body lengths apart. The nest in which the eggs were ultimately laid was the seventh nest begun Chamher/in • RED-NECKED (DiEBE IN MICHIGAN 37 by the pair ( excluding the nest with the 7 eggs j and was among the bulrushes at the edge of the inner open water area in 1.1 m of water (Fig. Ij. It had an inside diameter of 15-16 cm and an outside diameter of 38-42 cm. The depth of the depression was 2.5 cm and the top of the rim was only 5 cm above the water level. The first nest (found on 16 June) was floating in 0.5 m of water, 25 m from the shore, and only 30 m from a road. I suspect it became detached at its anchorage, drifted into the shore, and was consequently aban- doned. The nest was a sodden mass of bulrushes and water milfoil with a 33- cm outside diameter above water and a 61-cm diameter under water. The top of the nest was 6 cm above water and the depression containing the eggs was 15 cm in diameter. COPULATION I observed copulation 6 times from 30 June to 11 July (Fig. 2). The pro- cedure for all copulations was essentially the same: (1) The female climbed onto the nest platform and Invited. On 2 occasions the female uttered a faint, plaintive call. (2) Within 0.5-2.0 min the male mounted the female and copu- lated, while on the nest. Copulation was 3-7 sec in duration and accompanied each time by the copulating call ( “Rattern” ) described by Wobus (1964). (3) Immediately after copulation tbe male walked over the female’s head and shoulders and into the water at which time both birds raised their heads. This was followed by (4) Head Shaking by one or both birds or both birds. Slow Swaying ( “Wegsehen” ) , and (5) either both birds preened briefly or the female preened while the male briefly collected nest material. My observations were in accord with those of Wobus (1960, 1964). The faint vocalization of the female in the Inviting posture may correspond with the platform call of the Horned Grebe described by Storer (1969). EGG LAYING Three eggs were laid. The first egg was laid on 11 July and had a bluish matrix which became, by the time it hatched, dark brown due to staining from wet vegetation. I don’t know the exact dates of the laying of the second and third eggs. Wobus (1964) found that the average clutch size for July-nesting birds was 2.5. ! INCUBATION I observed incubation for 151 h and 39 min from 11 July to 9 August (Fig. ' 2). Incubation was shared by the sexes, the male incubating 41% of the time and the female 59%. For comparison I divided the day into two 8-hour I blocks; one representing mid-day (10:00 to 17:59) and the other morning ! and evening (06:00 to 09:59 and 18:00 to 22:00). In the morning and evening intervals the female incubated 65% of tbe time, whereas during the 38 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 o oooooooooooooooo o oooooooooooooooo O h-OOOO— CVirO^LOOl^OOOO — C\j O OOO— — — — — — — — — — C\JC\JC0 Time Of Day Fig. 4. Average percent of time spent on the nest by sex and time of day. Each time block represents a minimum of 5 liours of observation. middle of the day she was on the nest only 41%. During all observations be- fore sunrise and after sunset the female was on the nest, suggesting that she lierformed most of the night-time incubating. From day to day each sex tended to incubate at approximately the same times (Fig. 4). As the birds adjusted to the incubation routine, the average duration of their individual attentive i)eriods increased. The average attentive period lasted 87 min during the first week but increased to 132 min during the second with a corresponding decrease in the number of periods per day from 10 to 7. File last few days of incubation were similar to the first days of incubation in that the mean duration of the attentive periods decreased to 78 min and the freciuency of change-overs increased back up to 10 per day. Wobus ( 1961 1 found the average duration of attentiveness to be 1-2 h. An all-day observation on the third day of incubation revealed that incuba- tion was continuous, or nearly so, during the egg-laying and early incubation periods. During the first 3 days incubation was infreiiuently interrupted by brief periods of nest building and coi)ulation but from the fourth day on the eggs were very rarely and briefly left unattended. During the nest reliefs, or change-overs, the returning bird’s behavior ap- peared dependent not only on the strength of its own urge to incubate but also on the incubating lord’s readiness or reluctance to leave the nest. Occa- sionally during extremely hot weather (e.g. 32°C) and after unusually long periods of attentiveness (e.g. 3M.'-4 h), the mere presence of the returning Chamberlin • RED-NECKED GKERE IN MICF1K;AN 39 bird was sufficient to induce the incubating l)ird to leave the nest. Head Shaking by tbe returning bird was the dominant component of nest reliefs and in most cases induced its mate to leave the nest. Head Shaking was part of 49 (68%) of the 72 nest reliefs I observed. During the first week’s nest reliefs the returning bird Head Shook as many as 6 times and often the incu- bating bird also participated in Head Shaking. By the second week Head Shaking was primarily by the returning bird and only done once or twice per nest relief. If Head Shaking failed the returning bird often sat next to the nest (usually to the rear of the incubating bird) and performed comfort movements or poked at the nest for several min. When its mate still remained on the nest the returning bird made brief nest building actions. Twice the male simply “gave up” after these attempts and left for a while; once the female jumped onto the nest forcing the male off. Herring Gulls {Larus argen- tatus) demonstrate a similar behavioral progression during nest reliefs (Tin- bergen 1960, pers. observ.j. The returning gull’s inducements ranged from its mere presence on the territory, to Mewing, to Choking, to bringing nest material to the nest, to physically evicting its mate from the nest. Three times during the first 2 days of incubation the female, on leaving the nest. Reared and Wing Quivered, thus “coaxing” the male onto the nest. These 3 occurrences were the only times I saw Wing Quivering. Storer ( 1969) discussed this display as the most intense form of soliciting. The Inviting posture, a milder form of soliciting, was assumed by the incubating bird as its returning mate swam towards the nest. Inviting remained as part of the nest relief pattern through the seventh day, after which I no longer saw it. Prior to egg laying the nest platform had been used primarily as a copulation platform. Thus the occurrence of soliciting postures during the first nest reliefs suggested they were a carry-over from copulation, and possibly such actions on the part of the female encouraged the male’s transition to incu- bating behavior. Also during the first 2 days of incubation, nest reliefs were twice initiated by the female (as the returning bird ) carrying nest material to the nest but not depositing it thereon. Instead she swam back and forth in front of the male as if to entice him off the nest by an activity in which he had, until recently, been vigorously involved. Carrying nest material, as well as solicit- ing, may have reflected the ambivalence present in the birds as they changed from one behavior pattern to another. I saw none of these activities as part of nest reliefs after the first week of incubation. The nest reliefs gradually be- came less complex ( i.e. fewer movements and postures ) as various components were “phased out.” Several change-overs occurred in which I saw none of the usual cues, but rather they appeared to be initiated by impatience, rain damage to the nest, 40 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 or disturbances. On the third evening of incubation the male abandoned the nest after a long period of incubating. Both birds returned 27 min later and the female climbed onto the nest. Once after a 44-min downpour, the male slid off the nest and began nest repairing. The female appeared 2 min later and climbed onto the temporarily abandoned nest. The male continued to repair the nest for an additional 18 min. Change-overs occurred twice when the incubating bird was frightened off the nest by fishing boats and once by one of my nest checks. The birds approached the nest cautiously from the open water, rather than through the bulrushes. During the first 10 days the birds swam on the surface to the nest, usually with considerable head-bobbing movements. Starting on the eleventh day the female approached the nest underwater, diving from 20-30 m away. Seven times when the female surfaced right next to the nest and face to face with the incubating male they both raised their necks up as tall as possible, crest plumes erect, and called loudly and simultaneously. The same display occurred twice when the incubating bird left the nest at the sight of the returning mate. The birds displayed and called as they swam towards each other, and turned face to face as they passed. This mutual upright l)osturing and vocalizing ai)parently was a greeting of mate recognition. According to Storer ( i>ers. comm. I this vocalizing corresponds to the Triumph Ceremony of the Horned Grebe and the greeting trills of the Pied-billed { Podilyinbus podiceps) and Least grebes iPodiceps dominicus) . After each nest relief the departing bird spent 3-20 min (mean = 8) preen- ing before it swam out through the bulrushes to forage in the channel. The male was markedly more vocal than the female and often called during his returns to the nest, although by the 6th day his returns had become silent. I rarely heard vocalizing in the immediate vicinity of the nest after the first week of incubation, except during the aforementioned change-overs and after disturbances. LATE IX CUB ATI ON During the last 5 days of incubation, nest building was freciuent and occurred in conjunction with 7 of the 18 nest reliefs of this period. Except for one instance of nest repair, I had not observed extensive nest building since the second day of incubation when rim construction was completed. Since token nest building was a strong nest relief cue, such behavior may have indicated strong drives to incubate or possibly it was displacement activity reflecting frustrations caused by the sounds of chicks within the eggs and internal changes in the birds’ drives from incubating behavior to broodiness. Two days before hatching a new behavior, which I call Lunging, was incorporated into 6 of the 8 observed nest reliefs. Lunging consisted of a Chamberlin • RED-NECKED GREBE IN MICHIGAN 41 slabbing motion of the bill towards the inculiating bird’s back and was made by the returning bird as it sat next to the nest. Lunging was apparently an intention movement of feeding the young, during which the adult presents food in the bill to the chicks on the other parent’s hack. Several days prior to hatching the incubating bird freciuently stood up and either looked down at the eggs or rearranged the nest material around them. Such behavior was most likely stimulated by chick sounds within the eggs. During the earlier days of incubation the birds rarely stood up once comfortably settled on the eggs. During these last days of incubation the non-incubating bird spent considerably more time loafing in the vicinity of the nest than it had before and the birds started approaching the nest through the bulrushes, which they had not done previously. HATCHING The 3 eggs hatched on 6, 7, and 9 August I Fig. 2). Since the first egg was laid on 11 July, and incubation began on the same day and was con- tinuous throughout, the incubation period for the first egg was 26 days. Bent ( 1919 ) determined the period of incubation to be 22-23 days for eggs he hatched in an incubator. The eggs’ constant contact with wet vegetation and the possibility that, although the grebes were continuously on the eggs starting with the laying of the first egg, heat transfer may not have yet been complete might have accounted for the longer incubation period in the wild. Wobus ( 1964) gave the average incubation period as 23 days but added that it is often longer due to cold weather and/or disturbances. Hatching occurred in the mid-morning. From 09:49 to 10:24 on the morning the first egg hatched the incubating bird showed considerable un- easiness and stood and looked down at the eggs 6 times. The next day the second egg was intact at 07:30 and the second chick was first observed crawling out from under the incubating adult at 11:45. Two days later at 07:48 the third egg was still intact but during the change-over at 12:38 1 saw the chick in the bottom of the nest while the other 2 were in the water. Before settling onto the nest the male picked up the egg shell and dropped it over the rim of the nest. BROODING The chicks were brooded on the parents’ backs under their wings when the adults were on the nest as well as on the water. This undoubtedly had survival value considering the cold, wet state of the nest and the presence of acjuatic predators such as the northern pike iEsox lucius). Brooding was performed by both sexes and brooding periods ranged from 57-162 min (mean = 119). During change-overs on the nest the brooding bird stood up. 42 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 spread its wings, and shook the chicks off its back and into the nest. After the adult dismounted, its mate climbed onto the nest and raised its wings 4-5 cm off its back allowing the chicks to crawl up and under. During change- overs on the water the brooding bird raised itself into a nearly vertical position and shook off the chicks by wing-flapping. At 3 weeks of age the chicks were no longer brooded on the adults’ backs. After the hatching of the third and last chick on 9 August the nest was used during most of the following day, from 09:29 to 20:52, and then abandoned. I never saw the birds using the nest after 10 August. FEEDING THE CHICKS Both parents fed the chicks. During the first week the chicks were fed 2- to 5-cm minnows, small unidentified items ( probably insects), and an occasional feather. The age of the first chick I observed being fed a feather was 2 days. Wobus ( 1964) observed chicks being fed feathers, as well as insect larvae, during their first day of life. After the first week the food appeared to be almost exclusively fish in the 4-10 cm range. During the third week the fish were noticeably heavier bodied and once a medium-sized crayfish was fed to one of the chicks. As the parents swam towards the chicks with a fish they repeatedly dipped the fish in the w ater and appeared to be manipulating and pinching it in their bills, as described by Sim (1901 1 . I bis probably killed and softened the fish and made swallowing and digestion easier. Ihe food was held in the tip of the bill and presented to tbe clucks while they were on the other parent’s back. I he chicks freciuently dropped the minnows during the first several days’ feedings but the parents picked u]) the dropped minnow and presented it repeat- edly until the chick finally got it headfirst into its mouth. I he brooding parent frecpiently picked up any dropped items and fed the chicks on its own back. By the second week the chicks were fed on the water where they persistently begged for food and swam out to meet the parents each time they returned w ith food, and occasionally even pursued their parents underw ater. Feeding periods during the first week ranged from 23-113 min ( mean = 75) with a mean of 9 feedings per period (Table 1). Feeding intervals ( i.e. the time between individual feedings I ranged from 1-32 min (mean = 8). A 28- to 129-min loafing period (mean = 58 I, during which the non-brooding bird loafed and/or foraged for itself, immediately preceded or followed each change-over. During the second week feeding periods were one-third as long as during the first week while the number of feedings per period more than doubled due to the 6-fold reduction in the length of time between feedings (Table 1 ). Thus the chicks’ growing demand for food was met by decreasing the time interval between feedings. Chamberlin • KED-NECKEI) GKEliE IN MICHIGAN 43 A Weekly Gomparison Table 1 OF Feeding and Loafing Periods Week 1st 2nd 3rd Duration of feeding period s (min) 23-113 16-34 4-27 (75)=^ (25) (17) Number of feedings per period 2-15 10-31 7-32 (9) (19) (16) Feeding intervals (min) 1-32 0.5-12 0.25-7 (8) (1.4) (1.1) Duration of loafing period s ( min) 28-129 32-64 23-39 (58) (50) (31) Mean number of feedings per hour 3.5 12.5 18.0 * Means are given in parentheses. The mean duration of the feeding periods and feeding intervals continued to decrease through the third week ( Table 1) . The reduced number of feedings per period simply reflected the shorter duration of the periods. The very short time intervals such as 0.25 min between some feedings were probably the result of both parents simultaneously feeding the chicks. Short time intervals probably also occurred when the birds found their prey concentrated in large schools. Once when the chicks were fed 74 times in a 25-min period, every fish appeared to be the same size (4-5 cm) and while the birds fished they moved steadily along as if following a school. During the third week the mean duration of the loafing periods was 47% and 38% shorter than during the first and second weeks, respectively. Thus although the feeding periods were shorter they were also more frequent, as indicated by the reduced amount of time the adults spent loafing in between. As the chicks grew the mean number of feedings per hour increased steadily from 3.5 the first week, to 12.5 during the second week, to 18.0 during the third week (Table 1). 1 last observed the family on 28 September 3.5 km from the nest. The 3 chicks, at 51, 53, and 54 days of age, were still being fed by both parents. According to Wobus ( 1964) the family bonds break up after 8 to 10 weeks. INTERSPECIFIC RELATIONS Red-winged Blackbirds (Agelaius phoeniceus) were in frequent attendance of the grebes’ nest building. After the grebes’ departure from a nest platform, the Red-wings immediately dropped down to the nest and appeared to be snatching up insects, probably brought up with or attracted to the wet vegeta- 44 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 lion. Deusing (1939) watched a Longd3illed Marsh Wren (Telmatodytes palustris) catching insects on a Piechbilled Grebe’s nest. The incubating grebes tolerated the passing and activities of other marsh Hesters such as the Piedd^illed Grebe, American Coot [Fulica americana) , Black Tern (Chlidonias niger) and Red-winged Blackbird near the nest. The several species of ducks (Anas platyrhynchos, A. rubripes, A. discors, and Aix S])onsa) common in the marsh were tolerated as they fed near the nest prior to the hatching of the grebe chicks. The grebes assumed a defensive posture (neck and head upstretched with the bill directed towards the cause of the alarm) whenever Herring Gulls flew low and noisily over the nest. Once 2 immature Ring-billed Gulls ( Larus delawarensis I dived at one of the 3- week-old chicks and stimulated a half- running, half-flying attack by both adults. These same 2 gulls also tried to rob the male grebe of a crayfish, but only forced him to dive out of their reach. A female Marsh Hawk [Circus cyaneus) , gliding only 4-5 m directly over the nest, caused the male, with the chicks on his back, to leave the nest. At hatching the male became strikingly territorial towards other species. On one occasion the male drove off 2 immature Pied-billed Grebes that were foraging at least 40 m from the nest. The male approached one of the Pied-hills in a very i)ronounced threat attitude, hunched very low on the water. The male dived and in the same instant the Pied-bill half-ran, half-flew 3-4 m across the water. The male surfaced and continued to i)ursue the Pied-hill in a threat attitude. When the male dived a second time the Pied-bill flew off. I he male then turned his attention to the other young grebe and with a similar secpience of actions drove it away. According to Storer (1967) grebes seem to fear underwater attacks and so do not remain on the surface if an aggressive grebe dives. In aggressive situations diving is ‘‘understood” as a threat between different s|)ecies, and even genera, of grebes but apparently not between higher taxonomic groups. W hen the male approached a Mallard feeding 10 m from the nest in a threat attitude, the duck continued feeding, and as the male dived the duck remained oblivious to the grebe’s actions. However, several seconds after the male dived the Mallard si)iang into the air and, (juacking loudly, flew off. Apparently the duck did not “interpret” the dive as a threat and so fled only after (presumably) being physically attacked from underwater. I he pair’s interspecific territoriality appeared to be in defense of the brood and not the nest site or any fixed area of the marsh. This was suggested by the rarity of agonistic behavior prior to and during incubation and the sudden aggressiveness at hatching. Also, their aggressiveness extended far beyond the nest site and even after the nest had been abandoned. As the brood moved so did the territory. Both adults frequently chased away Pied-billed Grebes, Chamber/in • RED-NECKED GREBE IN MICHIGAN 45 Mallards, a female goldeneye {Bucephala clangula), and a Great Blue Heron [Ardea herodias) that came close to the brood as they traveled along the marshy shorelines of the channel. Twice during the hatching period a muskrat (Ondatra zihethica) swimming close to the nest was threatened by the male. When the muskrat dived the grebe immediately followed. Shortly, the muskrat surfaced and continued on its way and the male returned to the nest. The only other interaction w ith muskrats was that the abandonment of nest site #4, after 4 days of use by the grebes, coincided with muskrats starting to use it for one of their feeding platforms. SUMMARY A pair of Red-necked Grel)es iPodiceps grisegena) was studied in a northern Lake Huron marsh from 20 June to 29 August 1975. Observations were of a renesting attempt and courtship l)ehavior was brief and infreciuent. Copulation occurred on the nest platform from shortly after nest building began into the egg-laying period. The nest site was selected by the male although both sexes built the nest. The pair constructed 9 nest platforms, one of which ultimately became the nest in which 3 eggs were laid. Incubation began with the laying of the first egg and both sexes incubated, although the male incubated more during the mid-day and the female more in the morning and evening hours. Nest reliefs were initiated primarily by Head Shaking by the returning bird. The first egg had a 26-day incubation period. Hatching occurred in mid-morning. The nest was abandoned 2 days after the hatching of the last chick. The chicks were brooded on the adults’ backs under their wings. Both sexes brooded and fed the young. Food items consisted of minnows, crayfish, and probably insects. The mean number of feedings per hour increased from 3.5 to 12.5 to 18.0 during the first, second, and third weeks, respectively. Three chicks were successfully raised to over 7 weeks of age. ACKNOWLEDGMENTS I thank Dr. Robert W. Storer for his helpful comments and Dr. Nicholas L. Cuthbert for proofreading an early draft of the manuscript. Appreciation is also extended to Jim Hammers for translating several articles. LITERATURE CITED American Ornithologists’ Union. 1957. Check-list of North American birds, 5th ed. Baltimore, Am. Ornithol. Union. Bent, A. C. 1919. Life histories of North American diving birds. U.S. Natl. Mus. Bull. 107. Deusing, M. 1939. Nesting habits of the Pied-billed Grebe. Auk 56:367-373. Jones, S. P. 1938. Holboell’s Grebe and American Brant in Wisconsin. Auk 55:666. Mahan, H. D. 1963. Michigan bird survey, summer, 1962. Jack-Pine Warbler 41:75. McAllister, N. 1958. Courtship, hostile behavior, nest-estahlishment and egg laying in the Eared Grebe iPodiceps caspiciis) . Auk 75:290-311. Sim, R. J. 1904. Notes on the Holboell (irebe (Colymbus holboellii). Wilson Bull. 16:67-74. 46 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 SpEiris, J. M., G. W. North, and J. A. Crosby. 1944. Holhoell’s Grebe nesting in southern Ontario. Wilson Bull. 56:206-208. Storp:r, R. W. 1967. Observations on Rolland’s Grebe. Hornero 10:339-350. . 1969. The behavior of the Horned Grebe in spring. Condor 71:180-205. Tinbergen, N. 1960. The Herring Gull's world. Doubleday & Co., Inc., Garden City. N.Y. WoBUS, U. i960. Nestbau und balz des Rotbalstaucbers. Die Vogelwelt 81:61-62. . 1964. Der Rotbalstaucher (Podiceps grisegena Boddaert ) . A. Ziemsen, Witten- berg Lutherstadt. Zimmerman, D. A. and J. Van Tyne. 1959. A distributional check-list of the birds of Michigan. Univ. of Mich. Mus. Zook, Ann Arbor. INTEltLOCHEN AUTS ACADEMY, INTERLOCIIEN, MI 49643. ACCEPTED 20 JAN. 1976. NEW LIFE MP:MBEK Mr. Hub(‘rt B. Ziunickow is a new life member of the Wilson Ornithological Society. Mr. Zernickow is an Advisory Systems Engim*(*r with the IBM (Corpora- tion at Lansing, Michigan. His ornitho- logical interests are primarily as an (d>- server and photographer, and he has a special interest in owls. .Mr. Zernickow is a member of the AOU and is presently presi- dent of the Michigan Audubon Society. He is married and his wife, Norene. shares bis enthusiasm for birding. In addition to his ornithological interests. Mr. Znnickow •‘iijoys botany, hiking, ami canoeing. GROWTH AND DEVELOPMENT OE THE Pr.AIN CHACHALACA IN SOUTH TEXAS Wayne R. Mauion Growth and development of many game birds have received thorough in- vestigation, but this is not true for the Plain Chachalaca ( Ortalis vetula ) . I studied growth and development of chachalacas as part of a larger study (Marion 1974) of the ecology of this species in Texas. METHODS Research was conducted from January 1971 to August 1972 at Santa Ana National Wild- life Refuge, and 3 other study areas in the Lower Rio Grande Valley, Hidalgo and Starr counties, Texas. Data reported in this paper were obtained from captive birds, live- trapped birds, and collected specimens. Captive chachalacas were reared in 2 weld-wire pens at Santa Ana National Wildlife Refuge headcfuarters. An attempt was made to keep the pens as natural as possible; they included several small trees and additional plant materials which provided cover, shade, and sites for perching and nesting. One pen (4.6 X 3.0 X 2.4 m) contained 3 adult birds fl^, 2$ $). Another slightly larger pen (6.1 X 7.2 X 2.4 m) contained 7-9 immature birds (2-3 9 9, remainder^ ^). Captive juveniles were hatched in an incubator from eggs collected from 4 nests during 1971. Fresh water and commercial foods were pro- vided ad libitum. Captive birds were fed commercial starter, grower, and maintenance rations, corresponding to stage of maturation. Natural foods were fre(juently provided to supplement commercial foods. Chachalacas were live-trapped at the main study area in 25 X 50 mm mesh weld-wire traps (1.2 X 1.2 X 0.6 m) with funnel entrances similar to those described by Taber and Cowan (1963:261). Twelve traps were operated at randomly selected sites during late winter and early spring of both years of this study. Traps were baited with fresh cabbage and grain sorghum placed within the enclosure and near the funnel entrances. To avoid excessive stress on handled birds due to overheating, trapping was restricted to the cool morning and evening hours. Trapped birds were marked with aluminum hands and colored leg streamers for subsequent individual field recognition. Chachalaca specimens w^ere shot in a nonselective manner on all study areas between September 1971 and August 1972. Many birds were collected in the morning and evening when they were more active; fewer were collected at midday. Data from carcasses of chachalacas found dead during this study also were recorded. There is no obvious sexual dimorphism in this species, hut all birds handled were sexed using at least 1 of 4 known methods. Two of these methods, convenient for sexing live birds, were related to the presence or absence of a looped trachea. The adult male has a trachea lengthened by a loop which is easily felt between ventral musculature of the breast and the skin; this looped structure is lacking in young males and females (Merrill 1878). During this study, inspection of gonads of sacrificed birds verified this sexual difference in tracheal development of adult birds. Determination of the presence or absence of the tracheal looj) by feeling the breast was the major technicjue used in sexing older juveniles and adults. 47 48 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Generally the longer and wider the trachea, the deeper the bird’s voice; the shorter and narrower the trachea, the higher the voice. Adult males have a longer trachea and their voice is a full octave lower than that of females and young males (Sutton 1951:127). The pitch of the voice was a second method used in sex determination when live birds were heard calling. Chachalaca males have a penis which can he readily observed by cloacal examination, hut this teehniciue was rarely used because it re(|uired more handling of birds than the previously described tracheal loop method. All collected specimens were sexed by exam- ination of the gonads. Definitive aging criteria for chachalacas have apparently not been reported. I recorded tracheal loop lengths, measured externally from the distal portion of the loop to the |)oint of entry into the thorax, for use as aging criteria for male birds. Total length and diameter (at each end and near the middle) of the trachea of all collected specimens were measured to determine sex and age differences. Vernier calipers, permitting readings to the nearest O.I mm, were used in measuring tracheal diameters. Postal scales were used to measure total body weight to the nearest 0.5 ounce. These data were later converted to ecpiivalent values in grams. Several standard length measure- ments for birds, including total body, wing chord, tail, exposed cuhnen, tarsus ( tarso- metatarsus), middle toe, and total extent of wings, as described by Baldwin et al. (1931) and Pettingill (1946:323-325), were taken to the nearest 1 mm using a pair of dividers and a ruler. These measur(*ments were recorded for captive (at intervals of 1 month or less), live-trapped, and colIectc 0.05 ) . External tracheal loop measurements were recorded for male chachalacas. Mean loop length of captive juveniles was 17.5 ±7.1 mm (range 13-23) at 9 weeks of age (N = 2) and 30.0 ± 7.2 mm (range 34—38 mm) at 10 weeks of age (N = 3). The tracheal loop of captive juveniles elongated slower than other body parts; measurements began overlapping those of adults when young males were approximately 9 months old (Fig. 1). Tracheal loop development was apparently slower in wild juveniles than in captive juveniles. Mean loop lengths for wild juveniles handled during February (N = 9) and March 1972 (N == 4) were 51.2 ± 4.4 mm (range 46-60 mm) and 54.1 ± 3.5 mm (range 49-57 mm), respectively. The aver- age loop lengths of 5 captive juveniles in February and March were 62.1 ± 4.3 mm (range 58-67 mm) and 64.8 ± 2.4 mm (range 63-68 mm), respec- tively. These measurements for captive juveniles were significantly larger than those for wild juveniles in February (t = 4.4, P < 0.01) and in March (t = 5.5, P < 0.01). Several variables, all related to the exact age of wild (unknown age) vs. captive (known age) juveniles, may he responsible for this difference hut the major cause remains unknown. These data generally indicate that wild juveniles may he distinguished from adults using tracheal loop lengths until at least 9 months of age. By 1 year, tracheal loop development was nearly complete and differences among males were subtle. Further increases in tracheal loop length as males aged were apparently minor. Four banded males, known to be more than 5 years old, had an average loop length of 78.0 ± 3.1 mm (range 75-82 mm). This mean value, although slightly larger than the average loop length of 44 other adult males (74.2 ± 4.4 mm, range 67-87 mm), was not significantly larger (t = 1.7, P > 0.05). Significance of color of the upper mandible was investigated as an indi- cator of age. Presence or absence of a dark tip on the upper mandible was recorded for 17 juvenile males (tracheal loop partially developed) and 17 adult males (tracheal loop well developed) handled between December 1971 and March 1972. Upper mandibles of all 17 juvenile males had dark tips. Only 2 of 17 (11.8%) adult males had dark-tipped upper mandibles. The remaining 15 ( 88.2%) had uniformly colored ( blue horn ) bills. These results indicate that dark markings near the tip of the upper mandible are characteristic of juvenile birds. 90 80 70 60 50 40 30 20 10 0 ir(*mei THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 N (Adult Males) = N (Juvenile Males) = 2355555555555 5 5 3 I I A S I 2 3 T‘ T""'l I I I I I " ONDJFMAM 4 5 6 7 8 9 10 II AGE (MONTHS) raclieal loop lengtlis of wild adult and captive juvenile cliachalacas. External s were taken and data presented are means (rtSD) and ranges. Marion • CHACHALACA GROWTH SI Strength of the lower mandible varies with age in many gallinaceous birds. Generally, if a dead bird is supported only by the lower mandible and it breaks, the bird is a juvenile; if the lower mandible does not break, it is an adult (Leopold 1933:166, Taber 1963:134). Lower mandible strength was determined for males of known age ( by tracheal loop development ) collected between December 1971 and March 1972. When subjected to the “lower mandible test,” all 4 juvenile males had mandibles that broke. Each of the 10 adults tested had a lower mandible that supported the bird’s weight. Strength of the lower mandible seems to be a valid technicjue for dis- tinguishing juvenile and adult chachalacas. Males of known age (determined by tracheal loop development) handled between December 1971 and March 1972 were used to investigate leg color differences between juveniles and adults. Of 18 juveniles, 8 (44.4%) had legs that were slightly orange. The others had darker ( blue horn ) legs. Of 22 adults, only 2 (9.1%) had slightly orange legs. A significant (P < 0.05) chi square value of 6.6 indicates that orange legs are more tyi)ical for juve- niles than for adults. Age determination in gallinaceous birds commonly involves plumage char- acteristics (Taber 1963:128), however, plumage differences between juvenile and adult chachalacas diminish rapidly as young birds mature. At 2 or 3 months of age, slight differences exist in width and shape of flight feathers. Rectrices and remiges of juveniles are relatively narrow and pointed, while those of adults are broad and rounded. Most of these plumage differences are lost with the postjuvenal molt before the juveniles are 6 months old. The outermost juvenal rectrices and primaries, however, may be retained slightly longer. Growth. — Chachalaca chicks are precocial and leave the nest within a few hours after hatching. Chicks are very active and agile in climbing through trees and shrubs within a few days of hatching and are able to jump and fly at least 1.3 m at 6 days of age. Weights and measurements of wild adults (males and females) and captive juveniles (1 week and 1 month old) are presented in Table 1. Adult males averaged significantly (P < 0.01) larger than adult females in weight, total length, wingspan or extent, wing chord, tail length, exposed culmen length, tarsus length, and middle toe length. Although statistical comparisons of mean values indicate that adult males average larger than adult females, much overlap exists in the ranges of these measurements. These overlapping values reflect subtle sexual differences in size which are not easily recognized in the field. Mean adult weights were highest during October and November (631 ± 87 g and 646 ± 97 g, respectively), but were relatively constant during other 1 52 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 2 ^ Hi o I u w « H-l ^ ca H a C tqU' H I c bt'3^ .£ c E C S W — — •£'^ i: t£ ? o c ^ H =:; = +1 N CO o CO CSl r-H + 1 ^ ^ 2 CSl CM o\ 1 1 +1 ^ O l-H ,-1 Os CM --H p CM 2 p p r^' CM UO NO LO II ^ II CM II CO 1 lO o LO p cl P 1 o lO CO CO ON CM CM CO CM LO p p p ON CO ^ . . LO o NO CO o NO s II 5 o II p CM ct II p CO c!> \0 ^ CO CM CM LO LO LO LO CM CM CO CM nH ^ CO , + 1 00 ON CM i-O MO lO 'W LO LO "5 ie ^ +1 t- o rH* On CM '— ' + 1 lO NO t- O +1 ^ <=> s O Cm CM ^ • lO +'S LO o CM ON VO ^ II ^ r- II ^ ,-H CO 2 ?? II ^ II 7 LO ^ Tt CM ^ I o < - II o p no’ CO ON +|3 CO rH CM rH II ^ O CM CO fO ^ — II 7 CM t- CO CM CM 00 LO ON 52 09) ; 30 C^ II t- 1! M II ^ II s lO d NO CM ON Ht p lO CO O LO 149 vC On Data shown are mean ± SD and (range). WEIGHT (grams) Marlon • CHACHALACA GROWTH 53 800- 700- 600- 500- 400- 300- 200- 100- N ( Adults) = 3 4 8 7 2 5 23 12 65 73 ADULTS ^ 1 j! 1 - CAPTIVE JUVENILES N (Juveniles) = 998868777774 I I I r I I I I I 1 ASONDJ FMAM I 234 5678 9 10 II AGE (MONTHS) Fig. 2. Weights of wild adults and captive juvenile chachalacas. Data presented are monthly means (± SD) and ranges. 51 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 months (Fig. 2). An abundance of natural foods following floods in the fall of 1971 and heavy fat deposition may account for the increased adult weights during October and November. In addition, 11 of the 15 adults measured during these 2 months were males and this may have also contributed to the increase in recorded weights. Average weights of captive juveniles increased rapidly and, at 4 or 5 months of age, approached the 550-600 g level of adults (Fig. 2j. Similarly, hut at only 3 or 4 months of age, captive juveniles ap- proached adult size in total length, extent or wingspan, wing chord, and lengths of the tail, exposed culmen, tarsus and middle toe. Growth rates for wild juveniles may he slightly slower than this due to the disproportionate sex ratio among the captive juveniles favoring the males (and thus, larger size ) . Plumages and nwltin^. — Kectrices begin to develop at less than 1 week of age and grow rapidly. Initial rectrices are uniformly colored and rather narrow with i)ointed tij)s which are easily worn and broken. Remiges (except for the outer 3 primaries I are well developed at hatching and continue rapid development for several weeks after hatching. At approximately 1 month of age, juvenal i)lumage begins to replace natal plumage and juvenile birds begin to resemble adults. Body feathers of the head ajid neck region are the last to he replaced by drab olive plumage characteristic of adults. I he juvenal plumage stage in this species is rela- tively short. Juvenile chachalacas replace rectrices and remiges during the |)ostjuvenal molt. Postjuvenal molting of rectrices in captive juveniles oc- curred between August and December, when the birds were 2-6 months old. Postjuvenal molting began with intermediate ])airs (Nos. 3 and 4) in each half of the tail and proceeded both iiiward and outward until all pairs were replaced. Pair No. 3 was usually replaced slightly before i)air No. 4, hut both pairs were replaced in late August when juveniles were 2 months old. Replacement of i)airs 2 and 5 occurred in late September when the birds were 3 months old. Rectrices 1 and 6 were molted over a longer interval; the central pair (No. 1 ) was replaced between September and December and the outer i)air (No. 6) was replaced between September and January. Observations of wild juveniles handled during si)ring handing operations indicated that outer rectrices are occasionally retained until March. These older rectrices are easily recognized since they lack white tips and are obvi- ously old and worn. Following the postjuvenal molt, rectrices of young birds were white-tipped and both remiges and rectrices were relatively broad with rounded tips (as in adults). After the breeding season each year, feathers are replaced during the post- nut)tial molt. This molting is gradual and flight is not inhibited. Postnuptial molting of rectrices began as captive juveniles approached 1 year of age and Marion • CHACHALACA GROWTH 55 the sequence was extremely irregular compared to the post ju venal molt which followed a definite sequence ( 3-4-2-5-1-6 ) . Postnuptial molting of adult rectrices was also irregular, with no obvious pattern or se(iuence. Most rectrices of adults were molted during August and September, but this oc- curred as early as May and as late as December. Molting of primaries was se(iuential (proximal to distal) for both juvenile and adult birds. Molting observations for remiges of captive juveniles were not recorded prior to the age of 4 months. In the first year, captive juveniles replaced the outer 2 primaries (IX and X) during the postju venal molt. In most gallinaceous birds, except Ring-Necked Pheasants { Phasianus colchi- cus) , the 2 outer primaries are not replaced during postjuvenal molting (Taber 1963:134). Captive juveniles began the postnuptial molting of proximal (I and II) primaries in February, 2 months prior to adults and this continued until all distal primaries were replaced in the late summer and fall. Postnuptial molt- ing of adult primaries occurred in an ascending pattern similar to that de- scribed for juveniles. Replacement of proximal (I and II) adult primaries began in April and continued until all distal primaries were replaced in the fall. The majority of primary molting in adults occurred during August and September. Molting of secondary wing feathers was not as distinctly seciuential as in primary wing feathers. Secondaries of captive juveniles were molted during all months of year and postjuvenal molt was not clearly distinguishable from the postnuptial molt. Likewise, postnuptial molting of secondaries in adults followed no definite pattern, but most were being replaced during August and September. SUMMARY Plain Chaclialaca growtli and development were investigated in 1971 and 1972 in the Lower Rio Grande Valley of Texas. Chaclialaca chicks are precocial and growth and development of juveniles is rapid. At 4 or 5 months of age, juveniles resemlde adults and field recognition of differences becomes difficult. Size measurements are valid age criteria only during the summer and fall when juveniles are less than 4 or 5 months old. Differences in tracheal loop development (males), molting of outer primaries and rectrices, color and strength of bills and color of legs are valid criteria for distinguishing juveniles from adults. Plumage changes also occur rapidly; postjuvenal molting begins in early fall when juveniles are nearly 2 months old. Postjuvenal molting of rectrices follows a definite sequence of pairs (from innermost to outermost) and is usually completed before .Janu- ary of the first year. Juvenal primaries are also molted sequentially from the innermost to the outermost. Postjuvenal molting of secondaries is not distinctly sequential. Adult rectrices are molted in an irregular pattern during the postnuptial molt ( August and September). Adult primaries are molted in a sequential pattern (innermost to outer- most), but postnuptial molting of adult secondaries follows no definite pattern. 56 THE WILSON BULLETIN • Vol. S9, No. 1, March 1977 ACKi\OWLEDGMENTS I acknowledge with thanks financial assistance from the Caesar Kleberg Research Program in Wildlife Ecology at Texas University. Sincere thanks go to W. H. Kiel, Jr. for his advice and encouragement during this study. I am also indebted to K. A. Arnold, J. I). Dodd, T. M. Ferguson, and J. G. Teer for their critical reviews of the manuscript. Agencies granting collecting permits were the Texas Parks and Wildlife Department and the U.S. Fish and Wildlife Service. The latter agency also granted permission to hand and color-mark birds at the main study area. This paper is part of a dissertation in partial fulfillment of requirements of the Ph.D. degree at Texas A&M University. This is Texas Agricultural Experiment Station Technical Article No. 12054. LITERATURE CITED Baldwin, S. P., II. C. Ohkkhoi.skr. and L. (L Worley. 1931. Measurements of birds. Cleveland Mus. Nat. Hist. Sci. Puhl., Vol. 11. Cleveland, Ohio. Leopold, A. 1933. Game management. Charles Scribner's Sons, New York. Marion, W. K. 1974. Ecology of the Plain (diachalaca in the Lower Rio Grande Valley of Texas. Ph.D. thesis, Texas A&M Univ., College Station. Merrill. J. C. 1878. Notes on the ornithoktgy of Southern Texas, being a list of birds observed in the vicinity of Fort Brown, Texas, from February 1876 to June 1878. Proc. U.S. Natl. Mus., Vol. 1, jip. 118 173. Pethn(;ill, O. ,S., Jr. 1946. A laboratory and field manual of ornithology. Burgess Puhl. Co.. Minnea|)olis, Minnesota. .Sutton, (L ,M. 1951. Mexican birds; first impressions. Univ. Oklahoma Press, Norman. J arer, R. 1). 1963. (Iriteria of s(“x and age. Pp. 119-189, in Wildlife Investigational Techniipies. (H. S. Moshy. ed. ) 2nd ed. The Wildlife Society, Washington, D. C. AND 1. McT. (Iowan. 1963. Capturing and marking wild animals. Pp. 250-283, in Wildlife Investigational Technicjues. (11. S. Moshy, ed.) 2nd ed. The W'ildlife .‘“'ociety, Washington, D. C. CAESAR KLERERG RESEARCH l»ROGRAM L\ WILDLIFE ECOLOGY, DEPT. OF WILDLIFE AM) FISHERIES SCIENCES. TEXAS A&M UNIV., COLLEGE STATION 77845. PRESENT ADDRESS: WILDLIFE ECOLOGY LABORATORY, SCHOOL OF FOREST RESOURCES AND CONSERVATION, UNIV. OF FLORIDA, GAINESVILLE 32611. ACCEPTED 1 OCT. 1675. SOCIAL DOMINANCE IN WINTER ELOCKS OF CASSIN’S FINCH Fred B. Samson Reports of social dominance by females in avian winter flocks are few but have been described in the Bullfinch (Pyrrhula pyrrhula; Hinde 1955, 1956; Nicolai 1956) and the House Finch iCarpodacus mexicanus; Thomp- son 1960 ). I have noted this dominance in the Purple Finch i Carpodacus p. purpureus ) , and it is evident in this study of Cassin’s Finch ( Carpodacus cassinii) . The significance of female dominance in winter flocks is not known nor is the importance clearly evident for any pattern of avian social domi- nance during the winter (Watson and Moss 1970). The purpose of this study of winter flocks in Cassin’s Finch was to (1) assess patterns of social domi- nance, (2) suggest their possible ecological significance, and (3) describe displays involving agonistic or anti-predator behavior. Cassin’s Finch is an irregular winter resident of the Cache Valley in north- ern Utah (K. L. Dixon, pers. comm.) where I studied flocks during the win- ters of 1972-73 and 1973-74. 1 found no flocks in the area in 1971-72 or 1974-75. Aside from fragmentary observations by those engaged in faunistic or winter surveys ( Orr 1968 and references cited therein ) , little is known of the winter behavior or biology of Cassin’s Finch. METHODS I observed the activity and social dominance of finches ahnost daily from January to April 1973 and an average of 2 days per week from November 1973 to February' 1974. Five banding stations were established during the winter of 1972-73 at different sites within Cache Valley. All were at least 1 km apart with stations 1 to 4 in residential areas and station 5 at the mouth of Green Canyon. Cassin’s Finch visited only stations 2 and 3 during the second winter. I caught few finches in mist nets, but ca{)tured most in drop or walk-in traps baited with sunflower seeds and millet. Color of plumage was noted and wing lengths measured for all but 6 of 353 birds captured. Each bird was banded and I marked 131 with distinctive combinations of plastic color leg bands to permit later recog- nition without recapture. Cassin’s Finch females and yearling males have a similar streaked gray-brown plumage, but all females during the breeding season exhibit an incubation patch and also can be distinguished by wing length (Samson 1976). Wing length measurements in 3 summer populations I studied in northern Utah and those obtained in this study are not signifi- cantly different either for older males or gray-brown birds ( Samson 1974) . A criterion based on wing length similar to that employed for summer populations is used in this study to separate females (wing lengths of 85.0 to 89.9 mm) and yearling males (wing lengths of 90.0 to 96.9 nnn). As discussed under head-forward display, feather arrange- ment also may be used to identify females during agonistic encounters. I studied patterns of social dominance at or near banding stations. Finches concen- 57 58 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Table 1 Location and Number of Cassin's Finches Banded AND Recaptured Banding Station Nnmljer Banded Mean Per Day Number Recaptured at Banding Stations’^ 1 2 3 4 5 1972-1973 1 131 9.4 26 26 33 17 2 2 42 14.0 12 11 11 9 3 64 5.3 8 20 19 10 4 51 17.0 14 18 23 11 1973-1974 2 38 4.8 29 26 3 21 2.6 18 19 ^Aii iiiclividuiil bird may have J)een recaptured at more than 1 location. Iraled tlu'ir activity near the hait and were not older males <.01 females > yearling males <.05 older males > yearling males <.01 Feliruary 1973 females > older males <.01 females > yearling males < .001 older males > yearling males < .001 March 1973 females > older males < .001 females > yearling males < .001 older males > yearling males < .05 Noveinhcr 1973-Fcl)ruary 1974 females > older males < .001 females > yearling males <.001 older males > yearling males < .001 1 Clii-S(iuare with dl = 1. Heterosexual encounters most often occurred when a yearliiift; male ap- proached a feeding female or, rarely, when an older male attempted to sup- plant a female. In neither case were males regularly successful. Encounters of older males and females appeared to involve mistaken sex identification hy the male. Females were tolerant of other females, and 1 noted few inter- actions in either winter. Many finches were captured at more than I location ( "Pahle 1). In both winters, observers at the different locations noted the temporal and spatial association of color-marked birds. Comi)arison of these records indicates that feeding flocks of Cassin’s Finch lack continuity in membership from day to day and from feeder to feeder on any specific day. Pairs did form in these flocks during late winter hut well after the establishment of patterns of social dominance, l^iir status could not have influenced social dominance exhibited hy unpaired females less than a year old over older and yearling males. Thus, the dominance of females as a grouj) appears independent of site, flock com- position, or mate status. Winter disappearance. — The significance of female dominance in Cassin’s Finch may relate to improving their survival from breeding season to breed- ing season. In the winter of 1972-73, 64 of 85 females, 40 of 63 older males and 53 of 140 yearling males were recaptured at least 1 day following the initial handing. Significantly more females (P < .001 I were recaptured than Samson • CASSIN’S FINCH 61 expected. Conversely, significantly fewer yearling males (P < .001) were recaptured than expected. Attempts to locate or observe marked individuals within Cache Valley or adjacent mountain terrain that were not among re- captures were unsuccessful, and I presumed they were dead or had moved from Cache Valley to seek another food source. Fewer finches were winter residents in 1973-74 (Table 1 ) and few (n = 3j disappeared. The winter of 1973-74 was mild in comparison to 1972-73. Considering that the energy needs of a homeothermic animal increase as tem- perature decreases, both the milder winter conditions and fewer finches pres- ent to exploit available food resources may have contributed to the disap- pearance of few finches during the 1973-1974 winter. DISPLAYS Head- forward. — This display in Cassin’s Finch varied in intensity and, as in other finches ( Hinde 1955, 1956; Dilger 1960; Coutlee 1967), is divided into 2 categories, the low intensity head-forward display and the high intensity head-forward display. The closed beak is directed toward the opponent, the neck partially extended, legs slightly flexed, with the body tending toward a horizontal posture in the low intensity head-forward display (Fig. lA). If the aggressor is a female, the feathers of the forehead, breast, and hack are “shuffled” (Fig. IB) as in the House Finch (Thompson 1960). With females and yearling males nearly identical in plumage, this shuffling of feathers serves as a visual cue for sex identification in agonistic encounters. Rarely did females employ any other display to maintain their dominance or pre- ferential access to food or roost. Vocalizations did not accompany this or any other display. Figure 1C depicts the high intensity head-forward display. The beak is usually but not always open, the head and body feathers are sleeked, and the long axis of the body is horizontal and in line with the opponent. If the opponent was above or below the attacker, the head was directed toward the opponent and the tail slightly raised. During the most intense head-forward displays, both wings were raised through rotation at the shoulder (Fig. ID). Although performed by females and older males, the high intensity head- forward display was especially evident in encounters between yearling males. Combat. — I rarely noted combat (Fig. IE) between older males, among females, or in inter-sex encounters and did not observe it in the milder winter of 1973-74. Combat when evident usually occurred between yearling males. If a high intensity head-forward display was insufficient to dislodge an op- ponent, the attacker would proceed directly at the opponent with wings raised. If the opponent failed to yield, combat resulted. Combat did not result in noticeable body damage, and in most cases it was of short duration. 62 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 (E) comhat ; (F» suhmissive ; and ((0 anti-predator. Often, the birds would fly up almost vertically continuing to engage in combat before one or both birds withdrew to separate perches. Beaks remained open and feet extended during the combat phase of these flights. Submission. — When approached by an aggressor, suhmissive birds often Samson • CASSIN'S FINCH 63 assumed an erect, stiff-legged posture leaning away from the attacker ( Fig. IF). If not directly approached hut in the presence of a dominant bird, sub- ordinate birds would flex their legs and assume a partially crouched posture similar to that described for other fringillids ( Hinde 1956, 1957; Thompson 1960; Coutlee 1967). This posture is similar to that observed when an avian predator was present (Fig. IG). Sharp-shinned Hawks (Accij)Lter striatus) , Cooper’s Hawks ( Accipiter cooperii ) , and Northern Shrikes ( Lanius excuhitor) were active and preyed on Cassin’s Finches near handing stations. Finches in this posture remained stationary moving only the upper throat until the predator departed. The legs were flexed so that the breast and abdomen nearly rested on the substrate. Supplanting and avoidance. — As in the House Finch (Thompson 1960), I did not see special behavior by an attacking finch prior to supplanting a second bird. The direct or frontal presentation described for other Frin- gillidae (Hinde 1955, 1956) is apparent in Cassin’s Finch. In nearly all at- tempted supplants, the attacked bird flew l)efore the attacker landed. When the attacked bird did not flee, a lateral body presentation, a submissive pos- ture, or a slight fluffing of the feathers were considered indicators of avoid- ance. Aggressive chases among finches associated with supplanting were not observed either winter. Displacement activities (i.e., hill wiping, head scratch- ing, breast preening) were rarely observed in free-flying flocks hut were common in the 2 captive flocks. DISCUSSION Social dominance is not uncommon in avian winter flocks (Brian 1949, Sabine 1959, Dixon 1963, 1965; Kikkawa 1961, Zcihavi 1971). In these studies, males or males and their mates are reported dominant. In the House Finch (Thompson I960), Purple Finch and Cassin’s Finch, the members of this genus which breed in North America, females in winter flocks are either as or more dominant than males in agonistic encounters. This social dominance in Cassin’s Finch is considered independent of loca- tion in contrast to the importance of site attachment in other species ( Brown 1963, Dixon 1963). It may he related to ( 1 ) their lack of annual fidelity to a winter area (Bailey and Niedrach 1965, Buckley 1973), (2) the lack of consistent flock organization as in certain other carduelines ( Newton 1972 ) , (3) the mobility of the species, or (4) the variable number of finches at a winter area which may range from none as in Cache Valley in 1971-72, 1974- 75 to over 5000 as reported in northern Colorado ( Chapin 1958 ) . Other studies of finch populations during the winter (Fretwell 1969, Pul- liam and Enders 1971, Davis 1973) point out that food is important in deter- mining population levels and that intraspecific competition may influence 64 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 patterns of mortality. Newton ( 1964 ) provided evidence in the Bullfinch and Murton et al. ( 1966) in Wood Pigeon iColumba palumbus) that the avail- ability of winter food influences subsequent breeding population numbers. In Cassin’s Finch (Samson 1976) as in 2 other montane finches with sex ratios favoring males, the Black Rosy Finch {Lecousticte atrata, French 1959) and the Gray-crowned Rosy Finch ( L. tephrocotis, Johnson 1965), the num- ber of females is considered the limiting resource for reproductive effort. The significance of female dominance in Cassin’s Finch appears to involve the pro- tection of this limiting resource during the non-breeding season. Survival of females is enhanced by preferential access to food and roost sites in winter, thus allowing for maximization of reproductive effort during the subsequent breeding season. Considering that Cassin’s Finch, lacking a strong fidelity to a wintering area or breeding area, must colonize new wintering and breed- ing areas annually, a reproductive strategy to maximize reproductive effort may represent an important correlate to their nomadic lifestyle and enhance the efficient use of an unpredictable environment ( i.e., food and weather). dTiese habitat and species correlates all pertain to an r-strategy ( Pianka 1970). Opportunism and reproductive strategy in North American birds have not, however, been intensively studied (Cody 1972). J he displays used l)y Cassin’s Finch in agonistic encounters are generally homologous to those of the House Finch and to other fringillids ( Hinde 1955, 1956; Coutlee 1967). Cassin’s Finch does differ from many fringillids in that vocalizations did not accompany displays. This was particularly evident in intersi)ecific encounters between the Cassin’s Finch and the House Finch, the latter regularly using vocalizations in association with certain intense agonistic displays. In nearly all phases of its annual cycle, Cassin’s Finch tends to flock. The flocks are characterized by an al)sence of agonistic encounters except in winter and in those of yearling males which remain at high altitudes in late summer after other Cassin’s Finches have departed. Except among yearling males, the lack of intense agonistic encounters observed in this study may contribute to the flocking tendency. Aggressive behavior did increase at a food source as in the House Finch (Thompson I960), but this increase was not as substantial as that observed in early 1973 when weather conditions were severe and finches numerous. Nor, was it as intense as in yearling male flocks in late summer (Samson 1976). Females and yearling male Cassin’s Finches are well camouflaged in their striped gray-brown plumage when roosting on woody branches or foraging under a forest or shrub canopy. This coloration combined with the motion- less anti-predator posture may enhance their survival from breeding season to l)reeding season. However, the explanation for the imbalance in the sex Samson • CASSIN’S FINCH 65 ratios, subadult male plumage, and possible hormonal factors influencing fe- male dominance in Cassin’s Finch remains to be resolved. SUMMARY Female Cassin’s Finches were determined socially dominant over older and yearling males in flocks during 2 winters. Few females disappeared either winter in contrast to males. With number of females limiting for breeding effort, the dominance of females in winter is interpreted as a behavioral modification to maximize reproductive effort. This species’ trait and the need to semiannually colonize a new and often unpredictable environ- ment are correlates of an r-strategy. Displays in agonistic encounters are considered homologous to other fringillids. Reasons for the observed disparities in sex ratio or hor- monal factors influencing female dominance are not known. ACKNOWLEDGMENTS I thank A. W. Stokes, M. H. Ralph, and, in particular, F. L. Knopf for their assistance in the field. K. L. Dixon’s assistance throughout the study and his comments as well as those of C. F. Thompson on an earlier draft of the manuscript are appreciated. S. Samson provided valuable help in preparing the manuscript. J. Sidelinger prepared the drawings. I Financial support came from Sigma Xi and the Chapman Fund, American Museum of Natural History, New York. LITERATURE CITED Bailey, A. M. and R. J. Niedracii. 1965. Birds of Colorado, Vol. 2. Denver Mus. of Nat. History, Denver. Brian, A. D. 1949. Dominance in the Great Tit Pants major. Scott. Nat. 61:144-155. Brown, J. L. 1963. Aggressiveness, dominance and social organization in the Steller Jay. Condor 65:460-484. Buckley, P. A. 1973. The changing seasons. Am. Birds 27:578-585. Chapin, J. 1958. The 1957 Christmas counts in Colorado. Colorado Bird Notes 5:29-31. Cody, M. L. 1972. Ecological aspects of reproduction. Pp. 462-512, in Avian biology, Vol. 2. (D. S. Farner and J. R. King, eds.). Academic Press, N. Y. Coutlee, E. L. 1967. Agonistic behavior in the American Goldfinch. Wilson Bull. 79: 89-109. ' Davis, J. 1973. Habitat preferences and comjietition of wintering Juncos and Golden- i crowned Sparrows. Ecology 54:174-180. Dilger, W. C. 1960. Agonistic and social behavior of captive Redpolls. Wilson Bull. 72:115-132. Dixon, K. L. 1963. Some aspects of social organization in the Carolina Chickadees. Proc. 13th Int. Ornithol. Congr. : 240-258. . 1965. Dominance-subordination relationships in Mountain Chickadees. Condor 67:291-299. French, N. R. 1959. Life history of the Black Rosy Finch. Auk 76:159-180. Fretwell, S. 1969. Dominance behavior and winter habitat distribution in juncos (Junco hyemalis) . Bird-Banding 40:1-25. Hinde, R. a. 1955, 1956. A comparative study of the courtship of certain finches (Fringillidae) . Ibis 97:706-745; 98:1-23. : Johnson, R. E. 1965. Reproductive activities of Rosy Finches, with special reference to Montana. Auk 82:190-205. 66 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Kikkawa, J. 1961. Social behavior of the White-eye Zosterops lateralis in winter flocks. Ibis 103:428-442. Murton, R. K., a. J. Isaacson, and N. J. Westwood. 1966. The relationships between Wood-pigeons and their clover food supply and the mechanism of population control. J. Appl. Ecol. 3:55-96. Newton, I. 1964. Bud-eating by Bullfinches in relation to the natural food supply. J. Appl. Ecol. 1:265-279. . 1972. Finches. Taplinger Publ. Co., N. Y. Nicolai, J. 1956. On the biology and ethology of the Bullfinch iPyrrhula pyrrhula L.). Z. Tierpsychol. 13:93-132. Ohh, R. T. 1968. Cassin's Finch Carpodacus cassinii Baird. Pp. 280-289, in Life his- tories of North American cardinals, grosbeaks, buntings, towhees, finches, sparrows, and allies (A. C. Bent and collaborators), U.S. Natl. Mus. Bull. 237. PiANKA, E. R. 1970. On r- and A--selection. Am. Nat. 104:592-597. Pulliam, H. R. and F. Enders. 1971. The feeding ecology of five sympatric finch spe- cies. Ecology 52:. 557-566. Sarine, W. S. 1959. The winter society of the Oregon Junco: intolerance, dominance, and the pecking order. Condor 61:110-135. Samson, F. B. 1974. On determining sex and age in the Cassin's Finch. Western Bird Bander 49:4-7. 1976. Territory, breeding density and fall departure in Cassin’s Finch. Auk 93:477-497. d'lioMPSON, W. L. 1960. Agonistic behavior in the House Finch. Part 2: Factors in aggressiveness and sociality. Condor 62:378-402. Watson, A. and R. Moss. 1970. Dominance, spacing behavior and aggression in rela- tion to poj)ulation limitation in vertebrates. Pp. 167-220, in Animal populations in relation to tlu*ir food resources (A, Watson, ed.l. Blackwell Scientific Publ., Oxford. Zaiiavi, a. 1971. Tin* s<(cial behaviour of the White Wagtail Motacilla alha alba win- tering in Israel. Ibis 113:203-211. DKI’T. OF HIOLOGV, UTAH STATE UMV., LOGAN, UTAH (IL322. ( PItESENT ADDRESS: COOPERATIVE WILDLIFE RESEAltGII UNIT, STEPHENS HALL, UNTV. OF MIS- SOURI, GOLUMRIA 6.5201.) AGGEPIED 20 .JAN. 1976. INTER-BROOD MOVEMENTS OF JUVENII.E SPRUCE GROUSE f)ANIEL M. KePPIE Juvenile tetraonids form broods during most of their first summer of life. Undoubtedly this contact between parent and offspring has survival value for the chicks, particularly during development of thermoregulation and per- haps for acquisition of learned behavior. Presumably, a brood consists of a single family, yet a survey of 26 theses and published reports pertaining to brood size and behavior of tetraonids revealed 10 in which the author sus- pected broods contained chicks from different hens (Lehmann 1941, Wing et al. 1944, Bump et al. 1947:293, Patterson 1952:135, Bendell 1955, Chambers and Sharp 1958, Dalke et al. 1963, Bendell and Elliott 1967, Braun and Rogers 1971, Harju 1974). But the evidence is based only on observations of juveniles of different sizes or estimated ages or, supposedly, an excess number of chicks. Inter-brood movements among tetraonids apparently do not occur on the massive scale reported for aquatic birds ( Beard 1964, Gorman and Milne 1972). There is little known about the freciuency of occurrence and circumstances surrounding inter-brood movements of grouse. Herein I document inter-brood movements of juvenile Spruce Grouse (Canachites canadensis)^ calculate their frequency of occurrence, and briefly (luestion the function of broods remaining as individual units over a long period of time. METHODS Data were gathered from marked l)irds incidental to a population study at Gorge Creek (GC) and Blue Rock Creek (BRC), 27-32 km west of Turner Valley, Alberta from 1970 through 1973. Grouse were located hy repeatedly searching the study areas with trained pointing dogs. All hens with broods known to he on the study areas were captured and marked. Numbers of juveniles and their survival were determined from counts of brood size and records of marked individuals. Young chicks were marked with numbered wing tags (size no. 1, National Band and Tag, Newport, Ky. ) ; leg hands were used after about 40 days of age. Juvenile age was determined hy growth of primaries ( McCourt and Keppie 1975). RESULTS The efficiency of tagging grouse is shown in Table 1. In 3 of the 4 summers, over 50% of the maximum number of juveniles seen were marked hy 14 days of age and about 80% hy 42 days of age. All juveniles that survived until the end of the brood period were marked hy that time. Broods were seen on 428 occasions in 4 years; in only 8 instances (2%) 67 68 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Table Percentages of Juvenile Spruce 28, AND 42 Days of 1 Grouse Marked Age, 1970-73 BY 7, 14, Year and Area 1970 1971 1972 1973 GC BRC GC GC GC Age of bird (66)* (30) (103) (88) (89) ( days ) (53)** (24) (97) (84) (77) 7 6 0 35 43 39 14 9 3 56 58 54 28 32 13 70 68 71 42 52 23 84 81 79 * Maximum number of different juveniles sijihted. ** Number juveniles marked in brood period. were 2 found within 50 m of each other (Table 2), but brood ranges were not mutually exclusive. Nineteen % of the occupied habitat (all years) was included in overlap, i.e., within home ranges of different hens. Some habitat was within the home range of up to 4 families. In one instance (GC, 1971 j, 6 broods were present on a 108 ha plot, and although 37% of the area was used by 2 or more broods only 2% (1/44) of the sightings were of 2 broods together. In all years, overlap of brood ranges probably was greater than recorded because sightings of broods were limited. Among the 8 occasions that 2 broods were together, 33 juveniles were already tagged and 32 of these were seen with the same hen at a later date. The remaining juvenile joined the second hen when the 2 broods came together. In one sighting of 2 broods, both hens simultaneously called to their chicks from adjacent trees yet the 4 marked juveniles were seen later with the “correct” adult. riiere were 12 instances of 11 marked juveniles (7 females, 4 males) moving from a total of 7 broods. Four of the 7 hens were known or assumed to have died. Ihree juveniles moved from 3 other hens that remained alive, hut one of these juveniles was i)reviously orphaned. In all cases in which the hen remained alive (3) only one juvenile moved to a new brood; the only instances (2) of siblings moving together to a new family occurred when the original hen died. All 11 juveniles were at least 11 days old when mixing occurred and all joined broods that occupied overlapping or adjacent home ranges. 4Tn of these juveniles (91%) survived until at least the end of summer. Survival beyond summer was not determined because many juveniles dispersed in autumn. Seven hens were known or assumed to have died while with juveniles; 6 Keppie • INTER-UROOI) MOVEMENTS OE (;R0USE 69 Table 2 Summary of Brood Sightings and Inter-brood Movements of Juvenile Spruce Grouse, 1970-73 Year and Area 1970 GC BBC 1971 GC 1972 GC 1973 GC All Yrs Size of area (km^) 5.2 1.6 6.8 6.8 6.8 - No. broods^ 15 6 20 20 18 79 broods/km^ 2.9 3.8 3.9 3.9 3.5 - No. contacts with broods 57 28 118 121 104 428 No. sightings of 2 broods/contact 1 2 1 2 2 8 % of total 1.8 7.1 0.8 1.7 1.9 1.9 No. juveniles alive after about 2 weeks of age 56 29 69 50 62 266 No. marked juvs. known to change broods 1 1 4 1 4 11 % of total 1.8 3.5 5.8 2.0 6.5 4.1 1 Broods are excluded if aU chicks were lost early. deaths were at least 11 days after hatch (11-22, 17-22, 17-30, 40, 40-43, and 54 days). In this sample, the probable number of chicks alive when the hens died was 27, of which 20 (74%) were still alive at the end of summer (2 j others died from handling) . By contrast, a hen died at 4-9 days post-hatch and her 3 chicks were not seen again. This brood was rather isolated and if the juveniles did not die at the same time as the hen they may have had difficulty locating another family. Although data are limited and not clear-cut, the “need” for orphaned j juveniles to seek out a new brood may have varied with age. In 3 broods the I hen died before 30 days post-hatch; juveniles in 2 of these joined new families within 9 days, and juveniles in the third brood were captured with a new hen 28 days after the death of the hen. A brood count suggested juveniles in this latter brood were present in a new brood 10 days after the ' original hen died. Three broods were orphaned after 40 days of age and II the juveniles were seen later as intact units without a hen. Juveniles of one of these latter broods were without a new hen for at least 26 days, and only one I of the 7 chicks then joined a new family; juveniles in the other 2 broods I remained together for 8 and 11 days until they dispersed. The frequency of brood interchange was calculated from the number of II juveniles remaining after about 2 weeks of age (Table 2) because this probably excludes most of the high losses from natural mortality (Zwickel I 70 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 and Bendell 1967). Four % of the juveniles surviving beyond about 2 weeks of age were known to join other broods. At GC, slightly greater mixing occurred in 1971 and 1973 and resulted from several siblings, rather than one chick per brood, moving to other families. DISCUSSION Although brood ranges of Spruce Grouse were not mutually exclusive, there were few documented exchanges of juveniles between broods. Proportions of juveniles recorded changing broods probably were overestimates for the cohort hatched unless considerable mixing occurred before chicks were marked. I he freciuency of 2 broods coming together likely was greater than recorded, hut when considered on a temporal basis such gatherings probably still constituted a small iiroportion of the brood period. Call notes seemingly function to maintain contact between hen and chicks ( Zwickel 1967, this study), and individuality of call notes might facilitate proper reorganization when broods come together. Although there were few records of 2 broods together, the observation that juveniles reunited with their respective parent, coupled with individuality of sound, generates the <}uestion of whether survival of a juvenile is enhanced by staying with its respective mother. At least short term survival was good for juvenile Spruce Grouse that changed broods; survival also was good for juveniles that were orphaned. Survival of orphaned juveniles might he age related, recpiiring the full development of thermo- regulation, and for young chicks (<2 weeks old) the proximity of another brood might he critical to survival. Several authors have speculated on causes of inter-hrood movements of grouse, such as the death of a hen (Bump et al. 1947:293), a loose feeding formation and lack of an efficient rallying call (Lehmann 1941), and a concentration of broods (Wing et al. 1944, Bendell 1955). Bendell (1955) further suggested weather as the ultimate cause, by its influence on plant growth and distribution of preferred feeding sites. Death of hen Spruce (house seemed to he a cause for juveniles switching broods, hut perhaps only when juveniles were young. There was no evidence of a weak cohesion among family members, concentration of broods, nor preferred or localized feeding sites. I do now know whether densities of broods in this study were high for Spruce Grouse; they were generally as high or higher than densities recorded by others (Ellison 1974, McCourt 1969). It is open to (juestion whether higher densities might reduce the effectiveness of calling for maintaining brood organization, resulting in greater exchange of juveniles. There was no effect of movements between broods in summer on mean brood size. Many juveniles temporarily join other families while dispersing in autumn Keppie • INTEK-HKOOl) MOVEMENTS OE (;H()IJSE 71 j (Keppie, unpubl. data) and biases on counts of brood size would be greatest i at that time. Whether a specific mechanism or simple chance accounts for separation of I grouse broods ( Bendell and Elliott 1967, Zwickel 1973, Godfrey 1975; this study) is unknown. Dispersion of broods may result from other factors oper- . ating earlier during courtship and nesting. Perhaps brood dispersion enhances survival of the chicks, but present data on survival until autumn for juveniles I switching broods do not support this idea. If juveniles that move to a different brood survive, and if juveniles of a certain age can live without a hen, we should focus attention on the purpose of the dispersion pattern and why hen and chicks maintain contact longer than seems necessary. SUMMARY Inter-brood movements of juvenile Spruee (irouse were reeorded in Alberta from 1970 j through 1973. Although brood ranges were not mutually exclusive, broods generally I maintained their original constituency. Only 4% of the marked juveniles changed j broods; they moved from 7 broods and in 4 cases the hen had died. All juveniles that ' moved were at least 11 days old and all joined a family in the immediate vicinity. ; Juveniles that changed hroods or which were orphaned survived well until autumn. Although data are limited on the fate of juveniles that mix or which are orphaned, the I question arises as to why hroods exist as individual units for perhaps longer than necessary to ensure survival of the chicks. i ACKNOWLEDGMENTS Financial support was provided by the National Kesearch (Council of Canada and the National Wildlife Federation; I). A. Boag, Univ. of Alberta, provided funds for logistical support. I acknowledge assistance in the field by students in the Department of Zoology, Univ. of Alberta: W. E. Etherington, A. Garbutt, M. Henderson, R. Salter, K. Smith, and || D. Thompson. I am thankful for advice on the manuscript from F. C. Zwickel and J. I, Kristensen, Univ. of Alberta, C. E. Braun, Colorado Division of Wildlife, and T. Dilworth, 1 Univ. of New Brunswick. The University of New Brunswick has supported the costs of I publication. I LITERATURE CITED Beaki), E. B. 1964. Duck brood behavior at the Seney National Wildlife Refuge. J. j. Wildl. Manage. 28:492-521. Bendell, J. F. 1955. Age, breeding behavior and migration of Sooty Grouse, Dendrag- apus obscurus fuliginosus (Ridgeway). Trans. N. Am. Wildl. Conf. 20:367-381. AND P. W. Elliott. 1967. Behaviour and the regulation of numbers in Blue I Grouse. Can. Wildl. Serv. Rep. Ser. 4. i Braun, C. E. and G. E. Rogers. 1971. The White-tailed Ptarmigan in Colorado, j Colorado Div. Game, Fish and Parks, Tech. Publ. 27. 1 Bump, G., R. W. Darrow, F. C. Edminster, and W. F. Crissey. 1947. The Ruffed ■ Grouse — life history, propagation, management. New York State Conserv. Dept. 72 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 CiiAMBEits, R. E. AND W. M. Sharp. 1958. Movement and dispersal within a population of Ruffed Grouse. J. Wildl. Manage. 22:231-239. Dalke, P. D., D. B. Pyrah, U. C. Stanton, J. E. Crawford, and E. F. Schlatterer. 1963. Ecology, productivity, and management of Sage Grouse in Idaho. J. Wildl. Manage. 27:811-841. Ellison, L. N. 1974. Population characteristics of Alaska Spruce Grouse. J. Wildl. Manage. 38:383-395. Godfrey, G. A. 1975. Home range characteristics of Ruffed Grouse broods in Minnesota. J. Wildl. Manage. 39:287-298. Gorman, M. L. and H. Milne. 1972. Creche liehavior in the Common Eider, Somateria m. mollissima L. Ornis Scand. 3:21-25. Harju, H. j. 1974. An analysis of some aspects of the ecology of Dusky Grouse. Ph.D. Thesis, Univ. Wyoming, Laramie. Lehmann, V. W. 1941. Attwater’s Prairie Chicken. Its life history and management. North Am. Fauna 57. McCourt, K. H. 1969. Dispersion and dispersal of female and juvenile Franklin’s Grouse in southwestern Alberta. M.S. Thesis, Univ. Alberta, Edmonton. AND 1). M. Keppie. 1975. Age determination of juvenile Spruce Grouse. J. Wildl. Manage. 39:790-794. I^ATTERSON, R. L. 1952. The Sage Grouse in Wyoming. Sage Books, Inc., Denver. Wing, L., J. Beer, and W. Tidyman. 1944. Brood habits and growth of Blue Grouse. Auk 61:426-440. ZwiCKEL, F. C. 1967. Early behavior in young Blue Grouse. Murrelet 48:2-7. . 1973. Dispersion of female Blue Grouse during the brood season. Condor 75: 114-119. AND .1. F. Bendell. 1967. Early mortality and the regulation of numbers in Blue Grouse. Can. J. Zool. 45:817-851. DEPT. OF ZOOLOGY, UNIV. OF ALDEKTA, EDMONTON (PRESENT ADDRESS, DEPTS. OF FOREST RESOURCES AND BIOLOGY, UNIV. OF NEW BRUNSWICK, FREDERIC- TON, NEW BRUNSWICK, CANADA EBB 5A3). ACCEPTED 7 APR. 1976. BREEDING BIOLOGY OE YEAR-OLD AND OLDER FEMALE RED-WINGED AND YELLOW-HEADED BLACKBIRDS Richard J). Crawford Age of male Recl-wingecl (Agelaius phoeniceus) and Yellow-headed black- birds {Xanthocephalus xanthocephalus) is known to influence their breeding behavior (Orians 1961, Willson 1966), but little effort has been devoted to the comparative breeding biology between age classes of females. Both species are polygynous (Verner and Willson 1969) ; how age might affect the females within this system, however, is unknown. The objective of this study was to compare data on selected parameters of breeding between yearling and older adult females of both species. Field work was conducted from 1972-1974 on Dewey’s Pasture and Dan Green Slough, 2 glacial marshes in northwestern Iowa that have been described by Bennett (1938) . METHODS Red-wings were aged by using the methods presented hy Nero (1954, 1961) and Meanley and Bond (1970) ; yearling females have a pink or salmon epaulet and light pink chin and face, while older females show a more crimson epaulet and dark pink chin and face. To verify this aging technique, 1 initiated a handing program in 1972. Eighteen females were recaptured in years after their handing, 7 of which were yearlings. Both yearlings and older adults showed patterns similar to those described. No similar aging technique exists for Yellow-heads, hut Bent (1958:112) described the first-year female as “much like the adults, hut colors are more veiled.” By examining 21 returning marked females (11 yearlings and 10 older adults), I was able to establish that yearlings have lighter breasts, throats, and facial regions than older adults. A major advantage of the aging techniques for both species is that an observer can readily distinguish ages in the field. Observations on the nesting activities of both species began in late May of each year. Most females were captured, handed, and classed as either year-old or older. The few females not trapped were aged in the field hy the methods described. After incubation had begun, nests were checked only every 2 or 3 days to minimize disturbance; the nestlings used for growth rate studies, however, were checked daily. The date the first egg was laid was used as an indicator of nesting chronology. If not known precisely, this date was estimated, considering 12 days to he an average incubation period for both species (Nero 1961, Willson 1966). Only first nests were used for analysis. Statistical comparisons were made with Student’s t-test (Steel and Torrie 1960). RESULTS Nesting chronology, clutch size, and egg size. — Yearling Red-wing and Yellow-head females began nesting an average of 15 and 16 days later, respec- 73 74 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Dates of Table 1 Nest Initiation (1972-1974) N Mean ±; 2 S.E. (days) Range KW yearling’ 67 8 June ± 5.2 25 May-18 June KW older adult 41 25 May ± 5.7 19 May- 5 June YII yearling 71 7 June ± 4.9 22 May-15 June Yll older adult 49 23 May ± 5.4 18 May-29 May 1 RW = Red-winged Rlackhird, YH = Yellow-headed Blackbird. lively, than older adults (Table 1). ITie yearlings also showed a greater range in nest initiation dates. The differences between nest initiation dates of yearlings and older adults were highly significant for both species ( P < 0.01). Means and frecjuencies of clutch sizes observed are given in Table 2. Yearling females of both species had significantly smaller clutches than older females (P<0.01 I. ddie mean values are similar to those of other studies. 1 he average clutch size was 3.1 for 501 Yellow-head nests in Iowa ( Ammann 103o); 3.7 for I IB nests in I tah ( Fautin 1041); and 3.6 for 371 nests in Washington (Willson 1066). For the Ked-wing, average clutch sizes reported were 3.5 for 026 nests in New York (Case and Hewitt 1063) ; 3.4 for 243 nests in Oklahoma ( Goddard and Board 1067); and 4.2 for 13 nests in Missouri ( Craw ford 1070 ) . Yearling females of both species laid significantly shorter eggs than older females (P<0.01) (Table 3). Fgg width did not differ significantly for either species (P > 0.05). The mean values for both length and width are similar to those found in other studies (Bendire 1B05. Reed 1065). Table 2 Ui.r'Kai Size and Fl.EDCINt; Sl'CCESS ( 1972-1974) Clutch Size Mean Fledged Young Per Nest 2 3 4 .5 Mean B\\' yt'arling 4’ 37 34 3.40 0.73 (71)- HW older adult 1 40 6 4.11 1.63 (40) HW all ages 4 .38 74 6 3.67 1.05(111) ’i H yearling 8 49 31 3.30 0.87 (83) ^ 11 older adult 51 6 4.11 1.81 (52) H all ages 8 49 82 6 3.62 1.30(135) 1 X uni her of clutches. ~ Sample size in parentheses. Crawford • HLACKUIRI) HHEE1)IN(; mOLOGY 75 Table 3 Egg Sizes (1972-1974) Mean ( mm ) N Length Width RW yearling 165 22.1 17.0 RW older adult 145 26.8 18.2 YH yearling 181 23.2 17.3 YH older adult 157 27.1 18.9 Nestling; growth and fledging success. — At hatching;, nestling;s of yearling I females averaged only slightly smaller than those of older adults; weights at 10 3 days of age, however, were significantly lower for nestlings reared by ^ yearling females (P <0.()1) (Table 4). Male nestlings of both species have I been reported to grow faster and attain greater weights at fledging than females • of the same age ( Ammann 1938, Williams 1940, Holcomb and Twiest 1971). ! I assumed in this study that the sex ratio was constant throughout the nestling i period and that differences in sex-specific weights would have no net effect. Fledging success is given in Table 2. Yearling females of both species I fledged significantly fewer young than did older females (P<0.01). The fledging successes reported in this study are similar to those reported else- where (Wood 1938, Willson 1966, Goddard and Board 1967). Pairing status of age classes. — Data on pairing status and its relationship to age were collected on 30 Red-wing and 20 Yellow-head territories during 1'able 4 Growth IN Weight (g) of Nestlings (1972-1974) Age (days) Red-wing Yellow-head yearling older adult yearling older adidt 1 3.6 (65)' 4.0 (44) 3.9 (66) 4.1 (49) 2 5.8 (48) 6.1 (37) 6.8 (50) 7.0 (40) 3 8.8 (47) 9.5 (30) 10.4 (49) 10.8 (36) 4 12.7 (39) 13.7 (27) 16.5 (43) 16.8 (35) 5 16.6 (36) 18.9 (23) 22.0 (40) 22.5 (29) 6 21.1 (28) 24.2 (22) 28.7 (33) 29.9 (24) 7 25.9 (19) 27.3 (20) 33.1 (24) 35.8 (19) 8 28.3 (15) 31.6 (16) 37.3 (16) 40.7 (17) 9 30.4 (11) 34.3 (13) 40.5 (13) 45.1 (15) 10 32.1 (10) 37.4 (12) 43.7 (10) 49.3 (11) 1 Sample size in parentheses. 76 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Table 5 Mean Clutch Size anu Fledging Success in Status (1973-1974) Relation TO Pairing Primary Female Secondary Female N CSi FSi N CS FS KW yearling 6 4.0 1.2 29 2.9 0.6 KW older adult 24 4.2 1.8 11 4.1 1.3 YH yearling 7 4.0 2.0 17 2.8 0.8 YH older adult 13 4.1 2.0 4 4.0 1.2 1 (]S = clutch size, FS - fledging success. 1973-74. Herein, I use the terminology of Martin (1974) ; the first female to nest in a male’s territory is referred to as the primary female, and all those nesting suhsetiuently are termed secondary females. Pahle 5 i)resents data for females where pairing status was determined. For the Red- wing, most primary females were older adults. Some older adults were secondary females, hut in all such instances, the primary female was also an older adult. Only 6 yearlings were primary females, 4 of which mated monogamously, and the other 2 mated i)olygynously where the secondary females were also yearlings. In no instance was a yearling female the primary mate and an older female secondary within the same territory. A similar situation existed for the Yellowdiead (Table 5). All yearling primary females mated monogamously, and all older adult secondary females were secondary only to other older adults. Further evidence to suggest that age is an important factor influencing pairing status was gained hy examining data from returning females of known age rkahle 6). In most instances, females were secondary as yearlings and primary when 2 years old; 2 females, however, were secondary both as yearlings and as 2-year-old birds, and 1 female was primary when a yearling as well as when she was 2 years old. Po determine if differences in breeding biology existed in relation to pairing status, data were compared between yearling and older females of both species ( Pahle 5). Older adult primary females did not show significantly larger clutches than older adult secondary females (P > 0.05), hut primary yearling females of both species laid significantly larger clutches than did secondary yearling females (P<0.01). Trends similar to these also were found in the clutch sizes of known-age females (Tal)le 6). Yearling females fledged signifi- cantly ( P < 0.01 I fewer young than did older adults for all pairing situations, except for primary Yellow-heads (Table 5). Yearling and older adult primary females of both species fledged significantly more young than did secondary Crawford • BLACKBIRD BREEDING BIOLOGY 77 Pairing Status AND Clutch Size Table 6 OF Returning Known-age Females (1973-1974)^ Species Female No. Yearling Age 2 years old Red-wing DP74 IP (3) P (4) DP97 IP (3) P (4) DP 189 IP (2) IP (3) DP191 IP (4) P (4) Yellow-head I)P67 IP (3) P (5) DP96 IP (3) P (4) DP157 IP (3) IP (3) DP181 P (4) P (5) DP192 IP (3) P (4) DP197 IP (3) P (4) 1 1° = primary female, 11° = secondary female, number in parentheses is clutch size. 1 females (P<0.01). Egg size within age groups was not influenced by I pairing status. Sample sizes were inadetjuate to analyze differences in nesting J chronology or fledgling weight between primary and secondary females. DISCUSSION i ' Lighter-colored females have been noted several times in breeding popula- l[ tions of both species (Nero 1954, Bent 1958, Strosnider I960), but little I comment has been made regarding breeding success of these females. Data Ij presented in this paper suggest that these females are yearlings and that they 'I contribute less to total population production than older females. Why yearling females breed later than older adults is unclear, but appar- »| ently Red-wing yearling females migrate later in the spring than older females ! (Allen 1914, Nero 1956a j. Also, females of both species actively defend their sub-territories against trespass by other females (Nero 1956b, Willson 1966). ' Possibly one or both of these factors may act to delay breeding by yearling females. Goddard and Board (1967) noted that early Red-wing nests had larger clutches, were more successful, and fledged more young than later nests, but no indications were given as to causative factors involved. Holm (1973) ! stated that late arriving females and some early arriving females may he forced to occupy territories in poorer habitats. No data were collected on territory quality in this study, but it is possible that late arriving females ' (apparently yearlings) are forced to occupy sub-optimum territories and, thus, produce smaller clutches and fewer young. 78 THE WILSON BULLETIN • Vol. S9, No. 1, March 1977 In a study of the adaptations for polygynous breeding in Bobolinks {Dolich- onyx oryzivorus ) , Martin ( 1974 ) found that yearling females laid smaller clutches than older females, and primary females received more assistance in nestling care by the male, laid larger clutches, and fledged more young than did secondary females. He hypothesized that primary females laid larger clutches than secondary females mainly because males feed nestlings of l)rimary females more often than they do of secondary females. Yellow-head males are known to feed young in primary nests more often than in secondary nests (Willson 1966). Data from other studies suggest that Red-wing males do not feed their young (Nero 1956a, Holm 1973 j, hut some exceptions are noted (Bent 1958, Case and Hewitt 1963). Why this difference exists is unknown. Why older adult females did not show a significant difference in clutch size relative to pairing status is not clear. Possibly older adult females, being more experienced in nestling care, would he able to raise more young without help from a male than would yearling secondary females. Data presented here show that age has a pronounced effect on the breeding biology of Red-wings and Yellow-heads. Although a few studies of other species ( e.g. Leinonen 1973, Koskimies 1957) indicate that age has little or no effect on some parameters of breeding, I believe that most species will show age-related differences worthy of study. Other studies (e.g. Laskey 1943, Snow 1958, Lack 1966, (irawford 1971, 1975a, 1975b ) have discussed other ways in uhicL age may influence reproduction in birds. Further studies should he conducted so that a more complete understanding of reproduction in relation to age may he attained. SUMMAin riic relationships lK‘tvv(*(*n age and breeding biology of female Ked-winged and ^ ellow- In'acb'd blackbirds were sludi«‘d in north w«*stern Iowa during 1972-1974. \ earling females of both speei«*s bc'gan m'sting later, laid shorter eggs, and fledged fewer and slightly smalb*r young than did (dder females. I’rimary (first-nesting) females were mostly older adults while yearlings were typically of secondary status. Y earling primary females laid larger elutelu's than did yearling secondary females, but both yearling and older adult l)rimary femab's fledged more young than did secondary females of the same age. Pos- sibb‘ factors afft'cting didayed bret'ding and subs(‘(|uent reduced production of yearlings are diseusst'd. A ( ; K N O ^V L E I )G M E \ TS 1 wish to thank Milton W’. Weller for early observations relating to the study and for reading the manuscri]>t. Louis B. Best, William L. Hobman, and Dennis G. Jorde com- mented on various drafts of the manuscript. This is journal j>aper .1-8265 of the Iowa Agriculture ami Home Economies Experiment Station, Ames, b)wa. Project No. 1969. Crawford • BLACKBIRD BREEDING BIOLOGY 79 LITERATURE CITED Allen, A. A. 1914. The Red-winged Blackbird: A study in the ecology of a cattail marsh. Proc. Linn. Soc. N. Y.:43-128. Ammann, G. a. 1938. The life history and distribution of tbe Yellow-beaded Blackbird. Ph.D. thesis, Univ. of Michigan, Ann Arbor. Bendire, C. E. 1895. Life histories of North American birds. U.S. Natl. Mus. Bull. 3. Bennett, L. J. 1938. The Blue-winged Teal. Iowa State Univ. Press, Ames. Bent, A. C. 1958. Life histories of North American blackbirds, orioles, tanagers, and allies. U.S. Natl. Mus. Bull. 211. Case, N. A. and 0. H. Hewitt. 1963. Nesting and productivity of the Red-winged Blackbird in relation to habitat. Living Bird 2:7-20. Crawford, R. D. 1970. Mourning Dove and blackbird pntduction in a Missouri pine planting. Iowa Bird Life 40:65-67. . 1974. Age-related spatial distribution of the Adelie Penguin. Notornis 21: 334-336. . 1975a. Breeding biology of American Coots in relation to age. Ph.D. thesis, Iowa State Univ., Ames. . 1975b. Post-incubation activity of Adelie Penguins. Notornis 22:54-57. Fautin, R. W. 1941. Incubation studies of the Yellow-headed Blackbird. Wilson Bull. 53:107-122. Goddard, S. V. and V. V. Board. 1967. Reproductive success of Red-winged Blackbirds in north central Oklahoma. Wilson Bull. 79:283 289. Holcomb, L. 1). and G. Twiest. 1971. Growth and calculation of age for Red-winged Blackbird nestlings. Bird-Banding 42:1-17. Holm, C. H. 1973. Breeding sex ratios, territoriality, and reproductive success in the Red-winged Blackbird i Agelaiiis phoeniceus) . Ecology .54:356-365. Koskimies, J. 1957. Polymorphic variability in clutch size and laying date of the Velvet Scoter. Ornis Fenn. 34:118-128. Lack, 1). 1966. Population studies of birds. Clarendon Press, Oxford. Laskey, A. R. 1943. The nesting of Bluebirds banded as nestlings. Bird-Banding 14:39-43. Leinonen, M. 1973. Comparisons between the breeding biology of year-old and older females of the White Wagtail Motacilla alba in central Finland. Ornis Fenn. 50:126-133. Martin, S. G. 1974. Adaptations for polygynous breeding in the Bobolink, Dolichonyx oryzivorus. Am. Zool. 14:109 119. Meanley, B. and G. M. Bond. 1970. Molts and plumages of the Red-winged Blackbird with particular reference to fall migration. Bird-Banding 41:22-27. Nero, R. W. 1954. Plumage aberrations of the Redwing (Agelaius phoeniceus) . Auk 71:137-155. . 1956a. A behavior study of the Redwinged Blackbird. I. Mating and nesting activities. Wilson Bull. 68:5-37. . 19561). A behavior study of the Redwinged Blackbird. II. Territoriality. Wilson Bull. 68:129-150. . 1961. Red-winged Blackbird. In Bird banding manual. U.S. Dept. Interior, Laurel, Maryland. Orians, G. H. 1961. The ecology of blackbird i Agelaius) social systems. Ecol. Monogr. 31:285-312. 80 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Rked, C. a. 1965. North American birds eggs. Dover Publ., Inc., N. Y. Sn(jw, D. W. 1958. The breeding of the Blackbird Tardus merula at Oxford. Ibis 100:1-30. Steel, R. G. D. and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Co., N. Y. Str(jsnider, R. 1960. Polygyny and other notes on the Red-winged Blackbird. Wilson Bull. 72:200. Verner, J. and M. F. Willson. 1969. Mating systems, sexual dimorphism, and the role of male North American passerine birds in the nesting cycle. Ornithol. Monogr. No. 9. Williams, .1. F. 1940. The sex ratio in nestling Eastern Redwings. Wilson Bull. 52:267-277. Willson, M. F. 1966. Breeding ecology of the Yellow-headed Blackbird. Ecol. Monogr. 36:51-77. Wood, H. B. 1938. Nesting of Red-winged Blackbirds. Wilson Bull. 50:143-144. DEPT. OF ANIMAL ECOLOGY, IOWA STATE UNIV., AMES 50011. ( PRESENT ADDRESS: DEPT. OF BIOLOGY, UNIV. OF NOItTII DAKOTA, GRAND FORKS 58201). AC- CEPTED 22 .JAN. 1976. FOOD HAIUTS OF OLDSQUAWS WINTERING ON LAKE MICHIGAN Steven R. Peterson and Rodekt S. Ellarson Several trophic levels in Lake Michigan have been shown to be contaminated with organochlorines (Hickey et al. 1966), and contaminants that are not biodegradable are passed through the food chain from one organism to the I next. Knowledge of the food habits of bird species wintering on the lake would I be useful in measuring the movement of these contaminants from the aquatic ! environment and should help us interpret the distribution and activity of the birds on the lake. This report summarizes data on the food habits of Old- i squaws (Clangula hyemalis) collected during 2 periods: 1951-1954 and 1969- f 1972. Cottam (1939), Lagler and Wienert (1948), and Zimmerman (1953) i have previously reported on the gizzard contents of a few 01ds(iuaws collected 1, from commercial fishermen on Lake Michigan. METHODS We collected Oldsquaws from commercial fishermen who found them caught in gill nets i between November and May. These nets are suspended along the bottom in water 18-46 m •> deep, with the leads holding the lower part of the net on the substrate and the floats t keeping the mesh upright off the bottom. Presumably, since the nets are only about 1.5 m |i high, any birds captured would he a(‘tively feeding at or near the bottom. Even though * some birds may have remained in the ruTs for up to 1 week before removal, the cold ' water prevented extensive decomposition of the esoi)hageal material. After the specimens ' were removed from the nets, they were frozen until dissection and examination could he •j completed. I In the 1951-1954 study, 10 birds were taken from each catch of Oldsquaws and only those which had significant {{uantities of food were saved. Samples were collected from I southeastern Lake Michigan (South Haven, Saugatuck, Holland, Muskegon), southwestern |j Lake Michigan (Kenosha, Racine), and northern Lake Michigan (Port Washington, Two Rivers, Gills Rock, Washington Island). From this part of the study we analyzed 192 specimens: 41 were of gizzards only and 151 were of gizzards and gullets (esophagus plus proventriculus) . Food items were sei)arated and volume to the nearest 0.1 ml and frequency of occurrence of items were recorded. In the 1969-1972 study, we only recorded material from the esophagus because Dillon (1957), Bartonek and Hickey (1969), and Swanson and Bartonek (1970) demonstrated differences in the contents of the eso])hagus, proventriculus, and gizzard. Ellarson’s (1956) analysis of the 1951-1954 material indicated a rapid decomposition of the material once it reached the proventriculus. In the 1969-1972 study period, specimens were obtained only from Gills Rock and Washington Island (northern Lake Michigan). We examined as many specimens as possible (956) and recorded those specimens with ; nothing present in the gullet as well as those having fed. Only the total volume of food present in the esophagus and the frequency of occurrence of food items were recorded. The lipid content of specimens was determined from a 25 g sample of the homogenized 81 82 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 carcass (head, wings, feet, feathers, and gastrointestinal tract removed). After a sample was dried in a 40°C oven for 72-96 hours, it was ground with 100 g NajS04 and extracted 8 hours on a soxhlet extractor using 70 ml ethyl ether and 170 ml petroleum ether. An aliquot of the extract was then reduced to dryness, weighed, and the amount of ether- soluble lipid determined. RESULTS Ingested Material Grit content oj gizzards. — The volume of grit in the 1951-1954 sample was constant among gullets. The ratio of grit to total gizzard and gullet content, for all samples, was 24.7%, and the ratio of sand (<2 mm) to gravel (>2 mm j was consistently about 9:1. Besides sand and gravel, grit was composed of coal cinders and ash from steamships as well as limonitic oolites. In the northern part of the lake, % of the total grit content was oolite material, d he constant ratios of food to grit and sand to gravel suggest some internal mechanism regulating grit retention in the gizzard. Excessive (luantities of sand are i)ossihly ingested with the food organisms and voided into the intestinal tract Because the fecal material is gritty and fluoroscopic examination (n = 2106) indicated the lower intestinal tract was laden with sand. Sand often occurred in the esojdiagus when no animal or vegetable matter was present. In the 1969-1972 study, sand occurred in 83% of all esophagi examined and 90% of all esophagi with food organisms present, (iravel occurred in 28 and 30% of these samples respectively. Crustaceans. — In the 1951-19.54 sample, animal matter constituted 99% of the footl volume in 151 gizzards and was present in all 192 gullets and gizzards examined (Table 1). Amphipods { Pontojwreia affinis) made up 82% of that volume. In the 1951—19.54 sample, amphipods rated between 85 and 100% frecjuency of occurrence, depending on the area where taken, while in the 1969-1972 sample, they were found in 88% of all gullets examined and 95% of all gullets with some material present (Table 2l. One gizzard con- tained 2 chelae of a small crayfish in the 1951—19.54 sample. Mollusks. — Gastropods (snails) and i)elecypods (clams) constituted a rela- tively small proportion of the volume of gizzard contents (3.9% in 1951- 1954), hut their freciuency of occurrence was high (51 to 79% in 1951-1954), second only to amphii)ods. Clams ( Sphaeriidae) were present in 51 to 76% of the 1951-1954 esophagus and gizzard samples and in 61 to 65% of the 1969-1972 esophagus samples, while snails {Cyraulus and Amnicola) were present in 9 to 20% and 11 to 12% of the respective samples. The clams were generally very small; 25 average-size shells of Pisidiuin had a volume of 0.1 ml. Insects. — Adult insects and their larvae accounted for 0.4% of the total food Peterson and FA! arson • FOOD HABITS OF OLDSQUAWS B3 Tahle 1 Percent Frequency of Occurrence and Percent Volume of Ingested Material IN Oldsquaws Collected on Lake Michigan, 1951- 1954 Item Northern Lake Michigan n = 35 (15)* Southwestern Lake Michigan n=34 (34) Southeastern Lake Michigan n = 123 ( 102) Crustaceans ( Amphipoda" ) 97 (52) 100 (96) 85 (82) Mollusks 51 56 79 Gastropoda 11 9 20 Pelecypoda 51 (1) 56 (3) 76 (5) Fish 46 (4.3) 24 (trace'') 33 (12) Skeletal remains 11 (trace) 3 (trace) 24 (7) Eggs 40 (43) 24 (trace) trace (trace) Insects 51 15 28 Coleoptera 9 trace trace Trichoptera 31 (4) 15 (trace) trace (trace) Diptera 6 trace 25 Unidentified 9 t race trace Vegetable matter 74 59 85 * Numbers in parentheses refer to % volume. “ Pontoporeia affinis. » < 0.5 %. volume in the earlier sample, but in both this and the later sample, the frequency of occurrence was high (15 to 51% in the 1951-1954 sample). Diptera and Trichoptera larvae occurred most often hut they were never abundant in any 1 specimen. Diptera larvae occurred in 6 to 25% of the earlier sample and in 15 to 17% of the later sample, while Trichoptera were recorded in 15 to 31% and 3 to 4% of these samples, respectively. Diptera consisted almost exclusively of midge larvae ( Tendepedidae ) , hut a few adult forms were found from this family as well as from the order Coleoptera. Fish. — Fish and fish eggs were the second most important item in the 1951- 1954 sample, hut relatively infreiiuent in the 1969-1972 sample. In the earlier sample, fish remains (primarily Cottidae and Percidaej and eggs constituted an average of 13% of the food volume and the frequency ranged from 24 to 46%. In the later sample the frequency of occurrence was 3 to 4%. The variation in fish and eggs found in Oldsquaws collected during the 2 periods appears to be related to the feeding habits of this species on Lake Michigan. In some groups of Oldsiiuaws, fish remains or eggs constituted the hulk of the food present, indicating that individual flocks fed on whatever food was readily available in the area. In one small sample of birds from northern Lake Michigan in the 1969-1972 period, practically all the gullets contained large chunks of decayed alew ife ( Alosa ])seudoharengus ) , hut it was 84 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Table 2 Percent Frequency of Occurrence OF Oldsquaws Collected OF Ingested Material in ON Lake Michigan, 1969- THE Esophagi 1973 Item % of all esophagi examined n = 956 % of all esophagi with material present n = 884 Animal matter Crustaceans (Amphipoda") 88 95 Mollusks Gastropoda II 12 Pelecypoda 61 65 Fish Skeletal remains 4 4 Eggs 3 4 Insects Trichoptera 3 4 Diptera 15 17 Mysidacea 5 5 Oligochaetes 4 4 Isopods 7 8 Unidentified 3 4 Vegetable matter (detritus) 31 34 Mineral matter Sand (<2 mm) 83 90 Oolites 75 81 (Iravel (>2 mm) 28 30 “ Pontoporeia affinis. found in few other samples from the area. In the 1951-1954 sample, Ellarson (1956) recorded a catch of Birds from Saugatuck that had Been feeding on perch [Perea jlavescens) about 5 cm long, while other catches of birds from the area at that time were feeding on amijhipods. In northern Lake Michigan, Lllarson found 43% of the volume in 15 gizzards was composed of fish eggs (Table 2), and eggs were present in 40% of 35 birds examined from that area. One of the birds collected in this sample had ingested 35 ml of fish ova, or approximately 2600 eggs. In the 1969-1972 sample from the same area, we found 4% of 884 birds to have eaten fish eggs. No birds in the later sample contained large volumes of ingested ova. We suspect, in view of the selectivity on fish, that our earlier sample contained a few birds feeding exclusively on eggs. Since we did not find many Oldsquaws with ingested ova in the large sample from the same area in 1969-1972, the data suggest Oldsciuaws will feed on fish ova when they find them, but that this occurs in isolated localities. Peterson and Ellarson • FOOD HABITS OF OLDSQIJAWS 85 • Food I Lipids Fig. 1. Seasonal relationship between food ingested and carcass fat in Oldsquaws collected on Lake Michigan, 1969-1972. Curves are hand-fitted. Vegetable matter. — Pieces of vegetation occurred frequently in our samples (59 to 85% in 1951-1954, and 31 to 34% in 1969-1972), but the volume was generally less than 1%. We doubt that this material is taken as food because most of the vegetation consisted of decomposed fragments characteristic of bottom debris in deep water. Seasonal Changes in Ingested Volume Data we have on changes in body weights of Oldsquaws suggest a seasonal rhythm in the lipid levels of this species. Changes in these lipid levels could be due to the availability of food on Lake Michigan : as the ice extends farther out from shore in the winter, Oldsquaws are forced to dive in deeper water for their food, and when either the food is deeper than their diving capability or the amount ingested per dive is less than the energy expended, the birds have to draw upon stored energy reserves. In this section we examine this relation- ship between food and stored energy in more detail. Fig. 1 illustrates monthly changes in the volume of food present in Old- squaws and the relationship to the percent lipids in carcasses. Oldsquaws are readily obtained throughout the winter except in February, when the ice is usually so thick on the northern end of the lake that fishermen cannot get out to tend their nets. The food volume data are not normally distributed about 86 THE WILSON BULLETIN • Vul. 89, No. 1, March 1977 the mean: 42% of 956 esophagi had less than 1 ml present and the range was from 0 to 25 ml. We used the nonparametric Mann-Whitney U-test to check for significant differences among months. Since the sample sizes used to compute U were greater than 20, U was converted to the z distribution and the probabilities determined (Siegel 1956:116). Our lipid data indicate a peak in deposition occurring in January, followed by a decline through April (p<.01), while the volume of food ingested declined from December through February (p<.07). In late February, the ice edge begins to retreat and larger average volumes of food ingested were recorded between February and April (p < .01) . At the same time, lipid levels continue to decline through April, and when lipid deposits are lowest, food ingestion is at a maximum. Between April and May, there is an increase in lipid deposition (p < .01), especially in adults, while there is a drop in food intake (p < .01 ) . During the second week of June 1971, we collected a series of Oldsfiuaws on an arctic breeding ground in northwest Hudson Bay. These breeding pairs had not been on the tundra more than a few days. Lipid analyses indicated these birds had about 17% lipid content, as opposed to an average of 22% lipid content in a similar sample of birds on Lake Michigan prior to migration. However, the Lake Michigan and arctic samples may not be from the same populations. There was no food present in the esophageal and gastrointestinal tracts in the arctic sample, and the livers as well as the intestinal tracts were much smaller than the average size on Lake Michigan prior to migration. We concluded that the birds in the arctic sample had fed little during migration, and the data in Fig. 1 suggest Lake Michigan birds had started to reduce food intake prior to spring migration. DISCUSSION Two important conclusions can be drawn from data on the composition of food in the diet of the Oldsciuaw. First, animal matter constitutes the bulk of the ingested material. Our observations confirm what Dement’ev et al. (1952) recorded in the Soviet Lhiion, and Polunin and Eklund (1953) found in an 01ds(iuaw stomach collected in Ungava. Cottam’s (1939) analysis of 190 stomachs from various parts of North America and Madsen’s (1954) data from the Danish coast indicate crustaceans and mollusks are the favored food items. On Lake Michigan, our data as well as that collected by Lagler and Wienert (1948) and Zimmerman (1953) indicate this animal matter is primarily amphipods and to a lesser extent clams and snails. The second important conclusion concerning the composition of the diet is that the Oldsquaw is an opportunist that will take whatever animal matter Peterson and KHarson • FOOD llAlilTS OF OFDSOUAWS {VJ is most abundant or most available in its feedinp; area. Mackay commented on this in 1892: ‘"dhey [01ds(iuaws] do not seem to be particular in regard to their food, eating various molluscs, fish and sandfleas.” Cottam (1989:76) noted this tendency from his collections when he stated: “The | Oldsciuavv | , in general, do not seem to show any species preference for either small molluscs or crustaceans, if ecjually easily obtained;” and Madsen (1954:204) pointed out: “On the whole . . . this Diving Duck is very adaptable in its food selection. . . Eggleton (1937) and Alley (1968:18) found the benthic poi)ulation of Lake Michigan dominated by 4 groups: Pontoporeia, Tubificidae, Sphaeriidae, and Chironomidae, in order of decreasing abundance. The first 3 groups account for approximately 94% of the bottom population, with Pontoporeia comprising about % of the total. Alley (1968) cites Merna (1960) as stating amphipods constituted about 70% of the macrobenthos. Eggleton (1937) and Alley (1968:66) found a zone of concentration for Pontoporeia at 35 m, with 8420/m^ being recorded at 40 m. Marzolf (1962:32) found a maximum of about 14,221/m“ in Grand Traverse Bay, Lake Michigan. Amphipods are not only the most abundant bottom organism and the most prevalent in our food samples, but also the region of concentration of this organism is the most frequent depth at which Oldsquaws are taken (Ellarson 1956). The lower limit of the thermocline in Lake Michigan is generally about 35 m, and this coincides with the junction of the sublittoral and profundal zones (Alley 1968). Environmental extremes in the sublittoral zone cause a lower density of amphipods, while cold temperatures and less food cause decreased numbers of amphipods in the profundal zone (Alley 1968). Field tests indicated a thin detrital film (<5 mm) was generally present on the bottom of Lake Michigan, and laboratory experiments suggested amphipod densities were positively correlated with the density of bacteria in this organic matter (Marzolf 1962). The sublittoral or inshore areas are constantly subjected to wave action which causes the organic matter to be resuspended and deposited elsewhere, thereby lowering the productivity of the area for bacteria and amphipods ( Alley 1968) . The decrease in the food base contributes then to less attractive and constantly changing feeding areas for Oldsquaws. Eggleton (1937) reported Tubificidae to be abundant in the benthic zone of Lake Michigan, but we found relatively few in our samples. Oldsquaws will take members of the Tubificidae, as Rofritz ( 1972:57) noted: “Sludge worms were overwhelmingly the most important food source for Oldsquaw in the Milwaukee Harbor.” Milwaukee Harbor has been dredged to 9 m and the bottom is a hard clay. The sew age treatment plant in the center of the harbor THE WILSON BULLETIN • VoL 89, No. 1, March 1977 produces conditions suitable for an abundance of “sludf?e worms” (335,000/ m“ maximum, Rofritz 1972:56), and this food organism is readily available to Oldsquaws owing to the hard bottom. The substrate in Lake Michigan where Oldsquaws are most often taken is largely a mixture of sand and silt. Tubificidae can burrow into these materials and be essentially unavailable to Oldsquaws even though these worms are one of the most prevalent organisms in the benthic zone. Other data indicate the Oldsquaw is an opportunistic feeder. Several of our samples revealed that a few small groups of birds had been feeding on fish even though the majority of the flocks taken at the same time in the same area had been feeding on amphipods. Ellarson (1956:215) and Hull (1914) have observed Oldsquaws feeding on locally abundant schools of minnows, and Oldsquaws have been seen diving for fish offal discarded by the com- mercial fishermen on Washington Island. Madsen (1954) noted that the high incidence of mollusks in Oldsciuaws collected along the Danish coast reflected the abundance of these bivalves in the marine waters, whereas Cottam’s (1939) material showed a preponderance of crustaceans from birds collected largely from freshwater habitats. An important relationship of Oldsquaws to the gill-net fishery on Lake Michigan exists through the common amphipod food base. Our data indicate this invertebrate is the i)redominant food organism in 01ds(iuaws, and Koelz (1927:528), Rawson (1953), and Ellarson (1956) found the principal food of the whitefish (Coregonus clupeajormis) was amphipods. Since this bird and this fish are both largely dependent on the same food, it is understandable why both fish and ducks tend to concentrate in areas where amphipods reach maximum density, and why fishermen set their nets in these waters. Old- s(iuaws are also related to the gill-net fishery by their occasional predation on fish eggs. 3 he seasonal relationship between tbe volume of food ingested and the percent of lipids found in the bird is more difficult to explain. Because we do not know the energy reciuirements of Oldsquaws during the winter, one must interpret the data in Eig. 1 with caution. Surely the decline in lipids between January and April is caused by some stress factor, but whether or not this is due to a decrease in the volume of food ingested, as implied in Fig. 1, is (luestionable. Oldsquaws need a minimum winter food intake to maintain an energy balance, and when this energy demand is not met, stored energy must be used. As winter temperatures drop and shore ice forces the birds into deeper water, more energy is needed ; a decrease in energy intake could cause the observed decline in lipid reserves. King (1961) observed in White-crowned Sparrows (Zonotrichia leuco- phrys) that just prior to spring migration an increase in lipid deposition was Peterson and Elhirson • FOOD IIAIUTS OF OM)S(,)lIA\VS DO accompanied hy an increase in feeding activity that created a positive energy balance. Fig. 1 suggests this is not so for ()lds(iuavvs wintering on l.ake Michigan because lipid mobilization does not begin to level off for a month after food intake rises, and an increase in lipid reserves does not occur until about 2 months after the food intake rises. This suggests that, although the difference in the volume of food ingested between February and April may he statistically significant, there is no simple biological relationship between food intake and fat deposition during late winter and spring. As King and Farner (1966) pointed out, a change in lipid deposition is usually the result of changes in metabolism rather than changes in food availability. The April-May decline in food intake concurrent with increased lipid de- position suggests a negative feedback relationship between tbe level of fat reserves and appetite, as asserted by Dolnik and Blyumental ( 1964) for small migratory birds. These authors state that the characteristics of this system are seasonally variable. We believe additional controlled studies are necessary before we can fully interpret the relationship between energy con- sumption and lipid metabolism in Oldsquaws. SUMMARY The food habits of Oldsquaws wintering on Lake Michigan were examined from material collected in 2 periods: 1951-1954 and 1969-1972. Grit averaged 25% of total contents in the earlier sample, and 90% of all grit was sand. Animal food constituted about 99% hy volume of the food organisms present in the 1951-1954 sample. The principal food item was an amphipod, Pontoporeia affinis, which occurred in 52-96% of the earlier sample and in 88-95% of the 1969-1972 sample. Clams occurred frequently in both samples hut the volume ingested was relatively low. The occurrence of fish and fish eggs in the diet varied with the locality and individual flocks. Oldsquaws are related to the coregonid fishery in Lake Michigan through the common food organism Pontoporeia affinis. Oldsquaws will also eat fish eggs when available. The decline in Oldsquaw carcass lipids during the winter may he related to a decrease in the volume of food ingested during that period, hut a rise in the volume of ingested food during early spring does not appear to be associated with a premigratory increase in lipid deposition. Oldsquaws apparently decrease their feeding activity just prior to migration and do not resume heavy feeding while on spring migration. ACKNOWLEDGMENTS We thank E. Ellefson and the many commercial fishermen on Lake Michigan who saved birds caught in their gill-nets for our examination. Our appreciation is extended to Dr. Robert G. Williamson and staff of the Institute for Northern Studies, Arctic Research and Training Center at Rankin Inlet, Northwest Territories, for allowing us to use their facilities during our 1971 survey. This study was funded, in part, hy the National Oceanic and Atmospheric Administration’s Office of Sea Grant, Department of Commerce, through an institutional grant to the University of Wisconsin. 90 THE WILSON BULLETIN • VoL S9, No. 1, March 1977 LITERATURE CITED Alley, W. P. 1968. Ecology of the burrowing amphipod Pontoporeia ajfinis in Lake Michigan. Ph.D. thesis, Univ. Michigan, Ann Arbor. Bartonek, J. C. and J. J. Hickey. 1969. Food habits of Canvasbacks, Redheads, and Lesser Scaup in Manitoba. Condor 71:280-290. CoTTAM, C. 1939. Food habits of North American diving ducks. U.S. I)ep. Agric. Tech. Bull. 643. Dement’ev, G. P., N. a. Gladkov, Y. A. Isakov, N. N. Kartashev, S. V. Kirikov, A. V. Mikheev, and E. S. Ptuchenko. 1952. Birds of the Soviet Union. Vol. 4. Trans- lated 1967 Israel Prog, for Sci. Trans., Jerusalem. Dillon, 0. W., Jr. 1957. Food habits of wild ducks in the rice-marsh transition area of Louisiana. Proc. Annu. Conf. Southeast. Assoc. Game Fish Comm. 11:114-119. Dolnik, V. R. AND T. I. Blyumental. 1964. Bioenergetika migratsii ptits. Usp. Sovrem. Biol. 58:280-301. Eggleton, F. E. 1937. Productivity of the profundal benthic zone in Lake Michigan. Michigan Acad. Sci. Arts and Letters 22:593-611. Ellarson, R. S. 1956. A study of the Old-squaw Duck on Lake Michigan. Ph.D. thesis, Univ. Wisconsin, Madison. Hickey, J. J., J. A. Keith, and F. B. Coon. 1966. An exploration of pesticides in a Lake Michigan ecosystem. J. Appl. Ecol. 3 ( Supp.) : 141-154. Hull, E. D. 1914. Habits of the Old-squaw (Harelda hyemalis) in Jackson Park, Chicago. Wilson Bull. 26:116-123. King, J. R. 1961. On the regulation of vernal premigratory fattening in the White- crowned Sparrows. Physiol. Zool. 34:145-157. AND D. S. Earner. 1966. The adaptive role of winter fattening in the White- crowned Sparrow with comments on its regulation. Am. Nat. 100:403-418. Koelz, W. 1927. Coregonid fishes of the Great Lakes. Bull. U.S. Bur. Fish. 43:297-643. Lagler, K. F. and C. C. Wienert. 1948. Food of the Old-squaw in Lake Michigan. Wilson Bull. 60:118. Mackay, G. H. 1892. Haliits of the Oldscjuaw (C/angula hyemalis) in New England. Auk 9:330-337. Madsen, F. J. 1954. On the food habits of diving ducks in Denmark. Danish Rev. Game Biol. 2:157-266. Marzolf, G. R. 1962. Substrate relations of the burrowing amphipod Pontoporeia af finis Lindstrom. Ph.D. thesis, Univ. Michigan, Ann Arbor. Merna, j. W. 1960. A henthological investigation of Lake Michigan. M.S. thesis, Michigan State Univ., East Lansing. Polunin, N. and C. R. Eklund. 1953. Notes on food habits of waterfowl in the interior of Ungava Peninsula. Can. Field-Nat. 67:134—137. Rawson, 1). S. 1953. The bottom fauna of Great Slave Lake. J. Fish Res. Board Can. 10:486-520. Rofritz, D. j. 1972. Ecological investigations on waterfowl wintering in the Milwaukee Emhayment. M.S. thesis, Univ. Wisconsin, Milwaukee. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill Book Co., New York. Swanson, G. A. and J. C. Bartonek. 1970. Bias associated with food analysis in gizzards of Blue-winged Teal. J. Wildl. Manage. 34:739-746. Peterson and EUarson • FOOD IIAIJFI'S OF ()U)SOlIAWS ^)\ Zimmerman, F. R. 1953. Waterfowl liuhitat surveys and food liahit studies, 1910 43. Final Rep. l^ittman-Rohertson IVoj. 6-R, Wisconsin (^ons. De{)t. (A copy has l)(;en deposited in the Van Tyne Library.) DEPT. OF WILDLIFE ECOLOGY, UNIV. OF WISCONSIN, MADISON 53706. ( PKESENT ADDRESS SRP: DEPT. OF WILDLIFE ItESOURCES, UNIV. OF IDAHO, MOSCOW 83843.) ACCEPTED 10 MAY 1976. SEXUAL DIMORPHISM IN THE WHITE IBIS James A. Kushlan Sexual dimorphism in body size has been variously accounted for in different species of birds ( Selander 1972), though the phenomenon has been little studied in ciconiiforms. In this paper I analyze dimorphism in the White Ibis (Eudocimus albus)^ suggested by the few measurements in Palmer (1962 ) and briefly noted by Rudegeair (1975). I also discuss the potential significance of this dimorphism, especially related to food use, feeding behavior, and nesting. The morphological analysis is based on a sample of 36 specimens collected from a single population. METHODS All White Ibis specimens were collected in the Everglades and Big Cypress Swamp of southern Florida during the 1972 and 1973 nesting seasons. Standard measurements were made of tarsus, middle toe, wing arc, 8th primary, and central rectrix on the left side. Bill was measured as straight line distance (chord) from tip to anterior edge of nares. Body weight was taken after removal of stomach contents. Stomach contents were identified, separated, dried, and expressed as percentage of total dry weight. RESULTS AND DISCUSSION Both juvenile and adult males averaged significantly larger than females in all characteristics for which I have adetiuate samples (Tables 1, 2). Adults were most dimorphic in weight, females averaging 74% the weight of males, and most similar in measurements of the flight structures, females averaging 91% of males. Juveniles differed significantly from adults in several char- acteristics because of slow post-fledging growth, hut sexual dimorphism was still apparent in these as well as adults. Rills of females averaged 78% as long as hills of males, and also differed in shape and massiveness (Fig. 1). Bill dimorphism is probably the easiest sexual difference to detect in wild birds and is apparent to experienced ob- servers when the sexes are together and in many cases when they are not. The ratio of male to female bill length is 1.25. Such a pronounced difference suggests that the sexes are using different food resources ( Schoener 1965, Selander 1966). Differences in leg length, and weight (Hespenheide 1971) have also been related to differential food consumption. Hutchinson (1959) showed that ratios of trophic appendages of different species exhibiting char- acter displacement were about 1.28, while Schoener (1965) found smaller character difference ratios, about 1.14, among species of congeneric sympatric birds and suggested (Schoener 1970) that for animals of similar morphology 92 Mean Size (g or mm), Standard Error of the Mean, Range, and Coefficient of Variation of Meristic Characteristics Kushlan • WHITE IIHS DIMORPHISM CO lO ro O Tf< LO ON csj Tj; t-H r-; CO lO LO lO CO C^i CS CO CO lO CO CO o o O CM C^ r— I CM CO CM LO sC CO a^ CM i-H CO + 1 +1 +1 +1 +1 +1 +1 lO r'- I— I o CO CO O On C\ CM i-H CM >— I O O CO CO I-H r-H 00 CM O CO CM O CO (-5 LO CM CO CO ^ 2 LO LO LO CO I— I CM I— I CO CM r— I ''A I-H lA LO _ C CO 0\ LO 0\ o o ^ I-H CM CM I-H CO +1 II II II II II II Tji CM CM O CM LO CO Tf O O I-H O ^ I-H I-H CO CM ,-H CM CM CM CM CM CO 3 ^ ^ S H S ^ H LO CO LO Cl CM LO 00 LO I-H O CL O CM cm' lO CO i-i' CO CO r- CM I-H I-H OL CL LO f-H Cl lO CM I-H OL I 1 I I I I LO CO CO lO OL O CO LO LO CO 00 I-H CM I-H CM r-H CM LO CM I-H II II II II II II O CO LO O CO CM I-H CO lO LO CO Cl r-H CM r-H CO CO CO CO CO CO HjH CM r- LO O CM CM CO LO CO (Cl r-H LO CM co' CO i-h' i-h' ■rf O Tjl CM LO -I O r-H O ^ CO CM I-H I I I I I 1 Cl LO CM cO CO Cl O Cl O O I-H CM CM I-H II II II II II II II LO O CM 00 CO I-H hJi O Hji O O CL O O CL I-H I-H CM CM I-H Hfi CO Tfl ^ ^ PH E ^ -S ^ 2 ?C ‘h '3 ^ m H ^ Ch H Weights are minus stomach contents. 94 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Differences Table 2 OF Characteristics of White Ibises in Shown in Table 1 Southern Florida Adult vs Juvenile Juvenile vs Juvenile Adult vs Adult cf vs d $ vs $ cf vs $ d vs ? DS Weight * * - 73.8 Bill - - ** ** 78.3 Tarsus - - ** ** 85.8 Middle Toe - * * * ** 88.0 Wing arc - * ** * 91.7 Primary 8 - * * ** ** 90.7 Tail - - ** * * 91.7 DS = Sexual difference = 100 (mean of femaies/mean of males), see Selander (1966). * = Significantly different by t test at P < 0.05. ** = Significantly different by t test at P < 0.01. - = Not significantly different. and feeding liehavior, smaller ratios than found by Hutchinson should result in the taking of different foods. The large liill-size ratio of the W^hite Ibis indicates that selection for character dimorphism within this single species has resulted in a difference of similar magnitude to that for competing species. However, I found little evidence of resource partitioning. In nearly all instances, I observed no discernible difference in the feeding behavior of male and female White Ibis. An alternative method of resource division would he for ibis to forage allopatrically. Because they possess larger bills and legs, males could forage in deeper habitats than females. My limited data show that they do this in only 2 habitats, canal-edge marshes ( 19 observa- tions of S 2 oi $ 9) and offshore on coastal shoals (lo observations of S3, none of 9 9). However neither habitat is heavily used by ibises. I found no differences between sexes in the mixed species flocks feeding in the Everglades, the primary feeding habitat in southern Florida (70 observa- tions). As feeding site data are somewhat ecjuivocal, 1 collected males and females feeding in the same locations on 3 occasions in the heavily used fresh water marshes. In no case are large differences in the food taken by the sexes apparent (Table 3). The 2 sets of Everglades samples show overlap between sexes of 98 and 99% and the Cypress Swamp samples have a 75% overlap (Morisita’s index of overlap, Horn 1966). Although Earhart and Johnson (1970) suggested that consumption of numerous and relatively small prey items, such as is the case in ibises, results in the lack or reduction of dimorphism, White Ibis appear not to follow this generalization. It remains possible, of course, that the sexes could he taking different Kiish/an • WHITE IIHS DIMOKITHSM 95 Fig. 1. Sexual dimorphism in the l)ill length and hill shape of the White Ihis. Male left, female right. foods, particularly during the non-breeding season when water levels in the interior marshes of southern Florida are deeper and most birds feed along the coast in tidal situations. Male ibises may also take their prey from deeper in the sediment than females and thus take the same prey type without competition. Sexual dimorphism in the White Ibis, as in many species ( Selander 1972 ) , is likely adaptive in reproduction. The larger size and aggressiveness of males is important because males spend much of their time defending the nest. In the first 2 weeks after hatching, the male broods during most of the daylight hours ( Kushlan 1976) when predators and other colony members are active and thus protects the nestlings until they are large enough to he left alone. Dimorphism may also be of advantage to females in nesting. A smaller Table 3 Food of Male and Female White Ibis Collected at the Same Time and Location THE WILSON BULLETIN VoL 89, No. 1, March 1977 o o On CO ca c/D O p cm' i-h t'- csi lO On rH On "d CM CO O NO LO CM CO o o ON O On Ch- p CO p Os lO ON w it 0) w "D O o o 8 ^5 CM o P CM P LO o' ON LO NO NO pN CO' l-H P lO r-H LO CM o' O CO CM p r- p cm' cm’ o' P --H cm' o 5 5 a a. ^ 1:;. R3 03 c: '2 ^ ^ c Lh — * Data expressed as % of dry weight of stomach contents. Kushlan • WllITP: II5IS I )IM()KIM 1 ISM 97 female may be better able to enter the male’s territory durinji; i)air formalion. Dimorphic enlargement of the gular sac of the female also functions during pair formation (pers. obs., Rudegeair 1975). A primary pairing behavior consists of intertwining downward head thrusts described by Meyerriecks [in Palmer 1962). In early pair formation, the female when bringing her head upward turns it sideways to the male and bolds that posture rigidly, thus displaying in profile the small hill, bright red facial skin and enlarged gular pouch. Rudegeair (1975) suggested a similar function during another display. Thus dimorphic development of the pouch and the female’s smaller size probably function together in pair formation. During the first 2 weeks after hatching, the burden of food gathering falls primarily on the female. This is similar to the situation in some raptors in which the male (which is smaller) is the major food provider through the early stages of nesting. It has been argued (e.g., Reynolds 1972) that smaller size is adaptive in increasing foraging efficiency on numerous and agile smaller prey. Such reasoning is not transferable from actively pursuing predators to searching predators, such as White Ibis, that eat passive food items as they are en- countered. However, the smaller size of the female ibis does suggest that the amount of food needed for her own metabolism may he less than that of the male (Mosher and Matray 1974) and so a greater percentage of the foraging effort can he allotted to obtaining food for the young. This may he a selective force maintaining the smaller size of the female. Body size differences may also permit the promiscuous mating behavior that characterizes this species ( Kushlan 1973 ) . Irrespective of the adaptive- ness or maladaptiveness of promiscuity, die smaller size of the female may make it advantageous for her to permit promiscuous copulation rather than ineffectually attempt defense. The larger size of the male confers advantage in both dominance and mating interactions, much as the case in polygynous systems. Thus sexual dimorphism in size is a recognizable characteristic of the White Ibis that probably serves several functions within the adaptive complex of the species. My field observations, examination of small numbers of museum specimens of known sex, and comments in the literature suggest that other ciconiiforms including the Roseate Spoonbill {Ajaia ajaja). Glossy Ibis ( Plegadis falcinellus), Wbite-faced Ibis { Plegadis chihi). Scarlet Ibis \Eudo- cimus ruber). Sacred Ibis {Threskiornis aetJiiopica) , American Wood Stork (Mycteria americarm) , and the Marabou { Leptoptilos crumenijerus) are also sexually dimorphic in body size. Herons appear to be less obviously dimorphic, if at all, although statistical differences exist in some species (Browder 1973). For species in which dimorphism is recognizable in the field, body size difference becomes a promising tool in behavioral and 98 THE WILSON BULLETIN • Vol. G9, No. 1, March 1977 ecological study. Because of this, comparative study of character dimorphism in other ciconiiforms is desirable, ACKNOWLEDGMENTS The study was conducted at the University of Miami, specimens cited in this paper are deposited in the reference collection there. It was supported by the Maytag Chair of Ornithology. I thank Marilyn S. Kushlan, John C. Ogden, Oscar T. Owre, Ralph W. Schreiher, and William B. Robertson. Jr. for Comments. Marilyn Kushlan also assisted in data gathering and drew the figures. LITERATURE CITED Browder, J. A. 1973. Studies on the feeding ecology and morphological variation of the Cattle Egret, Bubulcus ibis (Linnaeus) (Aves: Ardeidae). M.S. thesis, Univ. Miami, Coral Gables, FI. Earhart, C. M. and N. K. Johnson. 1970. Sex dimorphism and food habits of North American owls. Condor 72:251-264. Hespenheide, H. a. 1971. Food preferences and the extent of overlap in some insectiv- orous birds, with special reference to the Tyrannidae. Ibis 113:59-72. Horn, H. S. 1966. Measurement of “overlap” in comparative ecological studies. Am. Nat. 100:419-424. Hutchinson, G. E. 1959. Homage to Santa Rosalia, or why are there so many kinds of animals? Am. Nat. 93:145-149. Kushlan, J. A. 1973. Promiscuous mating behavior in the White Ibis. Wilson Bull. 85:331-332. . 1976. Feeding rhythm in nestling White Ibis. Wilson Bull. 88:656-658. Mosher, J. A. and P. F. Matray. 1974. Size dimorphism: a factor in energy savings for Broad-winged Hawks. Auk 91 :325-341. Palmer, R. S. 1962. Handbook of North American birds. Yale Univ. Press, New Haven, Conn. Reynolds, R. T. 1972. Sexual dimorphism in accipiter hawks: A new hypothesis. Condor 74:191-197. Rudeceair, T. 1975. The gular pouch of the female \^'hite Ibis. Auk 72:168-169. ScHOENER, T. W. 1965. The evolution of hill size differences among sympatric congeneric species of birds. Evolution 19:189-213. . 1970. Size patterns in West Indian Anolis lizards II. Correlations with the sizes of particular sympatric species. Displacement and convergence. Am. Nat. 104: 155-174. Selander, R. K. 1966. Sexual dimorphism and differential niche utilization in birds. Condor 68:113-151. . 1972. Sexual selection and dimorphism in birds. Pp. 180-230 in Sexual selection and the descent of man. ( B. G. Campbell, ed. ) Aldine Puhl. Co., Chicago. DEPT. OF BIOLOGY, UNIV. OF MIAMI, C0R.\L GABLES, FL 33124. (Present address: U.S. NATIONAL PARK SERVICE, EVERGLADES NATIONAL PARK, HOMESTEAD, FL 33030). ACCEPTED 4 DEC. 197.5. EGGSHFJJ. THIGKNESS VAKIABIEn Y IN THE WHITE-EACED IBIS David E. Cai’en Many recent papers have reported the occurrence of persistent chemicals, such as DDT, and the associated thinning of eggshells of birds (Cooke 1973). Eggshell thinning is commonly documented by comparing recently-collected eggs with eggs found in museum collections. There are a number of factors, other than persistent chemicals, which may contribute to the variation in the thickness of eggshells. Some of these factors are geographic location ( Ander- son and Hickey 1970, 1972 1 ; the genetics, physiology, and diet among females (Romanoff and Romanoff 1949); the stage of incubation (Vander- stoep and Richards 1970, Kreitzer 1972) ; the order in which the eggs are deposited (Romanoff and Romanoff 1949, Berg 1945); and the size of the clutch ( Rothstein 1972). Hence, when collecting eggs to study shell thickness, one must be aware of these factors to determine the proper composition of the sample. I considered it important to conduct a study of eggshell variability of a species nesting in the wild, since most related investigations have been con- ducted in the laboratory, usually with the domestic chicken {Callus gallus). The purpose of this study was to evaluate variability in the shell thickness of eggs of a population of the White-faced Ibis ( Flegadis chilli) . The parameters studied were (1) the length of incubation, (2) the order in which the eggs were deposited, and (3) the clutch size. METHODS During May 1974, White-faced Ibis eggs were collected from active nests in the Bear River delta, Box Elder Co., in northern Utah. More than 1000 pairs of Ibises comprised the study colony. Eggs were taken from 112 of 220 nests where egg deposition was synchronous. The nests were visited each afternoon during the egg-laying period, and all eggs were marked to indicate the order of deposition. Complete clutches most com- monly contained either 3 or 4 eggs. I selected a random sample of 56 nests when egg-laying ceased. From each nest I took 1 egg, also randomly, so that the completed sample contained 8 replicates of the first egg deposited in a nest, 8 replicates of the second egg deposited, etc. Corresponding samples (8 replicates of the first egg, second egg, . . .) were collected from both 3-egg and 4-egg clutches. Thus, the 56-egg sample included 24 eggs from 3-egg clutches and 32 eggs from 4-egg clutches. A second sample of 56 eggs was collected after incubation had progressed to an average of 17 days, or about 4 days prior to hatching. All eggs contained embryos. This sample was selected in the same manner as before. After the eggs were removed from the nests, they were refrigerated for -4-6 days, then 99 100 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Table 1 The Collection Design and Mean Eggshell Thickness (mm) for Samples of White-faced Ibis Eggs Oi Order of Egg Deposition O2 O2 O4 Sample Means 6-days Incubation (L)^ 3-egg Clutches (Ci) .309" .331 .325 .322 .324 4-egg Clutches (C2) .321 .314 .341 .325 .325 17-days Incubation (U) 3-egg Clutches (Ci) .308 .324 .321 .318 .310 4-egg Clutches (C2) .302 .307 .306 .300 .304 Sample Means .310 .319 .323 .312 .317 1 Mean incubation for early eggs 2 Each figure represents a mean = 5.8 days; for late eggs for 8 eggs. = 16.6 days. cut around the equator, the contents removed, and the eggshells rinsed with warm tapwater. Eggshells were dried on ahsorlient paper for 24 h in a low-humidity laboratory. Eggshell thickness measurements were made with a Starrett No. lOlOM dial micrometer, read to the nearest .01 mm. Six measurements were taken, equally spaced around the equator of each eggshell. To obtain a desired precision of 95% for the eggshell thickness measurements, I calculated the variance for the 6 measurements of each eggshell, then computed the number of measurements necessary' to he 95% confident of remaining within a 5% limit of error (Steel and Torrie 1900:86). These calculations showed that, on the average, 5.3 measure- ments from each eggshell were necessary for the desired precision. Since I had taken a sufficient number of measurements, only the mean thicknesses for the eggshells, expressed to the nearest .001 mm, were used in the analyses. A 2 X 2 X 4 (Incubation [II X Clutch size [Cl X Order of deposition [01) factorial design (Steel and Torrie 1960:194) was tlie basis for the collection of eggs (Table 1). A freijuency distribution of the data exhibited no kurtosis. E(jr the statistical analyses, the collection design was partitioned into 3 separate factorial designs: a2x2x3, a2x4, and a 2 X 3. This was necessary because 2 cells ( LCiOi and LC1O4) do not exist, i.e. there are no fourth eggs deposited in nests with 3-egg clutches. The data were tested by standard analy sis of variance teclmiiiues. RESULTS Sources of variation. — The analysis of variance results are presented in Table 2. The effects of 3 variables on eggshell thickness were investigated in the 2X2X3 analysis: the length of incubation, the order of egg deposition, and clutch size. The effect of incubation was significant ( P < .01). Eggs which were incubated for an average of only 5.8 days had shells that were 4.3% thicker than shells from eggs which had undergone 16.6 days of incuha- Capcn • inis K(;(;S1IE[.L THICKNESS Analysis of Variance Table 2 for Shell Thickness OF WlIITE-FACEU Ibis Ecus Designs and Sources of Variation Degrees of freedom (df) Mean squares ( MS ) F 2X2X3 Incubation (I) 1 .00373 8.88*** Order (0) 2 .00149 3.55** Clutch Size (C) 1 .00048 1.14 I X 0 2 .00034 <1 Ixc 1 .00159 3.79* 0 X C 2 .00094 2.24 I X 0 X c 2 .00048 1.14 Error 84 .00042 Total 95 2X4 Incubation 1 .00753 18.82*** Order 3 .00056 1.40 I X 0 3 .00055 1.38 Error 56 .00040 Total 63 2X3 Incubation 1 .00023 <1 Order 2 .00163 3.70** I X 0 2 .00002 <1 Error 42 .00044 Total 47 Nested factorial Incubation 1 .004643 13.66*** Order 5 .000987 2.37** Clutch Size 1 .000738 <1 I X 0 5 .000340 <1 Ixc 1 .002065 6.08** Error 98 .000416 Total 111 * P < .10; **F<^ .05; *** P < .01. tion. The order of deposition also significantly (P < .05) affected the eggshell thickness. In general, the first and last eggs had thinner shells than those in between. The 2 different clutch sizes showed a non-significant (P > .10) effect. 102 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Three 2-vvay interactions and one 3-vvay interaction were tested. All were non-significant (P > ,10) except one. A significant (P < .10) effect was obtained with the incubation X clutch size interaction. The nature of the implied relationship is illustrated in Fig. 1. Apparently, there was a greater decrease in shell thickness during incubation for eggs in 4-egg clutches than for eggs in 3-egg clutches. Only the effects of incubation and order of deposition were tested in the 2-factor designs. Data from 4-egg clutches were used in the 2X4 analysis. The only significant (P < .01) factor was the length of incubation. In the 2X3 analysis, however, where only data from 3-egg clutches were used, the significant factor was the order of deposition (P < .05), The length of incubation did not significantly affect the shell thickness of eggs from 3-egg clutches. This, to some degree, confirms the interpretation of the significant interaction discussed above. The data were also analyzed using a nested, factorial design ( Sokal and Rohlf 1969:256). Sums of stjuares for the analysis of variance were obtained using regression techniques for an unbalanced design. Though not as straight- forward, this is actually a more powerful analysis since the design used all available data. The results were essentially the same as those of the previous analyses, though more convincing (Table 2). Again, the effect of incubation length was significant (P < .01), as was the order of deposition (P < .05). The incubation X clutch size interaction was significant (P < .05) and more noticeable than before. DISCUSSION dhere was no apparent interference from pecticides in the eggs collected in this study. Eggshells did not differ in thickness from shells of eggs collected prior to 1940 and preserved in museums. The mean thickness for the pre-1940 museum eggs collected in Utah was 0.324 mm (N = 29) (unpublished data, Denver Wildlife Research Center, U.S. Fish and Wildlife Service). For a comparable sample of eggs in this study, shell thickness also averaged 0.324 mm ( N = 56 ) . Incubation. — I was not surprised to observe the significant decrease in shell thickness between the eggs collected soon after incubation had begun and those taken just prior to hatching. A developing embryo obtains calcium from the eggshell. Simkiss (1967) estimated that 5% of the shell calcium may he used by the chicken embryo. Kreitzer (1972j reported a 7.3% decrease in shell thickness between incul)ated and unincuhated eggs of the Coturnix Quail {Cotuniix japonica). Rothstein (1972), who studied a species nesting in the wild, the Cedar Waxwing (Hoinhycilla cedrorum) ^ demon- strated a similar association between incubation and eggshell thickness. Capen • HUS EGGSHELL TiHGKNESS U)?y w/' ^ ' ' ^ 0 8 12 16 20 Incubation (days) Fig. 1. The effect of incubation on shell thickness in White-faced Ibis eggs from 3- and 4-egg clutches. The data in this study, showing a 4.3% decrease in shell thickness over approximately 11 days incubation, correspond with other published informa- tion. However, most of the decrease is a function of eggs from 4-egg clutches, which showed a 6.5% decrease in shell thickness, while eggs from the 3-egg clutches dropped only 1.2% (Fig. T) . On the average, eggs from 4-egg clutches were incubated only 0.7 days longer than those from 3-egg clutches. There is only a slight probability iP < .02 ) that the relationship is due to sampling error, hence there may be cause for additional study. Order of deposition. — The relationship between eggshell thickness and the order in which the eggs are laid has apparently not been investigated in wild birds. In the chicken, shell thickness usually changes throughout the laying 104 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 cycle. Generally, the first and last eggs have the thickest shells (Wilhelm 1940, Berg 1945). My data with the eggs of the White-faced Ibis suggest the opposite relationship, i.e. the first and last eggs are usually the thinnest. Nevertheless, there is noticeable variation (Table 2), and the order of egg deposition should not be overlooked when collecting eggs for determination of shell thickness. Clutch size. — Complete clutches of both 3 and 4 eggs are common in the White-faced Ibis, and many other species (Lack 1968:330). Thus, I believed that clutch size should be evaluated as a variable which might affect shell thickness. The analyses indicated that clutch size did not have a significant effect on thickness, despite the different responses to incubation discussed above. Rothstein (1972) reported that shell thicknesses were different in eggs of the Cedar Waxwing collected from 3-egg clutches and those collected from 4- and 5-egg clutches. However, Rothstein could not account for variation in the sequence of egg deposition which, as he pointed out, may have confounded the variation attributed to different clutch sizes. The shell thickness-clutch size relationship is worthy of investigation in other species. Sampling. — There are some practical implications from these results. When collecting eggs to measure shell thickness, one should attempt to take as few eggs as possible, yet sample enough to detect a desired difference, with confidence. Hence, one should select a sampling scheme which provides as little variation as possible among eggs. Klaas et al. (1974) analyzed variation in eggshell thickness in 5 species and concluded that the most efficient sampling procedure was to collect entire clutches. These authors emphasized that this scheme allows for the most efficient use of time and resources, while minimizing the impact on the reproductive success of the species. Their recommendations are valid for many species. For the White-faced Ihis, I favor a different sampling design. Intuitively, from the data presented in this pa])er, the most efficient design would be one where the two significant sources of variation in shell thickness are eliminated. Thus, I recommend: (1) collecting eggs before inculcation has progressed, and ( 2 ) collecting one egg per nest, sampling only the first eggs or only the second eggs laid. Of course, this plan is practical only when the egg laying seciuence can be determined. riie following comparison illustrates the efficiency of my recommendations. Klaas et al. (1974) presented eggshell thickness data for the Black-crowned Night Heron {Nycticorax nycticorax) and reported the sample estimate of the coefficient of variation (C.V. ) as 6.55. For White-faced Ibis eggs (this paper ) taken during early incubation, the C.V. was 6.54, thus our samples are clearly comparable. Klaas et al. (1974:162) used a formula (Sokal and Rohlf 1969:247) for estimating the sample size necessary to detect a minimum Capen • IIJIS E0(;SI1ELL THICKNESS 105 difference in mean eggshell thickness. These authors estimated that it would be necessary to collect 38 clutches (112 eggs) to show a 5% diffrence in shell thickness with eggs of the Black-crowned Night Heron. Using a different design, collecting one egg at random from a clutch, the authors calculated that 51 eggs would be required. 1 used the same formula, significance level, and power as Klaas et al. ( 1974) and calculated that only 34 eggs from the White- faced Ibis would be needed, providing the eggs collected were the first laid in each clutch. My sample for this calculation consisted of 16 eggs. With a similar sample, but taking the second eggs laid in each nest, it w as estimated that 41 eggs would be needed. I must (lualify my recommendations where shell thickness comparisons involve museum eggs. The sequence of egg deposition in clutches preserved in museums is rarely, if ever, known, thus eggs in museums are samples which represent all orders of deposition and clutch sizes. The most comparable field sample then, would be either one egg at random from many clutches or the collection of entire clutches. However, for a comparison of shell thickness differences between years or geographical areas, it may be more efficient to eliminate the variability due to the sefjuence of egg deposition. My suggested sampling scheme may be the most practical for colony-nesting species. For birds such as the White-faced Ibis, proper synchronization of nesting activities may influence reproductive success; therefore renesting, if it occurs at all, may have much poorer success (unpublished data). Also, there may be more nestlings than adults can feed, and many do not survive. Thus, taking one egg from a nest may not hinder reproductive output. In species possessing these characteristics, the impact of egg collecting may be minimized by selecting one egg per clutch rather than entire clutches. In other species, it may he desirable to collect entire clutches, rather than single eggs. Hence, when collecting wild bird eggs to detect differences in eggshell thickness, one should consider the biology and behavior of the species as well as the factors wTich contribute to eggshell variability. SUMMARY Eggs of the White-faced this were examined for natural variahility in shell thickness. Eggs collected soon after they were laid had thicker shells than those collected just prior to hatching. Eggshell thickness also varied with the order in which eggs were laid. Different clutch sizes (3 or 4 eggs) did not contribute to the variability in shell thickness. A design for collecting eggs to determine shell thickness is suggested. The most efficient sampling scheme for eggs of the White-faced Ibis, and perhaps other species, involved collecting only one egg per nest rather than entire clutches. ACKNOWLEDGMENTS I thank J. B. Low, D. W. Anderson, E. G. Bolen, J. 0. Keith, and E. E. Klaas for their advice and critical review of the manuscript. H. C. Romeshurg and D. V. .Sisson made 106 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 helpful suggestions on statistical procedures. Financial support was provided hy a fellowship from the Rob and Bessie Welder Wildlife Foundation, and this paper represents Scientific Contribution No. 182 from tbe Foundation. W. I. Jensen kindly provided laboratory facilities at the Bear River Research Station. LITERATURE CITED Anderson, D. W. and J. J. Hickey. 1970. Oological data on egg and breeding character- istics of Brown Pelicans. Wilson Bull. 82:14-28. . 1972. Eggshell changes in certain North American birds. Proc. XV Int. Ornithol. Congress: 514-540. Berg, L. R. 1945. The relationship of clutch position and time interval between eggs to eggshell quality. Poult. Sci. 24:555-563. Cooke. A. S. 1973. Shell thinning in avian eggs by environmental pollutants. Environ. Pollut. 4:85-152. Klaas, E. E., H. M. Oiilendorf. and R. G. Heath. 1974. Avian eggshell thickness: variability and sampling. Wilson Bull. 86:156-164. Kreitzer, j. F. 1972. Tbe effect of embryonic development on the thickness of the eggshells of Coturnix Quail. Poult. Sci. 51:1764-1765. Lack, D. 1968. Ecological adaptations for breeding in birds. Metheun and Co., Ltd., London. Romanoff, A. L. and A. J. Romanoff. 1949. The avian egg. John W iley and Sons, N. AL Rothstein, S. I. 1972. Eggshell thickness and its variation in the Cedar Waxwing. Wilson Bull. 84:4^)9-474. Sim KISS, K. 1967. Calcium in reproductive physiology. Reinhold, N. Y. SoKAL. R. R. AND F. J. Roiilf. 1969. Biometry. The principles and practice of statistics in biological research. W'. H. Freeman and Co., San Francisco. Steel, R. (L 1). and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill, N. Y. Vanderstoep, j. and j. F. Richards. 1970. The changes in eggshell strength during incubation. Poult. Sci. 49:276-285. Wilhelm, L. A. 1940. St)ine factors affecting variations in eggshell (juality. Poult. Sci. 19:246-253. UTAH COOPERATIVE WILDLIFE RESEARCH UMT, UTAH STATE UMV., LOGAN 84322 (PRESENT ADDRESS: SCHOOL OF NATURAL RESOURCES, UNIV. OF VERMONT, BURLINGTON 05401. 1 . ACCEPTED 29 OCT. 1975. CHARACTERISTICS OF A WINTERING EORUEATION OF WHITE-TAILED PTARMIGAN IN COLORADO Richard W. Hoffman and Clait E. Braun Information concerning White-tailed Ptarmigan (Lagopus leucurus) in winter is lacking primarily because of poor access, harsh environmental conditions, and insufficient effort hy investigators to study them during this season. Some data on wintering populations of this species have been pub- lished ( Braun and Schmidt 1971, May and Braun 1972, Moss 1973, 1974, May 1975), but most studies have pertained to behavior (Choate I960, Schmidt 1969) and general population biology (Weeden 1959, Choate 1963, Braun 1969, Haskins 1969). In order to further understand the biology of this alpine grouse, we collected data opportunistically from 1966 through 1971 and intensively from 1972 through 1974. Objectives were to: (1) document departure and arrival from and to wintering areas, (2) describe sex and age composition of flocks, (3) examine flocking behavior and flock size, and (4) ascertain the affinity of marked birds for individual wintering sites. STUDY AREA AND METHODS Guanella Pass, situated in the Front Range of the Rocky Mountains in nortli central Colorado, was selected as the primary winter study area ])rincipally due to the abundance of wintering ptarmigan, accessil)ility. and availaldlity of information on this area from prior studies (Schmidt 1959. Braun and Schmidt 1971, May and Braun 1972). The area studied encompasses approximately 9 knr. an unusually large area compared to most known wintering grounds. In addition to studies at Guanella Pass, limited winter investigations were conducted near Waldorf. Naylor Lake. Steven’s Gulch, and Horseshoe Basin where small numbers of ptarmigan winter. These sites are located within 3 to 13 km of Guanella Pass and vary in size from 2 to 4 knP. Winter use sites can he categorized as: (1) those at or above treeline generally at the head of a drainage, and (2) those below treeline along stream courses. Guanella Pass is an example of the former type of winter use site. Topography of this area is rolling with slopes ranging from 5 to 30% and elevations from 3475 to 3655 m above sea level. The vegetation is complex and may best be described as a mosaic of several communities, mostly dominated hy willow (Salix spp.). which subtly intergrade with each other. Dominant communities are Salix-Carex wet meadows, Salix marshes, SaUx-Picea Krumm- holz. and Carex-Trijolium meadows. Most of the area is snow covered during winter, with snow depths varying considerably. While snow cover is normally in excess of 95%, portions of the area are exposed to prevailing westerly winds. Consequently, bushes of willow are rarely completely snow covered. Winter groups of ptarmigan were located hy concentrated searching on foot with periodic stops to scan surrounding areas for tracks, snow roosts, and or birds. Upon locating a flock, we tried to count all birds present. Some bias resulted when all birds present could not be counted. This bias was most serious d) when 30 or more birds 107 108 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 were dispersed over a large area sueh tliat all individuals eould not be seen, (2) when the birds were in snow roosts, and (3) when the flock flushed before all members could be counted. In such instances, only minimum estimates of flock size were obtained. F'ollowing Koskimies ( 1957), 2 or more grouse constituted a flock. After obtaining counts of flock size, birds were observed through 7 X 50 binoculars in order to ascertain l)anded to unbanded ratio and to identify marked individuals. Some unbanded ptarmigan and all birds with unreadable bands were pursued until caught or flushed and not relocated. Observations were recorded at the time on standardized cards and locations were subsequently plotted on 7.5 minute U.S. Geological Survey topographic maps. Ptarmigan were captured with a 5 or 7 m telescoping noose pole as described by Zwickel and Bendell (1967) for Blue Grouse ( Dendragapus obscurus) and modified by Braun and Rogers (1971) for White-tailed Ptarmigan. Captured birds were banded and classified to age and sex following methods presented hy Braun and Rogers (1971). Age categories used were adult ^including subadults fl + l), and juvenile Hess than 1 yr old). RESULTS AND DISCUSSION Arrival and departure. — Winter on the alpine was arbitrarily designated as the period Iteginning in late October when ptarmigan became entirely white through mid- to late Ai)iil when the prenuptial molt of males was initiated. In 1971-72 and 1973-74, most Itirds started arriving on the wintering area hy 20 to 26 October and remained until 19 to 30 April when they departed for breeding areas. Birds did not arrive or dejiart simultaneously; movements to and from winter use sites were gradual extending over a 2-week period. How- ever, climatic conditions had a pronounced influence upon arrival and (lei)arture. Prolonged mild weather during the fall of 1972 delayed arrival until 3 to 5 November while extended winter conditions the following spring (1973) delayed deiiarture until 3 to 10 May. Sex and age composition. — We identified 799 j)tarmigan at Guanella Pass from from 1966 through 1971. I here were significantly more ( P < 0.005 ) adults than juveniles identified, hut the luoiiortion of the total birds handled in each age class was not consistent among winters (Table 1 ). This variation was attributed to differences in nesting success among years. During the summer of 1972, 37 (82.2%) of 45 females encountered on breeding areas surrounding Guanella Pass were accomi)anied hy chicks, whereas in 1973 only 19 (54.2%) of 35 hens located were with broods (Hoffman 1974). Juveniles comprised 12.3 and 27.2% of the birds identified at Guanella Pass during the 1972-73 and 1973-74 winters, respectively, supporting the hypothesis that more juveniles (percent of total identified) occur on the wintering area following summers of good production than after summers of poor production. Partial segregation of sexes has l)een documented at Guanella Pass (Braun and Schmidt 1971 ) ; however, both sexes are not eciually represented on the wintering area. Significantly more (P< 0.001) females than males were identified each winter (Table 1), hut the proportion of males and females Hoffman and Braun • WINTEKINC; I’TAKMK;AN l()<; Table 1 Sex and Age Composition of White-tailed Ptarmigan Wintering at (Juanella Pass Winteri N2 % Adults % Juveniles % Females % Males 1967-68 66 54.5 45.5 72.7 27.3 1968-69 43 70.0 30.2 81.4 18.6 1969-70 111 78.4 21.6 78.4 21.6 1970-71 125 61.6 38.4 80.8 19.2 1971-72 142 56.3 43.7 78.9 21.1 1972-73 153 57.7 42.3 77.4 22.6 1973-74 125 72.8 27.2 84.0 16.0 Total Mean 765 64.5 35.5 79.0 21.0 1 The winter of 1966—67 was excluded due to small sample size (34 birds identified). ^ Total number of birds identified. occupying the area between winters was not significantly different ( P > 0.50 ) . Nearly 80% of the estimated 200 to 300 grouse annually wintering at Guanella Pass were females. Suitable wintering areas for female ptarmigan, such as Guanella Pass, appear to be limited in number; consequently, females from considerable distances are attracted to the few suitable sites ( Hoffman and Braun PJ75 j . Areas used by males during winter are scattered throughout the alpine region and each area need only support a small number of males breeding in the immediate vicinity. Winter studies in Colorado indicate that partial segrega- tion of sexes is not unique to the Guanella Pass area. This phenomenon is not only characteristic of White-tailed Ptarmigan as Weeden (1964) reported evidence of sexual segregation in both Rock (L. inutus) and Willow ptarmigan ( L. lagopus). Additional evidence of sexual segregation in Willow Ptarmigan was reported by Irving et al. ( 1967). Segregation of sexes of White-tailed Ptarmigan may only be spatial, but generally a difference in habitat preference occurs. Extensive stands of willow do not grow in some locations while in others, the willow may be completely snow covered in winter. Freiiuently this situation and/or poor snow conditions (lack of available roosting sites) forces both sexes to move below treeline along stream courses. Under these circumstances females were found to winter farther down the same drainage than males; a purely spatial separation. In segregation by habitat, males winter in Krummholz areas alternately dominated by clumps of willow and Engelmann spruce iPicea engelmannii) . Food availability in these sites is largely dependent upon wind action. Females winter at lower elevations near or at treeline where dense, no THE WILSON BULLETIN • VoL 89, No. 1, March 1977 tall stands of willow grow. Regardless of whether the separation was spatial or an actual difference in habitat preference, males showed a strong tendency to winter closer to breeding areas than females. Available information concerning segregation of sexes is primarily oriented towards the descriptive aspects of this behavior pattern with no explanation for its occurrence. Obviously there must be an adaptive advantage for sexes to live separately during the winter. Weeden (1964) suggested that another process, migration, must be investigated in detail before segregation of sexes in Rock and Willow ptarmigan could be understood. He also suggested that due to the more sedentary nature and apparently non-migratory behavior of White-tailed Ptarmigan, possibly the 2 processes could be studied indepen- dently of each other with this species. However, studies of White-tailed Ptarmigan in Colorado reveal they are migratory and that the processes are very closely related (Hoffman and Braun 1975). The fact that most females traverse long distances to seek out suitable winter habitat and return to the same area year after year is a strong manifestation of migration and that conditions are better for winter survival in these limited areas to offset increased mortality during migration. Principal advantages of wintering in these limited sites appear to be the presence, abundance, and availability of willow which comprises about 89% of the winter diet (May and Braun 1972) and secondarily proper snow corulitions for roosting. It is unclear what advantages are gained by males wintering in exposed sites adjacent to breeding areas and not undergoing long migrations typical of females. No marked differences occur in winter diets of the sexes to necessitate separation (May and Braun 1972). However, food resources are not suffi- cient at high elevations to sui)port large numbers of wintering birds. Thus it would appear advantageous for survival that a segment of the population migrate to areas where food is abundant. Lrider Colorado conditions, ptarmigan not only gained weight throughout the winter, indicating no food shortage (Braun and Schmidt 1971, May 1975), but also more efficiently used the winter resources that were available. Possibly it is advantageous for male ptarmigan to winter adjacent to areas where they breed in order to successfully compete for territories the following spring. Flocking behavior. — Like most grouse, ptarmigan are gregarious, associ- ating in flocks most of the year. Flocks are temporarily fragmented from late April until mid-July due to breeding activities. Flocking tendencies are most evident from late October to late Ai)iil when ptarmigan are con- centrated on winter use sites. From 1966 through 1974, 172 winter flocks and 13 lone ptarmigan totaling over 3178 birds were observed on the Guanella Pass wintering area. Winter flock size variefl from fewer than 5 to over 80 birds. Number and Uoiiman and Braun • WINTEKINC ETAHMICAN 111 Table 2 Yearly Trends in Flock Size of White-tailed I’tariviican at (Aianella Pass Yeari No. flocks observed No. birds involved Mean yearly flock si/.e Yearly range in flock si/.e 1969-70 29 342 11.8 2-50 1970-71 23 340 14.9 3 42 1971-72 29 432 14.9 2 40 1972-73 33 689 20.9 2-75 1973-74 21 316 15.0 2-80 1 Data from winters of (< 15 flocks observed pe 1966-67, r winter). 1967-68, and 1968-69 were excluded due to small sample sizes frequency of encounters of 172 winter flocks and 13 lone birds were as follows: (1) 13 (7.0%) lone birds, (2) 127 (68.7%) flocks with 2 to 25 individuals, (3) 35 (18.9%) flocks with 26 to 50 individuals, and (4) 10 (5.4%) flocks with more than 50 individuals. Flocks of males were small (< 15 birds) while females typically congregated in flocks of larger size (20 to 30 birds). Only infreciuently were ptarmigan observed singly, indicating their highly social nature during the winter period. Mean flock size was not significantly different (P >0.05) among winters for the 5 winters of 1969-74 (Table 2). The 135 flocks observed over the 5 winters averaged 15.5 birds. Similarly, mean flock size did not differ significantly (P>0.05) among the 7 winter months, as average flock size varied from 16.0 to 23.8 birds (Table 3). Flocks tended to be smaller in October and April being 16.0 and 16.2 birds per flock, respectively. Both months represent transition periods when birds are either arriving on (October) or departing from (April) winter use sites. Since birds do not Monthly Trends in Flock Table 3 Size of White-tailed PtAR MILAN AT CuANELLA P.ASS Months^ No. flocks observed No. birds involved Mean monthly flock size Monthly range in flock size Oct. 8 128 16.0 12-30 Nov. 27 644 23.8 3-60 Dec. 30 664 22.1 2-80 Jan. 30 510 17.0 2-70 Feb. 18 316 17.6 4-75 Mar. 28 491 17.5 2-75 Apr. 31 501 16.2 2 60 1 Pooled data for winters of 1966-67 through 1973-74. 112 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 arrive or depart simultaneously, fewer birds were present at these times to form flocks. Koskimies (1957) suggested increased population numbers of Capercaillie [ Tetrao urogallus) and Black Grouse {Lyrurus tetrix) and good nesting success led to larger flocks, and to a lesser degree, more flocks. Considering the failure to detect a significant difference in mean flock size among years, even though nesting success varied from year to year, there was no indication that nesting success influenced winter flock size of White-tailed Ptarmigan. However, 2.37 flocks were observed per day on the wintering area following the good production in 1972, while only 1.29 Hocks were located per day following the poor production in 1973. Assuming more birds are present on a wintering area after a summer of good production, they apparently form more flocks instead of gathering into larger flocks. Winter flocks were not composed of family units nor did the same indi- viduals associate together throughout the winter. Ptarmigan broods disperse in late September or early October (Braun 1969) prior to their arrival on wintering areas. No females nor any of their chicks banded together during the summer were relocated in the same winter flock. Considerable interchange of birds occurred between flocks with no noticeable aggressive behavior exhibited towards new flock members. Although exact numbers cannot be cited, many individual birds banded at Guanella Pass were relocated several times througbout the winter associating with flocks of various sizes comprised of different individuals at different locations. Ajluiity for wintering areas. — Data obtained from 90 birds initially banded at Guanella Pass and subseciuently relocated on breeding or summering sites were used to ascertain the affinity of individual birds for the wintering area, dhe sample included 62 females and 2o males of which 45 (50.0%) returned in succeeding winters. A bird needed only to return once to be included in ibe samjde, but some individuals returned uj) to 4 consecutive winters. Occasionally there was an intervening vs inter when a marked bird was not observed. Birds in this category were probably ])resent on the wintering area but not located. Fifteen of tbe 45 birds failing to return were known to be lost from tbe poj)ulation ( 12 hunting mortalities, 2 collections, and 1 trapping mortality) before having a chance to return. 3 o obtain a more accurate estimate of tbe percentage of birds returning that were available, the 15 birds having no opportunity to return were excluded. Conse(|uently, the sample consisted of 75 birds (22 adult females. 32 juvenile females, o adult males, and 13 juvenile males I with at least 45 ( 60.0%) returning. This estimate must be considered conservative since it was assumed that all birds not returning were either dead or wintering elsewhere: whereas, possibly some l)irds returned that were Hoffman and Braan • WFNTEKINC; ITAKMKIAN 113 not located. Thus, ptarmigan demonstrate a high fidelity lo wintering areas similar to their attachment to breeding sites (Schmidt 1969). Using a 2 X 2 contingency table, differences between the number of l)irds returning were tested for adults vs juveniles, males vs females, juvenile females vs adult females, and juvenile males vs juvenile females. Comparisons involving adult males were not possible because of small sample size ( expected frequencies less than 5; Simpson et al. 1960). Number and fre(}uency of each age and sex class returning were as follows: 16 (72.7%) adult females, 17 (53.1%) juvenile females, 6 (75.0%) adult males, and 7 (46.2%) juvenile males. No comparisons were statistically significant (P > 0.05) ; however, for the pooled sample of both sexes the comparison between adults vs juveniles closely approached significance (< 0.055). In all cases, proportionally more adults (73.3%) than juveniles (51.1%) returned. Juveniles suffer higher mortality than adults (Braun 1969) ; consequently, fewer are available to return. In addition, juveniles have no prior attachment to breeding sites and upon leaving the wintering area in spring they frequently travel long distances in search of vacant territories where they can become established as breeding birds (Hoffman 1974). Some of these birds probably winter closer to their territories and do not return. Of 40 birds identified on other nearby winter use sites, 4 (10%) were originally banded at Guanella Pass. All were banded as juveniles. Although the sample is small, it does indicate that ptarmigan banded as juveniles are more likely to disperse from wintering areas where they were initially banded and subsequently winter elsewhere. No differences were apparent in the proportion of adult males (75.0%) and females (72.7%) returning to Guanella Pass. Based on the return of marked birds to breeding territories, Braun (1969) reported a 25 and 30% annual turnover of adult males and females, respectively. Annual turnover of adult males and females estimated by Braun (1969) is similar to tbe mortality rates for birds wintering at Guanella Pass indicated by data collected in this study. Since adults show a high fiflelity to wintering sites, it can be reasonably assumed that most not returning are dead ; consequently, percentage of non- returning adults approximates annual turnover rates. A similar assumption cannot be made for juveniles because non-returning juveniles are not neces- sarily lost from the population. Also, percent of birds not returning cannot be considered an approximation of annual turnover for the entire population as loss of chicks prior to their arrival on the wintering area is not included. SUMMARY Studies of the characteristics of a wintering population of White-tailed Ptarmigan were conducted in alpine areas in north central Colorado, primarily at Guanella Pass. Of 799 Ptarmigan identified on the wintering area, 80% were females and 65% were adults. Numbers of juveniles identified varied with year depending upon production success. 114 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Climatic conditions had a pronounced influence upon the timing of arrival and departure to and from winter use sites. Partial segregation of sexes coincided with the arrival of birds on wintering areas with males usually remaining closer to breeding areas. Large concentrations of females wintered in areas at lower elevations near treeline where dense, tall stands of willow occurred. At times sex separation was only spatial, but usually habitat separation occurred. Winter flocks ranged in size from 2 to over 80 birds with about 69% of all flocks (172) encountered consisting of 2 to 25 members. Flock sizes did not change significantly among years or months. Following a summer of good production, ptarmigan formed more flocks instead of gathering into larger groups. Due to greater mobility and higher mortality, fewer juveniles than adults returned in subsequent winters, but all age and sex classes exhibited a high fidelity (60% return) to the wintering area. ACKNOWLEDGMENTS We acknowledge the support and direction given this study by R. A. Ryder, Department of Fishery and Wildlife Biology, Colorado State University, and H. D. Funk of the Colorado Division of Wildlife. T. A. May, Institute of Arctic and Alpine Research, University of Colorado, and A. E. Anderson, Colorado Division of Wildlife, accompanied us and helped gather data during several field periods. The National Science Foundation supported this study in part from 1966 through 1969 through a traineeship to the junior author while the Colorado Division of Wildlife supported it throughout its duration. This is a contribution from Colorado Federal Aid in Wildlife Restoration Project W-37-R. LITEKATUItE CITED Braun, C. E. 1969. Population dynamics, habitat, and movements of White-tailed Ptarmigan in Colorado. Ph.D. thesis, Colorado State Univ., h'ort Collins. AND C. E. Rogkhs. 1971. The White-tailed Ptarmigan in Colorado. Colorado Div. Came, Fish and Parks. Tech. I^uhl. 27. AND R. K. Schmidt, Jr. 1971. Effects of snow and wind on wintering populations of White-tailed Ptarmigan in Colorado. Pp. 238-250, in Proceedings Snow and Ice Symposium (A. O. Haugen, ed.). Iowa Cooperative Wildl. Res. Unit, Iowa State Univ., Ames. Choate, T. S. 1960. Observations on the reproductive activities of White-tailed Ptarmigan (La^opus leitcurus) in Clacier Park, Montana. M.A. thesis, Univ. Mon- tana, Missoula. . 1963. Ecology and population dynamics of White-tailed Ptarmigan iLagopus leucurus) in Clacier Park, Montana. Ph.D. thesis, Univ. Montana, Missoula. Haskins, A. C. 1969. Endoparasites of White-tailed Ptarmigan (Lagopus leucurus) from ('olorado. M.S. thesis, Colorado State Univ., Fort Collins. Hoffman, R. W. 1974. Characteristics and migration of wintering populations of Colorado White-tailed Ptarmigan. M.S. thesis, Colorado State Univ., Fort Collins. — AND C. E. Braun. 1975. Migration of a wintering population of White-tailed Ptarmigan in Colorado. .1. Wildl. Manage. 39:485-490. Irmng, L., C. C. West, L. .1. Peyton and S. Paneak. 1967. .Migration of Willow Ptarmigan in arctic Alaska. Arctic 20:77-85. Koskimies, .1. 1957. Flocking behavior in Capercaillie. Tetrao uroga/lus (L.). and Blackgame Lyrurus tetrix (L.). Finnish Papers Came Res. 18:1-32. Hoffman and Braun • WINTERING PTARMIGAN 115 May, T. a. 1975. Physiological ecology of White-tailed lharmigan in Golorado. Ph.l). thesis, Univ. of Colorado, Boulder. — AND C. E. Braun. 1972. Seasonal foods of adult White-tailed Ptarmigan in Colorado. J. Wildl. Manage. 36:118CG1186. Moss, R. 1973. The digestion and intake of winter foods hy wild ptarmigan in Alaska. Condor 75:293-300. . 1974. Winter diets, gut lengths, and interspecific coiupetition in Alaskan ptarmigan. Auk 91:737-746. Schmidt, R. K. 1969. Behavior of White-tailed Ptarmigan in Colorado. M.S. thesis, Colorado State Univ., Fort Collins. Simpson, G. G., A. Roe, and R. C. Lewontin. 1960. Quantitative zoology (Rev. Ed.l. Harcourt, Brace and Co., New York. Weeden, R. B. 1959. The ecology and distribution of ptarmigan in western North America. Ph.D. thesis, Univ. British Columbia, Vancouver. . 1964. Spatial separation of Rock and Willow ptarmigan in winter. Auk 81: 534^541. ZwiCKEL, F. C. AND J. F. Bendell. 1967. A snare for capturing Blue Grouse. J. Wildl. Manage. 31 :202-204. WILDLIFE RESEARCH CENTER, P.O. BOX 2287, FORT COLLINS, CO 80522. AC- CEPTED 17 FEB. 1976. BREEDING DENSITIES AND MIGRATION PERIODS OE COMMON SNIPE IN COLORADO Bruce R. Johnson and Ronald A. Ryder Common Snipe (Capella gollinago) occur seasonally throughout most of North America ( Bent 1927). Despite their widespread occurrence, little is known about their breeding status and timing of migration outside of northern and southern localities. Knowledge of snipe in Colorado is limited with published records pertaining to distribution and seasonal occurrences (Anon. 1886, Bailey and Niedrach 1965, Niedrach and Rockwell 1939). Because of the paucity of data on this common wetland species in the central portion of its range, this investigation was initiated in late 1973. Primary objectives were to estimate densities of breeding snipe and timing of migration in representative habitat types in Colorado. METHODS AND STUDY AREAS Three study sites were selected in each of 4 locations in Colorado. These locations were the Fort Collins (sites 1-3) and North Park (sites 4^6) areas in north central Colorado; Yanipa Valley (sites 7-9) in the northwest, and San Luis V^alley (sites 10-12) in south central Colorado. Total area studied was 245.1 ha. Size of sites studied in each location ranged from 25.2 ha near Fort Collins, 63.5 ha in the Yampa Valley, 63.6 ha in the San Luis Valley to 92.8 ha in North Park. Elevation extremes were from 1480 to 2510 m above sea level. Vegetation of each site was describt'd as to dominant species based on frequency of occurrence and percent coverage. AH sites were seasonally wet through either irrigation flow or proximity to permanent water sources such as streams or ponds. Seven sites were dominated by species of Carex. Other important species on these sites were Typha spp., Scirpiis americanus, Hordeiim juhatum. Trifolium spp., S. laciistris. Lemna spp., and Taraxacum spp. in decreasing frecjuency of occurrence. The other 5 sites varied in composition of dominant vegetation from Carex spp. and Salix s])p. to Eleocharis spp. and Juncus arcticus. Ten of the sites were seasonally grazi'd by livestock varying from late March to early July into October. One of the remaining sites was mowed for hay in late summer. Climatic conditions varied with location depending upon elevation and proximity to mountains. All locations are semi-arid (<^60 cm of moisture per year; range 18-59 cm) and relatively cool (mean annual temp. = 2-9°C) with short (56 days. North Park; 81 days, Yampa Valley) to average (103 days, San Luis Valley; 139 days, near Fort Collins) frost-free seasons (Colorado State Planning Division 1964). Soils on study sites in 3 of the 4 locations are similar (unpuhlished data from U.S.D.A. Soil (Conservation Service). They are deep, poorly drained, often alluvial in origin with excellent moisture holding capacity and slow to fair air, water, and root penetration. Surface layers ranged from clay loam to fine sandy loam often covered with a thin organic mat. Soils in the Yampa Valley were more variable ranging from deep, well- drained loam or clay loam t(» saline-alkali rich media with silty clay loam textures. 116 Johnson and Ryder • SNII’E POPIILATIONS IN COLORADO 117 Measurements of soil compaction on each site were made with a Soiltest penetrometer following standard procedures. Selected water characteristics were measured on each study site using a Hach Model AL-36-B test kit and ranged from 50 to 510 mgd fas CaCOs) for total alkalinity, 30 to 2100 mg/1 (as CaCOaj for hardness, and 7.2 to 9.5 for pH. Systematic strip censuses were used to estimate snipe numbers on each study site. Censuses consisted of walking linear transects at 20 to 25 m intervals with census routes and locations of snipe flushed, alighting, or displaying recorded on field maps of each site. Censuses were conducted at 2-week intervals from early May to mid-October in 1974 and from late March to late June in 1975 on all sites except near Fort Collins where censuses were conducted on a weekly basis. Nests encountered during field investigations were inconspicuously marked for later relocation to record nesting progress. Dates of onset of incubation were estimated using the water flotation technique (Westerskov 1950) based on a 19-day incubation period. Egg measurements were taken with vernier calipers graduated to the nearest 0.1 mm. RESULTS A^D DISCUSSION Breeding density. — The relation between numbers of snipe recorded during censuses and actual number breeding on a given area proved difficult to evaluate. Censuses prior to early May may have overestimated breeding densities due to the possible presence of migrants. Censuses after early May possibly underestimated breeding densities by excluding nesting snipe which were difficult to flush. In order to minimize these problems, the estimated breeding density for each site was derived by using the mean value for all censuses conducted on the site during May. Densities of snipe using study sites during May varied from 0.2 to 2.1 per ha (Table 1). Of the 4 locations studied, highest densities were recorded near Fort Collins. This appeared to be the result of smaller sites (average = 8.4 ha) and uniformly suitable habitat on these sites. Sites studied in the other locations were generally larger (average size = 24.4 ha), and contained more diverse habitats, some of which were not suitable for snipe. The slightly lower densities recorded in 1974 may be an artifact resulting from conducting more censuses later in May. Numbers of snipe varied with water depth and coverage, vegetation height and density, and soil conditions. Areas providing most suitable habitats for snipe contained shallow, stable, discontinuous water levels. Vegetation was low (10-30 cm in late May), often grazed or mowed, and sparse. Soils were moist to saturated and frequently characterized by hummocks. Areas with highest densities of snipe generally had ground compactions from less than 0.1 to 0.75 kg/cm^ and occasionally to 1.5 kg/cm“ in areas of dense vegeta- tion. Areas having ground compactions of 2.5 kg/cm“ and greater were slightly moist or dry and seldom provided feeding sites for snipe. Seven study sites were partially flooded pastures and had densities from 0.4 118 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 Estimated Snipe Breeding Densities Table 1 ON Selected Study Sites, Colorado, 1974-1975 Location Size (Ha) Snipe/ha ( range )i 1974 1975 Fort Collins 25.2 1.3 (0.8-1.7) 1.7 (1.2-2.1) North Park 92.8 0.6 ( 0.4-0.9 ) 0.6 (0.3-1.2) Yampa Valley 63.5 0.5 (0.4-1.1) 0.7 (0.5-1.2) San Luis Valley 63.6 0.5 (0.2-0.7) 0.5 (0.3-0.8) Total 245.1 Averages 0.6 0.7 1 Estimated densities are averages of the 3 sites studied per location. to 1.9 snipe/ha (average = 0.8). Of the 5 remaining sites, 2 in the Yampa Valley had relatively high densities (1.2 and 1.0 snipe/ha). These sites had stable water levels through May and were ungrazed. Vegetation heights on these 2 sites ranged from 5 to 30 cm in late May. Three study sites ( 1 in North Park, 2 in the San Luis Valley ) provided few suitable habitats for snipe and had low breeding densities (0.3 to 0.5 snipe/ha). None of the 3 sites were grazed prior to late June. Heights of vegetation in early June at these sites ranged from 20 to 70 cm. V^ater levels on 2 of these sites were relatively stable hut decreased on the third throughout May into June. Nesting. — During the course of field investigations, 28 snipe nests (18 in 1974, 10 in 1975) were located. Nest sites were typically in grasses or sedges 20 to 40 cm in height on moist hut unflooded ground near water. One atypical site was at the base of 3-m high willow iSali.x spp. ) in the center of a stand of willows approximately 30 m wide. Estimated onset of incubation ranged from 2 May through 4 July, with the hulk of nesting activity occurring in May. Incubation of 12 of the 18 nests located in 1974 (67% ) had l)egun in May. Additional nesting records for Colorado are limited, ranging from 1 May to 1 July with most nests located in May and early June (Niedrach and Rockwell 1939, Bailey and Niedrach 1965). In Lhah, nesting dates range from 29 April to 24 July ( Johnson 1899, Rent 1927). In California, a late nest was located on 1 September (Bryant 1915). Complete clutches of 4 eggs were found in 26 nests (93%) and 3 eggs in 2 nests (7%). Forty-nine eggs were measured and had a mean length of 38.5 mm (SI) = ± 1.3 mm, range - 36.6-41.6 mm) and width of 28.4 mm (SI) = ± 0.7 mm, range = 27.2-29.7 mm). Nesting success and chick sur- vival rates were not determined. Johnson and Ryder • SNIPE IMIPULATIONS IN (’OLOKADO Table 2 Maximum Densities of Snipe and Dates of Occurrence During Spring (1975) AND Fall (1974) Migration in Colorado Location Spring 1975 ( Snipe/ha ) Dates Fall 1974 ( Snipe/ha) Dates Fort Collins 5.6 10-13 April 2.9 17 September North Park 1.2 21-23 April 3.4 21 September Yampa Valley 1.1 22-23 April NC NC' San Luis Valley 1.0 14-16 April 0.3 15-16 September ^ Censuses were not conducted. Spring migration. — From data collected on the 12 study sites, the extent and peak of the 1975 spring migration appeared indistinct. During censuses in late March and early April, a few snipe, probably migrants, were observed on most study sites which were snow and ice-free. Previously published arrival dates for snipe in Colorado include 10 March for Denver, 19 March for Boulder, 26 March for Sweetwater Lake (Bent 1927 1 and 17 April for Rocky Mountain National Park (Packard 1945). By mid-April, highest numbers of snipe were recorded in all locations with Fort Collins having a peak of 5.6 snipe/ha on 10-13 April, North Park with 1.2 snipe/ha on 21-23 April, Yampa Valley with 1.1 snipe/ha on 22-23 April (omitting one site which was snow covered until late May), and the San Luis Valley with 1.0 snipe/ha on 14-16 April (Table 2). Numbers of snipe declined slightly through early May probably because of continued northward migration. A second decline occurred from early to late May which may have been caused in part, by a continuing northward migration. Although migrating snipe begin arriving in Canada during April and early May (Tuck 1972), breeding populations may not peak until late May ( Arnold 1976 ) . Local movements to areas such as irrigated haylands and pasture which had been dry previously and the onset of nesting and incubation making snipe more difficult to flush probably added to the apparent decrease. After late May, numbers of snipe stabilized on most study sites. Fall migration. — Numbers of snipe were monitored during September and October 1974 to document timing of the fall migration. Data collected indicated that the fall migration was in progress by early September. During fall censuses, highest numbers of snipe were observed in the San Luis Valley with 0.3 snipe/ha on 15-16 September, near Fort Collins with 2.9 snipe/ha on 17 September, and in North Park with 3.4 snipe/ha on 21 September (Table 2). Study sites in the Yampa Valley were omitted because of unfavorable habitat conditions. These data indicate that the peak fall migra- 120 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 tion period for 1974 occurred about the third week of September. Numbers of snipe declined markedly after late September and by mid-October the fall migration was near completion. In Colorado, the majority of fall migrants evidently passed through during September; however, substantial increases in numbers of snipe were observed in August. Censuses near Fort Collins and in the San Luis Valley during early August indicated increases on 4 study sites. Censuses in late July in North Park and the Yampa Valley did not reveal increases. Densities of snipe in the San Luis Valley reached 0.9 snipe/ha on 9-10 August, exceeding the September maximum of 0.3 snipe/ha. This increase may indicate that the fall migration was underway by early August or that juvenile snipe had concentrated on favorable feeding grounds. Tuck (1972 1 reported that in Newfoundland juveniles aggregated in groups of a few birds in late July to flocks of 100 or more in mid-August and were likely to migrate together prior to adults. SUMMARY Breeding densities and migration periods of Common Snipe in Colorado were investigated in 1974-75. Sites studied were near Fort Collins and in North Park, both in north central Colorado; in the Yampa Valley in northwestern Colorado; and in the San Luis Valley in south central Colorado. Estimated densities of breeding snipe based on censuses conducted during May 1974 and 1975 were, by region: 1.3-1. 7 snipe ha near Fort Collins; 0.6 snipe/ha in North Park; 0.5-0.7 snipe/ha in the Yampa Valley; and 0.5 snipe/ha in the San Luis Valley. Overall mean densities were 0.6 and 0.7 snipe ha in 1974 and 1975 respectively. On indi- vidual study sites, densities of snipe ranged from 0.2 to 2.1 snipe ha. Areas with shallow, stable, discontinuous water levels, sparse, short vegetation, and soft organic soils had the highest densities. Twenty-eight nests were located having a mean clutch size of 3.9 eggs. Estimated onset of incubation ranged from 2 May through 4 July. Most nests were initiated in May. Si>ring migration extended from late March through early May. Highest densities of snij)e were recorded in all regions during 10 23 April. Fall migration was underway by early September and was completed by mid-Octmczak and R. Walter. C. Br>ant and C. Donner and their staffs of the Monte Vista and Arapaho National W'ildlife refuges, respectively, are acknowledged for their many courtesies. C. E. Braun of the Colorado Division of Wildlife helped direct this study and critically reviewed the manu- script. His interest and attention to detail are sincerely appreciated. Johnson and Ryder • SNIPE POPULATIONS IN COLORADO I2I LITERATURE CITED Anonymous. 1886. Winter snipe in Colorado. Forest and Stream 26:5. Arnold, K. (Chairman). 1976. Snipe {Capella gallinago). In Management of migra- tory shore and upland game birds in North America (G. C. Sanderson, ed.) Int. As- soc. Game, Fish and Conserv. Comm. Washington, D. C. In Press. Bailey, A. M. and R. J. Niedracii. 1965. Birds of Colorado. Denver Mus. Nat. Hist. Vol. 1. Bent, A. C. 1927. Life histories of North American shorehirds. U.S. Natl. Mus. Bull. 142. Bryant, H. C. 1915. Two records of the nesting of the Wilson Snipe in California. Calif. Fish Game 1:76-77. Colorado State Planning Division. 1964. Colorado year book: 1962-1964. Denver. Johnson, H. C. 1899. Nesting of Wilson’s Snipe in Utah. Condor 1:26. Niedrach, R. J. and R. B. Rockwell. 1939. The birds of Denver and mountain parks. Colorado Mus. Nat. Hist. Denver. Pop. Ser. No. 5. Packard, F. M. 1945. The birds of Rocky Mountain National Park, Colorado. Auk 62:371-394. Tuck, L. M. 1972. The snipes: a study of the genus Capella. Can. Wildl. Serv. Monogr. Ser. No. 5. Westerskov, K. 1950. Methods of determining the age of game bird eggs. J. Wildl. Manage. 14:56-67. DEPT. OF FISHERY AND WILDLIFE BIOLOGY, COLORADO STATE UNIV., FORT COL- LINS, 80523. (PRESENT ADDRESS OF JOHNSON; UNIV. OF NORTH DAKOTA SCHOOL OF MEDICINE, GRAND FORKS 58201). ACCEPTED 30 JULY 1976. PRINCIPAL COMPONENT ANALYSIS OF WOODPECKER NESTING HABITAT Richard N. Conner and Curtis S. Adkisson Biologists have long been able to associate species of birds in a general way, with their characteristic habitats. Yet, for most species few such studies of a quantitative nature have been published. James (1971) used principal component and discriminant function analyses to ordinate breeding habitats of 46 species of breeding birds in Arkansas on vegetational continua. These kinds of analyses enable habitat relationships among a set of different species of birds to be detected and expressed more readily than do univariate tech- nicfues. They emphasize the detection of relationships among species rather than attempting to achieve the fine resolution possible in evaluating single species. We have applied principal component analysis to the nesting habitats of 5 species of woodpeckers: The Downy (Picoides pubescens]^ Hairy {P. villosus ) , Pileated ( Dryocopus pileatus I , and Red-headed ( Melanerpes erytliro- cephalus) woodpeckers and the Common Flicker {Colaptes auratus). Red- bellied Woodpeckers {Melanerpes carolinus) were not abundant in our study area and were not included in the analysis because of an insufficient number of nests. We selected a set of habitat variables that we felt were pertinent to these cavity nesting species. Woodpeckers are uniciue among the cavity nesters in that they can exercise a choice as to where they excavate. Most other cavity nesters use cavities where they find them. METHODS 'I'lu' study area (20 knr’) was located mainly on tin* upper (’raig and Poverty creek drainages, BlaeksOurg Ranger District, Jefferson National Forest in southwestern Virginia. A small |)art of the area was on the Virginia Polytechnic Institute and State University- farm and consisted of large mature woodlots. We searched intensively for active woodpecker nests during the springs of 1972, 1973, 1974 to locate as many nests as possible. Stand condition maps of the Ranger District were us(‘d to assure that all habitat types were searched. Vocalizations and drumming of wood- peckers were used initially to locate territories. Suhseciuent movement of the birds was observed to locate nest trees. We felt that the actual location of the nest would yield a more accurate representation of nesting habitat reejuirements than measurements of only the nesting territorv. At each active nest tree 8 variables were measured: ( macrohahitat ) basal area and density of stems greater than 7 cm DBH (diameter at breast height) within a 20-m radius of the nest tree, canopy height to crown top, distance from the nest tree to the nearest clearing, ( microhahitat ) DBH of the nest tree, diameter of the nest tree at the 122 Conner and Adkisson • WOODPECKER NESTINC IIARITAT 12;’) Table 1 CoFUiELATlON MATRIX (r) FOR 8 HaRITAT VARIABLES MEASURED AT Woodpecker Nest Trees BA* DOS CH DTC DNT PTA DAN DOS* 0.27 CH 0.72** 0.03 DTC 0.17 0.48** 0.08 DNT 0.41** -0.40** 0.57** 0.18 PTA 0.27 -0.31 0.52** -0.28 0.52** DAN 0.47** -0.11 0.37 -0.05 0.61** 0.17 NH 0.59** -0.24 0.70** 0.03 0.68** 0.51** 0.39** * Variable abbreviations are: BA = basal area, DOS = density of stems, CH = canopy heijiht, DTC = distance of nest tree to nearest clearing, DNT = DBH of nest tree, PTA = percent of nest tree alive, DAN — diameter of tree at nest cavity, NH - nest hei^bt. ** Significant at = 0.01. cavity, height of the nest, and a subjective estimate of percentage of live wood in each nest tree. A correlation matrix was calculated for the 8 habitat variables (Table 1). As would he expected, basal area was highly correlated with canopy height, and DBH of the nest tree was highly correlated with the diameter of the nest tree at the cavity and with height of the nest. Diameter of the nest tree was significantly correlated with almost everything and distance from the nest tree to the nearest clearing was correlated with almost nothing. Variation within and among these variables was analyzed using the principal component analysis available in Biomedical Computer Programs BMDOIM (Dixon 1974). ItESULTS We found 19 Pileated, 20 Downy, 13 Hairy, 11 Ked-headed, and 29 flicker nests. Over % of the flicker nests were found in trees left within clearcuts; most of the remaining; nests were found on the edges of the old mature woodlots. The 11 Red-headed Woodpecker nests were found in old mature woodlots on the University campus. Nests of the remaining species were more widely distributed. More than 86% of the cumulative total variance was accounted for hy the first 4 principal components (Table 2). The first component accounted for 44.9% of the total variance. Most habitat variables were positively correlated with the first component; density of stems and distance to a clearing were the exceptions. The highest correlations were with basal area, canopy height, DBH of the nest tree, and height of the nest. High values on the first component correspond to habitat with high basal area, tall canopy, large DBH nest trees, and nest cavities that are high above the ground. Thus the first component represents, with increasing values, a trend from clearcuts to old mature forests. 124 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 Table 2 Results of the Principle Components Analysis of 8 Nesting Habitat Variables for 5 Species of Woodpeckers Component I II III IV Percentage of total variance accounted for Cumulative percentage of total variance accounted for Correlations of components to original variables BA* DOS CH DTC DNT PTA DAN NH 44.9 22.8 11.0 8.1 44.9 67.7 78.7 86.8 0.71 0.53 0.01 0.25 -0.25 0.83 -0.08 0.35 0.84 0.28 -0.24 0.07 -0.11 0.77 -0.02 -0.58 0.85 -0.23 0.21 -0.15 0.66 -0.32 -0.50 0.11 0.64 0.06 0.70 0.11 0.86 0.05 -0.14 -0.26 * Variable abbreviations as in Table I. I he second component accounted for an additional 22.8% of the total variance (Table 2). This component was negatively correlated with DBH of the nest tree and percent of the tree that was alive, and positively correlated with the remaining 6 variables. Density of stems and distance to a clearing Mere the variables most correlated with the second component. High values on the second component correspond to a high density of stems and great distances from clearings. The second component emphasizes the relationships between dense forest (weighted on stems, hut not on maturity factors such as canopy height and basal area) and cleared areas. 1'he third component accounted for 11.0% of the total variance. The diameter of the nest tree at the nest cavity (positive correlation) was highly correlated with the third Component. The fourth component accounted for an additional 8.1% of the total variance liut no single factor made a prominent contribution. Habitat relationships among the 5 species of Moodpeckers can be observed when mean values for each species are plotted on the first 3 components ( Fig. 1 1 . As can be seen on the first component axis. Red-headed Wood- peckers preferred to nest in areas of high basal area and tall canopy and to nest relatively bigh above the ground in trees with great DBH and large diameter at the nest. The Downy \^'oodpecker preferred to nest in areas with Conner and Adkisson • WOODPECKER NESTIN(; II A HIT AT 125 P I Fig. 1. Three-dimensional ordination of nesting habitat relationships among 5 species of woodpeckers on the first 3 principal components. Contributions of variables to each component are summarized in text. The first component, left to right, represents a change from less mature forest to mature forest. The second component, front to hack, represents a change from open areas to dense forests. The third component, low to high, represents a change from small diameter nest cavities to large. Total variance explained by this ordination is 78.7%. (Dots indicate means, D — Downy, H — Hairy, F — Flicker, P — Pileated, and R — Red-headed.) lower basal area and lower canopy height than the other 4 species of wood- peckers. The Pileated and Hairy woodpeckers and Common Flicker nested in habitat intermediate to the Downy and Red-headed woodpeckers. On the second component the Pileated, Downy, and Hairy woodpeckers have high values, indicating a preference for nesting areas of high density of stems, while the Red-headed Woodpeckers and the Common Flicker preferred to nest near clearings in areas with a low density of stems (Fig. 1). On the third component, as the size of the woodpecker increased, so did the diameter of the tree at the place where the nest cavity was excavated ( Fig. 1) . Table 3 Matrix of Similarity Values IS) for Nesting Habitat Bp:tween Each Pair of Woodpecker Species* Flicker Downy Hair>' Pileated Downy .196 Hairy .269 .379 Pileated .192 .207 .257 Red-headed .189 .000 .026 .090 * Higher values represent greater ecological average Euclidian distance in hyperspace). similarity between species ( S = complements of 126 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 -3-2-1 O 1 2 3 Fig. 2. Two-dimensional ordination of nesting habitat overlap among 5 speeies of wood- peckers on the first 2 principal components. (Contribution of variables to each component is summarized in the text. The first component, left to right, represents a change from less mature to mature forest. The second component, low to high, represents a change from op(‘ii areas to dense for»*st. ( .S(>e Fig. 1 for synd)ol code.) A matrix of ecological similarity of nesting habitats for each woodpecker species was calculated using the method described by Power (1971) (Table B). Higher values in the matrix represent greater similarity among nesting habitats. Red-headed and Downy woodpeckers have the least similar nesting habitats of all the species. The Downy and the Hairy woodpecker had the most similar nesting habitat. d he nesting habitat of each species \\as plotted on the first 2 principal components and circled to obtain a visual estimation of overlap (Fig. 2). Extensive overlap between the Downy and Hairy woodpeckers is obvious. There is no overlap between the Downy or Hairy and Red-headed wood- peckers. The Pileated Woodpecker and the Common Flicker overlapped with all other species. The habitat area used by the Red-headed Woodpecker was Conner and Adkisson • WOODI’ECKER NEST1N(; HABITAT 127 much smaller than the areas used by the other species. Ihis may reflect the limited availability of Red-headed Woodpecker nesting habitat in southwestern Virginia rather than specific nesting habitat reciuirements. A rough index of the nesting versatility of each woodpecker species was calculated by summing the variances of each species on each component ( the vectors for each component were solved for each species and the variances of these families of values calculated for the respective species) over the first 3 components: Flicker = 5.24, Hairy = 4.14, Pileated = 3.21, Downy = 2.71, and Red-headed = 2.51. The Common Flicker was the most versatile species, by this index, reflecting its ability to nest in conditions varying from mature woodlots to clearcuts, provided that nearby access to open ground was avail- able for foraging. The Red-headed Woodpecker had the lowest versatility and was only found in mature woodlots that lacked a shrub layer and were near clearings. A short-coming of this technique is that one species might show a great range for one component but he very narrow for one or both of the other components. For example, the Hairy Woodpecker had relatively high variance values on all 3 of the components, while the flicker had high values only on the first and third components, indicating its low tolerance of uncleared areas. The Red-headed Woodpecker had a high variance on only the third compo- nent. The Downy and Pileated had high values on the first 2 components and average values on the third component. DISCUSSION We believe that the principal component analysis is a valuable tool in evaluating multivariate habitat relations for the 5 woodpecker species. Many of the results were in accord with what is known of the natural histories of these species. General descriptions of Red-headed Woodpecker nesting habitat are abundant. Our results, which indicate that this species prefers areas with high basal area, tall trees, a low density of stems, and an open understory, tend to agree with these previous habitat descriptions ( Bent 1939, Stewart and Robbins 1958, Bock et al. 1971, Reller 1972 ). The open understory and nearness to a clearing ( Fig. 1 ) is compatible with the foraging requirements of this species. Open areas above and on the ground are needed since Red- headed Woodpeckers flycatch and forage on the ground extensively in the summer (Bent 1939, Reller 1972). Past descriptions depict nesting habitat of the Common Flicker as being diverse (Burns 1900, Bent 1939, Stewart and Robbins 1958). Dennis (1969) thought flickers well adapted to any relatively treeless situation. Our study agrees with all of these observations. Downy Woodpecker habitat in Maryland was reported as wood margins. 128 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 open woodland, and forest edge habitat (Stewart and Robbins 1958). Although many of our nests were found in edge type habitats, many were also found in dense stands far from clearings ( Fig. 2). Lawrence (1966) thought that Hairy Woodpeckers could nest in any place where sufficient foraging habitat and a suitable nest tree were present. We found this species to nest over a wide range of basal areas, canopy heights, densities of stems, and distances from cleared areas. Several instances have been reported of Hairy Woodpeckers nesting and foraging in clearcuts ( Kilham 1968, Conner et al. 1975, Conner and Crawford 1974) . Hoyt (1957) described Pileated Woodpecker nesting habitat as heavy timber sometimes on mountain slopes, but mainly in moist lowlands such as valleys or bottomland. Kilham (1959) reported Pileated Woodpeckers nesting in swamps in Florida and Maryland. Pileated Woodpeckers in our study typically nested within 75 m of a small stream in stands of high basal area, tall canopy, and usually far from cleared areas. Several reports exist of Pileated Woodpeckers nesting in clearcuts and in forest edge habitat (Bent 1939, Conner et al. 1975). The large amount of overlap of nesting habitat among some of the wood- peckers in this study (Fig. 2) could he misinterpreted as an indication of competition. Past observations, however, suggest a lack of competition. Law- rence (1966) reported that Llairy Woodpeckers ignored both Common Flickers and Downy Woodpeckers that came near their nest territories. Kilham ( 1969 ) reported no agonistic encounters between nesting Hairy and Downy woodijeckers, yet the similarity value between these woodpeckers was the highest ( LahleS). Competition between species might occur only if a resource recjuired by both species is limited. In the past selection favoring a divergence in the size of sympatric i)oi)ulations of Downy and Hairy woodpeckers may have been a factor in reducing competition for nest sites, if any competition existed. Other factors, however, such as foraging techni(iue probably also influenced the evolution of size differences in woodpeckers. A species that fed super- ficially might not need the larger size and mass of species that fed by excavating through several inches of sound wood to reach arthropod chambers. It would he difficult to determine if woodpecker nest sites are at present a limited resource. Woodpeckers cannot nest in any tree in a forest, even if the surrounding hal)itat and diameter and height of the tree are optimum. Lhey recpiire nest trees with fungal heart rots to soften the core of the tree (Conner et al. 1975). A low density of suitably infected trees, especially in forests that are clearcut on a short term rotation, might limit the nest site resource. No data are available at present on the prevalence of heart rots in southwestern Virginia. Conner and Adkisson • WOODPECKER NESTINC; HABITAT 129 ACKNOWLEDGMENTS The principal component analysis was calculated using the Biomedical Computer Programs (BMDOlM) and the SPSS program provided by the Virginia Polytechnic Institute and State University Computer Center. We would also like to thank Frances C. James, Jerome A. Jackson, and James 1). Rising for reviewing the manuscript and making many excellent suggestions. LITERATURE CITED Bent, A. C. 1939. Life histories of North American woodpeckers. U.S. Natl. Mus. Bull. 174. Bock, C. E., H. H. Hadow, and P. Somer. 1971. Relations between Lewis’ and Red- headed woodpeckers in southeastern Colorado. Wilson Bull. 83:237-248. Burns, F. L. 1900. A monograph of the flicker. Wilson Bull. 7:1-82. Conner, R. N. and H. S. Crawford. 1974. Woodpecker foraging in Appalachian clearcuts. J. For. 72:564-566. Conner, R. N., R. G. Hooper, H. S. Crawford, and H. S. Mosby. 1975. Woodpecker nesting habitat in cut and uncut woodlands in Virginia. J. Wildl. Manage. 39: 144^150. Dennis, J. V. 1969. The Yellow-shafted Flicker {Colaptes auratus) on Nantucket Island, Massachusetts. Bird-Banding 40:290-308. Dixon, W. J., ed. 1974. BMD Biomedical Computer Programs. Univ. of Cal. Press, Berkeley. Hoyt, S. F. 1957. The ecology- of the Pileated Woodpecker. Ecology 38:246-256. James, F. C. 1971. Ordinations of habitat relationships among breeding birds. Wilson BuU. 83:215-236. Kilham, L. 1959. Behavior and methods of communication of Pileated Woodpeckers. Condor 61:377-387. . 1968. Reproductive behavior of Hairy Woodpeckers. II. Nesting and habitat. Wilson Bull. 80:286-305. . 1969. Reproductive behavior of Hairy Woodpeckers. HI. Agonistic behavior in relation to courtship and territory. Wilson Bull. 81:169-183. Lawrence, L. de K. 1966. A comparative life-history study of four species of wood- peckers. Ornithol. Monogr., No. 5. Power, D. M. 1971. Warbler ecology: diversity, similarity, and seasonal differences in habitat segregation. Ecology 52:434-443. Reller, a. W. 1972. Aspects of behavioral ecology of Red-headed and Red-bellied woodpeckers. Am. Midi. Nat. 88:270-290. Stewart, R. E. and C. S. Robbins. 1958. Birds of Maryland and the District of Columbia. North Am. Fauna 62. DEPT. OF BIOLOGY, VIRGINIA POLYTECHNIC INSTITUTE AND STATE UNIV., BLACKS- BURG 24061. ACCEPTED 11 NOV. 1975. PHENETIC ANALYSIS OF THE SUBFAMILY CARDINALINAE USING EXTERNAL AND SKELETAL CHARACTERS Jenna J. Hellack and Gary D. Schnell The subfamily Cardinalinae includes 37 species of cardinals, buntings, and grosbeaks, which have been divided into from 9 ( Paynter 1970) to 15 genera ( Hellmayr 1938 ). Previously using skeletal variables, Hellack (1976 I inves- tigated phenetic relationships of the subfamily with cluster analysis. In that study 3 species in the genus Saltator clustered differently from that suggested in previous classifications (Hellmayr 1938, Paynter 1970 j. The 3 Cardinalis species grouped together only in analyses using 14 skull characters, and all 31 species included in the study were very similar in relative measurements of the pelvic region. In this paper, we examine further the phenetic affinities of the subfamily by analyzing an additional set of external characters. MATERIALS AND METHODS We used 75 external morphologic characters in 10 analyses; in 2 of these Hellack’s (1976) 49 skeletal characters were included. Table 1 lists the species, the number as- signed to each, and common names (nomenclature follows Paynter 1970). Table 1 of Hellack (1976) indicates the number of skeletons measured. The 49 mea- surements are from all skeletal regions. Due to lack of skeletal materials, only 31 of the 37 species were compared. In the analyses of external morphologic characters, similar problems of obtaining ma- terial ■ occurred. The Appendix lists the 75 external morphologic characters, which can he separated into 3 categories: (1) 33 study skin measurements of the tail, wing, toes, and hill; (2) color measurements (dominant wave length) from 8 body regions; (3) contrast characters in which 33 comparisons were made between various regions of the bird (e.g. contrast between the nape and the crown; 0 = no contrast, 1 = contrast). All measurements were taken from adult specimens; the means for each species are in Appendix IV of Hellack (1975). Hellack measured study skin characters on 10 males and 10 females of each species if specimens were available. W hen more than one race was involved, measurements were taken from specimens (A the nominate race. Study skins were available for females of all 37 species, hut only 36 are included in the analysis of males ( the only known speci- men of Saltator cinctus is a female). Color was measured using the Munsell Book of Color ( Munsell 1973 ) which specifies a given color in terms of 3 characters — hue. value, and chroma. e converted these to dominant wave lengths, excitation purity, and % reflectance using tables supplied by the Munsell company (Anonymous 1970); these conversions are discussed by Newhall et al. (1943). Only the dominant wave length of each region was included in the analysis. Color measurements were obtained for males of 34 species and females of 33. Caryo- thraustes humeralis. Saltator cinctus. and 5. albicollis were not included in color analyses of the males. These species plus S. niaxillosiis were not included in color analyses of females. 130 Table 1 Species Included in the Study Hellack and Schnell • ANALYSIS OF CARDIN ALINAP: \?A ^2 CAl CO CO cA) m co o OJ an rB 3 CQ •X D3 c/i hS CQ ioC 3 I ^ I « C ’■§ 3 § "S 3 3 3 CQ ;r; CQ PQ M ^ i: o .-H "C a; -3 9 C 3 3 3 3 hJ > 3 cu 2 9 O P3 cj « c/i CO ’C .2 c ^ Oj o .o ^ 9 CO CO a a 9 a a C a a a 9b t/3 a Cf) CD ^ 2 "§ s 3s o 2 -“ 2< o x: 3 2 2 0^0 33 "S 33 aj 3 !U as to 33 "oo Tt O o u o CO OJ ^ i_ ij 'Z ■% J o J S ca S c ^ aH o -Q 3 S 3 -z: 3 a; •fh nj o O '-^ ^ ^ 2 V 9 o O bU c« g -Q I i i I " i t; 33 33 .3 33 t; o; V ;h tu J2 CL, >- >H CJ Pi CO CO 3 3 ™ 9933 CO CO CO 33 33 33 9 0) oj o; 3 3 9 ^ Sb ^9 ^ 3 § 3 3 CO _a § i ^ 33 CO u 33 "S 9 'i 9 .S 2 pppppPOOOHC a >0 ” a 05 ,£ -3 -O a a. a ^ Ce on c 03a r- 3 05 a rt ^I’g . ■3 c/5 ^ c .c; 05 a ^ a a _ 90^3 .2 3 c a >- i jn rt a --- o > w a ^ ? E c ^ ^ a a I- e e ® [o C C 1:1° s i a _i o o c/5 "5 u u rt a j, ^ 3: U 132 THE WILSON BULLETIN • Vol. 89, No. 1, March 1977 When all available characters were used (skeletal, study skin, and color) we had com- plete data for only 30 of the 37 species. Therefore the analyses of combined data include neither the 4 above mentioned species nor Periporphyrus erythromelas, S. rufiventris, and Passerina caerulescens. To assess phenetic similarity, we used multivariate statistical programs from the Nu- merical Taxonomy System (NT-SYS, developed by F. James Rohlf, John Kishpaugh, and David Kirk) . Both Q- and R-type studies were conducted. In the Q-type analysis, characters were standardized so that each had a mean of 0 and a standard deviation of 1. Then a product-moment corielation coefficient or an average distance coefficient was calculated for all pairs of species (Sneath and Sokal 1973). Species were clustered by the unweighted pair-group method using arithmetic averages (UPGMA, Sneath and Sokal 1973) and the results summarized in phenograms. We extracted 3 principal components from a matrix of character correlations in the R-type analysis (Sneath and Sokal 1973), and phenetic relationships are presented as 3-dimensional models of species projected onto these components (Rohlf 1968). A short- est minimally connected network (Rohlf 1970) computed from the original distance matrix is superimposed on the 3-D models to point out possible distortions. To eliminate or reduce the size factor, study skin characters were used as ratios (see Appendix), and skeletal measurements were divided by the first principal component extracted from a matrix of unstandardized skeletal characters. Skeletal data were handled this way because the method produced the “best” phenetic classification from the skeletal data (see Hellack 1976). Ten phenetic classifications were produced using the various combinations of the 4 data sets (study skin, contrast, color, and skeletal characters) and 2 similarity coefficients (correlation and distance). Males and females were analyzed separately to; (1) see if there were major differences among the resulting classifications, and (2) include all species in some analyses without having to compare species with complete data with those for which some information was lacking. Various data combinations were made so as to include all the characters available for any one species in an analysis. When all available data were used they were handled as follows: study skin characters of both males and females were averaged; for contrast and color characters male and female averages were inclu(h*d separately; and skeletal characters were averaged for a species without regard to sex (as done in Hellaek 1976). This resulted in 168 “characters” per species. Matrices were produced from the classification systems of Paynter (1970) and Hell- mayr (1938; see Hellack 1976). These 2 matrices, the 10 from the various combinations mentioned above, and 2 from the analyses of skeletal characters (SKEL/COMP I ALL CORR and SKEL/COMP 1 ALL DLST, Hellaek 1976) were compared by computing the coefficient of correlation between each pair of basic similarity matrices. Similarities were summarized as a dendrogram that indicates which basic similarity matrices are most alike; phenograms were compared in a similar manner. The following abbreviations are used. CORR or DIST refer to the use of correlation or distance to analyze similarity among species. SKIN denotes the use of study skin mea- surements and contrast characters. COLOR refers to the use of 8 color characters of dominant wave length. SKEL indicates the use of skeletal characters divided by un- slandardized ])rincipal component I (SKEL/COMP I ALL of Hellack 1976). BSM is the abbreviation for basic similarity matrix. Hellack and Schnell • ANALYSIS OF CAKDINALINAE 133 CORRELATION 0 50 0.75 I 00 ' I ' ' ' ' I SKIN CORR 66 * SKIN + COLOR CORR <5($ SKIN DIST 66 SKIN + COLOR DIST 66 SKIN CORR SKIN + COLOR CORR ?? SKIN DIST $$ ♦ SKIN + COLOR + SKEL CORR SKEL/COMP I CORR SKIN + COLOR DIST SKIN+COLOR + SKEL DIST* SKEL/COMP I DIST HELLMAYR { 1938) PAYNTER (1970) 0 25 rr CORRELATION 0.50 0.75 T 100 n SKIN CORR 66.792 SKIN DIST<5<5 761 SKIN + COLOR CORR <5(5.739 SKIN CORR ?$.758 SKIN + COLOR CORR $?738 SKIN DIST ?$.8I4 SKIN + COLOR + SKEL CORR .727 SKEL/COMP I CORR. 764 HELLMAYR (1938) PAYNTER (1970) SKIN + COLOR DIST <5(5.788 SKIN + COLOR DIST ?$.8I2 SKIN + COLOR + SKEL DIST8I9 SKEL/COMP I DIST. 855 Fig. 1. Dendrograms showing relationships among: (A) basic similarity matrices; (B) phenograms. Letters indicate groups of very similar BSMs. Asterisks indicate the phenogram chosen to represent each of these groups— the one with the highest cophenetic correlation. These representative phenograms are shown in Figs. 2a-b. RESULTS Phenograms. — In Fig. lA, which is a dendrogram of similarities among BSMs, 9 groups are labeled. The 4 BSMs of group A (in which only males are compared) differ in similarity coefficient and/or the number of charac- ters (the BSMs also differ in the number of species included, although the dendrogram. Fig. lA, is comparing placement of only those species each pair of analyses has in common). Group B has 3 BSMs (where only females were compared) which like those of group A, differ in similarity coefficient and/or the number of characters. The 2 BSMs of group E differ in character set but are alike in the similarity coefficient used. The 5 remaining groups contain 1 BSM each. The main difference between the dendrogram showing similarities among phenograms (Fig. IB) and Fig. lA is that 1 BSM of group A (SKIN + COLOR DIST $ $) clusters in group E. Also distance analyses of groups E and F show less similarity to the other clusters than they did in Fig. lA. BSMs within groups A, B, and E are very similar (Fig. lA). We have depicted only 1 from each — the phenogram with the highest cophenetic cor- 134 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 ■0 bO -0. p-' ' ' I CORRELATION aOp 0.25 0 50 TT 1 I I T I 0.7L r-n-r ,00 1 5PIZ. 31 PASS 32 PASS 34 PASS 2 PHEU 3 PHEU 5 PHEU 4 PHEU. 30 PASS 6 CARD 8 CARD. 7 CARD I I RHOD 17 SALT AMERICANA CYANEA AMOENA CIRIS CHRYSOPEPLUS AUREOVENTRIS MELANOCEPHALUS LUDOVICIANUS CAERULEA CARDINAL-1S SINUATUS PHOENICEUS CELAENO SIMILIS 21 SALT AURANTIIROSTRIS 13 PITY GROSSUS 18 SALT COERULESCENS 14 SALT ATRICEPS 16 SALT ATRIPENNIS 15 SALT MAXIMUS 20 SALT MAXILLOSUS 19 SALT ORENOCENSIS 23 SALT ATRICOLLIS 24 SALT RUFIVENTRIS 25 SALT ALBCOLLIS 9 CARY CANADENSIS 12 PERI ERYTHROMELAS 27 PASS CYANOIDES 28 PASS BRISSONII 37 PASS CAERULESCENS 26 PASS GLAUCOCAERULEA 29 PASS PARELLINA 35 PASS ROSITAE 36 PASS LECLANCHERII 33 PASS. VERSICOLOR 10 CARY HUMERALIS I SRZ AMERICANA 31 PASS CYANEA 33 PASS VERSICOLOR 34 PASS CIRIS 35 PASS ROSITAE 36 PASS LECLANCHERII 32 PASS AMOENA 9 CARY CANADENSIS 26 PASS GLAUCOCAERULEA 27 PASS CYANOIDES 28 PASS BRISSONII 29 PASS PARELLINA 2 PHEU CHRYSOPEPLUS 3 PHEU AUREOVENTRIS 4 PHEU LUDOVICIANUS 5 PHEU MELANOCEPHALUS 30 PASS CAERULEA 6 CARD CARDINALIS 8 CARD SINUATUS 7 CARD PHOENICEUS I I RHOD CELAENO 13 PITY GROSSUS I 4 SALT ATRICEPS I 5 SALT MAXIMUS i SALT COERULESCENS salt SIMILIS SALT ATRIPENNIS 23 salt ATRICOLLIS 19 salt. ORENOCENSIS Fig. 2 a. Plienogram representatives of groups A and C of Fig. 1: (A) study skin characters of males with correlations; (C) all characters and correlations. HeUack and Schnell • ANALYSIS OK CAHDIN ALINAE i:’,5 r-rn-r-1-p B SKIN DIST r=0.8l4 TT DISTANCE I 5 “P 0 5 T~l — pr r 00 n I SPIZ AMERICANA — 4 PHEU LUDOVICIANUS — 5 PHEU MELANOCEPHALUS — 30 PASS CAERULEA _ 31 PASS CYAN E A — 32 PASS AMOENA — 34 PASS CIRIS — 35 PASS ROSITAE — 2 PHEU CHRYSOPEPLUS — 3 PHEU AUREOVENTRIS — 9 CARY CANADENSIS _ 10 CARY HUMERALIS — 26 PASS GLAUCOCAERULEA — 28 PASS BRISSONII — 29 PASS PARELLINA — 33 PASS VERSICOLOR — 36 PASS LECLANCHERII — 27 PASS CYANOIDES _ 37 PASS CAERULESCENS — 6 CARD CARDINALIS — 7 CARD PHOENICEUS — 8 CARD SINUATUS — I 1 RHOD CELAENO — 17 SALT SIMILIS — 21 SALT AURANTIIROSTRIS — 13 PITY GROSSUS _ 15 SALT MAXIMUS _ 19 SALT ORENOCENSIS — 1 8 SALT COERULESCENS — 16 SALT ATRIPENNIS — 20 SALT MAXILLOSUS — 12PERI ERYTHROMELAS — 14 SALT ATRICEPS — 23 SALT ATRICOLLIS — 25 SALT AL0ICOLL1S — 24 SALT RURVENTRIS — 22 SALT CINCTUS D SKIN + COLOR + SKEL r = 0.819 27 PASS 14 SALT 23 SALT _ I 9 SALT. AMERICANA CHRYSOPEPLUS MELANOCEPHALUS CAERULEA AUREOVENTRIS LUDOVICIANUS AMOENA CARDINALIS SINUATUS PHOENICEUS CELAENO GROSSUS MAXIMUS COERULESCENS SIMILIS ATRIPENNIS AURANTIIROSTRIS CANADENSIS GLAUCOCAERULEA BRISSONII PARELLINA CYANEA VERSICOLOR CIRIS ROSITAE LECLANCHERII CYANOIDES ATRICEPS ATRICOLLIS ORENOCENSIS Fig. 2 b. Phenogram representatives of groups B and D of Fig. 1: (B) study skin characters of females with distances; (D) ail characters and distances. I 136 THE \^TLSON BULLETIN • Vol. 89, No. 1, March 1977 relation coefficient (see Fig. IB for these values). Any substantial differ- ence in placement of species in phenograms within each group will be de- scribed below. Group A consists of 4 very similar BSMs, and is represented by SKIN CORK S $ (Fig. 2A), which differs little from SKIN DIST S $ (not fig- ured). Adding 8 color characters (SKIN + COLOR CORR $ S , not figured) caused 2 species to cluster differently from that shown in Fig. 2A. Passerina versicolor grouped with P. ciris, and Periporphyrus erythromelas showed little similarity to any species cluster. SKIN + COLOR DIST $ $ (not fig- ured) is the most divergent, but major clusters are much the same. Adding color characters resulted in Passerina amoena not being in the cluster of buntings; and Rhodothraupis caelaeno, Periporphyrus erythromelas, Saltator orenocensis, and S. atriceps showing little similarity to the other species. The 3 BSMs of group B, all resulting from analyses of females only, are represented by SKIN DIST $ 9 (Fig. 2B). In the 2 phenograms not figured, 5. rujiventris clusters with the other saltators, and Caryothraustes humeralis and C. canadensis are not as closely affiliated as indicated in Fig. 2B. Group C includes only the analysis with all characters (SKIN + COLOR + SKEL CORR. Fig. 2C). It connects with the BSM of SKEL/COMP I ALL CORR (described in Hellack 1976, not figured here). The cluster bounded by Passerina glaucocaerulea and P. parellina (Fig. 2C) is not found in SKEL/ COMP I ALL CORR ( the members of the genus Passerina form 1 cluster with the exception of P. caerulea and P. cyanoides). Saltator orenocensis and Caryothraustes canadensis cluster with the genus Pheucticus in SKEL/C0j\1P I ALL CORR. Group E, containing 2 BSMs, is represented by SKIN + COLOR + SKEL DIST (Fig. 2D). SKIN + COLOR DIST 9 9 (not figured) differs in the placement of several species. The buntings {Passerina) cluster much the same as they do in SKIN + COLOR + SKEL CORR ( Fig. 2C), and not as in SKIN + COLOR + SKEL DIST (Fig. 2D). Saltator atripennis shows little similarity to any other species in SKIN + COLOR DIST 9 9. Group F contains the BSM for SKEL/COMP I DIST. Its phonogram (fig- j ured in Hellack 1976) was considered the “best” classification when only skeletal characters were analyzed (Hellack 1976) and differs from those pre- sented here mainly in the placement of the species in the genus Cardinalis (i.e., they do not cluster together). J Principal component analyses. — Four representative 3-D models from R-type I analyses are shown in Fig. 3. Character loadings for the first 3 principal I components of each are in appendices I, II, and HI of Hellack (1975). ! Fig. 3A is the analysis of males using study skin and contrast characters. j The principal components explain 21.2, 11.7, and 9.0% of the total character j Hellack and Scime/l • ANALYSIS OF CARDIN ALIN AE 137 n I SKIN 6S I Fig. 3 a. Representative models of species projected onto the first 3 principal com- ponents based on (A) male study skin characters and (B) female study skin characters. Species names corresponding to the numbers on the models are in Table 1. Components 1 and II are labelled, III is the height. The shortest minimally connected network is pro- jected onto each of the models. 138 THE WILSON BULLETIN • VoL 89, No. 1, March 1977 c 4 Fig. 3 1). Representative models of species projected onto the first 3 principal com- ponents based on (C) all available characters and y studying the phenograms and 3-D models in which these species had been included. This phenogram was then used to look at similarities and compare these similarities with the classifications of Hellmayr (1938) and Paynter (1970). Based on phenetic groupings, several saltators (S. ruHventris, S. albicollis, S. cinctus, S. atricollis, S. auranti- irostris, and S. orenocensis) were found to have little similarity to the remaining saltators. In the case of S. rufiventris, S. albicollis, and S. cinctus, insufficient data may be the reason for their lack of similarity to the saltator cluster. However, 5. atricollis, S. oreno- censis, and S. aurantiirostris are clearly distinct. The genus Pheucticus clusters much as one would expect from Hellmayr’s (1938) classification. The species placed in the genus Passerina by Paynter (1970) could be grouped according to either former classification. ACKNOWLEDGMENTS We are grateful to the curators of a number of collections who allowed us to use material in their care (they are listed in Hellack 1976). We thank Harley P. Brown, Charles C. Carpenter, and James R. Estes for critically reviewing the manuscript, and Elizabeth A. Bergey who assisted with the measurement of specimens. Ginna Davidson and Sharon Swift aided in the preparation of figures. This study was supported in part by a travel grant from the Smithsonian Institution and a Sigma Xi Grant-in-Aid. LITEKATUItE CITED Anonymous. 1970. Dominant wavelength and excitation purity for designated Munsell color notations. Munsell Color Company, Inc., Newburgh, N.Y. Blake, C. H. 1969. Notes on the Indigo Bunting. Bird Banding 40:133-139. i)E ScMAUENSEE, R. M. 1970. A guide to the birds of South America. Livingston Publ. Co., Narbert, Pa. Hellack, J. J. 1975. A phenetic analysis of the subfamily Cardinalinae, Aves, Ph.D. thesis. IJniv. of Oklahoma, Norman. — . 1976, Phenetic variation in the avian subfamily Cardinalinae. Occas. Pap. Mus. Nat. Hist., Univ. Kans. 57:1-22. Hellmayr, C. E. 1938. Catalogue of birds of the Americas. Field Mus. Nat. Hist. Publ. Zool. Ser. 13, pt. 11. Mayr, E. and L. L. Short. 1970. Species taxa of North American Birds. Publ. of the Nuttall Ornithol. Club No. 9. Cambridge, Mass. Munsell, A. H. 1973. Munsell book of color. Macbeth Color and Photometry Division, Kollmorgen Corporation, Newburgh, N.Y. Newiiall, S. M., 1). Nickerson, and 1). B. Judd. 1943. Final report of the O. S. A. sub- committee on the spacing of the Munsell colors. J. Opt. Soc. Am. 33:385-418. Paynter, R. A. 1970. Subfamily Cardinalinae. Pp. 216-245 in Check-list of birds of the World. Vol. 13 (R. A. Paynter and R. W. Storer, eds.), Mus. Comp. Zook, Heffernan Press, Worcester, Mass. Peterson, R. T. and E. L. Ciialif. 1973. A field guide to Mexican birds. Houghton Mifflin Co., Boston. Hellack and Schnell • ANALYSIS OF CARDINALINAE 147 Phillips, A. R., J. Marshall, and G. Monson. 1964. The birds of Arizona. Univ. of Arizona Press, Tucson. Ridgway, R. 1901. The birds of North and Middle America. U.S. Natl. Mus. Bull., 50, pt. 1. Robins, J. D. and G. D. Schnell. 1971. Skeletal analysis of the Ammodramus-Ammo- spiza grassland sparrow complex: a numerical taxonomic study. Auk 88:567-590. Rohlf, F. j. 1968. Stereograms in numerical taxonomy. Syst. Zool. 17:246-255. . 1970. Adaptive hierarchical clustering schemes. Syst. Zool. 19:58-82. Schnell, G. D. 1970. A phenetic study of the suborder Lari (Aves) II. Phenograms, discussion, and conclusions. Syst. Zool. 19:264^302. Sneath, P. H. a. and R. R. Sokal. 1973. Numerical taxonomy. W. H. Freeman and Co., San Francisco. Sokal, R. R. and C. D. Michener. 1967. The effects of different numerical techniques on the phenetic classification of bees of the Hoplitis complex (Megachilidae) . Proc. Linn. Soc. London 178:59-74. DEPT. OF ZOOLOGY AND STOVALL MUSEUM, UNIV. OF OKLAHOMA, NORMAN 73019. ACCEPTED 15 DEC. 1975. Appendix DESCRIPTION OF STUDY SKIN, CONTRAST, AND COLOR CHARACTERS Study skin. — (1) Rectrix length, distance from where skin joins shaft of middle pair of rectrices to tip of longest rectrix. Five characters represent shape of tail and are divided by rectrix length to reduce size factor; measurement is coded as negative until longest feather is measured then positive from longest feather. Characters are as follows: (2) distance from tip of outer rectrix to tip of 2nd, (3) distance from tip of 2nd rectrix to tip of 3rd, (4) distance from tip of the 3rd to tip of 4th rectrix, (5) distance from tip of 4th rectrix to tip of 5th, (6) distance from tip of 5th rectrix to tip of 6th. Two measures of feather widths (from center of feather), each divided by rectrix length to reduce size factor. Characters are: (7) outer rectrix width, and (8) outer vane of outer rectrix. The relative amount of tail covered hy coverts was measured by the following 2 characters (divided by rectrix length) : (9) distance from tip of under-tail coverts to tip of longest rectrix, (10) distance from tip of the upper-tail coverts to tip of longest rectrix. Wing length 111), distance from carpal joint (bend of wing to tip of longest primary). Five characters represent shape of wing and are divided by wing length to reduce size factor, coded as negative numbers until longest feather is measured then a positive num- ber. Characters are: (12) distance from tip of 9th primary to tip of 8th, (13) distance from tip of 8th to tip of 7th, (14) distance from tip of 7th primary to tip of 6th, (15) distance from tip of 6th primary to tip of 5th, (16) distance from tip of 5th primary to tip of 4th. Ten characters represent the widths of wing feathers and are divided by wing length in order to reduce size factor (all measurements were taken at the center of the feather). Characters are: (17) width of the 9th primary, (18) width of outer vane of 9th primary, (19) width of 8th primary, (20) width of outer vane of 8th primary, (21) width of 7th primary, (22) width of outer vane of 7th primary, (23) width of 6th primary, (24) 148 THE WILSON BULLETIN • Vol 89, No. 1, March 1977 width of outer vane of 6th primary, (25) width of 1st secondary, (26) width of outer vane of 1st secondary, (27) distance from the tip of longest secondary to tip of longest primary; measurement divided by wing length. (28) Hallux length, measured without claw. Three toe lengths divided by hallux length to reduce size factor are: (29) length of middle toe, (30) length of 2nd toe, (31) length of 4th toe. Two angles were recorded from bill: (32) angle of commissural point relative to tomia, and (33) an angle measurement of arc of mandibular ramus. Contrast characters. — Thirty-three 2-state characters were used. They were recorded as either present or absent characters, or contrast or no contrast characters. They are: (34) white spots in tail, (35) under-tail coverts contrasting to belly, (36) white present at apex of primaries, (37) white at base of primaries, (38) white on primary coverts, (39) white on secondary coverts, (40) marginal coverts contrasting to other coverts, (41) malar region contrasting to auricular, (42) lore region contrasting to forehead, (43) forehead contrasting to crown, (44) occiput contrasting to nape, (45) occiput contrasting to crown, (46) nape contrasting to hack, (47) chin contrasting to gular, (48) gular contrasting to jugulum, (49) eye ring, (50) breast streaking, (51) hack streaking, (52) side of body streaked, (53) flanks streaked, (54) abdomen contrasting to breast, (55) rump con- trasting to hack, (56) presence of a crest, (57) color sexual dimorphism, (58) middle wing coverts contrasting to other coverts, (59) superciliary line contrasting to crown, (60) auricular white, (61) white spot at base of lower mandible, (62) stripes on throat, (63) upper-tail coverts contrasting to rump, (64) streaking on crown, (65) flanks contrasting to abdomen, (66) sides contrasting to breast. Color.- — Color characters of the bird were recorded using the dominant wave length as the measurement of color. Color measurements were taken from 8 regions of the bird: (1) crown, (2) back, (3) rump, (4) upper-tail coverts, (5) gular, jugulum region, (6) breast, (7) abdomen, (8) crissum. GENERAL NOTES Why Ospreys hover. — Ospreys (Pandion haliaetus) fish from a perch or while in flight. Birds hunting from the air interrupt gliding or flapping progression with inter- mittent 2-10-sec hovering bouts in which they hold themselves stationary in the air column. They dive on fish directly from a gliding “interhover,” but more often from a hover (pers. obs. in Florida, Maine, New Jersey, and New Brunswick, Canada). Here I examine the adaptiveness of hovering. In March of 1974 and 1975, I found the fishing success of Ospreys working Lake George, Lake Co., Florida to be dependent on weather. When the sun was clouded over or the lake surface rippled by wind, the birds’ capture rate and dive rate were both significantly depressed. I attribute these effects to reduced visibility into the water (Grubb, Auk 94:146-149, 1977). To pursue the adaptiveness of hovering, I noted the success of each dive from a hover or interhover (Fig. 1). Although I have no information on energetic costs of gliding and hovering in Ospreys, I presume the latter to be generally more expensive. An exception occurs in strong winds when an Osprey can “glide” with zero ground speed. As it is more costly, for hovering to be adaptive it should result in a large increase in capture rate. Details on methods are in Grubb (op. cit.). Briefly, I watched Ospreys seeking fish in a 0.2 km^ rectangular area off Lake George’s western shoreline. Numerous parameters of foraging behavior and concomitant weather were recorded. WEATHER OF ALL I SUN SHINING I I SUN OCCLUDED 1 CONDITIONS DIVES I WIND VELOCITY 0.1 - 8,0 M/SEC 1 Fig. 1. Success rates of Ospreys diving on fish from a hover or from an interhover under various weather regimes, and advantage of hovering. Hovering advantage is not calculated for sample sizes less than 20. Smooth = calm water surface; L. Rip. =: lightly rippled water surface; H. Rip. heavily rippled water surface. 149 150 THE WILSON BULLETIN • VoL 89, No. 1 The success rate of dives from hovers and interhovers under various weather regimes is shown in Fig. 1. Under all weather conditions combined, dives from hovers were 50% more successful than dives from interhovers, a significant difference (p < .05; x“ = 5.90, df = 1). A complete picture of the adaptiveness of hovering should account for the energetic cost of hovering vis-d-vis gliding flight. Unfortunately, I lack the information necessary for such a comparison. The size distributions of fish appeared similar under all weather regimes, whether caught from a hover or a glide. Thus, Ospreys which never dove from a hover would have to save approximately 50% of the energy expended by those diving from hovers to compensate for the hoverers’ increased energy intake. I thank G. H. Grubb and particularly W. M. Shields for field assistance, and K. Bildstein and W. M. Shields for commenting on an earlier draft.— Thomas C. Grubb, Jr., Dept, of Zoology, Ohio State Univ., Columbus 43210. Accepted 13 Nov. 1975. Storage of pinon nuts by the Acorn Woodpecker in New Mexico. — The food habits of the Acorn Woodpecker ( Melanerpes formicivorus) have attracted considerable attention (MacRoberts, Condor 72:196-204, 1970, and references therein). The most distinctive aspect of this behavior is the species’ extensive dependence on stored mast. The nuts, generally acorns, are harvested by groups of birds in the fall, and placed in holes that are especially excavated for this purpose in dead trees, dead limbs of live trees, power poles, fence posts, etc. The stored mast is then communally used and defended from competitors by the groups during the rest of the year. However, most of the infor- mation about this and other aspects of the food habits of the Acorn Woodpecker has been obtained from studies of California populations. Relatively little is known about its l)ehavior elsewhere. As part of a study of the behavior and ecology of this species in the American South- west, we periodically observed groups of Acorn Woodpeckers from December 1974 to August 1975 in Water Canyon, New Mexico. This canyon, located in the Magdalena Mountains near Socorro, contains riparian vegetation. Gambel’s oak {Quercus gambelii) and the gray oak ( griseo) are present, and pihon-juniper i Pinus edulis and Juniperus spp.) forests are found along the sides of the canyon. When our observations began, 7 out of the 10 groups of Acorn Woodpeckers which were studied held relatively large stores. Much of this mast consisted of acorns. However, in addition to acorns, we found that many pinon nuts also had been collected. When we examined a section of the storage tree of one of the groups in January, 84 out of 128 holes counted were found to contain pinon nuts, while 17 held acorns and 21 were empty. A recently fallen limb from the storage tree of another group had the remains of pinon nuts in 11 out of 33 holes, while the others were empty. In addition, during the winter and spring, woodpeckers often foraged among the pinon pines along the sides of the canyon, and birds consumed pinon nuts both on the storage trees and at “anvils” located among the pines. While the storage of acorns by southwestern populations of Acorn Woodpeckers has been reported by a number of authors ( cf. Bent, U.S. Natl. Mus. Bull. 174, 1939), we know of no published reference to the use of pinon nuts by this species. These observations suggest that pinon nuts, when available, form an important part of the diet of Acorn Woodpeckers in Water Canyon. California populations of the species have been observed to store other types of nuts l)esides acorns, including almonds, pecans, and walnuts obtained from orchards. However, according to Ritter ( The California Wood- pecker and I, Univ. of Calif. Press, Berkeley. 1938), the use of these nuts generally occurs March 1977 • (;ENEKAL NOTES 151 when acorns are in short supply. That this was prohahly not the case in Water Canyon is suggested by the fact that wliile pihon nuts were plentiful, the acorn crop also was good. The majority of the groups did not deplete their mast stores over the winter, and many acorns and pihon nuts remained in the storage trees through the summer. Bock and Bock lAm. Nat. 108:694-698, 1974) recently proposed that the distribution and abundance of the Acorn Woodpecker are affected not only by the abundance of oaks within a habitat, but also by the oak species diversity present. Because there may be an occasional failure of the production of acorns in each oak species, fewer species in an area would increase the probability of a total acorn crop failure. The use of pihon nuts by the woodpeckers in Water Canyon would be significant since there are only 2 oak species present. If the storage and consumption of pihon nuts which we observed is common, it would suggest that the diet of the species in this area has been expanded to regularly include an additional resource. This in turn would increase both resource abundance and diversity, and if the Bocks’ hypothesis is correct, would allow the popula- tion to both reach and maintain a higher size than would be possible with acorn storage alone. We thank Carl Bock for his many helpful suggestions, and also Melisse Reichman, who assisted with the field work. L. Kilham and P. L. Stallcup gave valuable comments on the manuscript. Financial assistance was provided by the Chapman Fund of the American Museum of Natural History and the Kathy Lichty Fund of the University of Colorado. — Peter B. Stacey and Roxana Jansma, Dept, of Environmental, Population, and Organis- mic Biology, Univ. of Colorado, Boulder 80302. Accepted 8 Dec. 1975. Flocking and foraging in the Scarlet-ruinpcd Tanager. — Efficiency in foraging may be an important factor in the evolution of bird flocks (Cody, Theor. Pop. Biol. 2: 142-158, 1971). In order to test this suggestion, it is desirable to have field data showing an association between flocking and foraging behaviors. This note reports data for these behaviors taken on the Scarlet-rumped Tanager ( Ramphocelus passerinii) in Costa Rica. Investigations were carried out during August, 1973 at the Tropical Science Center research station near Rincon on the Osa Peninsula in southwestern Costa Rica. Areas of forest edge along roadsides and river banks were searched for Scarlet-rumped Tanagers. The bushes and dense vegetation of these areas are the favored habitat of this tanager, and the close proximity of the forest provided opportunities to occasionally observe Scarlet-rumped Tanagers flocking with species of the forest interior. All observations were made between 05:30 and 11:30. Data were taken only on adult males as their striking plumage made them easier to follow than females. It is important to differentiate between flocks and aggregations. A flock was defined as a multi-individual group of birds moving in an integrated fashion, i.e. birds moving together as a unit from place to place. An aggregation was a multi-individual group with individuals in close proximity to one another, but which did not move in an integrated manner. For each flock, data on 4 variables were taken: (1) group size— the total number of individuals in the flock. Groups with at least 2 species present were desig- nated mixed-species flocks, groups with only Scarlet-rumped Tanagers were designated single-species flocks; (2) foraging rate — the number of feeding attempts in 15 sec intervals were counted. A feeding attempt was defined as a peck at fruit or insects. Use of an electronic timer and tape recorder allowed continuous observations. It was not possible to obtain an indication of success in these attempts; (3) foraging height — the height of the bird from the ground was estimated in categories of 5 m; (4) group move- 152 THE WILSON BULLETIN • VoL 89, No. 1 Table 1 Correlation CoEPnciENTs for Pairs of All Variables for All Flocks (A) AND Single-species Flocks (S) Group Size Foraging Rate Foraging Height Foraging A 0.91* Rate S 0.55* Foraging A 0.78* 0.76* Height S -0.09 -0.05 Group A 0.86* 0.90* 0.77* Movement S -0.03 -0.05 -0.01 * Indicates significance at the p = 0.05 level. merit — the net rate of movement in one direction, i.e. “doubling-back” results in zero net movement. When possible, all data were taken on the first male seen in each flock. An effort was made to see as many flocks as possible. During the course of the study, 34 flocks were encountered; 29 of these flocks were single-species flocks. Species seen in at least 2 of the 5 mixed-species flocks with Scarlet- rumped Tanagers included the Short-billed Pigeon (Columba nigrirostris) , Red-capped Manakin (Pipra mentalis) , Black-crowned Tityra iTityra inquisitor) , Masked Tityra (T. semifasciata) , Green Honeycreeper {Chlorophanes spiza) , Masked Tanager {Tangara larvata) , White-shouldered Tanager (Tachyphonus luctosus) , and Variable Seedeater (Sporophila aurita) . Relationships between all possible pairs of the 4 variables were examined by correlation analysis. Two sets of correlations were calculated: one set for all data and one set for single-species flocks. When all flocks are considered, there are significant positive correlations between all possible pairs of the 4 variables (Table 1). This suggests that Scarlet-rumped Tanager males alter their behavior as flocks become larger, showing increased foraging rate, increased group movement, and an increase in the height in the canopy at which the activities are performed. Foraging would seem therefore, to be a functional correlate of flocking. In single-species flocks foraging rate is still positively correlated with group Table 2 A Comparison of Scarlet-humped Tanagers in Mixed and Single-species Flocks ON THE Basis of Group Size and 3 Behavioral Parameters* Flock t>ije Observation time sec Group size (S.D.) Foraging rate pecks /min (S.D.) Foraging height m (S.D.) Group movement m/min (S.D.) Mixed-species 855 21(7.8) 6(2.8) 21(7.4) 8(5.6) Single-species 4770 4(2.3) 1(0.7) 7(3.9) KLO) All 5625 7(6.9) 1(2.2) 9(6.7) 2(3.5) * Values are means with 1 standard deviation in parentheses, except for observation time which is total time. March 1977 • (;ENEKAL NOTES J5:’> size (Table 1), but no other pair of variables is significantly correlated. Social factors may be important in single-species flocks, and family members may stay together even when not foraging. The preceding analysis indicates that it is the combination of the data on mixed-species flocks and those on single-species which produce most of the significant correlations for the data on all flocks. Mixed-species flocks may be functionally distinct from single- species flocks, a point also suggested by examining the mean values for all variables (Table 2). All means are greater for tanagers in mixed-species flocks than for those in single-species flocks (Mann-Whitney U-test, p 0.01), indicating that tanagers move at a faster rate and forage faster and higher in the mixed-species flocks. Scarlet-rumped Tanagers may gain any number of several advantages from joining large mixed-species flocks. Some possible advantages are: flocks may help the tanagers locate fruiting trees; flocks may “flush” insects; competition may be reduced by monitoring other species with similar food habits; there may be increased protection from predators in a flock which allows the tanagers to increase foraging. These and other ideas pertain- ing to flocking as an adaptation are discussed along with the pertinent literature else- where (Moriarty, Biologist 58: in press, 1976). Some or all of these benefits of flocking may also apply to single-species flocks, but if family relationships are an important aspect of single-species flocks, then it may not be surprising to find the foraging-related aspects of flocking occur more regularly and intensely in mixed-species flocks. J. R. Karr, D. W. Schemske, C. E. Schnell, and M. F. Willson kindly reviewed the manuscript. Financial support was from the National Science Foundation through the Organization for Tropical Studies.^ — David J. Moriarty, Dept, of Ecology, Ethology and Evolution, Vivarium Building, Univ. of Illinois, Champaign 61820. Accepted 15 Dec. 1975. Yellow Warbler nest used by a Least Flycateber. — While checking Yellow Warbler {Dendroica petechia) nests near the Delta Marsh, Manitoba, I observed a Least Flycatcher (Empidonax minimus) making use of a deserted Yellow Warbler nest. The nest, 94 cm above the ground, was placed next to the trunk of a small maple {Acer negundo) . On 5 June 1975 it contained 5 Yellow Warbler eggs but by 8 June only 1 egg was present and the nest’s interior had been disturbed. An active Yellow Warbler nest was located several meters from the deserted nest; the first egg in this nest was laid about 12 June. On 14 June the first nest contained the single Yellow Warbler egg and 2 Least Flycatcher eggs. The nest was shallower now and there was no evidence that the flycatcher had added ma- terial to it. The flycatcher clutch was completed by the following day with the addition of a third egg; the Yellow Warbler egg was gone. One Least Flycatcher egg disappeared 6 days later but by 28 June, 2 nestlings were present. An empty nest on 29 June suggested predation had occurred. Interest in old nests by the Least Flycatcher during the period of nest site selection has been noted by Mumford (unidentified nest, Wilson Bull. 74:98-99, 1962) and de Kiriline (Rose-breasted Grosbeak’s nest: Audubon Mag. 50:149-153, 1948). No occupation oc- curred in either case. Use of nest material from a previous year’s Yellow Warbler nest by a Least Flycatcher (pers. observ.) indicates that old nests may be a source of nest material. It is possible that the Least Flycatcher was physiologically ready to lay, but since its own nest had been destroyed, it took over the available Yellow Warbler nest. I noted Least Flycatcher nest building in the area on 28 May. The present observations were therefore 154 THE WILSON BULLETIN • Vol. 89, No. 1 made mid-way through its breeding season. It is possible that this was a renesting at- tempt. Yellow Warbler nests resemble Least Flycatcher nests (Bent, U.S. Natl. Mus. Bull. 179, 1942) ; thus the stimulus of a familiar nest structure may have enhanced this opportunistic behavior. As the breeding season progresses, it may be more advantageous to occupy an existing nest and to channel the energy saved into egg production rather than expending it on building another nest. This behavior may also indicate the initial stages of nest parasitism. Davis (Auk 57: 179-218, 1940) defines a nest parasite as one which builds no nest of its own but rather raises its young in old nests of other birds. The use of deserted or old nests may not illicit competition for the nest site with the original owner thereby allowing egg-laying to occur with minimal disturbance. This work was funded by grants from the National Research Council of Canada and the University of Manitoba Research Board to S. G. Sealy, whom I thank for assistance in preparing this note. This is paper number 34 of the University of Manitoba Field Station (Delta Marsh). — J. Paul Goossen, Dept, of Zoology, Univ. of Manitoba, Winnipeg R3T 2N2. Accepted 26 Mar. 1976. Pintail reproduction liainpererl by snowfall and agriculture. — The reproductive ( strategy of the Pintail (Anas acuta) show's several adaptations to the semi-arid variable j climate of the prairie pothole region of north central North America where the species is a common breeder. By nesting early and using temporary and seasonal water areas | replenished by snow' melt waters or early spring rains, the species has successfully occu- j pied broad areas containing limited permanent and semi-permanent water. The Pintail is | prone to select new breeding grounds during periods of drought. Smith (J. Wildl. Man- age. 34:943-94b, 1970) has shown that part of the population moves northward from the prairies and parklands when widespread drought conditions occur there. Though well- : suited for the natural prairie pothole environment, this reproductive strategy makes the j Pintail vulnerable to spring snowstorms and modern agricultural practices. In recent years, high cereal grain prices have caused most of the prime Pintail breeding areas of eastern North Dakota to he placed under annual cultivation. Because the Pintail is prone to nest on cultivated lands, it is particularly vulnerable to spring farming operations. j The magnitude of direct nest loss attributable to agriculture varies with the chronology j of planting operations, size of the nesting population, and timing of nesting. These factors are affected by precipitation patterns. A recent study indicated few Pintail and other duck nests survive when nests arc initiated on cropland prior to spring planting opera- tions (K. Higgins, J. Wildl. Manage, in press). Field observations of Pintail hens and examination of reproductive tracts of sampled specimens during the spring of 1970 in | eastern North Dakota provided an opportunity to identify nesting patterns and to study ^ their relationship to precipitation, including snowfall, and to agricultural operations. Pintails began arriving in substantial numbers in Stutsman County, North Dakota, on 4 April. Pair dispersal was initially slowed by a lack of water in shallow wetland basins, ^ but in mid-April water conditions improved dramatically following a wet snowfall on 12 and 13 April and additional precipitation on the 15th. Pintail pairs soon occupied the newly flooded habitat and egg laying was in progress by 18 April at the onset of a 10 cm snowfall. Precipitation at Jamestown, North Dakota during April 1970 totaled 5.77 cm (U. S. Dept. Commerce, Climatological Data 79:43-57, 1970). I March 1977 • (;ENEKAL NOTES 155 I collected 5 paired hens (P-45, P-47, P-49, P-50, and P-51) in Stutsman, Barnes, and LaMoure counties from 18 to 23 April while the ground was covered with snow. Examina- tion of the reproductive tracts of the sampled hens indicated all had continued to lay despite apparent disruption of nesting by snow cover. P-45 was collected shortly after the onset of snowfall on 18 April and was nearing completion of its clutch at that time based on number and stage of regression of ruptured follicles. The bird had laid on the date of collection, had an egg in the oviduct, and had 1 enlarged follicle weighing 10.0 g still to be ovulated. P-47 was collected during heavy snowfall on 19 April. Five recently ovulated ovarian follicles, an egg in the oviduct, and normal follicle gradation indicated continued laying activity. Ovary weight was 32.6 g. P-49 had laid 1 egg and its ovary weighed 33.8 g; the largest follicle weighed 15.8 g. Ovulation had not occurred on 20 April but appeared imminent. A series of 6 recently ruptured follicles and an egg in the oviduct indicated P-50 was continuing to lay on 22 April although no enlarged ovarian follicles remained to be ovulated. The ovary of P-51 contained a series of recently ovu- lated follicles and an enlarged follicle (6.3 g) . Three larger follicles had collapsed and were regressing, presumably because of weather-related factors. Continued laying activity by these hens in areas of heavy snow cover suggests eggs were either being dumped or new nest sites chosen. I doubt that many hens relocated their nests the morning of 19 April considering that snowfall began in late afternoon of 18 April and formed a uniform mantle several centimeters deep by the following morning; heavy snowfall continued through the 19th. Relatively mild temperatures presumably were a major factor contributing to the con- tinuance of laying despite snow cover. Dane and Pearson (Pp. 258-267 in Proceedings Snow and Ice in Relation to Wildlife and Recreation Symposium, Iowa Cooperative Wild- life Research Unit, Iowa State Univ., Ames, 1971) indicated that following a snowstorm on 1 May 1967 in North Dakota, all Mallard (Anas platyrhynchos) and Pintail pairs in the severe storm area abandoned their territories but some Mallards continued laying in the less intense storm areas. The 1967 storm was accompanied by high wdnds and tem- peratures fell to -9°C in the severe storm area causing ice formation of sufficient strength to carry a man (Dane and Pearson, op. cit.). Following the 1970 snowfall, the lowest temperature recorded at Jamestown w'as -5°C on 22 April. During the 1967 storm, -9°C readings lasted for 3 days whereas in 1970 the temperature dipped to -5°C for only 1 night. Because many wetlands remained at least partially open, invertebrates continued to be available and were fed upon by hens. The 5 hens collected during the snowfall period were feeding and esophagi of 4 contained invertebrates (earthworms or snails, and/or dipteran larvae) . Earthworms were obtained from newly flooded shallow wetlands. In- vertebrates are the dietary staple of Pintail hens during the period of egg formation (Krapu, Auk 91:278-290, 1974). The fact that food remained available at certain wetland sites during the period snow covered the ground in 1970 may have contributed to continued laying activity. Nalbandov (Reproductive Physiology, 2nd ed., W. H. Freeman. San Francisco, 1964) indicated that withdrawal of food triggers follicular resorption in laying hens. Some interruption of laying apparently did result from the storm. In addition to ovarian follicular regression noted in P-51 taken towards the end of the snow'fall period, ovaries of 2 of 5 hens col- lected from 24 April through 30 April had enlarged collapsed follicles indicating inter- ruption of a previous nesting attempt. Three of these hens w'ere laying, 1 was about to initiate laying, and 1 showed no evidence of recovery. Heavy nest loss and/or inhibition of nesting as a result of the 18-20 April snowfall presumably contributed substantially to the high transect pairilone male ratio of 23:19 on 29 April in comparison to a pair: 156 THE WILSON BULLETIN • VoL 89, No. 1 lone male ratio of 8:24 on that date in 1969 (C. Dane, pers. comm.). Favorable weather conditions prevailed throughout late April 1969. Following an unusually wet and cold period in late April 1970, a second major nesting effort began on about 1 May and lasted for approximately 10 days. Of 10 hens collected from 3 to 9 May all had ovulated on the date of collection. The Pintail pairilone male ratios along a transect near Jamestown on 29 April, 5 May, and 12 May were 23:19, 12:17, and 4:20, respectively fC. Dane, pers. comm.). These data also suggest that a major por- tion of the hens had completed clutches and had initiated incubation by mid-May. Mild weather accompanied this laying effort. Evidence of substantial nest destruction appeared during late May 1970. Agricul- tural operations commenced on about 10 May and a massive planting effort was underway by mid-May. Sightings of hens increased on the Drift Prairie as agricultural operations progressed. I obtained a measure of the magnitude of nest losses from data collected during the annual breeding ground survey conducted by the North Dakota Game and Fish Department. Two transects (6 and 7) totaling 643.7 linear km and encompassing an area of 259.1 km^ occur in the intensively cultivated Drift Prairie of eastern North Dakota. Surveys were conducted on 18-19 May 1969 and 18-21 May 1970. Data from the 2 tran- sects were combined resulting in pair: lone male ratios of 105:364 (1969) and 165:352 (1970) (C. Schroeder, pers. comm.) indicating a marked difference in the proportion of non-incubating hens between years presumably caused by differing nest destruction rates. Favorable weather conditions during the sj)ring of 1969 reduced potential for heavy losses of incubated clutches from planting operations late in the nesting season. In 1970, 11 paired Pintail hens were sampled on the Drift Prairie from 20 to 27 May and all had brood patches suggesting nesting attempts had progressed into incubation. Behavior and activity patterns of these hens suggested nesting attempts had been terminated. Estimates derived by projecting these transeet data to the entire Drift Prairie area of North Dakota and interpreting reproductive status on the basis of sampled birds suggest that approxi- mately 43,725 hens had lost one or more nests on the Drift Prairie and were approaching a nesting failure for the season when censused in late May 1970; 15,900 more hens than during a similar period in 1969. The impact of nest loss during years of delayed planting is intensified because of the reduced probability that hens will renest following nest destruction late in the breeding season. Deteriorating water conditions in late May 1970 ])robably also diminished renesting prospects. In summary, because most of the prime breeding area of the Pintail in eastern North Dakota is now under annual cultivation, prolonged periods of precipitation in spring that in pristine times either enhanced Pintail reproduetion or, as with snowfall, acted only as a temporary setback, now delay agricultural operations and set the stage for destruction of nests over millions of hectares in a few weeks. The problem is magnified from past agri- cultural periods because few fields are now left idle each spring for a sufficient period to allow clutches to hatch. The Pintail will renest when suitable habitat conditions exist so is capable of partially compensating for these losses. Agriculture poses an additional threat to the Pintail through destruction of wetland habitat. The degree of success of efforts to preserve a substantial part of the remaining temporary and seasonal wetland habitat of the prairie pothole region will be a major factor determining whether prairie-nesting Pintail populations comparable to those of recent years will continue to exist there. Protection from drainage has been afforded sizable blocks of shallow wetland basins at widely scattered locations in North Dakota through perpetual easements under the Small Wetlands Program administered by the U. S. Fish and Wildlife Service. This continuing effort is an important step toward March 1977 • GENERAL NOTES 157 maintaining the extensive shallow wetland base required by Pintail breeding in the prairie pothole region. I thank Charles W. Dane and Charles H. Schroeder for use of unpublished data. Dr. Dane also provided editorial assistance and comments on the manuscript. — Gary L. Krapu, U. S. Fish and Wildlife Service, Northern Prairie Wildlife Research Center, Jamestown, NO 58401. Accepted 15 Mar. 1976. Ticks as a factor in the 1975 nesting failure of Texas Brown Pelicans. — Fewer than 100 Brown Pelicans {Pelecanus occidentalis) remain on the Texas Coast from a pop- ulation that once numbered 5000 birds. Only a small proportion of these have nested in recent years and most attempts have not been successful (King et ah, Southwest. Nat. 21: in press). Pesticides were responsible for mortality of Louisiana pelicans in 1975 (Winn, Audubon Mag. 77:127-129, 1975), but nest failure in Texas was attributed to natural causes. Seven pairs of Brown Pelicans nested on a low-lying island near Aransas Na- tional Wildlife Refuge in April, but all deserted their nests before the eggs hatched. The cause of desertion was either storm tides that nearly inundated the island, or more likely, an infestation of nest parasites. A later nesting attempt at Pelican Island in Corpus Christi Bay ultimately produced 9 young. Adults were first seen building nests near Aransas Refuge on 14 April. On 24 April, 6 nests containing 16 eggs were still active. A 7th had failed, probably due to wave action that partially buried the nest and its single egg. Eight days later, all nests were found deserted. When the eggs were collected on 7 May, an unusually heavy infestation of ticks was noted in and around the nests. On 25 May, 3 nests were collected which yielded a total of 2389 adult and nymphal ticks. Many thousand larvae were also present. All ticks were identified as Ornithodoros capensis by personnel of Naval Medical Re- search Unit-3 (NAMRU-3) Cairo, Egypt, and United States Public Health Service, Rocky Mountain Laboratory, Hamilton, Montana. O. capensis is a common argasid tick infesting many species of aquatic birds in tropical, subtropical, and south temperate climates (Kohls et ah, Ann. Entomol. Soc. Am. 58:331-364, 1965; Hoogstraal, in Viruses and In- vertebrates, A. J. Gibbs (ed.), American Elsevier Publishing Co. Inc., New York, 1973; Hoogstraal et al., J. Med. Entomol. 12:703-704, 1976). Although 0. capensis is found in many areas of the Old World, its occurrence in continental United States has been re- ported only once. Twelve specimens were taken from a Roseate Spoonbill {Ajaia ajaja) collected on an unnamed island off the coast of Texas in 1940 (Kohls et al., op. cit.). Closely related specimens in the 0. capensis group, but not true 0. capensis, have been recorded in Oregon (Clifford et al., J. Med. Entomol. 7:438-445, 1970), California (Radovsky et al., J. Parasitol. 53:890-892, 1%7), and Florida (Kohls et al., Ann. Entomol. Soc. Am. 58:331-364, 1965). Infestations of O. capensis have caused nest desertion and perhaps the death of nest- lings through the transmission of a lethal arbovirus in some sea-bird colonies. Converse et al., (Am. J. Trop. Med. Hyg. 24:1010-1018, 1975) and Feare (Ibis 118:112-115, 1976) documented the abandonment of 5000 Sooty Tern {Sterna fuscata) nests containing eggs and young in a colony of 400,000 pairs on Bird Island in the Seychelles. They found numerous ticks in the deserted portion of the colony but few or none in adjacent areas where reproduction was normal. Not only did the ticks cause desertion, they remained so abundant the following year that the terns did not reoccupy the area. Marshall (Wilson Bull. 54:25-31, 1942) reported nest abandonment of incubating Common Terns [Sterna 158 THE WILSON BULLETIN • Vol. 89, No. 1 hiriindo) at night and their return to the sites during the day. The cause of night deser- tion was not identified but it may have been related to a heavy infestation of ticks such as 0. capensis. A Soldado-like virus was isolated from our Texas tick samples sent to Rocky Mountain Laboratory. When 0. capensis ticks collected from sick Sooty Terns on Bird Island were allowed to feed on young domestic chickens, they transmitted a Soldado virus that caused the death of their host (Converse et ah, op. cit.). Feare (op. cit.) found an unusually high number of dead young terns and felt tbe Soldado virus transmitted by the ticks may have contributed to the die-off. In our study the occurrence of a spring storm 30 April within the period of desertion (24 April-2 May) complicated defining the cause of nest abandonment. Winds reaching 38 knots and rainfall of 0.41 cm w'ere recorded at the nearest National Weather Service station in Victoria 163 km). However, the 6 nests which had been active on 24 April showed no sign of damage by wind or high water when they were examined on 7 May. We therefore concluded that the infestation of ticks was the probable cause of nest desertion. In contrast to the complete failure of the Brown Pelican’s first nesting attempt, the later nesting on Pelican Island produced young; 9 fledged from 11 nests. We found no Ornithod- oros ticks associated with the young pelicans, their nests, or in the soil and litter beneath the nests on Pelican Island. Distribution of O. capensis is influenced by bird movements. Our preliminary investi- gations reveal tbe occurrence of O. capensis in several heronries on the central Texas Coast. Host species noted so far include the Brown Pelican, Roseate Spoonbill, Cattle Egret (Buhulcus ibis), Reddish Egret U)ichromanassa rufescens) , Black-crowned Night Heron (Nycticorax nycticorax) , Laughing Gull iLarus atricilla) , and Black Skimmer i Rynchops nigra). The long-term effects of ticks on pelicans and other colonial nesting birds remain to be determined. — Kiuke A. King, U. S. Fish and Wildlife Service, Patuxent Wildlife Research Center, Gulf Coast Field Station, P. O. Box 2506, Victoria, TX 77901; David R. Blankinship and Rich Aim T. Paul, National Audubon Society, 115 Indian Mound Trail, Tavernier, FL 33070; Robin C. A. Rice, Dept, of Fntomology, Univ. of Hawaii, 2500 Dole Street, Room 28, Honolulu 96822. Accepted 13 July 1976. Prairie W arbler feeds fr«»ni spider web. — A note in tbe March 1976 Wilson Bulletin described an incident of feeding from a spider web by a Cedar Waxwing {Bombycilla cedrorum) ( Burtt et al., Wilson Bull. 88:157-158, 1976). It was believed that this repre- sented the first account of such behavior in a passerine. I observed a somewbat similar incident involving an adult male Prairie Warbler {Den- droica discolor) in Everglades National Park, Monroe Co., Florida, on the afternoon of 3 July 1971. The bird was perched low' in mangroves about 20 cm from the vertically- oriented web of a golden silk spider iNephila clavipes) . Three times during a 30-sec period he flew briefly to the web and each time picked an insect from it with his bill. Upon alighting on his perch, he swallowed each insect and then w'iped his bill against a branch as if cleaning silk from it. No spider was seen on the web. Prairie Warblers are known to eat spiders and to use spider silk in nest-construction (Wetmore, U. S. Dept. Agr. Bull. 326:1-133, 1916; Bent, U. S. Natl. Mus. Bull. 203:1- 734, 1953). Webs of Nephila clavipes often persist for relatively long periods of time March 1977 • GENERAL NOTES 159 (Comstock, The spider book, Cornell Univ. Press, Ithaca, 1948), and abandoned ones might serve as effective insect traps for birds detecting them. — John F. Douglass, Arch- bold Biological Station of The American Museum of Natural History, Route 2, Box 180, Lake Placid, FL 33852 (Present address: Dept, of Ecology and Evolutionary Biology, Univ. of Michigan, Ann Arbor 48109). Accepted 27 Sept. 1976. Notes on the hummingbirds of Monteverde, Cordillera de Tilaran, Costa Rica. — Monteverde, a lower montane site on the Pacific slope of northwest Costa Rica’s Cordil- lera de Tilaran, supports a strikingly rich avifauna. During the periods October 1971- May 1973 and June-July 1975, I recorded 20 species of bummingbirds — over a third of Costa Rica’s total — at Monteverde. Many species were well outside ranges previously described by Slud ( Bull. Am. Mus. Nat. Hist. 128, 1964) and others. I present here information on avifaunal affinities of this diverse assemblage as well as data on local distribution, breeding, plumages, and behavior of species seldom studied in the field. Monteverde lies upon a bench, elevation 1320-1540 m, just below the continental divide, which reaches 1600-1700 m. The approximately 3 km wide belt bounded by the lower edge of the bench and the divide contains a steep gradient of biotic communities, cor- responding to the steep moisture gradient produced by trade winds which carry mist over * the divide during the November-May “dry season.” These communities range from a constantly wet, wind-sculptured elfin forest on the divide proper to a partly deciduous moist forest, subjected to severe dry-season moisture stress, on the lower edge of the bench. One may subjectively divide this gradient into “life zones,” though no obvious {boundaries exist and elevations are only approximate. Life zones were determined with the aid of Tosi (Mapa ecologico de Costa Rica, Centro Cientifico Tropical, San Jose, 1969) and Holdridge (Life zone ecology. Tropical Science Center, San Jose, 1967; pers. comm, to G. V. N. Powell). They are abbreviated below as follows: MF-WF = Lower I Montane Moist Forest-Wet Forest Transition, elevation 1200-1400 m; WF Lower Mon- tane Wet Forest, elevation 1400-1480 m or higher; WF-RF = Lower Montane Wet Forest- Rain Forest Transition, elevation 1480-1540 m or higher; RF = Lower Montane Rain Forest, elevation over 1540 m; EF = Elfin Forest, crest of divide. I Many of Monteverde’s bird species, especially those inhabiting the lower habitats, are j characteristic of the dry Pacific northwest or the subtropical belt (sensu Slud 1964). Many species of the Caribbean slope stray over the divide, however, and the wetter forests j' of the higher elevations contain many characteristic highland birds. At these elevations |j hummingbirds typical of tbe forests also exploit flowers in the limited second-growth j areas. The extensive pasture and scrub habitats of the lower life zones contain a distinct i group of species, however, though forest populations contribute scattered individuals j (Feinsinger, Organization of a tropical guild of nectivorous birds, Ph.D. thesis, Cornell Univ., 1974). ; Though I made observations in all life zones, most studies, mist-netting, and color- marking of hummingbirds — following the method of Stiles and Wolf (Condor 75:244-245, : 1973) — took place in MF-WF successional habitats. Within these habitats, hummingbirds i fed at flowers of 15 plant species, particularly the herb Lobelia laxiflora ( Campanula- ii ceae) and the tree Inga brenesii (Leguminosae) . Aggression was most pronounced during j flowering peaks of these species. Territorial species directed most aggression toward I flying or feeding birds; only Philodice bryantae consistently displaced perched conspe- i: ! 160 THE WILSON BULLETIN • Vol. 89, No. 1 cifics. Feeding birds often spread their rectrices as a display, a stabilizing maneuver, or both. Most species vocalized while fighting, feeding, perching, or flying. Additional in- formation on foraging and aggressive behavior is given by Feinsinger (1974) and by Feinsinger and Chaplin (Am. Nat. 109:217-224, 1975). Phaethornis guy. Green Hermit (resident) . — A bird of the humid sub-tropical and lower montane belts of both slopes (Slud 1964), at Monteverde this species resided in mature forest understory from MF-WF through EF. At least one lek site existed in RF. G. V. N. Powell (pers. comm.) discovered a nest with eggs in January 1971. Phaethornis longuemareus, Little Hermit (vagrant). — The Little Hermit is local in the dry Pacific northwest of Costa Rica and on the lower slopes of the northwestern cordil- leras (F. G. Stiles, pers. comm.; see Slud 1964). In April 1970, W. H. Buskirk (pers. comm.) sighted an individual in WF-RF understory. I observed a bird at the lower edge of the Monteverde bench (MF-WF) on 5 May 1972, and netted an adult in the same spot on 21 January 1973. Doryfera liidovicae. Green-fronted Lancebill (vagrant). — On 16 October 1972, I netted an immature near a flowering Inga brenesii tree in a MF-WF pasture. (Immatures were identified by their striated cuhnens — see Ortiz-Crespo (Auk 89:851-857, 1972).) Previous published records for Costa Rica are restricted to the Caribbean slope of the central high- lands (Slud 1964). Campylopterus hemileiicurus, Violet Sabrewing (resident). — A locally common highland species (F. G. Stiles, pers. comm.; Slud 1964), this large hummingbird commonly foraged among banana flowers at Monteverde. Forest-edge and forest understory at all elevations also supported a moderate density of sabrewings. F. G. Stiles (pers. comm.) recorded a lek at Monteverde; however, I noted only single singing males, usually near rich food sources. Males, which weighed more and had shorter bills than females (Feinsinger 1974), exhibited a range of plumages from solid green through mixed green and violet to solid violet. This secjuence was correlated with disappearance of culmen striations, indi- cating a distinct immature male plumage unlike that of the gray-bellied females {contra Wetmore, Smithson. Misc. Collect. 150(2), 1968). Colibrl delphinae, Brown Violet-ear (vagrant). — 1 netted an immature of this species near an Inga brenesii tree on 15 October 1972. On 17 October, at least one other Brown Violet-ear fed at a nearby /. brenesii. Slud (1964) cites several isolated records from the central and southern mountains and one from the Cordillera de Guanacaste (the next mountain range to the northwest) but none from the Cordillera de Tilaran. Colibri thalassiniis, Green Violet-ear (seasonal). — From October through June, non- forested habitats from MF-WF through RF supported large numbers of this widespread highland species. Singing, presumably by males, took place on exposed perches well away from food sources. Even at particularly rich food sources. Violet-ears made few efforts at defense and w^ere often displaced by smaller birds (Feinsinger 1974). Feeding birds often voiced a repetitious dry chatter. Chlorostilbon canivetii. Fork-tailed Emerald (seasonal). — From late January through late November, a number of Emeralds resided in the MF-WT study areas. Immature males resembled females but possessed dark-green patches of varying extent on breast and throat. Many individuals of both sexes possessed entirely black mandibles and thus resembled the race — or species (Wetmore 1968) — assimilis of southwestern Costa Rica and Panama. Others resembled the race salvini, a member of the dry northwest avifauna that is abundant on the lower slopes of the northw^estern mountains ( Slud 1964) . All individuals observed behaved similarly, however, exploiting scattered flowers while giving a soft, wren-like chatter. March 1977 • GENERAL NOTES 161 Panterpe insignis. Fiery-throated Hummingbird (resident). — This highland species is most typical of the Cordillera Central and Cordillera de Talamanca ( Slud 1964). The EF and upper RE at Monteverde also support a conspicuous resident population, dis- cussed by Stiles and Hespenheide (Condor 74:99-101, 1972). Individuals I observed for- aged in clearings and at epiphytes in the forest canopy but rarely entered forest under- story. Hylocharis eliciae, Blue-throated Goldentail (seasonal). — Common lower on the Pacific slope (Slud 1964), Goldentails entered Monteverde’s MF-WF non-forested habitats only. A few adults and many immatures converged on Lobelia laxiflora fields February-April 1972 and 1973. During October and November 1972, some individuals fed at Inga brenesii trees. Immatures were dull huff on the underside, with pale pink at the base of the hill and flecks of blue on the throat or upper breast. Adults and immatures alike defended territories, often fighting with each other or with Blue-vented Hummingbirds. Amazilia saucerottei. Blue-vented Hummingbird (resident). — Typical of the drier Pa- cific slope and central highlands of Costa Rica (Slud 1964), this species remained abun- dant year-around in the MF-WF study areas but was seldom seen above 1400 m or in wooded regions. Nearly every rich flower in the study areas supported a Blue-vented Hummingbird territory. Any other hummingbird that attempted to forage among defended flowers and attracted the attention of the territory holder was chased. These high-speed chases often entered other territories and became 3- or 4-bird affairs. Occasionally chase attempts were unsuccessful. Green Violet-ears entering Blue-vented territories in Lobelia laxiflora fields often ignored the residents and continued to feed. The numerous heli- coniine, ithomiine, and pierid butterflies attracted to large Hamelia patens shrubs also ignored resident Blue-vented Hummingbirds despite the birds’ continued efforts to eject them. On such occasions, defenders often abandoned their attempts in favor of irregular feeding, presumably a displacement activity. Two marked Blue-vented Hummingbirds in particular remained in the study areas throughout much of the 1971-73 research period. These moved their territories from one flower concentration to another over the months. Their “shifting territories” isensu Wolf, Condor 72:1-14, 1970) remained within a limited area which might be considered the home range. Singing and fighting that took place well away from rich flower clumps peaked from early August through October 1972, indicating a possible spurt of mating activity in late wet season. Immatures (as judged by culmen striations) appeared throughout the year, however, and there was a great influx of them in May 1972. I encountered 2 nests, 1 on 2 December 1971 (containing a well-grown juvenile) and 1 on 9 January 1972 ( containing eggs) . The latter nest was on a dead Cecropia obtusifolia branch at secondary forest edge, elevation 1200 m. The nest with the juvenile was on a Ficus vine along a pasture edge at 1370 m. On 2 December an adult fed the juvenile at the nest, but on 4 December the latter had fledged and was in the grass beneath the nest. The bird feeding it chased away another adult that landed on the rim of the empty nest and made side-to-side head move- ments suggestive of feeding. This juvenile possessed duller plumage than most adults or immatures. Primaries and secondaries were medium gray instead of black; coverts and contour feathers on head, neck, and back appeared quite buffy (probably due to buffy edgings) ; and the rectrices were a light iridescent bluish-green resembling the color of a Green Violet-ear tail. I estimated exposed culmen length at 8 mm, less than half that of an adult. During the May 1972 influx of immatures, several netted had especially deep bill striations as well as duskier plumage and tails lighter in color than adults. These may have recently fledged. 162 THE WILSON BULLETIN • Vol. 89. No. 1 Amazilia tzacatl. Rufous-tailed Hummingbird (resident or seasonal) . — Widespread at lower elevations and occurring in clearings to at least 1500 m (F. G. Stiles, pers. comm.; Slud 1964) , at Monteverde I observed this species only sporadically. Rufous-tailed Hum- mingbirds were especially conspicuous in the MF-WF study areas during June 1972, when they wrested control of many large Hamelia patens bushes from Blue-vented Humming- birds. This species breeds at Monteverde: F. G. Stiles (pers. comm.) has discovered nests, and on 13 and 15 November 1971 I observed an adult feeding a juvenile perched on a WF roadside tree. Though its tail was the same striking rufous-brown as the adult’s, the juvenile’s green plumage appeared much duller (perhaps due to gray or huffy edg- ings) and its culmen appeared somewhat shorter. Eupherusa eximia. Stripe-tailed Hummingbird (resident). — A typical mid-elevation spe- cies (Slud 1964), the Stripe-tailed Hummingbird was abundant throughout forest-edge and deep forest from MF-WF through EF. Stripe-tails often entered MF-WF clearings as well. They visited a wide variety of flowers, piercing those corollas adapted for longer- billed species. Foraging or fighting individuals often voiced loud buzzes and spread their striking tails for the duration of each buzz. Elvira cupreiceps. Coppery-headed Emerald (resident).- — This normally Caribbean-slope species (Slud 1964) was most abundant in forest-edge and forest habitats in the MF-RF and lower RF. Emeralds also entered MF-WF openings at times, usually to feed at Inga hrenesii flowers. G. V. N. Powell (pers. comm.) found 2 nests in WF-RF in November 1971 and October 1972. Larnpornis hernileucus, White-bellied Mountain-gem (vagrant). — On 14 February 1973, I netted an immature at 1320 m elevation in the MF-WF study areas. I also observed an immature or female in the same area on 29 May 1972. W. H. Buskirk (pers. comm.) ob- served this species on the Caribbean slope northeast of Monteverde during dry season. Previously, this species was not recorded further north than the Caribbean slope of the central highlands (Slud 1964). Larnpornis calolaema, Purple-throated Mountain-gem (resident). — I often encountered this common highland hummingbird in all habitats from WF through EF. Some Moun- tain-gems strayed into the MF-WF study areas to feed at Inga hrenesii. Although domi- nant over even Blue-vented Hummingbirds, Mountain-gems gave few displays. Nests with eggs or young were encountered during all seasons in understory trees or shrubs (January, March 1972 — WP"), on low^ forest-edge vines (October 1972, W'F ; December 1972, WF- RF), even in crevices in a clay roadbank (May 1972, August 1972, July 1975 — WF-RF and RF). Ileliodoxa jacula. Green-crowned Brilliant (resident). — W. H. Buskirk and G. V. N. Powell (pers. comm.) netted these birds year-around in mature forests of the WF-RF zone. Slud (1964) cites no records from the northwestern cordilleras for this species, more typical of the subtroj)ical-lower montane Caribbean slope. Ileliornaster constantii. Plain-capped Starthroat (seasonal). — A member of the dry- forest avifauna (Slud 1974). this species appeared in the MF-WF study areas. Females or immatures fre(piently foraged at the vine Mandevilla veraguasensis ( Apocynaceae) May-y\ugust 1972 and .luly-August 1975. Ehilodice hryantae, Magenta-throated Wood-star (seasonal) .-Wood-stars were con- sidered “very uncommon*’ by Slud (1%4), who listed several records from the central highlands but only 1 from the Cordillera de Guanacaste and none from the Cordillera de Tilaran. At Monteverde, however, these unique little birds appeared predictably and abundantly from September through April of each year. Though most common at Inga hrenesii and Lobelia laxiflora flower concentrations in the MF-WF study areas. Wood- stars also foraged in openings and forest edges through at least WF-RF. March 1977 • GENERAL NOTES 163 Intense intraspecific belligerence characterized both sexes of Wood-stars. Rich food sources were vigorously defended. Territory holders perched on dead branches high in trees near the defended flowers. Territorial males sang a complex weak, scratchy melody that often included loud snaps. Songs were often interspersed with preening bouts or with displays that consisted of rotating the head back and forth, presumably exposing the males’ brilliant gorgets. Conspecific intruders of either sex, whether feeding or perching, were immediately attacked. If its rapid approach failed to displace the intruder, the attacker hovered back and forth over the trespassing bird, uttering a variety of squeaks and a buzzy churrrr and occasionally darting at the other. The slender rectrices were kept spread throughout this action, producing a palmate appearance in contrast to the fanlike spread tails of other species with broader rectrices. These displays seldom failed to elicit either flight or a battle. Both “churrrring” loudly, fighting Wood-stars circled each other 5-30 cm apart and darted back and forth. Still circling and darting, the fight- ing pair sometimes rose high in the air. More often, however, opponents stayed near ground level; failing to eject one another, both would resort to displacement-feeding, still churrrring. Males performed an aerial display reminiscent of certain North American species. These displays were usually aimed at a male or female trespasser that the de- fender had been unable to displace, but were occasionally aimed at perched females and thus may have served a mating function as well. A displaying bird swept back and forth pendulum fashion in a shallow, 20-30 m wide arc that centered just above the target individual. At each endpoint, the displayer paused or flew at a tangent for 2-3 sec before sweeping down again. The visual display was accompanied by a loud, snipe-like whistle, undoubtedly produced by the wings, and at the bottom of the arc by 3 to 7 loud manakin- like snaps, perhaps produced by the rectrices. From 1 to 8 such displays were performed in a sequence, the plane of the arc changing all the while. The performer invariably concluded by flying to a nearby perch. I never observed females to sing or to engage in the pendulum display. Nevertheless, females often defended feeding territories, displaced both sexes from flowers or perches, and darted around intruders if necessary. Perched Wood-stars never uttered the churrrr. Foraging birds often did so, however, especially if agitated. A feeding bird never opened its rectrices unless under attack but rather kept the tail closed and pointed up at about a 60° angle from the body plane. Moving methodically from flower to flower, never chirping, wings beating so rapidly and smoothly that a steady loud hum was produced, these stocky, dull-colored birds resembled large hymenopterans. In fact, they may derive some benefit from that resemblance (Fein- singer 1974) . Archilochus colubris, Ruby-throated Hummingbird (seasonal). — A number of females and an occasional male of this species, wbich is most often observed at lower elevations on the Pacific slope (see Slud 1964, Wolf 1970), appeared in the MF-WF study areas from October-March. Culmen striations showed that all 6 birds netted were birds of the year. Nevertheless, all 4 males, even 1 caught on 14 October 1972, possessed full gorgets. Selasphorus scintilla, Scintillant Hummingbird (vagrant). — On 7 July 1975 I observed an individual of this tiny species feeding in a field of Rubus rosaeafolia (Rosaceae) in the WF-RF zone, and 2 individuals were netted on 10 July. Slud (1964) mentions no records of this mid-elevation species from the northwestern cordilleras. I especially wish to thank the people of Monteverde for allowing and indeed encourag- ing these studies on their land. I am also grateful to F. G. Stiles for copious comments on the manuscript, and to W. H. Buskirk and G. V. N. Powell for use of their field notes. Field work was supported by an Andrew D. White (Cornell University) Fellowship, a Cornell Graduate Fellowship, an NSF grant to the Section of Ecology and Systematics at 164 THE WILSON BULLETIN • Vol. 89, No. 1 Cornell, and a grant from J. S. Dunning. — Peter Feinsinger, Dept, of Zoology, Univ. of Florida, Gainesville, 32611. Accepted 1 Oct. 1975. Nest-site differences between Red-headed and Red-bellied woodpeckers in South Carolina. — Red-headed {Melanerpes erythrocephalus) and Red-bellied (M. carolinus) woodpeekers are potential competitors for nest-sites over much of their range. Parameters serving to lessen competition between them have been discussed by Reller (Am. Midi. Nat. 88:270-290, 1972) for Illinois and by Jackson (Condor 78:67-76, 1976) for Kansas. Reller states that “All Red-heads observed nested in trunks of dead trees. Red-bellies, on the other hand, favored dead limbs in live trees for nest sites,” her observations having been made in oak-maple-hickory woodlands. Jackson (op. cit.), studying the 2 species under differing ecological conditions, noted that while both species preferred to nest in dead trees, 50% of which were elms, the Red-headeds preferred nest trees with open spaces around them and Red-hellieds, ones located in woodlands. Other differences were that Red-headeds, in contrast to Red-hellieds, preferred dead limbs with no bark and ones with a crack in which to make entrance holes. The aim of this report is to describe nest- site differences under still other conditions, namely those of the coastal plain in South Carolina. Observations were made at a (juail shooting plantation in Luray in April and May 1973 to 1975. Pairs of Red-hellieds and of Red-headeds were more or less intermixed in terrain where strips of loblolly pines (Pinus taeda) , along with scattered oaks and other deciduous trees alternated with open fields. As shown in Table 1 the Red-bellied occupied holes carved originally by Red-cockaded Woodpeckers {Picoides borealis) in living pines or excavated ones of their own in pines that had recently died. The outstanding feature of these latter was that they still retained hark and branches. Pairs of Red-headeds, in contrast, excavated or occupied pines dead for some years. These were well-weathered, had almost no hark, and had only broken limbs remaining. Many, having lost their tops, were no more than stubs. One exceptional dead pine fell between the categories. It had, oddly enough, a pair of Red-hellieds trying to nest in an old hole made by Red-hellieds Table 1 Nest Trees Occupied by Red headed and Red-bellied Woodpeckers Early in the Breeding Season on A Plantation in South Carolina Location of Nest Hole ( completed or being excavated ) No. of Pairs Red-headed Red-bellied Hole of Red-cockaded, living 1 6* Recently dead pines 0 8 Old dead pines 10 0 Old pine stubs 13 0 Deciduous tree; dead trunk or limb 0 2 TOTALS 24 16 * One of tlie pines had died within the previous year. March 1977 • (;ENEKAL NOTES 165 the year before, 3 m from the ground; 4 m higher up a pair of Red-lieadeds were trying to start an excavation in the face of much harassing from other Ked-headeds (Kilham, Auk, in press) . Trees chosen by the 2 species differed also in that those used by Red-headeds usually contained numbers of old holes from previous years. As a result of this latter situation, Red-headeds on the plantation shared stubs in one case with Starlings (Sturnus vulgaris)^ once with Common Flickers (Colaptes auratus) , and once with a flying squirrel (Glau- comys volans) . It thus seemed that Red-headeds may he more prone to share nest trees with other species, an observation concurred in by Reller (pers. comm.) although she cites an exception fop. cit.). Jackson fop. cit.) in contrast, found that Red-hellieds charac- teristically nested in trees with more than one hole in Kansas. These discrepancies among observers are of interest in indicating that nest-site preferences can vary with underlying ecologic conditions. A main finding that seems to emerge is that wherever studied, whether in Illinois, Kansas, or in South Carolina, Red-headeds and Red-hellieds do ex- hibit differences in their choices of nest sites. Another parameter serving to lessen interspecific competition it would seem, is time of onset of breeding seasons; Red-headeds, being irregularly migratory and nesting later than the resident Red-hellieds f Jackson, op. cit.) and Kilham fAuk 75:318-329, 1958; Wilson Bull. 70:347-358, 1959). — Law^rence Kilham, Dept, of Microbiology, Dartmouth Medical School, Hanover, NH 03755. Accepted 8 Dec. 1975. Ground foraging and rapid molt in the Chuck-will’s-widow. — In a detailed study of the annual molt of the Chuck-will’s- widow ( Caprimulgus carolinensis) Rohwer fAuk 88:485-519, 1971) inferred that some individuals might be missing so many feathers in late stages of the molt that they would have trouble flying. When growing primaries 8 and 9, Chuck-will’s-widows lose all 10 of their rectrices, more or less simultaneously, and are missing up to bi of the primary surface of each wing fall at the critical tip), as well as nearly ^ of the secondary surface area. At this same time the rictal bristles are also lost simultaneously. Rohwer (op. cit.) felt it unlikely that Chuck-will’s-widows in such an intensive molt could forage aerially hut little more could he said of the matter at that time, partly be- cause of the also suggestive fact that only a single specimen molting either primary 8 or 9 had been preserved. This was a bird shot by Sutton fBull. Okla. Ornithol. Soc. 2:9-11, 1969) at the Oklahoma Biological Station. Students had flushed it from an earthen ledge near the bottom of a deep erosion gully tangled with shrubs, vines, roots, and dead branches. It was flushed again from the same area when Sutton collected it. He reported finding the area strewn with feathers, and was able to find 9 of the 10 molted rectrices, many remiges and a great number of smaller feathers. Mengel fWilson Bull. 88:351-353, 1976) recently collected the second known specimen in late stages of the molt. His bird was flushed 4 times before it was shot; he reported its flight as “direct and somewhat slow and labored,” a striking descriptive contrast to the normally buoyant flight of a Chuck-will’s-widow. The most remarkable fact con- cerning Mengel’s specimen was that it was virtually emaciated, weighing only 86.7 g, a value 27.5% below the normal summer weight of 119.6 g (mean of 12 specimens). Sut- ton’s (op. cit.) specimen was normal in weight (117.1 g). The question raised by Rohwer’s report on the intensity of the molt in its late stages and by the specimens taken by Sutton and Mengel is “How do Chuck-will’s-widows forage in this period of intensive molt?” One possibility, suggested both by the many feathers 166 THE WILSON BULLETIN • Vol. 89, No. 1 found at the secluded resting site of Sutton’s specimen and by the emaciated condition of Mengel’s specimen, is that they forage very little. Another possibility is that they forage terrestrially. In watching Chuck-will’s-widows walking about on roads swallowing pebbles, Jenkinson and Mengel 1 Condor 72:236-237, 1970) give the impression that they might easily forage on the ground. An extensive search of the literature, however, reveals no information on ground foraging by Chuck-will’s-widows; thus, we report the following observations. On the evening of 23 June, 1974, in a residential suburb of Fort Myers, Lee Co., Florida, Butler repeatedly observed a Chuck-will’s- widow capturing squirrel tree frogs 1 Hyla squirella) from a black-top road surface. The incident occurred in the light cast by a street lamp where the frogs were plentiful, presumably attracted to insects. On several occasions the bird alighted on the road near its intended prey and then captured a frog unaided by wdngs or feet and swallowed it. Once the initial attack was evaded by a timely series of leaps, but the bird again flew close to the frog and captured it. Simi- larly, in 1972 Clifford G. Richardson (pers. comm, to Butler) observed a Chuck-will’s- w'idow capturing frogs beneath a street light near his home on Pine Island, Lee Co., Florida. These observations of Chuck-will’s-widows foraging on frogs are significant, not so much because they add an unknown food item to the species’ diet, but because they prove ground feeding to he a fact. An apparent difficulty wdth the ground feeding hypothesis is the very short legs of Chuck-will’s-widows; but this may be resolved by the fact that both Sutton’s and Mengel’s specimens could, indeed, fly. Thus, while individuals in the most intense stages of the molt might be incapable of the sort of maneuvers required to capture flying insects, they could, perhaps, move to points of prey concentration where ground feeding, such as that reported here, might pay. Furthermore, terrestrial foraging would likely be facilitated by the absence of the rictal bristles, thus explaining their simultaneous replacement. — Sievert Rohwer, Dept, of Zoology and Washington State Mu- seum, Univ. of W'ashington, Seattle 98195, and James Butler, College of Forest Resources, Univ. of Washington, Seattle 98195. Accepted 8 Dec. 1975. Feeding responses of fall niigruiits to prolonged ineleinent weather. — September 1975 was unusually cold in northwestern Ohio. A light frost on 14 September was the earliest ever recorded, and temperatures remained 3 to 6°C below normal each day there- after until October. The migration peak for many passerines occurred between 23 and 27 September during a period of heavy cloud cover, gusty winds, frequent rain, and cool temperatures (range 8-16°Cl. Our home in a wooded area near Toledo, Ohio is sur- rounded by fruit-bearing shrubs including yews (Taxus sp.) and Tartarian honeysuckle (Lonicera sp.). During the fall migration many frugivorous species feed at these shrubs; between 23-27 September these species were joined by birds not normally noted for frugivory. The minimum number of normally non-frugivorous birds eating fruit and the fruits se- lected (H = honeysuckle. Y = yew) were as follow's: flycatcher i Empidonax sp.), KH); Tennessee Warbler il'ermivora peregrina) , 1n temperature. <“gg turning, the adaptiveness of the nest, nest tending in the Mega- podes, and antipredator behavior. He then summarizes these interrelated topics in a final discourse on the organization of incubation behavior. I believe this is Drent’s most successful section, but his review of the energetics of incubation, the final section in the chapter, covers some of the newest information to come out of research on incubation. March 1977 • ORNITHOLOGICAL LITERATURE IR3 He concludes that parental costs of incubation probably have upper limits of 20-25% of productive energy. Once again, 1975 “in preparation” citations in the reference section attest to the completeness of Drent’s up-to-date review of incubation, but the omission of Ricklefs’ review (Avian Energetics, Nuttall Ornithol. Club, No. 15:152-297, 1974) because of its late publication date is unfortunate. In the final chapter Vuilleumier reviews the current status of research in zoogeography. The fresh approach currently underway in recent work is immediately recognizable. New understanding in community and population ecology with reference to dispersal, inva- sion, competition, adaptation, and extinction has had a tremendous impact. This, coupled with the theor>' of plate tectonics and the acceptance of continental drift, has injected new vitality into the study of zoogeography. It is refreshing to read of r and K selection, equilibrium values, and extinction rates in a chapter devoted to zoogeography. Vuil- leumier, however, emphasizes that the break from more traditional zoogeography is only a beginning, and empirical studies are badly needed. His review will, I hope, provide the necessary impetus to get these studies underway. In an effort to make his review up- to-date Vuilleumier has included at the end of the reference section 7 additional references that were probably added when the galley proofs for his chapter were in hand. No review would he complete without some cursorv’ attention to errors. I found no glaring factual errors, but some minor ones were noticeable. For instance, in Penny- cuick’s chapter the Parrott reference should be 1971, not 1970, and in Berthold’s review the term “isepeptises” on page 82 should probably he isopiestices or isobars. In Emlen’s contribution those references from the Proceedings of a Conference on the Biological Aspects of the Bird 'Aircraft Collision Problem fAhle 1974b, Emlen 1974, Williams et al. 1974) contain several errors. In all those citations “Clemsen, North Carolina” should be Clemson, South Carolina, and whereas the Able and Emlen references correctly identify the Air Force Office of Scientific Research, the Williams et al. reference has Naval Office of Scientific Research. Although the latter errors were very noticeable to me because I edited these Proceedings, I wonder how many other errors, less obvious to me, are contained in the lists of references. These, however, are small points and should not detract from the overall significance of the contributions and the success of the volume. I feel it is the best of the series, and it is definitely the most expensive — approximately ten cents a page! — Sidney A. Gauthreaux, Jr. Avian Physiology: Symposia of the Zoological Society of London, No. 35. Edited by Malcolm Peaker. Academic Press, London and New York, 1975: 377 pp. $25.50 — The editor invited papers from those “working primarily on domesticated species. . . ,” but this should not turn away the ornithologist for over half of the chapters are comparative in scope, and the rest are coneerned with basic physiological problems. In “Recent advances in digestive physiology of the fowl,” K. J. Hill and P. J. Strachan begin with a brief review, then discuss their research on birds with re-estrant fistulae. “Motor activity of the digestive tract,” “ingesta flow along the duodenum,” and “inter- relationships between crop and gizzard” are the subtopics. M. J. Purves points out the general paucity of quantitative knowledge about “the control of the avian cardiovascular system” compared with mammals. Several gaps are identified where “important evidence . . . still requires to be obtained” especially response to exercise. He examines cardiovascular control during rest, in diving, in flight, and with regard to environmental temperature. Of the 3 purposes (gas exchange, temperature regulation, and vocalization), the first is emphasized by Knut Schmidt-Nielsen in “Recent advances in avian respiration.” Body 184 THE WILSON BULLETIN • VoL «9, No. 1 size is seen to exert a major influence on rates of oxygen consumption. A brief section on anatomy is followed by “breathing and ventilation,” again related to body size. The unique avian lung-air sac system is discussed lucidly in sections on gas flow and ex- change, followed by oxygen carrying capacities of avian blood. The aerodynamic control of airflow remains an unsolved aspect of resting avian respiration; virtually nothing is known about gas flow during exercise and flight. The next chapter, “Flight energetics,” is by Vance Tucker, whose contributions in experimentation and theory establish him as a foremost ornithologist of our time. The level of intellectual effort required to cope with 21 equations and a half-page table of symbols is higher than for the descriptive prose of most ornithological works, but the great potential for understanding the physical basis of natural history patterns should be sufficient motivation. “The theory” and Fig. 1 very lucidly set out the components of a bird’s metabolic rate, so that anyone can visualize the power terms in equation 1. Equation 2 is reached on faith or via the derivations in the appendix to Tucker (1974). The assumptions involved are such that one cannot see at a glance where Pojn^ pp went as components of eqn. 2 and what happened to Pip j,. Coefficients in eqn. 5 and 6 differ slightly from the same basal power relationship given on p. 35 of the previous chapter. Otherwise, the good logic is apparent and I could proceed intuitively, qualitatively to the reward. “Cost of transport,” “Approximation equations,” and “Energy budgets” bring us to the thresh- old of interpretation of migratory and time-allocation behavior. Tucker frankly acknowledges the limitations of the theory in applications to local activity vs. extended flight. This serves to remind us how over-simplified and tenuous are our “energy budgets.” If the reduction of natural history to equations seems esthetically lacking, ponder how the bird “knows” or integrates information on profiles of winds aloft, or solves Pi/wV in order to pick the optimum altitude and effort. This is awe-inspiring! In “Thermal homeostasis in birds,” S. A. Richards points out that “few experiments have been designed specifically for the difficult task of separating the two possible components” of peripheral and central temperature receptors. Perhaps the temperature sensors should be in the deep-body, brain, and spinal regions, and heat-flux rather than temperature-sensors should be in the periphery. Richards describes emerging patterns, but carefully points out the limitations of methods and the interpretation of results. Central integration is not based solely upon hypothalamic input, but on a weighing such tliat the hypothalamus had 60 to 80% of the votes, at least in controlling panting. Especially interesting is the work on hypothalamic control of feather position index. The existence of non-shivering thermogenesis is neither disproven nor established. “Renal and cloacal transport of salt and water,” is, because of many recent reviews, limited by E. Skadhauge to recent observations. These include renal concentrating abilities of Western Australian birds’ excretion of electrolytes, water, and nitrogen, cloacal resorption and interaction of renal and cloacal function, and the relationship between the concentrations of saline drinking water and renal concentrating ability. Skadhauge found an interesting correlation between weight-specific metabolic rate and maximum salinity of drinking fluid for water balance. A small bird such as the Zebra Finch can form a maximum urine concentration of 1000 mOsm, but fluids of about 1440 mOsm can constitute 10% of their water turnover, the difference being made up by metabolic water. M. Peaker discusses “Recent advances in the physiology of the salt glands” as follows: “The primary stimulus” (increase in plasma tonicity, but not necessarily in osmolality), “Location of the receptors” (most likely “on or near the luminal wall of the major March 1977 • OHNITIIOLOGICAI. LITEKATUKE blood vessels near the heart or the heart itself”), “The secretory reflex,” and “secretion and blood flow.” Salt glands are of interest not only in regard to homeostasis, hut for the insight they can furnish regarding the basic functioning of ceils and organs, the control of blood flow to secretory organs, and the cocvolution of an organ and the mechanisms controlling the blood flow rates, which are the highest measured in any organ. From such flow the salt glands can remove up to 80% of the chloride, 57% of the sodium, and 21% of the water. The % increase in plasma sodium concentration needed to stimulate salt-gland secretion is inversely related to body mass. In “Prolactin and adaptation.” D. M. Ensor discusses the complex effects of environ- mental factors and blood composition on prolactin release and the effects of prolactin on nasal-gland function, food and water intake, fattening and fat utilization, water balance, and urine output. Five chapters on reproductive physiology follow. R. K. Murton (“Ecological adaptation in avian reproductive physiology,” pp. 151-152), Lofts (“Environmental control of reproduction,” p. 189) and Follett and Davies (“Photoperiodicity and the neuroendocrine control of reproduction in birds,” p. 207) agree that daily photoperiod serves (1) to entrain a circadian oscillation and (2) to induce the photoperiodic gonad-stimulating response when light falls within the photoinductihle phase of that circadian rhythm. These are interesting accounts of ingenious attempts to understand the physiological clock and its linkage from long photoperiods outside to gonads’ trophin secretion within. These papers must have provided stimulating exchange of current knowledge when presented originally. However, the editor could have produced a more effective review for those not attending by appointing 1 spokesman to integrate the information from 7 species and 3 chapters. P. E. Lake reviews “Gamete production and the fertile jieriod with particular reference to domesticated birds.” The fertile period (prolonged survival of spermatozoa from insemination until fertilization or potential fertilization is said to have been “studied closely” and “to vary distinctly between species; the reasons for the inter-specific dif- ferences are not clear.” He did not consider that duration of fertile period is a unit of physiologieal time, which may be strongly influenced or scaled to body size. A symposium provides the opportunity for cross-fertilization of ideas and techniques. The influence of body size was noted in previous chapters. The potential for allometry in reproductive studies can be seen if one runs logarithmic regressions of fertile period as a function of body size in Lake’s Table 1, p. 236. There are differences not only correlated with size but with order. Anseriformes have shorter spermatozoan fertile periods than Calliformes. Mean duration of galliform fertile periods =: 1.09 m”®^^ (p 0.001), where m rr body mass in g (representative body weights from other sources; original references which I was able to consult list no weights of sperm-donors or recipients). Considering the limited sample size, the exponent is roughly similar to those relating breath and heart cycle durations to body mass in birds. This relationship seems worthy of further examination. Two chapters deal with the influence of hormones on nesting behavior, the authors reviewing their own experiments, using 2 species with long histories of adaptation to captivity and thus amenable to experimentation without undue alarm. In “The dual role of daylength in controlling canary reproduction,” R. A. Hinde and E. Steel review their interesting experiments on the influence of photoperiod on nest-building in response to estrogen. This is followed by “The physiological basis of a behavior pattern in the domestic hen” by D. G. M. Wood-Gush and A. B. Gilbert. They established that the post-ovulatory follicle plays a major role in nesting behavior such as nest examination. 186 THE WILSON BULLETIN • Vol. 89, No. 1 Injection of estrodiol benzonate and progestrone restores normal behavior after ovariec- tomy. Having considered the control of reproduction and reproductive behavior, we proceed to tlie products, the nestlings and eggs. R. J. O’Conner reviews “Growth and metabolism in nestling passerines.” A derivation from “Fourier’s Law” of heat loss attempts to relate tlie development of homeothermy to “the deerease in surface-volume ratio and hence the reduction of the heat lost from each gram of heat producing tissue.” The attempt goes awry when power functions of body weight are substituted for some but not all of the weight-dependent terms in the equation: d QL/'dt = W%/r (Tb-Ta) wlnne d Qf,/dt is rate of heat loss with time, W is nestling weight, r is specific insula- tion, and Tb and Ta are body and air temperature, respectively. However, it can be derived from Lasiewski et al. (Comp. Biochem. Physiol. 23:797-813, 1967) and Herreid and Kessel (Comp. Biochem. Physiol. 21:405-414, 1967) that the specific insulation (per unit surface area) is proportional to Thus, d Qr/dt is proportional to W'-^^ at Ta l)elow tlu'rmoneutrality, as Kendeigh (Auk 86:13-25, 1969) has demonstrated em- I)irically for adult birds. Whether one includes the allometry of r and sees heat loss rate as a function of W'-^ or us(‘s W’^ as in the text, tliere is still a positive allometry of heat loss, i.e. the larger the nestling, the greater the total heat loss, and for that matter, the larger the nestling, the more surface area it has for heat loss. These facts invalidate the conclusion: “The simph‘ change in surface-volume ratio inherent in a heavier body weight thus provides tlu! simplest (‘xplanation of the observed pattern of metabolism in blue tits” (p. 301). Thus, incr('ased metabolic-capacity, directly and through “Oio-effect” as Tb is maintained higlu^r, and incr(‘ased insulation, must he the foundations for homeothermy. This dead- end digression should not distract the reader from the virtues of O’Conner’s painstaking mcasureiiKMits and interesting data on growth in weight, plumage, and metabolism of the Blue Tit, House Martin, and House Sparrow. We may take aspects of the natural liistory of birds, such as the formation of eggshells, for granted and never stop to wonder how they work. K. Simkiss reviews just how marvelously complex and incompletely understood are the calcium metabolism and regulation in “Calcium and avian rei)roduction,” considering 8 current questions in the understanding of calcium metabolism. The subject is of interest not only in regard to this essential stage in perpetuation of birds, but as a system for basic cellular research. Ih^ concludes that “tlie control of avian calcium metabolism remains largely mysterious” but has ])in-pointed research ojiport unity and provided new awe for egg-laying. In “Pesticides and eggshell formation.” A. S. Cooke contributes a valuable inquiry into eggshell thinning, comparing the effects of calcium-deficiency and sulfanilamide- treatment (chickens), l)I)T-treatment (ducks) and examination of shells collected bnieath a heron rookery. Thinning seems to be caused both by reduction in availability of compoiK'nts and by normal deposition rates but premature termination. Cooke points out that these preliminary investigations must be followed up by detailed study of a single sj)ecies, in both lal) and field. A symposium cannot produce a complete textbook on avian physiology but the ornithologist desiring a current knowledge of how the bird functions should have this volume. The physical (piality of the volume' is good, with flat glare-free paper. I found only 2 typographical errors. If more birders were sufficiently interested in the objects of their life-lists to make an informative work like this sell widely, perhaps the price could be less. — William A. Calder HI. March 1977 • OKN ITHOLOC; ICA I. U'rEHATUKK I<‘i7 Avifauna of Nohtji wkstkkn Colombia, South Amkkica. liy Jurgen Haffer. Hoiiner Zoologisclie Monographien, No. 7, Zoologisclies Forscliungsinstitul und Museum Alexan- der Koenig, I^onn, 1975: 182 pp., paper, 7 plates, 51 text figs., 8 tables. Price 35 I)M.- This would appear to be the latest installment by tbe author in a series of repetitively overlapping analyses of Pleistocene refugia and tbe subseijuent dispersal of tbeir avian populations. In the introductory portion he deals “with general aspects of the ecology and zoogeography of the avifauna of northwestern Colombia as they relate toward elucidating the relationships of this forest fauna with other trans-Andean faunas and with the Amazonian fauna.” The second portion presents the results of his fieldwork from 1958 to 1967: “I summarize the information obtained and give a complete list of birds that I encountered. I specially emphasize the secondary contact zones of parajiatric species and hybridizing subspecies. . . .” The format and contents quoted above differ little from that by the author in an earlier paper (Haffer, Am. Mus. Novitates, 1967, no. 2294:1-2). The introductory portion has the greater significance as it synthesizes the secondary- contact itemizations which are scattered individually through the unevenly annotated list that is the second portion. On the positive side, the author clearly delineates his “core areas” and “suture zones.” He names the endemic species in each of 8 distribu- tional centers, amounting to a total of 195 in the forested lowlands of his entire trans- Andean “province.” Any species that is not one of the 195 trans-Andean endemics the reader must assume to he cis-Andean as he is given no alternative. On the negative side, the reader has no way of knowing which other birds — i.e. the great majority of the avifauna, including the hulk of the trans-Andean and most of the cis-Andean species — occur in or between any or all of the author’s distrihutional centers. For one thing, the information is mostly not there. For another thing, the author’s use of “trans-Andean” and “cis-Andean,” “forest” and “nonforest,” even “humid” and “dry” and “lowland” versus presumably cloud-forest highland, terms which must he understood in combination, can he confusing and contradictory. Take, for example, the explanatory footnote on the opening page: “The tropical low- lands and their faunas west of the Andes and in Middle America are designated as ‘trans- Andean’ or ‘Pacific’, and those east of the Andes as ‘cis-Andean’ or ‘Amazonian’.” Only- later does the reader realize that “trans-Andean” includes northern Colombia and Venezuela north of the Orinoco into Guyana, that “west of the Andes” may or may not include “north of the Andes,” and that in Middle America, “Pacific” means primarily “Caribliean.” He should keep in mind that a trans-Andean avifauna includes cis-Andean species; often it is not evident whether “trans-Andean” is intended in a restricted or in an unlimited sense. A trans-Andean species, we learn, is at least specifically distant, while a cis-Andean species is either undifferentiated from or is at most suhspecifically, distinct from a representative Amazonian population. This definition allows the cis-Andean element a short evolutionary life in which to remain what it is before transmuting into some- thing trans-Andean. Nevertheless the cis-Andean element dominates the trans-Andean element by a ratio of from 1.3:1 to 2:1 (following the author’s figures), while many trans-Andean endemics are not easily distinguishable at the specific level from cis- Andean representatives (p. 36). A paradoxical consequence is that in the cases of dis- agreement over the specific status of forms that are trans-Andean geographically, genetically they must he simultaneously trans-Andean to one authority and cis-Andean to the other. The author supplies no instance in which his 195 endemics may not have evolved from cis-Andean representatives. 188 THE WILSON BULLETIN • VoL 89, No. 1 Another consequence is that the reader is left to decide unaided the status of a con- tingent of Middle American species (e.g. Trogon melanocephalus, Piculus simplex, Celeus castaneus, Piprites griseiceps) belonging to South American genera that the author does not mention. By definition they cannot be cis-Andean, yet they are not listed among the trans-Andean. Similarly, the author does not include, for example, the Cathartidae, 2 members of which frequent forest, in a list of numbers of neotropical lowland forest birds 1 Table 3), perhaps because he tacitly considers them North American? The reader is ignorant not only of the specific composition of the trans-Andean avifauna in toto or per refugium but also of the species which the author for unstated reasons may or may not have included in his numerical totals. This may be the place to quote his view, shared by many, that “the faunal relationships of the densely forested Caribbean slope of Panama lie with the humid Pacific lowlands of northwestern South America,” a peculiarity that is equally conspicuous in the case of the herpetofauna (p. 67). Unfortunately, the faunal resemblance formula the author chose to select 1 Table 6) shows western Panama to be much closer to Caribbean Costa Rica than to central and eastern Panama and as close to southeastern Mexico as to the Choco lowlands on the Pacific side of Colombia. Central Panama and eastern Panama are both closer to the Cauca-Magdalena region at the drier northern end of the Andes than to the Choco lowlands on the wetter Pacific side. An illuminating comparison could perhaps have been attempted, whether by the reader or by the author, of the Choco and the Caribbean Costa Rican centers. The author assigns them respective forest-bird totals of 247 and 239, with 112 trans-Andean species included in the former and 104 in the latter, while the number of cis-Andean species, 135, is the same in both. Only 58 of the trans-Andean species are held in common; no figure is given for the cis-Andean species. The opportunity to analyze the replacement of large numbers of presumed ecological counterparts on a virtual one-to-one basis has been lost simply because of the lack of species lists. In regard to “forest” and “nonforest,” the author includes as forest birds a number of species — e.g. Laterallus albigularis, Jacana, Amazilia tzacatl, Synallaxis brachyura and S. erythrothorax, Geothlypis semi f lava — that I for one would not. Species the author lists here as forest birds he previously considered nonforest (Haffer, Hornero 10:315- 333, 1967). Some examples are Todirostrum sylvia, Thamnophilus doliatus, Galbula ruficauda, Basileuterus rivularis. The trans-Andean Ortalis garrula-cinereiceps complex, used by the author to illustrate speciation in Amazonian forest birds (Haffer, Science 165 (3889), 1969), he now specifies as nonforest on p. 67 but includes in a table of forest birds in secondary contact on p. 58. The reader need not face these gratuitous conceptual difficulties as he cannot, in any event, reconstitute the avifaunas in Tables 2, 3, and 6. As to “humid” and “dr> ,” the author states ( p. .30): “We are here concerned ex- clusively with an analysis of the avifauna of the trans-Andean humid lowland forests (Dry, Moist, Wet and Pluvial Forest in the Holdridge classification). . . .” Thus he sees fit to lump Holdridge’s Dry Forest with Holdridge’s humid to very wet forests, a largely unshared view. He does so, however, for the northern Colombian and western Venezuelan but not for the Middle American portions of his trans-Andean lowlands. Incidentally, the author uses the life-zone terminology and the numerical rainfall limits of Holdridge, yet Holdridge is not cited in the author’s very long list of References. Table 1 illustrates the author’s interpretation of the ecological distribution of 131 selected species in northern Colombia in dry forest on the east, moist forest in the center, and wet forest in the west. This table and the accompanying text effectively mislead the Murch 1977 • ORNITHOLOGICAL LITERATURE 1R9 reader early in his exeursion into “trans-Andea.” The reader can hardly avoid tin; erroneous inference that tlie species wliieli range throughout the Chocd wet forests, thence varying distances eastward into moist forest and sometimes into dry forest, are all Pacific or trans-Andean ; the species that range primarily in dry forest and varying distances in moist forest, rarely reaching the Choco wet forests, are all cis-Andcan, though it may occur to him later in retrospect that a number may he “dry” trans-Andean, an apparent contradiction in terms the author’s use of which makes it difficult to resolve. Another result of his use of the Holdridge terminology without appreciating the concepts is that, whatever may he the author’s idea of “lowland,” it is not that of Holdridge, who defines it by biotemperature values which, except perhaps in local pockets, do not reach the lower limits of the atmospheric manifestation known as “cloud forest.” A number of the author’s tropical lowland species — e.g. Bangsia arcaei, Popeluiria conversii, Proc- nias camnculata — are really subtropical inhabitants of bis Caribbean Costa Rican dis- tributional center (see Slud, Bull. Am. Mus. Nat. Hist. 128, 1964). Another species, Euphonia anneae, which is there subtropical, has also been found in the lowlands of eastern Panama. Thus the reader may wonder how the author may distribute other species which occur in the humid, sometimes in the “dry,” lowlands in South America and only under very humid, mountainous conditions in Central America. Finally, the wary reader should, among other items, take note that in Table 3 the head- ing “Caribbean Middle America” should be “Caribbean Costa Rica” and the use of “sympatric” for species known from areas the size of Surinam or southeastern Colombia is a misconception of the term; in Table 6, “southwestern Mexico” should be “south- eastern Mexico,” while “Guatemala” and the “Uraba region,” included in Table 2, are omitted here; in Table 8, the 2 species of Oncostonia and 1 of the forms of Geolhlypis semiflava are entered incorrectly. — Paul Slud. The Book of Birds. Five Centuries of Bird Illustration. By A. M. Lysaght. Phaidon Press, London, 1975: 208 pp., 142 illustrations (40 in color). $55.00. — A not infrequent dilemma for book reviewers in a specialized field such as ornithology, in which many of the authors are likely to be known personally to the reviewer, is whether or not to include “inside information” in a review, facts pertinent to some aspect of the publication that are not evident to other readers. This book is a case in point. Averil Lysaght is a good friend, and I was privileged to examine an advance copy of her book at her home in London. I drew up for her benefit, at that time, a list of typographical errors, misidcntifications, and other matters, so that nothing I may say in this review will come as a surprise to her. But 1 also had the benefit of a first-hand account of the problems she encountered in the course of getting out “The Book of Birds.” The first thing that needs to be said is that the utterly banal title was imposed upon her by the publisher — the subtitle (which appears on the title page but not on the binding or the dust jacket) is a more precise indication of the subject matter. After the publisher was committed to this project, with advance contracts, a deadline, and half a dozen color blocks already made, the original author withdrew. Ur. Lysaght was persuaded to take on the job, with much encouragement and assistance from Derek Goodwin of the British Museum (Natural History) and Gavin Bridson of the Linnean Society (London). The pressure to meet deadlines was such that the publisher actually told the author that she need not see proofs after she had handed in the typescript! 190 THE WILSON BULLETIN • Vol 89, No. 1 In view of these and other items recounted to me about the history of this book, its beauty and the obvious care and scholarship evidenced in the text, are even more impressive. As the subtitle indicates, it is basically a survey of bird illustration, up to the late 19th Century. Ironically, the oldest painting reproduced, from a 12th Century Chinese scroll in the British Museum Department of Oriental Antiquities, a Finch-billed Bulbul iSpizixos semitorques) perched and singing among flowers, is one of the most meticulously accurate renditions in the book. Many of the best-selling bird artists of today have less understanding of the proportions, scalation, and mechanics of a bird’s legs and feet than did the anonymous painter of this bulbul (possibly the Emperor Hui- tsung, 1082-1135). The book opens with a 22-page introduction, a rapid survey of man’s interest in birds from the most ancient times, gradually focusing on bird portrayal in particular. The emphasis at first is on birds in mythology and superstition, including the use of birds in ancient materia medica. There are such delightful tidbits as “Excrement of the cormorant mixed with lard was used lin ancient China] in the treatment of red noses resulting from too much wine.” And how frustrating to be told only that “in Turkey, erotic cults concerned with geese are still extant”! The text (introduction, notes to the plates, bibliography, index) is printed on a handsome dull-finished heavy gray paper, whereas the plates are on a white, faintly glossy stock. The margins are huge — the page measures 10%" by 14", but the type bed for the introduction is only 6" by 11M>”, printed toward the outer edge of the page. On the plate captions, the right margins are not justified. In a less expensive book, such design decisions might be considered extravagant, but the price of this book could scarcely have been significantly reduced had the 22 introductory pages been reduced by expansion of the type bed. And the visual effect is indeed striking. The plates are arranged in rough chronological order, but with many exceptions, |)robably a concession to layout design and to the mixture of color and black-and-white printing. Reproduction of the color plates, printed in Holland, seems excellent. Each caption includes the English and scientific names of the bird or birds portrayed (if identifiable), bibliographic details of the original (if an unpublished work, the library or collection in which the original is housed), the medium and size of the original, and a variable miscellany about the artist, the author of the book in which a plate was first published, something about the bird, and various historical notes. Dr. Lysaght is not an ornithologist, hut a specialist in the history of science, especially of the scientific and exj)loratory expt'ditions of the 18th and 19th centuries. This has meant that a few ornithological inaccuracies and misidentifications have slipped into the hook, but for this small price we are rewarded with information and anecdotes from a field of scholarshij) virtually unknown to most modern ornithologists. Identification of the rather crudely drawn birds in old woodcuts and engravings is sometimes almost impossible, and at other times uncertain. I believe “The Little Horn Owl” of plate 64 is probably a .^cops Owl iOtus scops) rather than a Short-eared Owl { Asio jlammeus) as identified in the caption, whereas the “Scops Owl” of plate 132 is definitely a Long-eared Owl i Asio otus) . On plate 104, only figure 1 is a manakin of the genus Chiroxiphia : figure 2, which one might deduce from the text of the caption to be a female manakin, appears to me to be a White-headed Marsh-Tyrant i Arundinicola leiicocephala) . The text for plate 127 is a mixture of facts about the Imperial Wood- pecker (Campephilus imperialis) and the portrayed Ivory-hilled Woodpecker iC. prin- cipalis). Contrary to the stat(Miient aecompanying plate 146, it is the New World and March 1977 • ORNITHOLOGICAL LITERATURE 1 91 not the Old World vultures that have a well developed olfactory sense. I found fewer than half a dozen typographical errors, and 3 or 4 obsolete scientific names. Most of the expected ornithological artists are represented — Gould, Wolf, Audubon, Catesby, Bewick, etc. — and there are unexpectedly beautiful or charming portraits by lesser known or even anonymous artists. One’s respect for 16th Century science is en- hanced by a 1555 engraving by one Pierre Belon, homologizing the skeletons of man and bird. An odd color drawing by J. D. Meyer of Nuremburg (1713-1754) shows a pair of Blue Tits iParus caeruleus) at the top of the plate and their skeletons, in the identical poses, at the bottom; the skeletons are done in far better detail than are the live birds. One could go on and on, listing the striking and the unexpected (the rather hackneyed Audubon Blue Jay, featured on the dust jacket, was not Dr. Lysaght’s choice but was one of the few plates imposed on her by the publisher). The somewhat overwhelming price of this book will unfortunately label it as a luxury item. Do not be misled by its oversize format and high price — it is emphatically not one of the virtually interchangeable and superfluous “coffee-table books” that have flooded the market in recent years ( at least half having originated in England, which seems odd in view of the straitened British economy of the 1970’s). Dr. Lysaght’s book fills admirably an unoccupied niche in ornithological literature. Any reader with an interest in avian iconography might well begin dropping hints to any affluent relatives about a suitable birthday present. — Kenneth C. Parkes. (N.B. — As of January 1977, this book appeared on at least one list of publisher’s overstocks, at less than half the original list price. This is a bargain worth hunting for.) Identification Guide to European Passerines, 2nd (revised) edition. By Lars Svensson, illus. by the author. Naturhistoriska Riksmuseet, S 104 05 Stockholm 50, Sweden, 1975: 184 pp., ca. 160 line drawings. Price not given. — This compact (10 X 19 cm) identification guide by one of Europe’s most noted ornithologists will be of use to a strictly limited number of people in the United States. It covers 180 species and 35 subspecies of European passerines and is primarily intended to present ageing and sexing characters for fledged birds in the hand. It is thus of little value for field ornithology but covers all the likely species a bird bander or museum worker might encounter. The first 10 pages of “directions for use” and 15 pages of “general techniques for ageing and sexing” present a detailed explanation and discussion of the different methods used in the “systematic list” of species that follows. These introductory pages require careful reading if one is not familiar with the symbols and abbreviations used in the first edition of this book; however, all methods are fully explained and the practical problems discussed. The author follows the nomenclature of C. Vaurie, "'The Birds of the Palaearctic Fauna,” Passeriformes (1959), and he presents characters to identify the species (in difficult cases), the age, and the sex, where possible on external charac- ters. There is also a summary of the molt regime, and wing lengths with sample sizes are usually given. Nearly every species description is accompanied by one or more text figures illustrating particular points of difference. A comprehensive 9 p. of important references are followed by an index of scientific names only. Three controversial methods are perhaps worth singling out from an otherwise good presentation. Firstly, Svensson elects to number primary feathers ascendantly from wing tip towards the body, while most modern molt studies employ the descendant order, i.e. that in which the feathers are usually replaced. Secondly, he gives the impres- sion that a live bird should be held with its head towards the bander’s wrist during 192 THE WILSON BULLETIN • VoL 89, No. 1 measurements, whereas most North American and British banders prefer an alternate grip with the head held between the first 2 fingers and the body in the palm of the hand. Finally, the tail shape character used in ageing some birds, e.g. genus Turdus, may be even more difficult in practice than the author states. Apart from these, and a few minor errors, the overall impression is of an excellent. specialist reference work embodying all the most recent information on ageing and sexing from a wide range of European sources. North American banders are doubtless eagerly awaiting a similarly complete and modern reference for nearctic passerines. — Trevor L. Lloyd-Evans. State Laws as They Pertain to Scientific Collecting Permits. By M. Houston McGaugh and Hugh H. Genoways. Museology No. 2, 1976: 81 pp. Order from the Museum Shop, The Museum, Texas Tech University, Lubbock, TX 79409. $2.00 — For each state as well as Puerto Rico and the Virgin Islands, the laws pertaining to collect- ing animals and plants are given, along with lists of protected species and game animals. The addresses of relevant state agencies are given so that one may write for further information or permit applications. These state laws are in addition to federal require- ments, which the collector must also take into account when planning research. — R. J. R. A Guide to the Birds of Panama. By Robert S. Ridgely, illus. by John A. Gwynne, Jr. Princeton University Press, Princeton, N.J., 1976: xv + 394 pp., 32 color plates, many black-and-white drawings. $15.00. — Although smaller than the state of South Carolina, the Republic of Panama has a remarkably large avifauna of some 883 species, mainly because of its geographical location as an area of overlap between North and .South American faunas. This hook is a one-volume compendium of information useful both as a field guide and general reference. The author gives full credit and appreciation to Eugene Eisenmann, who encouraged him to write the book, and provided extensive field notes that Ridgely has incorporated into his text. Following a foreword by Alexander Wetmore and an introduction, there are chapters on Climate, Migration and Local Movements, Conservation, and on the Plan of the Book. The main text consists of species accounts organized by family. For each species there is a concise description, and sections on similar species, status and distribution, habits, and range. This is fol- lowed by an appendix listing and briefly describing additional species of southern Middle America that are not known to occur in Panama. A second appendix tells where to find birds in Panama, and includes such useful information as where it is safe to drink the water. The book ends with a bibliography and an index. The color plates by Gwynne are conveniently grouped in the center of the book, and are a major contribution to the volume. The paintings are attractive, and the birds appear lifelike. These plates should be of great value in field identification, though it is obvious that some groups, such as the hummingbirds, woodcreepers, and flycatchers will pose special i)roblems to the inexperienced observer because of the many similar species in each group. Measuring about 16 X 23 cm the hook is a bit large to carry easily in the field, yet from the amount of information included and the efficient, compact design of the book, it is clear that there is no wasted space. Students of tropical American ornithology will welcome this attractive addition to the literature of the area. — Robert J. Raikow. 'I'liis issue of The If ilson Bulletin was published on 7 April 1977. The Wilson Bulletin Editor Jerome A. Jackson Department of Zoology P.O. Drawer Z Mississippi State University Mississippi State, MS 39762 Editorial Assistants Bette J. Sciiardien Patricia Ramey Bonnie J. Turner Martha Hays Wayne C. Weber Review Editor Robert Raikow^ Color Plate Editor William A. Lunk Department of Life Sciences 865 North Wagner Road University of Pittsburgh Ann Arbor, MI 48103 Pittsburgh, PA 15213 Suggestions to Authors See Wilson Bulletin, 87:144, 1975 for more detailed “Suggestions to Authors.” Manuscripts intended for publication in The Wilson Bulletin should be submitted in dupli- cate, neatly typewritten, double-spaced, with at least 3 cm margins, and on one side only of good quality white paper. Do not submit xerographic copies that are made on slick, heavy paper. Tables should be typed on separate sheets, and should be narrow and deep rather than wdde and shallow. Follow the AOU Check-list (Fifth Edition, 1957) and the 32nd Supplement (Auk, 90:411-419, 1973), insofar as scientific names of U.S. and Canadian birds are concerned. Summaries of major papers should be brief but quotable. Where fewer than 5 papers are cited, the citations may be included in the text. All citations in “General Notes” should be included in the text. Follow carefully the style used in this issue in listing the literature cited; otherwise, follow the “CBE Style Manual” (1972, AIBS). Photographs for illustrations should have good contrast and be on gloss paper. Submit prints unmounted and attach to each a brief but adequate legend. Do not write heavily on the backs of photographs. Diagrams and line drawings should be in black ink and their lettering large enough to permit reduction. Original figures or photographs submitted must be smaller than 22 X 28 cm. Alterations in copy after the type has been set must be charged to the author. Notice of Change of Address If your address changes, notify the Society immediately. Send your complete new address to the Treasurer, Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. He will notify the printer. The permanent mailing address of the Wilson Ornithological Society is: c/o The Museum of Zoology, The University of Michigan, Ann Arbor, Michigan 48104. Persons having business with any of the officers may address them at their various addresses given on the back of the front cover, and all matters pertaining to the Bulletin should be sent directly to the Editor. CONTENTS NOTES ON THE BIOLOGY AND POPULATION STATUS OF THE MONKEY-EATING EAGLE OF THE PHILIPPINES ..Robert S. Kennedy 1 COWBIRD PARASITISM AND EGG RECOGNITION OF THE NORTHERN ORIOLE .... Stephen I. Rothstein 21 OBSERVATIONS ON THE RED-NECKED GREBE NESTING IN MICHIGAN Michael L. Chamberlin 33 GROWTH AND DEVELOPMENT OF THE PLAIN CHACHALACA IN SOUTH TEXAS .... Wayne R. Marion 47 SOCIAL DOMINANCE IN WINTER FLOCKS OF cassin’s FINCH Fred B. Samson 57 INTER-BROOD MOVEMENTS OF JUVENILE SPRUCE GROUSE Daniel M. Keppie 67 BREEDING BIOLOGY OF YEAR-OLD AND OLDER FEMALE RED-WINGED AND YELLOW-HEADED BLACKBIRDS Richard D. Crawford 73 FOOD HABITS OF OLDSQUAWS WINTERING ON LAKE MICHIGAN Steven R. Peterson and Robert S. Ellarson 81 SEXUAL DIMORPHISM IN THE WHITE IBIS James A. Kushlan 92 EGGSHELL THICKNESS VARIABILITY IN THE WHITE-FACED IBIS David E. Capen 99 CHARACTERISTICS OF A WINTERING POPULATION OF WHITE-TAILED PTARMIGAN IN COLORADO Richard W . Hoffman and Clait E. Braun 107 BREEDING DENSITIES AND MIGRATION PERIODS OF COMMON SNIPE IN COLORADO Bruce R. Johnson and Ronald A. Ryder 116 PRINCIPAL COMPONENT ANALYSIS OF WOODPECKER NESTING HABITAT Richard N. Conner and Curtis S. Adkisson 122 PHENETIC ANALYSIS OF THE SUBFAMILY CARDINALINAE USING EXTERNAL AND SKELETAL CHARACTERS Jenna J. Hellack and Gary D. Schnell 130 GENERAL NOTES WHY OSPREYS HOVER Thomas C. Grubb, Jr. 149 STORAGE OF PINON NUTS BY THE ACORN WOODPECKER IN NEW MEXICO Peter B. Stacey and Roxana Jansma 150 FLOCKING AND FORAGING IN THE SCARLET-RUMPED TANAGER David J. Moriarty 151 YELLOW WARBLER NEST USED BY A LEAST FLYCATCHER J. Paul GoOSSen 153 PINTAIL REPRODUCTION HAMPERED BY SNOWFALL AND AGRICULTURE Gary L. KrapU 154 TICKS AS A FACTOR IN THE 1975 NESTING FAILURE OF TEXAS BROWN PELICANS Kirke A. King, David R. Blankinship, Richard T. Paul, and Robin C. A. Rice 157 PRAIRIE WARBLER FEEDS FROM SPIDER WEB John F. DoUglaSS 158 NOTES ON THE HUMMINGBIRDS OF MONTEVERDE, CORDILLERA DE TILARAN, COSTA RICA Peter Feinsinger 159 I NEST-SITE DIFFERENCES BETWEEN RED-HEADED AND RED-BELLIED WOODPECKERS IN SOUTH CAROLINA Lawrence Kilham 164 GROUND FORAGING AND RAPID MOLT IN THE CHUCK-WILL’s WIDOW Sievert Rohwer and James Butler 165 | FEEDING RESPONSES OF FALL MIGRANTS TO PROLONGED INCLEMENT WEATHER Elliot J. Tramer and Flora E. Tramer 166 ^ SOUTHBOUND MIGRATION OF SHOREBIRDS FROM THE GULF OF ST, LAWRENCE Raymond McNeil and Jean Burton 167 FLOCKING AND FORAGING BEHAVIOR OF BROWN JAYS IN NORTHEASTERN MEXICO Michael L. Morrison and R. Douglas Slack 171 DO MORE BIRDS PRODUCE FEWER YOUNG? A COMMENT ON MAYFIELD’s MEASURE OF NEST SUCCESS Richard F. Green 173 MORTALITY OF NESTLING MISSISSIPPI KITES BY ANTS James W. Parker 176 ORNITHOLOGICAL NEWS 177 179 ORNITHOLOGICAL LITERATURE Tfie WIson Bulletin PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY VOL. 89, NO. 2 JUNE 1977 PAGES 193-372 MUS. COMP. zool:. LIBRARV JUL 131977 HARVARD The Wilson Ornithological Society Founded December 3, 1888 Named after ALEXANDER WILSON, the first American Ornithologist. President — Andrew J. Berger, Department of Zoology, University of Hawaii, Honolulu, Hawaii 96822. First Vice-President — Douglas A. James, Department of Zoology, University of Arkansas, Fayetteville, Arkansas 72703. Second Vice-President — George A. Hall, Department of Chemistry, West Virginia Uni- versity, Morgantown, W. Va. 26506. Editor — Jerome A. Jackson, Department of Zoology, P.O. Drawer Z, Mississippi State Uni- versity, Mississippi State, Mississippi 39762 Secretary — James Tate, Jr., P.O. Box 2043, Denver, Colorado 80201. Treasurer — Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. Elected Council Members — Helmut C. Mueller (term expires 1977) ; Abbott S. Gaunt (term expires 1978) ; James R. Karr (term expires 1979). Membership dues per calendar year are: Active, $10.00; Sustaining, $15.00; Life memberships, $200 (payable in four installments). The Wilson Bulletin is sent to all members not in arrears for dues. The Josselyn Van Tyne Memorial Library The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of Michigan Museum of Zoology, was established in concurrence with the University of Michigan in 1930. Until 1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines from members and friends of the Society. Now two members have generously established a fund for the purchase of new books; members and friends are invited to maintain the fund by regular contribution, thus making available to all Society members the more important new books on ornithology and related subjects. The fund will be administered by the Library Committee, which will be happy to receive suggestions on the choice of new books to be added to the Library. William A. Lunk, University Museums, University of Michi- gan, is Chairman of the Committee. The Library currently receives 104 periodicals as gifts and in exchange for The Wilson Bulletin. With the usual exception of rare books, any item in the Library may be borrowed by members of the Society and will be sent prepaid (by the University of Michigan) to any address in the United States, its possessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers, as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to “The Josselyn Van Tyne Memorial Library, University of Michigan Museum of Zoology, Ann Arbor, Michigan.” Contributions to the New Book Fund should be sent to the Treasurer (small sums in stamps are acceptable). A complete index of the Library’s holdings was printed in the September 1952 issue of The Wilson Bulletin and newly acquired books are listed periodically. The Wilson Bulletin The official organ of the Wilson Ornithological Society, published quarterly, in March, June, September, and December. The subscription price, both in the United States and elsewhere, is $15.00 per year. Single copies, S4.00. Subscriptions, changes of address and claims for undelivered copies should be sent to the Treasurer. Most back issues of the Bulletin are available and may be ordered from the Treasurer. Special prices will be quoted for quantity orders. All articles and communications for publications, books and publications for reviews should be addressed to the Editor. Exchanges should be addressed to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, Michigan. Known office of publication : Department of Zoology, Mississippi State University, Mississippi State, Mississippi 37962. Second class postage paid at Mississippi State, Mississippi and at additional mailing office. Allen Press, Inc., Lawrence, Kansas 66044 Immature (above) and adult (below) Oahu 'Elepaio, Chasiempis sandwichensis gayi. Painting by Doug Pratt. THE WILSON BULLETIN A QUARTERLY MAGAZINE OF ORNITHOLOGY Published by the Wilson Ornithological Society VoL. 89, No. 2 June 1977 Pages 193-372 THE BREEDING BIOLOGY OF THE OAHU ‘ELEPAIO Sheila Conant The monotypic genus Chasiempis (Muscicapidae) is endemic to the Ha- waiian archipelago, and represented by 3 subspecies: Chasiempis sandwich- ensis sclateri on Kauai, C. s. gayi on Oahu, and C. s. sandwichensis on Hawaii. Commonly called the ‘Elepaio, Chasiempis is believed to be related to the genus Monarcha ( Mayr 1943, Amadon 1950), and to be of Melanesian origin by way of Polynesia or Micronesia ( Baker 1951) . C. s. gayi (see frontispiece) was described by Wilson (1891), and is endemic to the island of Oahu. Despite its wide distribution and relative abundance, this bird has received little attention from field ornithologists. Photographs of the nest and eggs of C. 5. sandwichensis and of the nest, eggs, and young of C. s. sclateri have been published only recently ( Berger 1969, 1972). METHODS AND STUDY AREA Breeding biology data on the Oahu ‘Elepaio were gathered by field observation begin- ning in November 1965 and ending in May 1968. Brief observations of the Kauai ‘Elepaio were made for 3 days in April 1967, 10 days in June 1967, and 7 days in July 1975. From October 1972 to August 1973, and for about 1 week per month from July 1974 to March 1975 I collected data on the Hawaii ‘Elepaio while conducting field studies of avian communities on the east flank of Mauna Loa on the island of Hawaii. Although this paper reports primarily the results of my studies of the Oahu population, I have drawn upon observations of the other 2 subspecies to add to my discussion of some topics. Specific information on a particular subspecies will be identified as such. The main study area consisted of about 70 ha on the northeast side of Manoa Valley on the island of Oahu, and I made occasional trips to other areas. The vegetation of the study area was mesic forest with grassy clearings. Most plants in the area were intro- duced species. I recorded song and call-notes with 2 portable tape recorders and 2 types of micro- phones: a Nagra HI with an Electrovoice 655 microphone and a Uher 4000L with a 514 microphone. To determine song patterns in relation to sex and phase of the breeding season I recorded the number of songs given by each member of a pair during the same 193 194 THE WILSON BULLETIN • VoL 89, No, 2, June 1977 hours of the day in each breeding season phase for a total of about 30 h. I also recorded the number of songs and nest material additions by the members of pairs for about 25 h during the first week of construction of several nests. Plumage, behavior, and egg-laying permited me to sex the birds while data were recorded. Only 3 birds were banded and color-banded during the study: 1 bird in immature plumage caught in a mist net in November 1966 and 2 nestlings in May 1967. A total of 32 nests were found, 11 of which were collected and measured. Fifty-three eggs in 26 clutches were observed, though none was measured. A small, moveable mirror attached to a length of pipe was used to observe the contents of some inaccessible nests. In nests accessible by using ladders and by climbing trees, I marked 12 eggs as they were laid. DIET AND FEEDING BEHAVIOR Although ‘Elepaio forage at all canopy levels in the forest, (Perkins 1903, MacCaughey 1919, this study) they were most often seen in the lower story. During 33.3 min of feeding behavior observations of the Hawaii ‘Elepaio, birds spent 23.7 min feeding between 0 and 3 m in the canopy, 8.3 min be- tween 3 and 6 m, and 1.3 min above 6 m. MacCaughey ( 1919) and Richard- son and Bowles ( 1964 ) mentioned the association of ‘Elepaio with various plant species, but Perkins (1903) felt, and I concur, that ‘Elepaio are op- portunistic feeders, that is, they are most likely to be found where insect den- sities are highest, regardless of the plant species involved. MacCaughey (1919) and the Hawaii Audubon Society (1967 — probably based on Mac- Caughey 1919) reported nectar feeding, but Munro (1944) and I never saw ‘Elepaio eat anything other than animal material, especially insects. After examining stomach contents and observing feeding behavior, Perkins (1903) reported that ‘Elepaio fed on a variety of insects as well as arachnids, chilo- pods, diplopods, and some molluscs. DISTRIBUTION AND MOVEMENTS The ‘Elepaio is sedentary (Henshaw 1902b), and was reported by Mac- Caughey (1919) to have the widest altitudinal range of any native forest bird. This is probably still true of all 3 subspecies, although since 1968 I have noticed that densities, though not range, of the Oahu ‘Elepaio have de- creased. The ‘Elepaio can still be found in the backs of most valleys and on ridges on Oahu, but, whereas 8 years ago one would hear or see several pairs on a 5 or 6 km ridge trail, only one, sometimes none, will appear today. The ‘Elepaio’s adaptability to a wide variety of foods (Perkins 1903), habitats, elevations, and weather conditions (MacCaughey 1919, Richardson and Bowles 1964, Berger 1972:112, 114) is undoubtedly responsible for the success of the species and its wide distribution in comparison to other native forest birds. Henshaw (T902a) predicted that it would persist in substantial numbers after other endemic passerines were rare or extinct. During the non-breeding season ‘Elepaio remained in or near the same area Conant • THE OAHU ‘ELEPAIO 195 Fig. 1. Territories occupied by breeding pairs of ‘Elepaio in Manoa Valley, island of Oahu, Hawaii. as their breeding or hatching territory the entire year, but I don’t know if they always defended territories. From July through December the birds traveled in pairs or family groups, presumably depending on whether or not any young had survived from the previous breeding season. These small flocks disintegrated as males established breeding territories (Fig. 1) by sing- ing in late December or in January. The 2 nestlings banded in May 1967 and their parents exhibited this pattern of behavior. The nestlings hatched and one (the second was not seen after July 1967) remained in or near terri- tory 6 with its parents after the 1967 breeding season. In January 1968 the banded bird established territory 5, pairing with an adult female (pair 5-68 — number of pair combines territory number and year of breeding). The pre- sumed parents of this bird (6-67 and 6-68) reestablished their former terri- tory, and bred there. Pair 5-68 renested 3 times. After the destruction of the 4th nest, the banded male did not renest, but remained in territory 5, and his mate was not seen again. VOCALIZATIONS According to Henshaw (1902a), the ‘Elepaio’s name is the Hawaiian trans- literation of its primary song, a short, melodious whistle. Perkins (1903) 196 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Fig. 2. Number of songs given by both sexes during each hour of the day, during the first week of nest building. observed that birds often broke into song after intense chatter. I have heard 4 types of vocalizations; chip, chatter, song, and the alarm call. The most common vocalization heard was chatter, which ranged widely in pitch, fre- quency, and volume. Chatter often preceded songs, which were usually uttered in series of threes. Chatter, unaccompanied by song, was frequent as the birds moved about feeding. Frequent, loud, and excited chatter indicated mild alarm. The alarm call, a short raspy cry, was given by adults and young only when the birds were extremely disturbed, as when the young in the nest were handled or photographed. A high-pitched, soft chatter served as a food-begging call given by nestlings and fledglings, and was similar to the vocalizations of a female indicating readiness for copulation. While feeding and moving about, ‘Elepaio often uttered a single, low-pitched chip, appar- ently as a contact note, maintaining flock or pair contiguity. This chip was also given immediately before birds exchanged places on the nest to incubate or brood, although the male often sang instead. Repeated chips appeared to function as alert notes when a bird was mildly excited or disturbed. The ‘Elepaio was invariably the first bird to sing in the morning, and the last to sing in the evening. During the breeding season I determined terri- torial boundaries by listening to morning song periods that began before 06:00 and lasted 30 to 45 min, and evening song periods that began at about Conant • THE OAHU ‘ELEI’AIO 197 COURTSHIP AND NEST SITE SELECTION NEST BUILDING INCUBATION NESTLING PERIOD FLEDGLING PERIOD AVERAGE NUMBER OF SONGS PER HOUR Fig. 3. Average hourly frequencies of songs during the different phases of the breeding cycle. 18:00 and lasted about 20 min. After the morning song period, singing was rather infrequent, except during nest construction. At this time both birds, especially the male, sang frequently throughout the day. Fig. 2 illustrates the average number of songs given by both sexes during a day of nest build- ing. Fig. 3 shows the hourly frequencies of song by both sexes during each phase of the breeding season. During incubation singing was infrequent, ex- cept that males often sang once or twice during a nest exchange. The female sometimes sang on the nest in response to a long, excited series of songs and chatter given by the male, and incubating males often responded by singing to the song or chatter of a mate or of an ‘Elepaio in an adjacent territory. The frequency of singing by the parents increased after the young hatched and until they became independent. When the birds were not breeding, song was rare. Although I did not re- cord daily song frequency during the non-breeding season, I noted that morn- ing and evening singing periods decreased in duration and song frequency in June, increasing again in December each year. ‘Elepaio responded to tape recordings of chatter with chattering and, occa- sionally, with singing, especially during the breeding season. TERRITORY AND NESTING HABITAT ‘Elepaio territory should be considered Type A territory (Berger 1961), that is, courtship, copulation, nesting, and feeding occurred within the terri- 198 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 tory. Territory was defined by the male, although the female’s movements, as she was followed by her singing mate, apparently influenced the location of boundaries. This was especially true in cases where the pair had remained together since the last breeding season. Territorial defense, primarily by song, was shared by both sexes, although the male was slightly more am- bitious in defense. Territorial encounters usually resulted in pursuit, the defender chasing the intruder chattering excitedly. I observed such terri- torial encounters more often in the Hawaii ‘Elepaio than in the Oahu ‘Elepaio. Red-billed Leiothrix (Leiothrix lutea) and Japanese White-eye [Zoster ops japonica) were silently chased from the area within about 8 m of the nest itself, but other species were usually ignored. I saw birds of these species remove material from ‘Elepaio nests twice and once, respectively. Seale (1900 ) found that ‘Elepaio often chased larger birds from feeding grounds, and MacCaughey (1919) reported that ‘Elepaio engaged in intraspecific chasing that was both territorial and sexual in nature. ‘Elepaio were quite disturbed at human intervention into their territory, particularly near the nest. Several times while nest 3 was being photographed with eggs or young in it, the male attacked the photographer, scratching face and arms with its claws, chattering, giving the alarm call, and fluffing its feathers. The female chattered and fluffed her feathers on these occasions. Fig. 1 illustrates 11 territories occupied by the breeding pairs studied. It shows habitat type and size of each territory. The size of the territory varied from 1.2 to 2.9 ha, averaging 2.0 ha. C. van Riper ( pers. comm.) reported territory size of 17 pairs of Hawaii ‘Elepaio to range from 0.65 to 1.46 ha averaging 1.08 ha. This difference between average territory sizes may be re- lated to the fact that his work was done in a sparse, savanna-like, dry forest, in contrast to the dense mesic forest of the present study. The shape of the territory appeared to be influenced by the distribution of clearings and unsuitable nesting habitat. Territories included small clearings ( < 50 m-), but never encompassed larger clearings, which provided a natural boundary between pairs (Fig. 1: territories 3, 4, and 5). The same reasoning applied to unsuitable nesting habitat, that is, either Eucalyptus sp. or paper bark ( Melaleuca leucadendron ) forest in the study area. Suitable ‘Elepaio nesting habitat consisted of dense, mesic forest with thick undercover. The forest was a mixture of several trees: Java plum [Eugenia cum ini) , kukui [Aleurites moluccana) , fiddlewood (Citharexylum spinosum) and guava [Psidium guajava). Undercover included a variety of shrubs and grasses: thimbleberry {Rubus rosaefolius) , ti plant iCordyline terminalis } , and palm grass [Setaria palmifolia ) , as well as tree saplings. This descrip- tion of habitat applies only to the study area — ‘Elepaio habitat varied through- out Oahu. Conant • THE OAHU ‘ELEI’AIO 199 COURTSHIP AND COPULATION Courtship began in January, consisting mainly of singing and chasing. As the male established the territory he attracted a female by singing, or re- peated pair formation behavior with his mate of the previous year. In the initial phase of pair formation, 2 birds began associating regularly. Later, patterns of vocal interchange became apparent. This started with chipping, and led to loud, frequent, excited chatter, after which one of the pair sang, the other usually responding by singing. As these singing bouts became more frequent, the birds chased each other in a series of short, swift flights. This chasing was the most characteristic part of courtship, was usually initiated by the male, and was always accompanied by excited chatter and usually by song. Chasing bouts lasted from 20-30 sec to several min, sometimes with short feeding breaks. The single copulation I observed occurred at 07:20 in the morning about 30-60 min before the female of pair 6-67 laid her second egg in nest 8. As the pair was feeding, the female began low-pitched, excited chatter. After about 30 sec of chattering, to which the male responded in like fashion, the female began to follow him as he hopped from branch to branch. She crouched each time she perched, spreading and quivering her wings and lower- ing her tail. The male eventually mounted. Mounting and copulation took about 5 sec. After copulation both birds chattered briefly, and the male sang twice. They resumed feeding until I left at 08:00. When the nest, which had contained 1 egg at 08:00, was checked at 08:35, it had 2 eggs in it. The single copulation I observed in the Hawaii ‘Elepaio suggested that the be- havior patterns of the 2 subspecies are essentially the same. NEST SITE AND NEST CONSTRUCTION Nest site selection. — After pairing, the birds began to examine prospective nest sites. The female perched in tree forks or on lateral branches among supporting branches. At each stop she chipped or chattered, and exhibited the nest-molding movement (described below) that both birds used later dur- ing nest construction to shape the nest cup. During this time the male usually remained within a few meters of his mate, feeding and answering her vocaliza- tions with similar ones. Occasionally he sang and the female answered with song or chatter. When she began the nest molding movement, singing and chattering became very intense and lasted up to 10 min. After the site had apparently been chosen, the pair repeatedly returned to it and exhibited this same behavior. The ultimate point in this nest site selection pattern was the beginning of nest construction by the female. Nest construction. — Nests were placed in forks or on lateral branches. The former type of placement was the most common: of the 32 nests observed 27 200 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 25 23 21 19 17 15 13 11 9 7 5 3 1 1 3 5 7 9 11 13 15 17 19 21 23 25 27 NUMBER OF NEST MATERIAL ADDITIONS Fig. 4. Number of additions of nest material liy lioth sexes during each hour of the day, during the first week of nest building. were in forks. Nest height, the distance from the bottom of the nest to the ground directly helow^ the nest, varied considerably. Heights ranged from 2.8 m to 15 m, averaging 7.6 m. Henshaw ( 1902a j stated that the ‘Elepaio habitually nested low. He recorded one nest at less than 0.6 m, hut noted that this was unusually low'. According to Perkins (1903) nests were placed at heights of 1.8 to 12.2 m. The nest was built by both sexes, although the female did most of the construction. During the first week of nest building both sexes added ma- terial freiiuently throughout the day. Fig. 4 shows the number of additions of nest material by both sexes during the day. An ‘Elepaio often chipped as it worked material into the nest, sometimes singing on the nest or on a perch nearby after the addition. This behavior was especially characteristic of the male, and accounted for the ease with which 1 often found nests under construction. ‘Elepaio used large amounts of spider web and some spider egg cases in the nest. This material strengthened the nest without lending rigidity to it. Web was added with a wiping motion of the head, which served to smooth and shape the outside of the nest. When the nest was nearly complete, the Conaiu • THE OAHU ‘ELEPAIO 201 Table 1 Species of Trees Used as Nest Sites by Chasiempis SANDWICHENSIS GAY I Species of Tree Number of Nests in Each Species Aleurites moluccana (kukui) 4 Citharexylum spinosum (fiddlewood) 7 Eugenia cumini (Java plum) 13 Grevillea robusta (silk oak) 1 Macadamia ternifolia (macadamia) 1 Mangifera indica (mango) 1 Psidium guajava (guava) 5 bird perched in the nest cup, stretched its head out over the edge and used the bill to smooth the entire surface of the nest. The birds shaped the nest cup by sitting in it, erecting the body plumage, and pushing against the sur- face of the material while shaking the entire body. Generally the birds did not seem disturbed when I was observing nest building at a distance; however, pair 5-68 deserted nest 27, which was under construction, on the day I spent 2 hours photographing the birds with a strobe light at about 4 m distance. Most nest construction was complete in a week, although the first egg was usually not laid until about 2 weeks after construction had begun. Small amounts of spider web, lichen, and fine lining material were added during the second week of construction. If the first nest of the season for a pair of ‘Elepaio were destroyed or deserted and several days of heavy rains and high winds delayed the construction of a new nest, the new nest was completed and the new clutch laid within 17 days of the date of destruction. In mild weather the time taken to complete the clutch of a renesting was about 11 or 12 days. The 32 ‘Elepaio nests I found occurred in 7 species of trees (Table Ij. Java plum was the tree most often used as a nest site. All nest trees were introduced species. An analysis of nest 30 revealed that it was composed primarily of the bark of the paper bark tree and the leaves of a grass. The territory from which this nest was taken, territory 5 (Fig. 1), was adjacent to a stand of paper bark trees. The outside of the nest was partly covered with lichen, a liverwort, and the “pulu” (soft, glossy “wool” on the bases of the fronds) of the tree fern iCibotium sp.). The nest was lined with fine rootlets, possibly of a grass, and animal hair. C. van Riper (1977) also found animal hair in Hawaii ‘Elepaio nests. I have found leaf skeletons in some nests. 202 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Nest measurements. — Nine of the 11 nests measured had been built in forks, 2 on lateral branches. Average dimensions were: total width, 7.2 cm; total depth, 7.8 cm; cup width 4.6 cm; cup depth, 4.1 cm; wall thickness 1.3 cm. Of previous workers, Bryan ( 1905) recorded the greatest variations in nest size; he also found more nests placed in upright forks than on lateral branches. Nests reported by Bryan (1905 ) averaged 7.6 cm in width and height with a cup depth and width of 4.6 cm. LAYING AND INCUBATION Eggs and egg laying. — Of the 16 C. s. gayi clutches I observed, 15 contained 2 eggs, nest 29 had 3 eggs. Two C. s. sclateri clutches I observed had 2 eggs. I have recorded three 2-egg clutches for the Hawaii ‘Elepaio and observed 5 pairs of this subspecies feeding 2 young each. Eddinger found eight 2-egg and one 3-egg clutches, and Berger observed 2 pairs feeding 2 young each of the Kauai subspecies (Berger 1972:111). According to Perkins (1903) and MacCaughey ( 1919) a 2-egg clutch was usual, but Henshaw (1902a) recorded one 3-egg clutch. The first egg laid in nest 32 was destroyed on the day it was laid, and I broke one of the eggs in nest 14 on the day the second egg was laid. Both pairs of birds began incubating and no more eggs were laid, although the pair at nest 14 deserted (for no obvious reason) 4 days after the second egg was laid. This indicated that ‘Elepaio are determinate layers. The eggs were white with reddish-brown spots concentrated at the larger end (Fig. 5). No eggs were measured in this study, but Newton (1897) re- ported length to range from 2.1 to 2.2 cm and diameter to be from 1.5 to 1.6 cm. Eddinger measured 2 Kauai ‘Elepaio eggs at 2.04 X 1.52 and 2.05 X 1.53 cm (Berger 1972:111). Rothschild (1893-1900) obviously erred in reporting that eggs measured 1.25 by 1.11 in. Such egg dimensions would be truly remarkable for a 16.4 cm passerine bird. Unfortunately this mistake was carried over at least once (MacCaughey 1919). Eggs were laid on consecutive days, 2 to 3 days after the birds had stopped adding material to the nest. I recorded the approximate time of laying of the first egg in 2 clutches and the exact time of laying of the second egg in 6 clutches. All these eggs were laid between 06:30 and 08:30. The first egg in nest 5 was laid before 07:30 on 20 April 1966, and the second egg between 06:38 and 06:45 on 21 April 1966; thus, the interval between layings was close to 24 h. The earliest observed clutch of a season was completed on 18 February 1966, the latest on 9 May 1966. The median date of completion was 23 March for 12 clutches recorded over 3 years. The clutch of nest 5 was the latest clutch laid in any year. This nest was deserted apparently because the male, which had begun to molt, did not Conan t • THE OAHU ‘ELEPAIO 203 Fig. 5. The clutch of nest 5. incubate. After several days of incubating by herself, the female deserted. Both eggs were fertile. The male came regularly to the nest, as if to incubate, remaining the entire time the female was gone, but did not sit on the eggs. Nest desertion by this pair was probably due to the physiological changes manifested by a decrease in reproductive behavior — to be expected as the breeding season drew to a close. Incubation. — The first egg in nest 16 was laid on 6 April 1967, the second on the following day. The eggs had not hatched by 08:00 21 April 1967, but were hatched by 17:00 23 April 1967. Thus, the incubation period, as defined by Nice (1953), was between 14 and 16 days. Eddinger recorded the incuba- tion period at 3 Kauai ‘Elepaio nests as 18 days (Berger 1972:111). More incubation period data are needed to clarify the reasons for this subspecific variation, if, indeed it is real. Incubation began immediately after the second egg was laid, and females spent the night on the nest only after laying the second egg. Occasionally a female sat on the nest for short periods, less than a total of 1 h, on the day the first egg was laid. Males and females incu- bated in shifts, and the eggs were covered almost constantly. Single attentive periods varied from 2 to 44 min. The average attentive period of males was 204 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Fig. 6. ‘Elepaio nestlings at nest 29, approximately 10 days old. 12 min, for females 18 min. For the first 3 days of incubation the attentive periods averaged less than 14 min, hut after this became longer. Throughout the incubation period the eggs were incubated 96% of the total time (6 h, 26 min) I observed nests. In most cases, when one bird replaced the other, the eggs were uncovered less than 30 sec. On 3 occasions I saw males of 2 dif- ferent pairs feed their incubating mates. CARE OF THE YOUNG At hatching the nestlings had dark pink skin with a sparse covering of black down on the head and dorsal body surface. The gape was orange, the rictus cream. Fig. 6 is a photograph of the 3 nestlings at nest 28 when they were about 10 days old. When the young were ready to leave the nest their wing, back, and head feathers were dark brown at the base and ochreous at the ends. The underparts were huffy white, and the short tail feathers were dark brown. Fecal sacs of nestlings were white with brownish-black tips. The fledgling’s head was brown and underparts whitish. There were black down feathers scattered about the head and body (Fig. 7). Conant • THE OAHU ‘ELEPAIO 205 Fig. 7. ‘Elepaio fledgling, approximately 28 days old. Both parents brooded and fed the young. I observed a pair brooding their day-old nestling for 5 h, 5 min. The nestling was brooded 96% of this time, and the average attentive periods were 18 min for the female and 10 min for the male. During this time the nestling was fed 5 times and the parents brought food to the nest an additional 5 times, but did not feed the nestling, apparently because it did not gape. The male fed his mate 4 times when she was brooding. The nestling died when it was 3 days old, but the adults con- tinued to brood, although the attentive periods rapidly became shorter. Feed- ing attempts continued until the nestling was missing from the nest 2 days after it died. The 2 nestlings in nest 16 hatched on 21 April 1967 and left the nest on 6 May 1967. The nestling period, as defined by Berger (1961), was 16 days. During a 69 min period, pair 6-68 at nest 16 led their two 9-day-old young 23 times at intervals of less than 4 min, and removed 4 fecal sacs. At nest 29, the parents fed 3 young, approximately 10 days old, about 10 times per hour per nestling during 6 h of observation. At each feeding the adults waited, and if a fecal sac was ejected they picked it up, flew away immedi- ately, and swallowed it; they removed about 3 fecal sacs per hour. In 2 other 206 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Fig. 8. Diagram of habitat and distance relationships among nestings of pair 1-66 and 5-68. Nests 3, 4, and 5 are those of pair 1-66. instances I observed 2 adults feed a fledgling 9 times during a 30 min period, and a different pair feed a fledgling once every 7 min for about an hour. The only previous mention of care of eggs or young was by Rothschild (1893-1900). His collector, Palmer, found a nest on which he collected an adult male that was incubating. Palmer also saw both parents feeding their young. RENESTLNG AND NESTING SUCCESS Most of the pairs I studied renested at least once, first nests having been unsuccessful. In 1966 pair 1-66 renested twice and in 1968 pair 5-68 renested 3 times. Fig. 8 illustrates habitat and distance relationships for the nests of these 2 pairs. Nest dimensions of renestings were similar to those of first nestings, but height, nest habitat, and time taken to build varied. Because it took a total of about 6 weeks from the beginning of nest con- struction until the young fledged, and because nesting occurred from Febru- ary to May, it would have been possible for one pair to raise 2 broods in one season under ideal conditions. Pair 10-68 was feeding Hedgings on 4 April 1968, and a pair in territory 10, possibly the same pair, was feeding 2 newly fledged young on 16 May 1968. This suggests that the same pair raised 2 broods in one season. The total nesting success for 27 nests with 53 eggs was 13%, while hatching success was 29%, and nestling mortality 42%. The figure for hatching success Conant • THE OAHU ‘ELEPAIO 207 takes into account both infertility and egg destruction. Clearly the egg stage was the most vulnerable: 34 of 53 eggs observed were destroyed before hatching. Rats, which I observed frequently in the study area both before and after dark, may have preyed upon eggs and young. The low percentages for hatching and nesting success and nestling mor- tality may not be typical of the entire species. The Oahu ‘Elepaio I studied nested in slender-stemmed trees that bent freely in the wind. The period of high winds and heavy rains, which usually occurred in March, destroyed many nests each year. If the birds had nested in the sturdy native ‘ohi‘a tree, as they often do in areas where native forest remains intact, fewer nests may have lost eggs or young due to tipping by the wind. In Kokee State Park on Kauai, where ‘ohi‘a is in abundance, nesting success seemed to be greater, in that nearly every pair of adult ‘Elepaio I observed was feeding fledglings in June 1967. Four of the 6 ‘Elepaio nests I found at Kokee were in ‘ohi‘a, and Berger ( 1972 1 reported that most Kauai ‘Elepaio nests observed were in ‘ohi‘a. LENGTH OF THE BREEDING SEASON No author has specified a time duration for the breeding season, although various authors have given all of the months from January through June as breeding months ( Bryan and Seale 1901, Bryan 1905, MacCaughey 1919, Richardson and Bowles 1964). There is evidence indicating that each sub- species begins nesting at different times I Berger 1972). In this study, the breeding season of C. s. gayi was from mid-January to mid- June. Courtship activities began in mid- January, although males sometimes began territorial singing during the last week of December. Nesting began during the first week of February and continued until May; the last observed young of the season became independent in June. AGE AT BREEDING ‘Elepaio breed in their first year. A male hatched in nest 16 on 21 April 1967 was banded on 1 May 1967. He began building a nest in territory 5 on 7 February 1968, after forming pair 5-68 with an adult female. The female completed her first clutch of the year on 21 February 1968 in nest 21. This pair renested 3 times without success, and the female completed 2 more clutches. Perkins (1893) and Bryan and Seale (1901) believed that breeding of the species in immature plumage was rare. However, Perkins (1903) re- ported that birds in immature plumage were breeding, and MacCaughey (1919) stated that ‘Elepaio paired and nested before assuming the adult plumage. 208 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 TAXONOMIC RELATIONSHIPS Comparisons between aspects of the breeding biology of Chasiempis and that of other muscicapids (Stead 1932, Barret 1945, Jack 1949, Vaurie 1953, Oliver 1955, Storr 1958, Kirkman and Jourdain 1966) indicate that its breed- ing biology is similar to all 4 groups of muscicapids (whistlers, typical fly- catchers, robins, and monarchs) ; however, the data gathered in this study indicate that the ‘Elepaio probably belongs to the monarchid group, as Mayr (1943) and Amadon (1950) suggested. Specifically, the roles of the sexes in territoriality, nest building, incubation, and care of the young in the ‘Elepaio are more similar to this behavior in monarchs than in whistlers, typical flycatchers, or robins. The appearance, constituents, and placement of ‘Elepaio nests are similar to those of monarch nests. In addition, the color of ‘Elepaio eggs and the plumage of their young most closely resemble the eggs and young of monarchs. SUMMARY The breeding biology of Chasiempis sandwichensis gayi was studied from November 1965 to May 1%8 in Manoa Valley, island of Oahu, Hawaii, Birds bred monogamously from mid-January to mid-June. The phases of courtship were territory establishment, attraction of females by singing males, and chasing. The territory, which averaged 2.0 ha, was defended against other C. s. gayi, Leiothrix lutea, and Zosterops japonica pri- marily by males. Nesting habitat in the study area was a dense forest of Eugenia cumini. Aleurites moluccana, Citharexylum spinosum, and Psidium guajava; whereas. Eucalyptus sp. and Melaleuca leucadendron forests were suitable for feeding, but not for nesting. There were 4 types of vocalizations: (1) chipping, given during feeding, or by one bird before it replaced the other on the nest to incubate or brood, or as a mild alerting note; (2) chatter, a prelude to song, a low-intensity alarm note, and, in modified form, a food- begging or copulation-solicitation call; (3) song, given by both sexes; (4) the alarm call, a raspy cry, given by the parents when their young were being disturbed or by the young themselves in the same situation. Singing reached peaks of daily frequency during nest building; it was infrequent during incubation, and increased again from the time the young hatched until they were independent. It decreased after May, and resurged in December. The nest site, selected hy the female, was a tree fork or a lateral branch. Nest building, carried out by both sexes, took about 2 weeks. Nests, placed at an average height of 7.9 m were made of spider web. bark, grass leaves, leaf skeletons, lined with rootlets and animal hair, and covered with lichen and liverwort. Nest width averaged 7.2 cm, depth 7.8 cm, cup width 4.6 cm, and cup depth 4.1 cm. Nest dimensions of renestings were similar to those of first nestings, but height, nest habitat, and time taken to build varied. Eggs were usually laid in clutches of 2, and were white with reddish-brown spots. The eggs were laid before 08:30 with an interval between layings of about 24 h. They were incubated 96% of the day and all night for at least 14 days. Both sexes incubated, brooded and fed young, and sang on the nest. Nine-day-old young were fed an average of 10 times per hour; the nestling period was 16 days. Total nesting success was 13%; hatching success was 29%; nestling mortality was Conant • THE OAHU ‘ELEPAIO 209 42%. Although high winds and heavy rains were the primary causes of nest failure, predation by rats may have been a factor. Many aspects of C. s. gayi s breeding biology indicate that it is related to the monarchid group of muscicapids. ACKNOWLEDGMENTS Charles H. Lamoureux identified nest constituents and plants in the study area. Patrick Conant and Carl F. Frings assisted with fieldwork and photography, and Andrew J. Berger, who advised me during the project, Harvey I. Fisher, and Robert J. Raikow offered comments and criticisms on the manuscript. This paper is part of a thesis submitted in partial fulfillment of the requirements for the Master of Science degree at the University of Hawaii. The author was supported by NSF Grant GB-5612 from January 1967 to June 1968. LITERATURE CITED Amadon, D. 1950. The Hawaiian honeycreepers (Aves, Drepaniidae) . Bull. Am. Mus. Nat. Hist. 95:155-262. Baker, R. H. 1951. The avifauna of Micronesia, its origin, evolution, and distribution. Mus. Nat. Hist., Univ. Kansas Publ. 3:1-359. Barret, C. 1945. Australian birdlife. Oxford Univ. Press, Melbourne. Berger, A. J. 1961. Bird study. John Wiley and Sons, Inc., N. Y. . 1969. The nest, eggs, and young of the Elepaio. Wilson Bull. 81:333-335. . 1972. Hawaiian birdlife. Univ. Press of Haw'aii, Honolulu. Bryan, W\ A. 1905. Notes on the birds of the Waianae Mountains. Occ. Pap. B. P. Bishop Mus. 2:229-241. AND A. Seale. 1901. Notes on the birds of Kauai. Occ. Pap. B. P. Bishop Mus. 1:129-137. Hawaii Audubon Society. 1967. Hawaii’s birds. Hawaii Audubon Society, Honolulu. Henshaw, H. W. 1902a. Complete list of the birds of the Hawaiian Possessions, with notes on their habits. Hawaiian Gazette Co., Thos. G. Thrum, Publisher, Honolulu. . 1902b. The Elepaio of Hawaii. Auk 19:225-232. Jack, N. 1949. Territory and nesting in the Rufous Whistler. Emu 49:26-34. Kirkman, F. B., and F. C. R. Jourdain. 1966. British birds. Thomas Nelson Ltd., London. MacCaughey, V. 1919. The Hawaiian Elepaio. Auk 36:22-35. Mayr, E. 1943. The zoogeographic position of the Hawaiian Islands. Condor 45:45-48. Munro, G. C. 1944. Birds of Hawaii. Bridgeway Press, Rutland, Vermont. Newton, A. 1897. On some new and rare birds’ eggs. Proc. Zool. Soc. Lond. 1897:890- 894. Nice, M. M. 1953. The question of ten-day incubation periods. Wilson Bull. 65:81-93. Olwer, W. R. B. 1955. New Zealand birds. A. H. and A. W. Reed, Wellington. Perkins, R. C. L. 1893. Notes on collecting in Kona, Hawaii. Ibis 1893:101-114. . 1903. Vertebrata (Aves). Vol. 1, part 4, pp. 368-465 in Fauna Hawaiiensis (David Sharp, ed.), The Univ. Press, Cambridge, England. Richardson, F. and J. Bowles. 1964. A survey of the birds of Kauai, Hawaii. B. P. Bishop Mus., Bull. No. 227. Rothschild, W. 1893-1900. The avifauna of Laysan and the neighboring islands. R. H. Porter, London. 210 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Seale, A. 1900. Field notes on the birds of Oahu, H. I. Occ. Pap. B. P. Bishop Mus. 1:33-46. Stead, E. F. 1932. The life histories of New Zealand birds. Search Publ. Co., Ltd., London. Storr, G. M. 1958. On the classification of the Old W'orld Flycatchers. Emu 58:277- 283. VAN Riper, C. 1977. The use of sheep wool in nest construction by Hawaiian birds. Auk 94: On press). Vaurie, C. 1953. A generic review of flycatchers of the tribe Muscicapini. Bull. Am. Mus. Nat. Hist. 100:453-538. Wilson, S. B. 1891. On the Muscicapine genus Chasiempis, with a description of a new species. Proc. Zool. Soc. Lond. 1891:164-166. DEPT. OF GENERAL SCIENCE, UNIV. OF HAWAII, HONOLULU 96822. ACCEPTED 10 MAY 1976. NEW LIFE MEMBER Dr. Norman F. Sloan is a new life mem- ber of the WTlson Ornithological Society. Dr. Sloan is presently a professor in the Department of Forestry at Michigan Tech- nological University in Houghton. His orni- thological interests include avian popula- tion dynamics as they relate to biological control of forest insects; he has done con- siderable work with White Pelicans and also has a research interest in ravens and Peregrines. Dr. Sloan is married and has two children. In addition to belonging to several ornithological societies and other professional organizations. Dr. Sloan has l)een editor of Inland Bird Banding News since 1970. LIFE HISTORY FACTORS AFFECTING THE INTRINSIC RATE OF NATURAL INCREASE OF BIRDS OF THE DECIDUOUS FOREST BIOME Richard Brewer and Lynda Swander The intrinsic rate of natural increase, rm, is an important population trait indicating the capacity of a population for growing (Lotka 1925). It is, specifically, the rate of increase per head when the population is suffering no adverse effects from crowding (see Caughley and Birch 1971 for a more com- plete definition and Cole 1954 and Andrewartha and Birch 1954 for discus- sion of the concept). A high r^ results from many offspring (in birds, large clutches and multiple hroods), low mortality through the reproductive ages, especially through the first reproduction, reproduction at an early age, and a long reproductive life. MacArthur and Wilson (1967) used the terms r and K selection to desig- nate, respectively, selection occurring in populations living in an uncrowded environment and selection in populations at their carrying capacity (that is, at K). One important contribution of the discussions of r and K selection (Cody 1966; MacArthur 1960; Pianka 1970, 1972; Southwood et al. 1974) was the realization that, contrary to the traditional view, natural selection will not necessarily tend to maximize r„i in populations which are generally at their carrying capacity. In a series of field studies of the organization of avian communities and life histories of the constituent organisms (such as Brewer 1967 and Robins 1971) differences between communities in life history traits influencing im became evident. The purpose of this study was to examine, by a literature survey, life history features bearing on rm for characteristic birds of different communities of the deciduous forest biome of eastern North America. More specifically, the purpose was to determine whether species of successional communities tended to have traits favoring a high r,„ in comparison with spe- cies of presumably more stable communities, such as the climax forest. METHODS Three basic vegetation types, forest, forest edge, and grassland-marsh, were subdivided as follows: forest into wet, mesic, and dry; forest edge into thicket, trees-shruhs, and open trees; and grassland-marsh into grassland and marsh. Thicket is dense, low, woody growth, generally set in herbaceous vegetation; tree-shrubs is an interspersion of low, woody growth with some trees and usually patches of herbaceous vegetation; open trees is a parkland category with good-sized trees numerous hut not forming a closed canopy. After preparation of lists of characteristic native species for each community, life history information was collected. Seventy-five species were used. Several other species breeding 211 212 THE WILSON BULLETIN • Vol. 89, No. 2, Jane 1977 in the biome were omitted because we could not locate minimal satisfactory life history information for them. Studies from the part of the biome between about 38° and 44° N latitude were favored; findings obtained elsewhere were ignored unless there seemed reason to believe they were applicable to the middle area of the biome. Features con- sidered were clutch size, number of broods per year, age at first reproduction, presence of non-breeding individuals, and nesting, hatching, and fledging success. For most species included we failed to locate satisfactory information on one or more topics. In cases of conflicting statements we exercised judgment, often based on our own experience with the species, to reject aberrant reports. The species used, by community, are listed in Appendix I. A complete listing of sources of life history information has not been provided because most are the same as those cited in the various volumes of Life Histories of North American birds by A. C. Bent and in the compilations by Kendeigh (1952), Verner and Willson (1966), and Ricklefs 11969). Most data were assembled so that significance of differences in a particular trait could he tested between communities using x^- I*^ the subsequent text, stated differences are significant at the 5% level. To save space we have not included values hut the data are presented in such a way that the reader can readily calculate them for himself. CLUTCH SIZE Modal clutch size was determined for each species for which satisfactory data were available. For double-brooded species, the first clutch was used. When 2 clutch sizes seemed equally prevalent, the species was counted as half in one category and half in the other. Modal clutch size in the deciduous forest biome overall is clearly 4 (Table I ) hut differences existed among the communities. Confining our attention to open-nesting passerine species (to eliminate 2 obvious additional sources of variation in clutch size), number of eggs is largest in grassland where there are as many species with clutches of 5 as of 4 and smallest in mesic forest where modal size is but 3. Comparing related birds in different communities tends to confirm the trend of smaller clutches in forest, particularly mesic forest. The 4 species of mesic forest having clutches of 3 are Red-eyed Vireo ( Southern 1958), Eastern Wood Pewee ( Bendire 1895, Mengel 1965), Acadian Flycatcher ( Mumford 1964), and Hooded Warbler (Mengel 1965). Most vireos (Bent 1950) and small flycatchers (Walkinshaw 1966a, h I in other communities of the biome have 4-egg clutches. Other warblers of the biome have 4- or 5-egg clutches (Bent 1953). Good comparisons within families are fewer for grassland-marsh versus other habitats. The grassland sparrows, nevertheless, seem to have 5 eggs more frequently than do sparrows of forest edge; the Bobolink ( Raim 1975) and Eastern Meadowlark ( Roseberry and Klimstra, 1970) have 5 eggs whereas the Northern Oriole has 4 (Bendire 1895); and the marsh wrens have larger Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 213 Table 1 Numbers of Species Having a Given Modal Clutch Size by Habitat (All Species AND, Second Line in Each Habitat, Open-nesting Passerines) Habitat Modal Clutch Size^ 2 3 4 5 6 7 8 9 10 11 15 Mesic forest 4 2.5 2.5 1 1 4 2 1 Wet forest 2 4 2 1 4 Dry forest 2 4 1 1 3 1 All forest 4 4 10.5 5.5 1 1 1 1 4 9 1 1 Grassland 1 3.5 3.5 1 1 3.5 3.5 1 Marsh 1 3 2 1 1 1 1 2 1 Grassland-marsh 2 6.5 5.5 1 1 1 1 1 1 5.5 4.5 1 Thicket 4 1 4 Trees-shrubs 1 4 6 4 3 6 3 Open trees 1 4 1 2 4 All forest edge 1 5 14 5 2 1 3 14 3 Total, all habitats 5 11 31 16 4 2 2 1 1 1 1 8 28.5 8.5 1 1 1 No species had modal clutch sizes of 1, 12, 13, or 14. clutches than do House or Carolina wrens (Bent 1948), even though the latter pair are hole-nesters. NUMBER OF BROODS PER YEAR The difference between forest and other habitats in the tendency toward multiple broods is striking (Table 2). Forest birds are single-brooded; the regular exceptions are the Wood Thrush (Brackbill 1958), Carolina Wren (Laskey 1948), and Acadian Flycatcher (Walkinshaw 1966c). A high pro- portion of grassland-marsh and forest-edge birds — probably well over half in both categories — regularly have multiple broods. 214 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Number of Species with Table 2 Single or Multiple Broods by HabitaT 1 1 brood. 1 brood 1, 2 or more Habitat brood sometimes 2 sometimes 2 broods Mesic forest 3 6 9 2 2 4 6 2 Wet forest 4 2 6 1 0 2 2 1 Dry forest 2 3 5 0 1 2 3 0 All forest 9 11 20 3 3 8 11 3 Grassland 1 2 3 5 0 1 1 3 Marsh 1 4 5 3 0 0 0 0 Grassland-marsh 2 6 8 8 0 1 1 3 Thicket 0 1 1 4 0 1 1 3 Trees-shruhs 3 4 7 5 1 1 2 1 Open trees 2 0 2 4 0 0 0 4 All forest edge 5 5 10 13 1 2 3 8 Total, all habitats 16 22 38 24 4 11 15 14 1 The category “1 brood, sometimes 2” includes species in which second broods have been re- ported but are rare and species said to have one brood in the north and two in the south. The first line in each habitat includes all species, the second, species on which information is based on banded populations. Most species in a family are consistent in being single- or multiple-brooded regardless of habitat; differences between communities, consequently, are mainly the result of differences in the distribution of families among the vari- ous communities. Differences within families, when they occur, generally are in the expected direction: although most warblers have only one brood, the Common Yellowthroat ( thicket) has 2 ( Stewart 1953) ; most fringillids raise more than one brood hut the Rose-breasted Grosbeak (dry forest) raises only one (Forhush 1929 hut cf. Rothstein 1973). Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 215 NON-BREEDING BIRDS As Cole (1954) and Lewontin (1965) have demonstrated, the age at which reproduction begins is an extremely influential determinant of r. Although some birds are known to delay reproduction until they are 2 or more years old (Lack 1968:295-305), with a consequent low value for r, most of the birds considered here probably breed as yearlings. The existence of “surplus” or non-breeding birds is known for several species, including birds of forest, forest edge, grassland, and marsh (e.g., Stewart and Aldrich 1951, Offutt 1965, Ficken 1962, Zimmerman 1963, 1966, Hardy 1961, Kendeigh 1941). Reliable information is not available for enough species to draw conclusions as to differences among the various habi- tats. Most of the surplus birds detected are males and in a few of the species considered here (such as the Red- winged Blackbird) some or most males regularly delay breeding until their second year. The failure of males to breed would not lower rm if all females were mated. The failure of females to breed would lower rm, and markedly if the birds which did not breed were yearlings. Although the percentages of birds of various ages which do not breed is difficult to determine, the topic is one to which more life history studies should address themselves. For a few species, mostly not those of habitats considered here, age at first breeding is clearly density dependent, being earlier when populations are low (Lack 1968:298). For these species r„i is higher than it would appear to be when based on the age of first reproduction in stable populations. MATING SYSTEMS Monogamy is the rule among species of the deciduous forest biome (Table 3). Our compilations agree with Verner and Willson (1966) that regular polygyny is a feature primarily of grassland and marsh. The influence of polygyny on rm is not completely obvious. We may con- sider the situation in which a number of birds greater than the carrying capacity is available to occupy a particular area. If, as often happens, males which fail to gain a mate derive less of their food from the area than those that do (for example, by failing to maintain a territory through the whole breeding season; Ficken 1962), then polygyny will increase rm by substituting females for males as members of the population existing on the limited resources. This implies that polygyny might raise rm for birds living at K but would be unlikely to do so for birds under uncrowded conditions. It has been sug- gested (Verner 1964, Orians 1969) that selection for polygyny occurs where resources important for nesting are patchily distributed such that polygamous 216 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Number OF Monogamous Table 3 AND POLYGYNOUS Species by Habitat^ Habitat Monogamous Monogamous with exceptions Monogamous -\- monogamous with exceptions Polygynous Mesic forest 7 2 9 0 Wet forest 3 3 6 0 Dry forest 4 1 5 0 All forest 14 6 20 0 Grassland 4 0 4 4 Marsh 1 0 1 3 Grassland-marsh 5 0 5 7 Thicket 2 2 4 0 Trees-shrubs 12 0 12 1 Open trees 3 3 6 1 All forest edge 17 5 22 2 Total, all habitats 36 11 47 9 ^ The category “Monogamous with exceptions” includes normally monogamous species for which any incident of polygyny has been reported. females on good territories are more successful than monogamous ones on poor territories. This hypothesis of the evolution of polygyny depends on local populations being at K; at population sizes which were low relative to K, sufficient favorable habitat would be available that all males could estab- lish themselves in suitable sites and monogamous females presumably would be more successful than polygamous ones ( because of the greater participation of the male in nest building, feeding young, etc., Martin 1974). OTHER FACTORS INFLUENCING rm For 2 species otherwise similar, the one producing more eggs in its first year has a higher r,„ and this is true even if the other species later has an increased number of eggs so that the lifetime egg production is the same ( Lewontin 1965). Some changes in number of eggs laid with age occur (Ricklefs 1973:368, van Balen 1973 I hut how important they are and whether there is a difference among habitats are unknown. The pattern of survivorship of birds under uncrowded conditions is another trait influential in determining I'm but poorly known. Survival to the first breeding season is, of course, particularly important. Information on survival to fledging is available for many species and some comments on this time I Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 217 period are given beyond; however, the period from fledging to first nesting is difficult to study owing to its being the main time of dispersal in birds. Also, differential mortality in this period is probably of importance in popu- lation limitation; consequently, the mortality figures for stable populations will be higher than under uncrowded conditions and will be of little value for comparisons designed to evaluate r„,. ESTIMATES OF rm It is worth dealing with some numerical examples to help visualize the effects of some of these life history features on rm- Uncrowded survivorship/ maternity data for a real calculation of r,„ do not exist for any bird as far as we know. For the following calculations we tried to use mortality values that would be realistic for a small altricial bird in an uncrowded environment. The assumed mortality rates were 15.5% from egg-laying to hatching (15 days), 7.5% for the nestling period (10 days), 40% per annum from fledging to the age of 12 months, and 35% per annum past this age. Potential natural longevity was assumed to be 10 years. The same survivorship schedule was used throughout. The maternity schedule was varied to include clutch sizes of 1-8 eggs and 1-3 broods per year. Egg laying was assumed to begin at 12 months of age for all birds in single-brooded species and for birds of first broods in multiple-brooded spe- cies but at 11 months for birds produced in second broods and at 10 months for third broods. Also, a one-year delay (first breeding at 24 months for single-brooded species) was examined for several clutch-brood combinations. Calculation of r was iterative, using the formula (Lotka 1925) CO 1 = 2 l\ nix x = o where U is survival rate at age x, mx is natality rate at age x, and e is the base of natural logarithms (see Birch 1948 lor additional details of calculations). Although we believe that the survivorship-natality model and the values used are realistic, some simplifications are involved (Ricklefs 1973:394-396). We think that these do not impair the model’s usefulness for examining the rela- tive effects on rm of the various maternity schedules. The results (Fig. 1) suggest that r,n lor mesic forest species (if represented by 1 clutch of 3 eggs per year) may he only about that of grassland species (if represented by 3 clutches of 4 eggs per year). The importance of this dif- ference lor population growth is more easily grasped when translated into population figures. If a male and a female of 2 species arrive on an island of favorable habitat and one species undergoes population growth such that 218 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 CLUTCH SIZE Fig. 1. Values of rm (per head per year) for birds having clutch sizes of 1 to 8 and 1-3 broods per year (mortality schedules as described in text). Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 219 r is .3 per head per year and the other such that r is 1.2 (Fig. Ij, the popula- tion size at the end of some period (t) during which the environment remains uncrowded is given by the expression for exponential population growth Nt = No e*-t If we take t as 3 years, then the population of the species with the r of .3 is, at the end of 3 years, 3 or 4 birds. The population of the species with the r of 1.2 is over 70 birds. Certain other features in Fig. 1 are worth pointing out. As has been men- tioned, age at first breeding is extremely important. A bird which lays a clutch of 8 eggs but delays breeding until 2 years of age has an rm well below that of a bird which lays a clutch of 4 eggs but begins breeding at age 1 ( Fig. 1). Delaying reproduction by a year reduces rm proportionately less for small clutches than for large but only birds with considerably lower mor- tality than that assumed in the model could adopt a life history of delayed breeding if their clutch size w as 1 or 2 ; rm in both these situations is negative. Finally, the assumed survival rates are not high enough to allows a bird such as the Passenger Pigeon (which was, of course, a mesic forest species) to exist; rm is negative for a bird laying a single 1-egg clutch per year. If, as we believe, the estimates of rm in Fig. 1 are in the right neighbor- hood, they indicate that rm for birds is low in relation to other animals of com- parable size (e.g., Leslie and Ransom 1940, Leslie 1945). The main reason is the periodic nature of avian reproduction. Birds of temperate regions have little opportunity for raising rm by lowering age of first reproduction below one year; the opportunity does exist for some tropical species and a few have taken it (Cody 1971:468). Most tropical birds, however, breed at about one year, have some sort of annual cycle of reproduction ( Immelmann 1971) and, owing particularly to their small clutches ( Ricklefs 1973), probably have low^ values of rm. DISCUSSION It has been postulated that r selection should occur in organisms of tem- porary or unstable ecosystems; species would be fitted to such environments if they could “(1) discover the habitat quickly, (2) reproduce rapidly to use up the resources before other, competing species could exploit the habitat, and (3) disperse in search of other new habitats as the existing one began to grow unfavorable” (Wilson and Bossert 1971:110). In stable environments (Wil- son and Bossert mention climax forest, coral reefs, and caves) “no longer is it very advantageous to have a high r. It is more important for genotypes to confer competitive ability, in particular the capacity to seize and to hold a piece of the environment and to extract the energy produced by it.” 220 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 4 Summary of Life History Traits Affecting rm by Habitat Habitat Trait Mesic forest Other forest Grassland marsh Forest edge Clutch size low medium high" medium Number of broods single single multiple multiple Mating systems Non-breeding females, uncrowded changes in monogamy monogamy polygyny frequent polygyny infrequent clutch size or number of broods with age 9 9 9 ? 1 Traits favoring a high rm are shown in boldface. Our compilations indicate that birds of forest, especially mesic forest which is generally climax, tend to have life history traits favoring a low rm (Table 4). Birds of forest edge and, especially, grassland and marsh tend to have traits favoring a high rm- We address, first, the question of whether forest represents a K-selecting environment and grassland, marsh, and edge, r-selec- ting environments and, later, the question of whether the low rm of forest birds is, or is not, maximal. Are breeding-bird populations at carrying capacity? — By the Clementsian criterion of potential replaceability (Clements 1916) all of the ecosystems dis- cussed except mesic forest are successional and, in this sense, temporary. The actual period of existence of a particular stand may vary from a few years for some pioneer herbaceous communities to thousands of years for some marshes and forests. For some kinds of organisms some of these habitats may he uncrowded for a significant period after establishment but we think this is not true for birds. We know of no studies of newly created grassland show- ing population growth curves for any bird species unrelated to successional changes in vegetation. In studying a new alfalfa field Harrison (1974) found the same species in almost the same numbers in the first full year of its existence (following planting the preceding year) as in its second. The high dispersal powers of birds in conjunction with habitat selection presumably account for the virtually immediate occupancy at carrying capacity of newly created habitats. We believe that the characteristic birds of all of these habitats tend to be near K. The existence of surplus birds in various life history studies and in the studies of Stewart and Aldrich (1951) and Hensley and Cope (1951) Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 221 suggest K conditions. Further, in those 2 studies, the re-establishment of most species in 1950 in about the same numbers as 1949, despite the almost com- plete annihilation of the 1949 resident population and the failure of the plot to produce any fledglings (3 species excepted) suggest that these numbers represent carrying capacities. Experiments in which nest-boxes have been added to habitats (cited in von Haartman 1971:401^04) have produced larger breeding populations of both migratory and sedentary species. The studies mentioned in the preceding paragraph deal mostly with forest species but surplus birds are known in marsh and forest edge (e.g. Kendeigh 1941, Nero 1956). The correlation of numbers of Red-winged Blackbirds with breeding site favorability reported by Davis and Peek (1972) suggests K conditions. As we have already pointed out, polygyny, common only in grass- land and marsh, is not an adaptation to uncrowded conditions. In the next section we present evidence that grassland species frequently do not occupy every area that seems to be suitable habitat. This may seem contradictory to the conclusion just reached but local populations and regional (or whole-species) populations must be considered separately. Birds tend to be at or near K in examples of their optimal habitat because of their tendency to occupy habitats according to their suitability (Fretwell 1972) . Even if some additional field, marsh, or forest exists which could support the species, the slightly suboptimal habitat tends not to be occupied (disregarding the effects of site tenacity and sociability which are complicating factors for some spe- cies) if the regional population of the species is low (cf. Svardson 1949, Brown 1969). Consequences of low habitat stability in grassland and marsh. — We suggest that a feature important in selection for life history traits affecting rm by which forest and grassland-marsh differ is stability (cf. Wilbur et al. 1974). Grasslands and marshes may show drastic differences from year to year, or even within years, even though successional changes are occurring slowly, if at all. The classic study of vegetational fluctuation (sensu Hanson and Chur- chill 1961) is of the Nebraska grasslands during the great drought of the 1930’s (e.g. Weaver 1961). Such fluctuations presumably affect K for the bird populations, both by altering the structure of the vegetation and by changing the kinds and amounts of food. Changes of similar magnitude occur in marshes (Weller and Spatcher 1965). Vegetational fluctuations have rarely been reported for forest. This is at least partly because of the buffering effect of the greater perennial biomass of forest and possibly also a greater capacity for integrating environmental fluctuations (e.g., by leaf production being based partly on energy stored in preceding years). Ricklefs (1969) has suggested that the high starvation component of nestling mortality of birds 222 THE WILSON BULLETIN • Vol. 89, Xo. 2, June 1977 of field and marsh indicates that the food supply in these habitats is more variable than in forest and edge. Fluctuations in avian populations in marshes were studied by eller and Spatcher i l965). Their study began with the marshes nearly dry and con- ^ tinned into a wet period in which much of the vegetation was inundated. Bird populations changed drastically; for example, Least Bittern nests at one marsh rose from 5 in 1958 to 62 in 1961 and dropped to 2 the following year (\^'eller and Spatcher 1965: Table 4). Reports of fluctuations in density of grassland species are numerous. iens (1974:397), in a paper which develops the theme of instability in climax grassland, cited large population changes in bird numbers on a Texas grass- land during drought. Most reports are not quantitative but consist of remarks about the "local and erratic” occurrence of the species, implying that not all seemingly suitable areas are occupied every year and that abundance at a given site varies greatly from year to year. Some typical citations are Barrows (1912), Pough (1946i, Smith (1963), Bull (1964i, and Mengel (1965) for : the Dickcissel, Short-billed Marsh ren, Grasshopper Sparrow, Henslow’s Sparrow, and Lark Sparrow, respectively. Our own observations on grassland birds and several accounts in the litera- ture I Nero 1956, Smith 1963, Robins 1971, Potter 1972, Raim 1975) suggest a shifting pattern of occupancy, including shifts in territory locations and j changes in population size, even within a breeding season. A forest-edge spe- | cies for which large-scale territory shifts have been reported is the Gray Cat- bird (Darley et al. 1971). Comparable changes between breeding seasons and, especially, within the season are difficult to find in the literature for forest | species. Populations of the more common species on a given area vary little 1 from year to year (Brewer 1963). The whole breeding population apparently j arrives within a few days and the arrival of a new male thereafter is a rare I event (e.g., Hann 1937). ! e suggest that grassland or marsh that remains stable tends to support populations around K. Because of habitat fluctuations altering K. popula- tions fairly frequently find themselves overcrowded or undercrowded. The shifting pattern of occupancy of grassland birds can be best understood as movements into and out of areas as suitability changes. Areas which, in fact, show large swings in numbers with essentially no change in vegetational or other environmental conditions are usually suboptimal for the species in ques- ; tion, occupied mainly by the overflow, large, small, or none, from more nearly optimal areas (cf. Kluyver and Tinbergen 1953, Brewer 1963, 1967, Fretwell I 1972). : \^’e presume that the main source of additional birds, when grassland popu- Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 22B lations build up over the course of the breeding season, is from nearby areas abandoned as unsuitable; however, this may not be the only source. It is possible, for example, that some of these species have an unusually long period over which they return from winter quarters. Although we suspect that grassland birds have evolved a more flexible sys- tem of habitat occupancy than forest birds, it is possible that forest birds if presented with habitat changes of comparable magnitude might also show a more shifting pattern. Kluyver (1951), for example, was able to cause adult Great Tits, Parus major, to abandon an area by plugging all the nest holes. When about half of a 10 ha spruce forest was blown down by hurricanes be- tween breeding seasons (Cruickshank and Cadbury 1954, 1955) nesting Golden-crowned Kinglets, Regains satrapa, decreased from 8 to 1; however, several spruce forest warblers showed no significant decline. Wilson and Bossert (1971) mentioned the abilities to discover habitat quickly and to disperse in search of new areas as the existing one began to grow unfavorable as traits to be selected for in species of temporary habitats. These traits are related to a high r„i only to the degree that a high rm favors greater or more effective dispersal. For most organisms, increasing offspring is a poor way of increasing distant dispersal. This is because the number of dispersing units, or disseminules, reaching a given distance tends to be directly proportional to the number produced (e.g., Kettle 1951, Andrewartha and Birch 1954:103). Consequently, committing the resources necessary to double production of offspring merely doubles whatever small fraction of an indi- vidual is expected to reach some distant site. Because an exponential relation- ship tends to exist between dispersal ability and the number of disseminules reaching a particular distance, increasing dispersal ability is a more efficient way of achieving increased dispersal. For birds (and some other kinds of organisms; Johnston 1961, French 1971), the pattern of dispersal is complex, with more individuals reaching greater distances than would be expected. Birds as different as the Song Spar- row (Johnston 1956) and the Pied Flycatcher, Ficedula hypoleuca, (Berndt and Sternberg 1968) have a basically similar dispersal pattern. Murray (1967) showed that this pattern could be reproduced by a model making the assumptions that dispersal is by sub-adults, the adults returning to their previ- ous breeding site where they are dominant to sub-adults; that the first sub- adult to reach an unoccupied site is dominant to subsequent arrivals; and, an important implicit assumption, that the environment is patchy such that indi- viduals that leave the modeled habitat rarely settle on an immediately adjacent site but instead go some appreciable distance. These assumptions seem real- istic for birds (Murray 1967; see also Pinowski 1965). Murray further found that increasing survival of young to breeding age 224 THE WILSON BULLETIN • Vol 89, No. 2, June 1977 from 23% to 62% increased the number of young settling near their natal site slightly but greatly increased the number of young leaving the area. It seems possible, then, that a high production of offspring, in fact, increases dispersal. The major affect is probably increased dispersal ability as a result of intra- specific antagonism. Although this paper deals primarily with life history traits affecting rm, this is just one aspect of the broader question of the adjustment of life his- tories to ecosystems. There are too few detailed studies to judge whether birds of different habitats differ in dispersal pattern. One behavioral feature, how- ever, that could increase dispersal ability, including the ability to leave an unfavorable area for a favorable one in the middle of the summer, exists in at least 2 grassland birds. In the Bobolink (Raim 1975) males throughout the breeding season take long flights in which they leave the breeding area; often they fly out of sight and may be gone for many minutes. Suitable areas could be found on such flights, either for switching locations in the same breeding season or for possible occupancy in a later one. Similar “distant flights” also occur in the Dickcissel (Schartz and Zimmerman 1971). The foregoing discussion would suggest that site tenacity (except on optimal sites) ought to be low in grassland and marsh and this seems often to be the case; however, males of one of the most characteristic marsh species, the Red- winged Blackbird, are known to show strong site tenacity ( Nero 1956) . Davis and Peek ( 1972 ) described a situation in which the number of territorial males ( apparently not individually marked) varied in one marsh only between 17 and 21 during an 8-year period that included a drought. The number of females in the same period varied from 7 to 42. Surplus females are rare or absent in the species (Orians 1969) so many of the females presumably shifted to other areas where males were occupying more favorable habitat. This sug- gests that polygyny is, in effect, an exceedingly flexible method of maintaining a high r„i where short-term fluctuations are common. The evolution of such a system is explainable by individual selection: females having the ability to shift would probably leave more offspring than if they remained in a deterio- rating habitat; males, on the other hand, might well contribute more to the next generation if they stayed put rather than trying to establish themselves in a strange area, often in competition with other, already-established, males. These disadvantages may help to explain the higher site tenacity of male birds generally; however, except in polygynous species, the result of some males occupying territories unacceptable to females might be to leave some females unmated, lowering r„i. Does the low r„i of forest birds represent K selection? — The alternative to K selection, in explaining the low rm of forest birds, is that, low or not, it is the best they can do. The most common version of this idea is Lack’s view of Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 225 the evolution of clutch size (or what might be called the “I can scarce main- tain two” hypothesis after the ancient nursery rhyme that goes [Opie and Opie 1951] : The dove says coo coo, what shall I do?/ I can scarce maintain two. pooh, pooh, says the wren, I have got ten,/ and keep them all like gentlemen). In its simplest form, this hypothesis states that “in species in which parents feed their young, clutch size corresponds to that brood-size from which the parents can, on average, raise most young, the limit being set by the amount of food which they can collect for them” (Lack 1968:307). Following Lack’s line of reasoning for number of broods per season would suggest that every bird raises as many broods as it can; the single brood of forest birds results from the period in which the food supply is sufficient for raising young being too short for another brood. There is no reason, for birds living generally under carrying capacity con- ditions, that clutch size or number of broods must be selected for in this way (although under r-selecting conditions they should be). Neither, however, is selection against a high rm under K-selecting conditions inevitable. For the genotype with fewer eggs or fewer broods to have the advantage over ones with more, the first must use savings of time or energy from the decreased birthrate to enhance its changes for recruitment into the breeding population above those of the second (cf. Cody 1966; Fret well 1969). Some of the ways in which this might be achieved would be by improved parental care that brought a larger fraction of the young through the vulnerable egg-nestling stage, by increased ability to locate appropriate territorial sites and to estab- lish and defend territories, and by increased ability to escape death from pre- dation, parasitism, and disease. If the low rm of forest birds represents K selection the saving of time or energy, or its result, should be detectable in comparisons with related species of different habitats. In comparisons not presented here we examined, first, various measures of nesting success. Nesting success is clearly higher in forest than in grassland- marsh (see Ricklefs 1969: Table 5) ; however, nesting success of forest-edge species is at least as good as forest species. Consequently, it is difficult to conclude that the higher success of forest species vs. grassland-marsh is the result of diversion of resources which would otherwise have been used in maintaining a high rm- We also compared birds of the various habitats as to egg size (relative to body size) and length of incubation and nestling periods. We detected no consistent differences in forest birds compared with their non-forest relatives. It is, however, true that differences need not be consistent from one species to another: the savings of time or energy might be used differently by dif- ferent species. For example, most members of the Tyrannidae in eastern North America delay molting until they reach the wintering grounds, unlike most 226 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 passerines which molt prior to fall migration. A lack of time or energy or both may tend to preclude molt prior to migration (Morehouse and Brewer 1968 j. We suggest that the ability of the Great Crested and Acadian fly- catchers to molt prior to migration is related to their inhabiting mesic forest. ( We note, further, that the other North American tyrannid which molts be- fore migration, the Eastern Phoebe, Sayornis phoebe, is probably K-selected. Although it occurs in a variety of habitats, it depended for nesting, under primeval conditions, largely on rock ledges which were probably stable in terms of appearance and disappearance.) Historical factors. — Many forest-edge species appear to have high values of rm yet forest-edge habitats do not show^ obvious vegetational fluctuations and may be long-lived (Niering and Egler 1955). The forest-edge group is, of course, heterogeneous. Particular species may have life history features favoring either a high or a low* rm (e.g.. Am. Goldfinch vs. Eastern Kingbird). Probably finer habitat subdivisions leading, ultimately, to a consideration of the different circumstances of every species would help to reconcile these differences (as would be true also for forest, grassland and marsh). Additionally, historical factors may be important. The deciduous forest hiome has undergone many changes during the Quaternary ( Kendeigh 1974: 303-310) ; however, the period from around 8000 years ago until the arrival of European man was one of relative stability. In this period many forest birds must have been common and many grassland and forest-edge birds rare. Much of the habitat for forest-edge species may have been along rivers which, owing to flooding, probably was unstable. Many other suitable sites, such as areas of w ind-thrown trees, must have been isolated and transient. Such con- ditions might have favored a high rm- The period from around 1650 to 1850 and a little later, in which man altered the landscape on a large scale, must have seen populations of forest-edge and grassland birds increase enormously. \*iTiether or not local breeding popula- tions were at K, the increasing amounts of habitat could have provided r- selecting conditions for many forest-edge and grassland species during this period. At some point, winter limitation of the size of the species populations might have become important (Fretwell 1972). It is conceivable that at least a part of the failure of some grassland species to occupy all seemingly suit- able breeding sites results from limitation of total population size on the wintering grounds. At the same time total populations of forest species must have shrunk greatly. Unless large changes in competitive relations occurred, it seems unlikely that most migratory forest species have been winter-limited. This situation may, of course, be changing wdth the heavy destruction of natural habitats that has recently occurred in the American tropics. Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 227 SUMMARY Forest birds, especially those of mesic forest, have life history traits favoring a lower intrinsic rate of natural increase (rm) than do birds of grassland, marsh, or forest edge. Specifically, forest birds tend to have smaller clutches than birds of grassland or marsh; several mesic forest species typically have clutches of only three eggs. Almost no forest species are regularly double-brooded but a high percentage of grassland, marsh, and forest- edge species have two or more broods per season. Forests can probably be thought of as K-selecting environments. Grassland and marsh probably are not, specifically because vegetational fluctuations make particular areas un- predictably uncrowded or overcrowded. Probably because of these fluctuations (and consequent changes of K) grassland birds show a more flexible pattern of occupancy, even within a breeding season, than do forest birds. Polygyny may be a particularly powerful method of maintaining a high rm where short-term fluctuations are common. Probably one advantage of a high production of young in unstable environments is en- hanced dispersal. The main effect seems to be through increased dispersal ability because of increased intraspecific antagonism. The low rm of forest birds may be partly the result of natural selection diverting re- sources from reproduction to other traits more advantageous for organisms living con- tinually around their carrying capacity; however, a conclusion to this effect should await a demonstration of what activities that time or energy otherwise used in reproduction is being diverted to. This demonstration may be difficult because the diversion may be to different activities in different species. The alternative explanation, that rm in the various habitats is a reflection of the amount of available energy and the time over which it is available, deserves continued attention. ACKNOWLEDGMENTS Some of the ideas developed in this paper arose from field studies of various species and communities by Brewer, Ted Gottshall, Arlo Raim, J. D. Robins, and J. D. Wenger. Helpful conversation on community organization with the last four is acknowledged. Ste- phen Fretwell, Robert Ricklefs, John Wiens, and John Zimmerman commented construc- tively on earlier drafts of the manuscript. Some of the field work was done while Brewer held Faculty Research fellowships and grants from Western Michigan University. Com- puter time was provided by the Western Michigan University Computer Center. LITERATURE CITED Andrewartha, H. G. and L. C. Birch. 1954. The distribution and abundance of ani- mals. Univ. of Chicago Press, Chicago. Balen, j. H. van. 1973. A comparative study of the breeding ecology of the Great Tit Pams major in different habitats. Ardea 61:1-93. Barrows, W. B. 1912. Michigan Bird Life. Michigan Agric. Coll., East Lansing. Bendire, C. E. 1895. Life histories of North American birds. U.S. Natl. Mus. Spec. Bull. 3. Bent, A. C. 1948. Life histories of North American nuthatches, wrens, thrashers, and their allies. U.S. Natl. Mus. Bull. 195:1-475. . 1950. Life histories of North American wagtails, shrikes, vireos, and their allies. U.S. Natl. Mus. Bull. 197:1-411. . 1953. Life histories of North American wood warblers. U.S. Natl. Mus. Bull. 203:1-734. 228 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Berndt, R. and H. Sternberg, 1968. Terms, studies and experiments on the problems of bird dispersion. Ibis 110:256-269. Birch, L. C, 1948. The intrinsic rate of natural increase of an insect population. J. Anim. Ecol. 17:15-26. Brackbill, H. 1958. Nesting behavior of the Wood Thrush. Wilson Bull, 70:70-89, Brewer, R. 1963. Stability in bird populations. Occas. Papers Adams Ctr. Ecol. Studies 7:1-12. . 1967, Bird populations of bogs. Wilson Bull. 79:371-396. Brown, J. L. 1969. Territorial behavior and population regulation in birds. Wilson Bull. 81:293-329. Bull, J. 1964. Birds of the New York City area. Harper and Row, New York. Caughley, G. and L. C. Birch. 1971. Rate of increase. J. Wildl, Manage. 35:658-663. Clements, F. E. 1916. Plant succession. Carnegie Inst. Washington Publ. 242:1-512. Cody, M. L. 1966. A general theory of clutch size. Evolution 20:174-184. . 1971. Ecological aspects of reproduction. Pp. 461-512 in Avian biology, vol. 1 ID. S. Earner and J. R. King, eds,). Acad. Press, New York. Cole, L. C. 1954. The population consequences of life history phenomena, Q. Rev. Biol. 29:103-137. Cruickshank, a. D. and J. M. Cadbury. 1954. Climax red and white spruce forest. Audubon Field Notes 8:381. . 1955. Hurricane-devastated spruce forest. Audubon Field Notes 9:427. Darley, j. a., D. M. Scott, and N. K. Taylor. 1971. Territorial fidelity of catbirds. Can. J. Zool. 49:1465-1478. Davis, D. E. and F. Peek. 1972. Stability of a population of male Red-winged Black- birds. Wilson Bull. 84:349-350. Ficken, M. S. 1962. Agonistic behavior and territory in the American Redstart. Auk 79:607-632. Forbush, E. H. 1929. Birds of Massachusetts and other New England States, pt. HI. Land birds from sparrows to thrushes. Massachusetts Dept, of Agric. French, N. R. 1971. Simulation of dispersal in desert rodents. In Statistical ecology, vol. 3 (G. P. Patil, E. C. Pielou, and W. E. Waters, eds.). Pennsylvania State Univ. Press, University Park. Fretwell, S. D. 1969. The adjustment of birth rate to mortality in birds. Ibis 111: 624^627. . 1972. Populations in a seasonal environment, Princeton Monogr. Pop. Biol, no, 5:1-217. Hann, H. W. 1937. Life history of the Ovenbird in southern Michigan. Wilson Bull. 49:145-237. Hanson, H. C. and E. D. Churchill. 1961. The plant community. Reinhold, New York. Hardy, J. W'. 1961. Studies in behavior and phylogeny of certain New World jays. Univ. Kansas Sci. Bull. 42:13-149. Harrison, K. G. 1974. Aspects of habitat selection in grassland birds, M.A. thesis. Western Michigan Univ., Kalamazoo. Hensley, M. M. and J. B, Cope. 1951, Further data on removal and repopulation of the breeding birds in a spruce-fir forest community. Auk 68:483-493. Immelmann, K. 1971. Ecological aspects of periodic reproduction. Pp. 341-389 in Avian biology, vol. 1 (D. S. Earner and J. R. King, eds.). Acad. Press, New York, loiiNSTON, R. F, 1956. Population structure in salt marsh song sparrows. Part I. En- vironment and annual cycle. Condor 58:24-44. Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 229 , 1961. Population movements of birds. Condor 63:386-389. Kendeigh, S. C. 1941. Territorial and mating behavior of the House Wren. Illinois Biol. Monogr. 7:91-123. . 1952. Parental care and its evolution in birds. 111. Biol. Monogr. 22:1-358. . 1974. Ecology with special reference to animals and man. Prentice-Hall, Engle- wood Cliffs. Kettle, D. S. 1951. The Spatial distribution of Culicoides impunctatus Goet. under woodland and moorland conditions and its flight range through woodland. Bull. En- tomol. Res. 42:239-291 (cited from Andrewartha and Birch 1954). Kluyver, H. N. 1951. The population ecology of the Great Tit, Parus m. major. Ardea 39:1-135. AND L. Tinbergen. 1953. Territory and the regulation of density in titmice. Arch. Neerl. Zool. 10:265-287. Lack, D. 1968. Ecological adaptations for breeding in birds. Methuen, London. Laskey, A. R. 1948. Some nesting data on the Carolina Wren at Nashville, Tenn. Bird- banding 19:101-121. Leslie, P. H. 1945. On the use of matrices in certain population mathematics. Bio- metrika 33:183-212. AND R. M. Ransom. 1940. The mortality, fertility, and rate of natural increase of the vole {Microtus agrestis) as observed in the laboratory. J. Anim. Ecol. 9:27-52. Lewontin, R. C. 1965. Selection for colonizing ability. In The genetics of colonizing species (T. G. Baker and G. L. Stebbins, eds.). Academic Press, New York. Lotka, a. J. 1925. The elements of physical biology. Williams and Wilkins, Baltimore. MacArthur, R. H. 1960. On Dr. Birch’s article on population ecology. Am. Nat. 94: 313. AND E. 0. Wilson. 1967. The theory of island biogeography. Princeton Univ. Monogr. Pop. Biol. no. 1:1-203. Martin, S. G. 1974. Adaptations for polygynous breeding in the Bobolink, Dolichonyx oryzivorus. Am. Zool. 14:109-119. Mengel, R. 1965. Birds of Kentucky. Ornithol. Monogr. 3:1-581. Morehouse, E. L. and R. Brewer. 1968. Feeding of nestling and fledgling Eastern Kingbirds. Auk 85:44-54. Mumford, R. E. 1964. The breeding biology of the Acadian Flycatcher. Univ. Mich. Mus. Zool. Misc. Publ. no. 125:1-50. Murray, B. G., Jr. 1967. Dispersal in vertebrates. Ecology 48:975-978. Nero, R. W. 1956. A behavior study of the Red-winged Blackbird. Wilson Bull. 68: 5-37, 129-150. Niering, W. a. and F. E. Egler. 1955. A shrub community of Viburnum lentago, stable for twenty-five years. Ecology 36:356-360. Offutt, G. 1965. Tufted Titmouse behavior. Wilson Bull. 77:382-387. Opie, I. AND P. Opie. 1951. The Oxford dictionary of nursery rhymes. Oxford Univ. Press, Oxford. Orians, G. H. 1969. On evolution of mating systems in birds and mammals. Am. Nat. 103:589-603. PiANKA, E. R. 1970. On r and K selection. Am. Nat. 104:592-597. . 1972. r and K selection or b and d selection. Am. Nat. 106:581-588. PiNOWSKi, J. 1965. Overcrowding as one of the causes of dispersal of young Tree Sparrows. Bird Study 12:27-33. 230 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Potter. P. E. 1972. Territorial behavior of Savannah Sparrows in southeastern Michi- gan. Wilson Bull. 84:48-59. Pouch, R. H. 1946. Audubon bird guide to eastern land birds. Doubleday, New York. Raim, a. J. 1975. Territority and mating system in the Bobolink. M.S. thesis. Western Michigan Univ., Kalamazoo. Ricklefs, R. E. 1969. An analysis of nesting mortality in birds. Smithson. Contrib. Zool. no. 9:1-48. . 1973. Fecundity, mortality, and avian demography. Pp. 366-435 in Breeding biology of birds (D. S. Earner, ed.). Natl. Acad. Sci., Washington. Robins, J. D. 1971. A study of Henslow’s Sparrow in Michigan. Wilson Bull. 83:39-48. Roseberry, j. L. and W. D. Klimstra. 1970. The nesting ecology and reproductive performance of the Eastern Meadowlark. Wilson Bull. 82:243-267. Rothstein, S. I. 1973. Extreme overlap between first and second broods in the Rose- breasted Grosbeak. Wilson Bull. 85:242-243. SciiARTz, R. L. and j. L. Zimmerman. 1971. The time and energy budget of the male Dickcissel {Spiza americana) . Condor 73:65-76. Smith, R. L. 1963. Some ecological notes on the Grasshopper Sparrow. W ilson Bull. 75:159-165. Southern, W. E. 1958. Nesting of the Red-eyed Vireo in the Douglas Lake region, Michigan. Jack-Pine Warbler 36:105-130, 185-207. SouTiiwooD, T. R. E., R. M. May, M. P. Hassell, and G. R. Conway. 1974. Ecologi- cal strategies and population parameters. Am. Nat. 108:791-804. Stewart, R. E. 1953. A life history study of the Yellowthroat. Wilson Bull. 65:99-115. AND J. W. Aldrich. 1951. Removal and repopulation of breeding birds in a spruce-fir forest community. Auk 68:471-482. SvARDSON, G. 1949. Competition and habitat selection in birds. Oikos 1:157-174. Verner, j. 1964. Evolution of polygamy in the Long-billed Marsh Wren. Evolution 18: 252-261. AND M. F. Willson. 1966. The influence of habitats on mating systems of North American passerine birds. Ecology 47:143-147. VON Haartman, L. 1971. Population dynamics. Pp. 391-459 in Avian biology, vol. 1 (1). S. Farner and J. R. King, eds.). Acad. Press, New York. Walkinshaw, L. H. 1966a. Summer biology of the Traill’s Flycatcher. Wilson Bull. 78:31-46. . 1966b. Summer observations of the Least Flycatcher in Michigan. Jack-Pine Warbler 44:151-168. . 1966c. Studies of the Acadian Flycatcher in Michigan. Bird-banding 37:227- 257. Weaver, J. E. 1%1. Return of midwestern grassland to its former composition and stabilization. Occas. Papers Adams Ctr. Ecol. Studies 3:1-15. Weller, M. W. and C. S. Spatcher. 1965. Role of habitat in the distribution and abun- dance of marsh birds. Iowa Agric. Home Econ. Exp. Stn. Spec. Rep. 43:1-31. Wiens, J. A. 1974. Climatic instability and the “ecological saturation” of bird communi- ties in North American grasslands. Condor 76:385-400. Wilbur, H. M., D. W. Tinkle, and J. P. Collins. 1974. Environmental certainty, trophic level, and resource availability in life history evolution. Am. Nat. 108:805- 817. Wilson, E. 0. and W. H. Bossert. 1971. A primer of population biology. Sinauer Associates, Stamford. Brewer and Swander • LIFE HISTORY AND BIRD POPULATIONS 231 Zimmerman, J. L. 1963. A nesting study of the Catbird in southern Michigan. Jack- Pine Warbler 41:142-160. . 1966. Polygyny in the Dickcissel. Auk 83:534-546. DEPT. OF BIOLOGY, WESTERN MICHIGAN UNIV., KALAMAZOO 49008. (PRESENT ADDRESS LS: JOHNSON COUNTY COMMUNITY COLLEGE, OVERLAND PARK, KS 66210). ACCEPTED 9 APRIL 1976. APPENDIX I Bird species assigned to each habitat: Mesic Forest: Melanerpes carolinus, Red-bellied Woodpecker; Picoides pubescens, Downy Woodpecker; Myiarchus crinitus. Great Crested Flycatcher; Empidonax virescens, Acadian Flycatcher; Contopus virens, Eastern Wood Pewee; Pams atricapillus. Black-capped Chickadee; Paras bicolor. Tufted Titmouse; Hylocichla mustelina, Wood Thrush; Vireo olivaceus. Red-eyed Vireo; Dendroica cerulea, Cerulean Warbler; Wilsonia citrina, Hooded Warbler. Wet Forest: Aix sponsa, Wood Duck; Strix varia. Barred Owl; Caprimulgus carolinensis, Chuck-will’s- widow; Thryothorus ludovicianus, Carolina Wren; Catharus fuscescens, Veery; Polioptila caerulea, Blue-gray Gnatcatcher; Parula americana, Northern Parula; Protonotaria citrea, Prothonotary Warbler; Setophaga ruticilla, American Redstart. Dry Forest: Bubo virginianus, Great Horned Owl; Caprimulgus vociferus, Whip-poor- will; Picoides villosus, Hairy Woodpecker; Sitta carolinensis, White-breasted Nuthatch; Mniotilta varia. Black-and-white Warbler; Seiurus aurocapillus, Ovenbird; Piranga olivacea. Scarlet Tanager; Pheucticus ludovicianus, Rose-breasted Grosbeak. Grassland: Eremophila alpestris. Horned Lark; Cistothorus platensis. Short-billed Marsh Wren; Dolichonyx oryzivorus. Bobolink; Sturnella magna. Eastern Meadowlark; Spiza americana, Dickcissel; Passerculus sandwichensis, Savannah Sparrow; Ammodramus savannarum, Grasshopper Sparrow; Ammodramus henslowii, Henslow’s Sparrow; Pooe- cetes gramineus. Vesper Sparrow. Marsh: Podilymbus podiceps. Pied-billed Grebe; Ixobrychus exilis. Least Bittern; Botau- rus lentiginosus, American Bittern; Anas platyrhynchos. Mallard; Rallus limicola, Vir- ginia Rail; Porzana Carolina, Sora; Chilidonias niger. Black Tern; Telmatodytes palustris. Long-billed Marsh Wren; Agelaius phoeniceus. Red-winged Blackbird; Melospiza georgi- ana, Swamp Sparrow. Thicket: Colinus virginianus, Bobwhite; Dumetella carolinensis. Gray Catbird; Geo- thylpis trichas. Common Yellowthroat; Spizella pusilla. Field Sparrow; Melospiza melodia. Song Sparrow. Trees-shrubs: Butorides striatus. Green Heron; Zenaida macroura. Mourning Dove; Coccyzus americanus. Yellow-billed Cuckoo; Tyrannus tyrannus. Eastern Kingbird; Empi- donax traillii. Willow Flycatcher; Cyanocitta cristata. Blue Jay; Toxostoma rufum. Brown Thrasher; Vireo griseus. White-eyed Vireo; V ermivora pinus, Blue-winged Warbler; Den- 232 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 droica petechia. Yellow Warbler; Dendroica pensylvanica. Chestnut-sided Warbler; Car- dinalis cardinalis. Cardinal; Passerina cyanea. Indigo Bunting; Carduelis tristis, Ameri- can Goldfinch; Pipilo erythrophthalmus, Rufous-sided Towhee. Open trees: Coccyzus erythrophthalmus. Black-billed Cuckoo; Colaptes aiiratus. Common Flicker; Troglodytes aedon. House Wren; Turdus migratorius, American Robin; Sialia sialis. Eastern Bluebird; Vireo gilvus, Warbling Vireo; Icterus galbula. Northern Oriole; Spizella passerina. Chipping Sparrow. NEW LIFE MEMBER Mr. William W. Cole, Jr. is a new life member of the Wilson Ornithological So- ciety. Professionally Mr. Cole has been a management consultant in the materials control field, although presently he is processing manager for a centrifugal pump manufacturer. He has been an amateur ornithologist since youth and has quite an extensive library of both books and periodi- cals. Mr. Cole conducts an ornithological thematic stamp business on the side and for the past year and one-half has been involved in preparing the rough draft of a comprehensive catalog of bird topical stamps of the world. He is married with twin daughters and in addition to his work and catalog project he enjoys tennis and classical music. NESTING HABITAT OF COMMON RAVENS IN VIRGINIA Robert G. Hooper The Common Raven [Corvus corax) inhabits a region 80-160 km wide extending along the Appalachian Mountains from northern Georgia to north- ern Pennsylvania. The raven is generally described in the regional literature as a wary species that avoids man and his activities by nesting on cliffs in remote mountainous areas at high elevations (Harlow 1922, Murray 1949, Stupka 1963:101—103, and others). A species with these characteristics could be seriously disadvantaged by intensive use of the southern Appalachians for recreation and raw materials. This study was conducted to better define the ecological latitude in which the raven exists in a portion of the southern Appalachians. STUDY AREA AND METHODS Study area. — The study was conducted northwest of Radford, Salem, and Lexington, and southeast of the West Virginia state line. Quartzite cliffs occur on ridges and in water gaps. Shale cliffs are usually adjacent to streams. Limestones and dolomites form bluffs on the Maury and New rivers. Valley elevations are 275-305 m on the James River and 470-538 m on the New River, while ridges rise to 1318 m above sea level. Over 85% of the study area is forest and the remainder is pasture and urban area. Mountain slopes are forested with mixtures of oaks (Quercus spp.) and other hardwoods. Xeric ridges are covered in Virginia, Table Mountain and pitch pine {Pinus virginiana, P. pungens, and P. rigida) . Mesic ravines are forested with hemlock {Tsuga canadensis), white pine (P. strobus) , yellow poplar ( Liriodendron tulipifera) , and other hardwoods (Braun 1967:225-242). The 7000 km^ study area has a population approaching 20 people per km^. Areas of 10-25 km^ with less than 1 person per km^ are scattered throughout. A 2000 km^ strip adjacent to the southeastern boundary supports over 90 people per km^. Ravens have inhabited the study area continuously since pre-Columbian times but no data exist on past abundance. A 518 km^ segment of the study area had 1 active nest per 30.6 km^ (Hooper et al. 1975). I consider this to be a moderate density based on densi- ties in Britain of up to 1 nest per 17.1 km^ (Ratcliffe 1962). Methods. — I searched for raven nests during March and April of 1972 to 1974. Nest building began in late January, and most nests had eggs by 8 March. Nestlings fledged in late April and early May. From 2 to 6 visits were made to each nest site to determine if it was active and the number of young fledged. Nestlings that lacked visible sheaths on their contour feathers as the birds sat in the nest fledged in less than 7 days and were at least 35 days old. I considered birds that reached this stage of development as “fledged.” Dorn (1972) found nestling periods in Wyoming to be 39-45 days. Twenty-eight territories were found. Because birds in 13 territories used alternate nest sites in different years, 41 nest sites were found. I determined the outcome of nesting attempts in 5 territories during all 3 years, in 10 territories during 2 years, and in 9 territories in only 1 year. 233 234 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 1 Productrity of Raven Nests in Relation to Altitude and Roads Altitude or Distance Total Nest Attempts Number Fledged Mean Number Fledged % Success 12 3 4 (% Successful Nests) pa Altitude <580 m 19 68.4 0.0 23.1 46.2 30.8 3.08 .001 >580 m 25 64.0 31.0 25.0 18.8 25.0 2.37 Distance to Road <.4 km 23 73.9 11.8 17.6 41.2 29.4 2.88 >.05’’ >.4 km 21 57.1 25.0 41.7 16.7 16.7 2.25 ® P — .05 was considered significant. Tabular U = 64 and calculated U = 66.5. The Mann-Whitney U-test was used to evaluate differences among and between means (Siegel 1956). Values for U with probabilities less than .05 were considered significant. RESULTS Characteristics of nest sites. — Eighteen nest sites were found between 335- 579 m above sea level, 12 between 580-879 m, and 11 between 880-1130 m. The lowest nest site at 335 m was just above the lowest point in the study area. Few cliffs existed above the highest site at 1130 m. Nest success was similar above and below 580 m, but significantly more young were fledged in suc- cessful attempts below 580 m than at higher altitudes (Table 1). All but 3 nest sites were on cliffs. The use of cliff sites of a particular type of rock was roughly proportional to the abundance of that type. Quartzite, the most common cliff forming rock, composed 27 of the nest cliffs. Seven sites were on shale cliffs, 3 on limestone cliffs, and 1 on a tufa deposit. Height of nest cliffs varied from 4.9 to 38.1 m and averaged 19.7 m. Nests that were always successful during the study were no higher on cliffs than nests that always failed (Table 2). Nest ledges faced south to west at 18 sites and north to east at 10 sites. The main ridges ran northeast-southwest and most cliffs were at the ends of ridges and in water gaps. The other 10 cliff sites were on the sides of ridges. Profiles of nest cliffs had several consistent characteristics. All cliff nests were sheltered by overhanging ledges of 2 cm to 600 cm, measured from the front of the cup. Cliffs were undercut below the nest ledge at 63.2% of the Hooper • RAVENS IN VIRGINIA 235 Comparison Table 2 OF Raven Nest Sites that had Successful Nest Attempts Sites that Failed'^ TO T^a Mean Range Site Number (m) (m) pb Height of Nest Ledffe Success 20 11.2 3.8-19.8 .32 Failure 8 9.9 5.2-18.3 Height of Cliff ...... Success 20 19.6 4.9-38.1 .44 Failure 8 18.6 9.7-27.4 Distanee to Road Success 23 513 60-1450 .26 Failure 8 865 80-2410 Distance to Dwelling Success 23 853 60-1920 .43 Failure 8 1366 320-4830 ® Sites with both successes and failures were not included. ^ .05 was considered significant. sites and nearly vertical at 31.6%. At 5.3% of the sites the rock face sloped away from the nest ledge. One nest was in a cavity in the side of a cliff and another was wedged in a vertical crevice. The other 36 nests were on ledges that were usually shielded by a rock buttress on one or both sides. Mean width of 13 ledges was 63.5 cm and the range 38-107 cm. Eight nests aver- aged 61.3 cm wide with a range of 41-91 cm. Ledges often sloped away from the cliff, and at 4 sites a small shrub helped hold the nest on the ledge. Rock faces were too steep and protected from weathering near nest ledges to sup- port the large foliose lichens that covered more exposed surfaces. One end of a nest ledge on a limestone cliff extended beyond the overhang and the exposed portion was covered with bleeding heart {Dicentra eximia) . This was the only instance where forbs or grasses were seen on a nest ledge. Search of a 104 km^ area without cliffs revealed no tree nests, but I could not be sure none existed. In a 518 km^ area that was intensively searched, one nest was found in a Virginia pine and 2 in shortleaf pines {Pinus echinata). 236 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Table 3 Distribution of Raven Nest Cliffs by Height and Distance FROM Human Dwelling Height of Nest Cliff Distance to Dwelling Number of Cliffs Mean'i Height (m) Range (m) < 15 m ^ 15 m <.8 km 5 13 21.6 4.8-38.1 >.8 km 7 13 18.0 9.1-30.5 Means were not significantly different (p = .14). The nests were built by ravens that occupied 1 territory. In 1972 they at- tempted to build on a cliff about 50 m from the nest tree but the ledge was too steep to hold the nest. A nest of previous years was found on a cliff about 220 m from the tree nest and within 50 m of a newly constructed road. In 1973 and 1974 the birds built within 250 m of the 1972 nest, but farther from the road. These birds fledged 4 young in 1972, 3 in 1973, and 2 in 1974. The percentage of the area that was forested within 0.8 km and 1.6 km of nest sites was obtained from 7.5-min topographic maps. I did not determine home ranges, but adult ravens on several occasions flew over 2 km from the nest. I saw 1 bird fly about 3 km from its nest. Thus 0.8 km and 1.6 km are probably within the home range of most nesting pairs. Within 0.8 km of the nests, the mean area in forest cover was 90% and the range 20-100%. The mean area in forest cover within 1.6 km of the nests was 86% and the range 28-100%. Relation of nest sites to human activity. — Nearly as many low nest cliffs (< 15 m) were found less than 0.8 km from a dwelling as were found in more secluded locations and an equal number of taller cliffs were found in both categories (Table 3). Successful nest attempts averaged closer to roads and dwellings than unsuccessful ones, but differences were not significant (Table 2). Successful nest attempts less than 0.4 km from a road averaged more fledglings than attempts farther away, but the difference was not quite sig- nificant (Table 1). Ravens exhibited 2 general behavior patterns when I was near their nests. In 16 territories the birds were evasive and seldom vocal. They would fly into sight then quickly disappear. Often they repeated this pattern several times but rarely flew close to the nest or intruder. Some birds would soar 400-800 m away but within sight. In 8 territories the birds appeared defen- sive and were vocal. A rapid ‘"‘‘kack-kack . . .” was usually given. This call was given on 3 occasions when ravens near their nests were diving at Red- Hooper • RAVENS IN VIRGINIA 237 tailed Hawks {Buteo jamaicensis) . A sharp single was given by some birds when I was at the nest. This call was given by a tame raven when de- prived of food for short periods or when threatened (D. R. Chamberlain, pers. comm.). Other calls were also heard near nests but these were the prominent ones when I was in direct contact with a nest. Defensive birds flew within 50 m or less of intruders. One bird landed on the cliff 7 m above a worker and another dove within 3 m. In both cases the man could touch the nest. Birds in a given territory usually reacted similarly during each visit. Strik- ing changes were noted in different years in 3 territories, perhaps indicating a replacement of mates. At 2 nest sites, birds that were normally evasive to humans reacted defensively when Red-tailed Hawks flew close to the nests. Unusually defensive birds in another territory became evasive when I found a prematurely fledged nestling at the base of the cliff. Residents of homes 60, 90, and 150 m from a nest did not know of its existence, although a former resident knew the nest had been active at least since 1961. The nest was on a river bluff and the houses were on top, 50 m above the river. The ravens were probably evasive to attract so little attention from people living above them. When I found the nest in 1974 the ravens were defensive and gave the call until I was 200 m downstream and out of sight. Since ravens cannot defend a nest against a human, an evasive reaction seems more adap- tive than a defensive one, particularly since defense advertises the presence of a nest. However, 50% of the pairs classified as defensive and 44% classi- fied as evasive nested within 0.4 km of a road or dwelling. In one case an extremely evasive pair of ravens may have jeopardized nest success through inattention to nestlings. The nest failed in all 3 years of the study and may have been unsuccessful for 3 years prior to the study, according to observa- tions of a local fisherman. The nest was on a 27 m cliff and was well-pro- tected from direct human intervention. However, the cliff was 150 m from a popular trout stream and a frequently traveled road. In 14 trips by the nest site, adults were seen only twice and were evasive both times. The oldest nestling I saw was about 10 days old. The nest most subject to human activity was on private land and the owner permitted interested groups to visit the nest for short periods. On one oc- casion in 1972, about 20 people were below the nest ledge and a climber rap- pelled to the nest. Two young were fledged that year. In 1973, a graduate student rappelled to the nest once a week throughout the nesting cycle, and 4 young were fledged. The site was not used for nesting in 1974 or 1975, al- though a pair of ravens frequented the cliff in January of both years. Adults at this site were evasive. At only one site did I suspect destruction of a nest by humans. The nest was on a 9.6 m cliff 40 m from a road and clearly visible to people passing in cars. 238 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 DISCUSSION Selection of nest sites. — Cliff profile — a suitable ledge with an overhang above and steep rock face below — was one of 2 factors apparent in the selec- tion of a particular cliff as a nest site. Similar sites have been described by others ( Harlow 1922, Bent 1946, Ratcliffe 1962, White and Cade 1971). The lack of deviation from this profile in the nest sites I found is important, be- cause cliffs with these attributes are relatively scarce in the southern Ap- palachians. The second factor was the proximity of other active raven nests. In a 518 km^ segment of the study area, the mean distance to the nearest active nest was 4.3 km and the closest nests were 2.2 km apart (Hooper et al. 1975). Although scarcity of potential nest cliffs probably limit some local populations in the southern Appalachians, a surplus of suitable cliffs was available in the above-mentioned portion of the study area. Ratcliffe (1962) thought the den- sity of nesting ravens, unless nest sites were scarce, was determined by a proximity tolerance limit of nesting pairs to each other and that the limit in a particular area was dependent upon the food supply. The regular spacing of active nests, the moderate density of nesting pairs, and the surplus of suit- able nest cliffs suggest that overall food supply may have been regulating the population to a large extent in my area. But, until a surplus of potential breeders is demonstrated and other facets of the population dynamics exam- ined, the above hypothesis lacks support. Because tree nests were difficult to find in the heavily timbered area in which I worked, I did not adequately assess their relative value. In the 518 km^ area previously mentioned, birds in 1 of 17 territories nested in trees. Since a surplus of cliff sites existed, there was little need to use trees. Ravens nest extensively in trees in other regions, and such sites may be more impor- tant in the southern Appalachians than my data indicate. In Pennsylvania, Harlow (1922) found cliff nests outnumbered tree nests about 8 to 1. Altitudinal relationships. — Only 27% of the nest sites I found were over 880 m above sea level and 44% were below^ 580 m. Although nesting at lower altitudes is probably not a recent adaptation, specific nest locations men- tioned in the regional literature were all above 880 m (Bailey 1913, Jones 1933, Hostetler 1938, Tyrrell 1945, Murray 1957). Murray (1957) men- tioned a nest near a hard-surfaced road that was probably below 880 m. About 1950 a resident of Giles County found a raven nest with young at 518 m near the New River. One site at 335 m on the Maury River was active as early as 1961. Thus, nesting occurred at low altitudes in the 1950’s if not sooner. The frequency of nest sites found at lower elevations could indicate an increase in the raven population. Although the difference I found in production of fledglings at high and low altitudes was significant (p ^ .001), the biological implications are not Hooper • RAVENS IN VIRGINIA 239 known. A similar but non-significant relationship between altitude and pro- ductivity was reported by Allin (1968) in Wales. I probably worked at the low end of the altitudinal range of the raven in the southern Appalachians. Jones (1933) reported a nest at 1500 m in Virginia. Most sightings of ravens in the Smoky Mountains of Tennessee and North Carolina were above 1070 m (Stupka 1963). More recently, W. D. Zeedyk (pers. comm.) has sighted birds primarily at 1070-1900 m in North Carolina. Nests have been reported at these elevations (Stupka 1963; B. A. Sanders, pers. comm.). Reduced pro- duction of fledglings with increasing altitudes, particularly if the trend con- tinues above elevations I worked, is potentially an important factor in the population dynamics of the raven in the southern Appalachians. Stupka (1963) thought ravens were more plentiful in the Great Smoky Mountains prior to creation of the National Park when livestock was pastured at the higher elevations. The distribution of food within my study area could have been a cause of higher productivity at the lower elevations. A greater food supply probably existed in the valleys, where railroads, houses, farms, and most roads were located. Investigation of raven food habits in the study area indicated considerable foraging was done at lower altitudes (Harlow et al. 1975). Ravens nesting at higher elevations may have to spend more time hunting for food, or have to fly farther to find it. If so, the efficiency of adults feeding their nestlings would be affected, perhaps contributing to in- creased mortality in nestlings through starvation. That starvation is a major cause of mortality in nestlings of other corvids was shown by Holyoak (1967) for Corpus frugilegus and C. corone, and by Mishaga (1974) for C. crypto- leucus. Interactions with humans. — Ravens in the southern Appalachians have a reputation of avoiding human contact by living in remote areas. However, many nested relatively close to human residences and probably relied on hu- man activities for a substantial part of their food. Ravens were persecuted in former times (Sprunt 1956), but I believe killing of the species by man is currently of minor importance in Virginia. I found no evidence that ravens in my area caused loss of sheep as reported in the West (Larsen and Dietrich 1970). The only nuisance reports I had on ravens were on a flock that dug grubs from a golf green and on a single bird that took balls from another course. Several birds were destroyed as a result, but these were isolated in- cidents. Thus, there is little reason for man to molest ravens in Virginia. The apparent restriction of ravens to higher and remoter areas in other parts of the southern Appalachians suggests that persecution may still be a problem in those sections. Hickey (1942) found that the minimum acceptable height for a Peregrine Falcon {Falco peregrinus) nest site was inversely related to its remoteness 240 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 and directly related to the amount of molestation by man. Ratcliffe (1962) found the same relationship for the raven and Peregrine in England, but a similar one was not apparent in my data. Only 17.2% of the raven nest cliffs in England within 0.8 km of a main road or dwelling were less than 30 m tall compared to 77.8% in this study. Also, nest success in Virginia was higher and number of young fledged greater within 0.4 km of roads than farther away. Raven nest cliffs probably have been used for centuries and occupancy of a given site may have continued as the landscape slowly changed. During years of inactivity, evidence of past nesting remained perhaps to serve, along with the general suitability of the cliff, as a stimulus to birds seeking a nest site. Of course, many former sites in the southern Appalachians are no longer in use, especially in Alabama and Kentucky where the species was extirpated (Imhof 1962, Mengel 1965), as well as throughout the region where present densities are low. Hickey (1942) thought the occupancy of a given site by Peregrines in face of human disturbance was determined by the personality of the resident birds. Personality differences in ravens were noted by Harlow (1922), Ratcliffe (1962), and Dorn (1972). The ability of ravens I studied to cope with human activity near their nest varied greatly and apparently played a role in nesting success of sites close to roads and dwellings. How- ever, the overt reaction of ravens to humans near their nests was not appar- ently related to the birds’ tolerance of disturbance. Human activity, in my judgment, should be curtailed near active nests, despite the tenacity of some nesting pairs. The actual distance to restrict activity depends on the terrain and type of activity. In general, pedestrians should not be permitted within 200 m of a nest if they are visible to birds on the nest cliff, or within 100 m if they are hidden from view. Vehicular traffic as close as 100 m to a nest would not create excessive disturbance if parking areas are not provided within 200 m. However, road construction within 200 m could cause desertion. Overlooks should not he built on top of nest cliffs. Rockclimbing should be discouraged on active nest cliffs from 15 January until the nestlings fledge in late April or early May. These guidelines are more liberal than I had originally thought possible. Although a few birds may he adversely affected by humans within 200 m of the nest, I believe most will not. Unless physically harassed, many birds would accept closer contact. SUMMARY Common Ravens in Virginia were primarily cliff nesters. The major factors apparent in selection of a nest site were cliff profile, determined by a suitable ledge with an overhang above and steep rock face below, and the distance to other active raven nests, the closest 2,2 km and the average 4.3 km. Nest cliffs averaged 19.7 m in height. No significant difference was found between heights of successful and unsuccessful sites. Hooper • RAVENS IN VIRGINIA 241 Nest cliffs close to human activity were not taller than those in remote areas. Observed proximity of roads and dwellings to nests had no significant effect on nest productivity. Nest sites were found between 335-1130 m above sea level, with 44% below 580 m. Suc- cessful nests below 580 m fledged a mean of 3.08 young compared to 2.37 at higher ele- vations. Starvation of nestlings, due to a loss of feeding efficiency in adults nesting at higher elevations, was suspected. ACKNOWLEDGMENTS I am especially grateful to the Harold H. Bailey Research Trust for providing the Rock- bridge Alum Springs Biological Laboratory as a base for the fieldwork in 1972-1974. The help of C. A. Dachelet, D. R. Chamberlain, H. S. Crawford, T. Ziegler, R. F. Harlow, T. C. Cutler, and R. N. Conner made the study possible as well as pleasurable. The cooperation of the Virginia Commission of Game and Inland Fisheries, George Washing- ton and Jefferson National Forests and many private landowners is appreciated. D. R. Chamberlain, H. B. Tordoff, C. M. White, and W. D. Zeedyk made helpful comments on the manuscript. LITERATURE CITED Allin, E. K. 1968. Breeding notes on ravens in north Wales. Br. Birds 61:541-545. Bailey, H. H. 1913. The birds of Virginia. J. P. Bell Co., Lynchburg, Va. Bent, A. C. 1946. Life histories of North American jays, crows, and titmice. U.S. Natl. Mus. Bull. 191. Braun, E. L. 1967. Deciduous forests of eastern North America. Hafner Publ. Co., N.Y. Dorn, J. L. 1972. The Common Raven in Jackson Hole, Wyoming. M.S. thesis, Univ. Wyoming, Laramie. Harlow, R. C. 1922. The breeding habits of the Northern Raven in Pennsylvania. Auk 39:399-410. Harlow, R. F., R. G. Hooper, D. R. Chamberlain, and H. S. Crawford. 1975. Some winter and nesting season foods of the Common Raven in Virginia. Auk 92:298-306. Hickey, J. J. 1942. Eastern population of the Duck Hawk. Auk 59:176-204. Holyoak, D. T. 1967. Breeding biology of the Corvidae. Bird Study 14:153-168. Hooper, R. G., H. S. Crawford, D. R. Chamberlain, and R. F. Harlow. 1975. Nest- ing density of Common Ravens in the Ridge-Valley Region of Virginia. Am. Birds 29:931-935. Hostetter, j. H. 1938. The raven. Raven 9:94-96. Imhof, T. 1962. Alabama birds. Univ. Ala. Press, University. Jones, F. M. 1933. Nesting of the Northern Raven in Virginia. Wilson Bull. 45:201- 202. Larsen, K. H. and J. H. Dietrich. 1970. Reduction of a raven population on lambing grounds with DRC-1339. J. Wildl. Manage. 34:200-204. Mengel, R. M. 1965. The birds of Kentucky. Ornithol. Monogr. No. 3. Mishaga, R. 1974. Notes on asynchronous hatching and nestling mortality in White- necked Ravens. Wilson Bull. 86:174-176. Murray, J. J. 1949. Nesting habits of the raven in Rockbridge County. Raven 20: 40-43. . 1957. The birds of Rockbridge County, Virginia. Va. Soc. Ornithol. Ratcliffe, D. a. 1962. Breeding density in the Peregrine Falco peregrinus and raven Corvus corax. Ibis 104:13-39. 242 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, N.Y. Sprunt, a., Jr. 1956. Is the raven coming back in the Southeast? Aud. Mag., July- Aug. 170-171, 182-183. Stupka, a. 1%3. Notes on the birds of Great Smoky Mountains National Park. Univ. Tenn. Press, Knoxville. Tyrrell, W. B. 1945. A study of the Northern Raven. Auk 62:1-7. White, C. M. and T. J. Cade. 1971. Cliff-nesting raptors and ravens along the Colville River in Arctic Alaska. Living Bird 10:107-150. USDA, FOREST SERVICE, SOUTHEASTER'S FOREST EXPERIMENT STATION, CLEMSON, SC 29631. ACCEPTED 8 DEC. 1975. GROWTH AND DEVELOPMENT OF KNOWN-AGE RING-BILLED GULL EMBRYOS John P. Ryder and Lynn Somppi This paper describes growth and development of Ring-billed Gull (Lams delawarensis) embryos. It provides a basis for estimating the age of eggs at previously unvisited colonies. The data also supply a way to determine, within a colony, the location of early and later nesting pairs by comparing, during the same sampling time, relative ages of eggs located in different parts of a colony. STUDY AREA AND METHODS There are no descriptions of embryo growth and development of Ring-billed Gulls in the literature. Dawkins et al. (1965) compared increase in body weight and development of California Gull {L. californicus) and Domestic Chicken {Callus gallus) embryos using known-age eggs. Drent (1970), using knowm-age embryos plus data from Paludan (1951) and Harris (1964) formulated a logarithmic body weight curve for the Herring Gull (L. argentatus) . Maunder and Threlfall (1972) described the growth and development of various parts of Black-legged Kittiwake (Rissa tridactyla) embryos. Gilbertson and Hale (1974) used Maunder and Threlfall’s (1972:800) body weight curve for the Black-legged Kittiwake embryos to age those of the Herring Gull. We consider such inter-specific comparisons weak because different species do not necessarily show the same develop- mental characteristics at equivalent age. We decided not to use the egg flotation tech- nique devised by Westerkov (1950) because the results vary depending on egg size and the age of the egg when incubation starts. Additionally, Schreiber (1970) noted that addled and infertile eggs of Western Gulls (L. occidentalis) show essentially the same flotation as viable eggs during the early stages of development. We conducted this study in 1975 at a colony of approximately 800 pairs of Ring-billed Gulls on Granite Island (48°43'N, 88°29'W), Black Bay, northern Lake Superior, On- tario. The island is a strongly undulating granite outcrop 402 m by 201 m with a summit 30 m above the surrounding water. Soil and vegetation occur in depressions of the rock surface. Each spring Ring-billed Gulls nest in the depressions especially near the summit away from wave action and possible flooding. Dominant plants in the depressions are Kentucky bluegrass (Poa pratensis) , rough cinquefoil {Potentilla norvegica) , and red raspberry {Rubus strigosus) . The remainder of the island is densely forested with balsam fir {Abies balsamea) , white cedar {Thuja occidentalis) , and white birch {Betula papyri- fera). On 16 May we marked, with a black felt pen, 31 1-egg clutches and 5 2-egg clutches in the center and 19 1-egg clutches and 13 2-egg clutches on the periphery of the Granite Island colony to determine if equivalent age embryos showed equal development in the two areas. Central and peripheral clutches were designated respectively as those near the geometric center of the colony and those forming the outside border (Dexheimer and Southern 1974) . Only nests which subsequently contained 3 eggs formed our sample so that we eliminated potential variation in development because of different clutch sizes. Our final sample was 29 central and 21 peripheral nests. 243 244 THE WILSON BULLETIN • Vol 89, No. 2, June 1977 Fig. 1. Measurements of Ring-billed Gull embryos. A, head length; B, head width; C, cuhnen; D, eye diameter; E, total length; F, back length; G, hand; H, forearm; I, tibia; J, tarsus; K, midtoe. Modified from Maunder and Threlfall (1972). The age of eggs found in the study nests on our first visit was determined by assuming a 1.9 day interval between laying of successive eggs in Ring-billed Gulls (Vermeer 1970: 20) . For example, if a 1-egg clutch on 16 May contained 2 eggs on 17 May, we assumed the first egg was laid on 15 May. If a second egg was not in the nest on 17 May but was by 18 May, we assumed the first egg was laid on 16 May. We considered the longest egg in the 2-egg clutches marked on 16 May the first laid (Vermeer 1%9, Ryder 1975). Because we assumed the 1.9 day interval between successive eggs, we aged embryos to an accuracy of ±24 h and grouped embryos according to age into 3-day intervals. On 17 May we collected 2 clutches each from the center and periphery of the colony in which the second egg was freshly laid and 2 clutches from each area in which the third egg w'as fresh. From these, we determined if any development occurs in first and second eggs by the time the third egg is laid. Sampling for the remainder of the study involved taking individual first, second, and third eggs of known age from different nests in each Ryder and Somppi • GULL EMBRYO DEVELOPMENT 245 Table 1 Extremes in Early Development of Known-age Ring-billed Gull Embryos FROM THE Center and Periphery of the Granite Island Colony, 1975 Age (days) Development Center Periphery 1 No development to blastodisc with diameter 0.66 cm. No development. 3 Head fold stage of primitive streak to embryo with 30 somites, heart beating, area vasculosa developed. No development to primitive streak. 5 Embryo with 18 somites to an embryo 0.82 cm in length, wing and leg buds visible. Embryo with 14 somites to 16 somites. 6 Embryo 0.79 cm in length, slightly prominent midbrain to embryo 1.0 cm in length, leg bud 0.30 cm and wing bud 0.42 cm, choroid fissure visible. Embryo with 23 somites to em- bryo 0.99 cm in length, promi- nent midbrain, choroid fissure visible. 7-9 Embryo 0.93-1.56 cm in length, area vasculosa 4.52 cm at sinus terminalis, body wt. 0.25-0.40 g, wing bud 0.36- 0.43 cm, leg bud 0.32-0.41 cm, well defined choroid fissure. Embryo 0.78-1.54 cm in length, area vasculosa 3.74-6.10 cm at sinus terminalis, body weight 0.30-0.50 g, wing bud 0.39-0.46 cm, leg bud 0.28-0.53 cm, well defined choroid fissure. sampling period so that each of the eggs collected was of equal age per sampling day. Each egg removed from a nest was replaced by an unmarked one in an attempt to elimi- nate any growth changes in the remaining study eggs which might have resulted from reduced attentiveness by the parents because of a smaller clutch (see Beer 1965). We opened eggs by the procedures outlined in Rugh (1962). This involved cutting around the widest diameter of the egg and emptying the entire contents into a petri dish without breaking the yolk. During the early stages of development we retained the em- bryo in the yolk and measured the diameter of the blastodisc and area vasculosa, number of somites, and other general developmental characteristics using a Wild M5 dissecting microscope. We removed embryos older than 6 days from the yolk and immediately weighed them to the nearest 0.1 g on a triple beam balance and measured them to 0.01 mm with vernier calipers. Embryos were preserved in 10% neutral buffered formalin. The measurements taken are illustrated in Fig. 1. Later development is defined here as that shown by an embryo 10 or more days old. By this time all of the various body parts are easily visible and can be measured accurately. RESULTS Early development (1-9 days). — The aging of these embryos presented problems because of considerable variation in development among embryos 246 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 2 Intra-clutch Variation in Early Development of Ring-billed Gull Embryos FROM THE Center and Periphery of the Granite Island Colony, 1975 Development Egg Age No. (days) Center Periphery 1 3 11 somites. Embryonic shield stage. 2 1 No development. No development. 1 3 Primitive streak stage. No development. 2 1 No development. No development. 1 5 18-19 somites; area vasculosa developing. 14 somites; initiation of area vasculosa. 2 3 7 somites; optic vesicle visible. Head fold stage of primitive streak. 3 1 No development. No development. 1 5 Embryo 0.82 cm in length. Limb buds visible. 16 somites, head turned. 2 3 Area vasculosa well developed. Heart visible and beating. Head fold stage of primitive streak. 3 1 Diameter of blastodisc 0.66 cm. No development. of the same age. Table 1 details our results of embryos from the center and periphery of the colony to 9 days of age. Embryos from the center of the colony were slightly advanced to those of equal age from the periphery (Tables 1 and 2). It was clear from our collections of complete clutches on 17 May that some development occurred in first and second eggs before the third egg was laid (Table 2). In all clutches the first egg showed more development than the second and the second more than the third. Similar variation in first, second, and third eggs of Herring Gulls was reported by Parsons (1972). The sig- nificance of this result is that based on the apparent differential development among eggs of a single clutch, individual eggs should be aged according to the day each was laid and not from the day on which the clutch was completed (see Drent 1970:80 and Parsons 1972:540). Later development (10 days to hatching). — Figure 2 illustrates typical Ring- billed Gull embryos in each of the 3-day groupings. Embryos collected from the center were slightly, but not significantly (P > 0.05) larger than their peripheral counterparts. Consequently, we grouped all embryos from both Ryder and Somppi • GULL EMBRYO DEVEI.OPMENT 247 Fig. 2. A series of known-age Ring-billed Gull embryos, Granite Island, 1975. A, 7-9 days; B, 10-12 days; C, 13-15 days; D, 16-18 days; E, 19-21 days; F, 22-24 days; G, 25- 27 days (pipping, note intact yolk sac). Embryos shown actual size. BACK LENGTH IN CM WEIGHT IN GRAMS 248 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 0 0 -n ^ ^ ^ ^ ^ ^ 4-6 ' 10-12 ' 16-18 22-24 28 7-9 13-15 19-21 25-27 0 0 4-6 “1 — ^ — I — ' — I — ^ — n ' 10-12 16-18 22-24 ' 28 7-9 13-15 19-21 25-27 AGE IN DAYS AGE IN DAYS Fig. 3. Increase in weight, total length, back length, and culmen length of Ring-billed Gull embryos, Granite Island, 1975. Vertical line is range; horizontal line is mean; rec- tangles enclose X ± 1 SD, and number above is sample size. Ryder and Somppi • GULL EMBRYO DEVELOPMENT 249 5-| 4- 3- 2- HEAD S i • ^0. UJ ^2 MIDTOE UJ o 1 — I — I — I — \ — ^ — r O 2-, HAND I- I 7-9 I Isi-IS I I9-2I I 25-27 I 4-6 I0-I2 I6-I8 22-24 28 TIBIA Tn — r-rn — i — r I I TARSUS 1 — ^ rn ^ \ r FOREARM I5 i • 0 I 7-9 I I3-I5 I I9-2I I 25-27 I 4-6 I0-I2 I6-I8 22-24 28 AGE IN DAYS Fig. 4. Growth of body parts of Ring-billed Gull embryos, Granite Island, 1975. Num- ber above each dot is sample size. areas to determine aging parameters useful to the field investigator. Figure 3 shows growth in terms of body weight, total length, back length, and culmen length. No overlap at 1 standard deviation occurred in any of the age group measurements except those recorded near the end of incubation. Some of the chicks in the 22-24-day and 25-27-day groups were in the process of pipping 250 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Table 3 Field Chart for Aging Ring-billed Gull Embryos Age (days) Characteristics 1- 3 No development through primitive streak stage to presence of somites, heart beating, blood vessels on area vasculosa. 4- 6 Embryo usually ^1.0 cm in length; wing and leg buds appearing; brain becoming prominent, eye pigmented; choroid fissure visible. mid- 7- 9 Embryo 1-2 cm in length; less than 1 g body wt. 10-12 Embryo 2-4 cm in length; 1-2 g body wt. Choroid fissure complete. 13-15 Embryo 4^5 cm in length; 2-6 g body wt. ; pterylae visible; feathering on spinal and caudal tracts. 16-18 Embryo 5-7 cm in length; 6-13 g body wt. All pterylae feathered; dorsal pterylae well feathered. 19-21 Embryo 7-8 cm in length; 13-20 g body wt. All pterylae feathered; ment appearing in upper and lower mandibles, claws, and legs. pig- 22-24 Embryo >8 cm in length; >20 g body wt.; mandibles, claws, and well pigmented; yolk compacted when close to hatching. feet and of similar size. The point at 28 days in Figs. 3 and 4 represents 1 pe- ripheral embryo which, along with others, tended to take longer to start pipping compared to embryos from the center. Figure 4 presents growth curves for various body parts. They are not as useful by themselves as an aging tool because of overlap at 1 standard devia- tion in most cases. However, they do provide aid to the field researcher as a supplement to data in Fig. 3. Table 3 summarizes the results. Based on 2 easily obtained measurements, weight and length of body, one can estimate the age of an embryo to within 3 days. Supplementary information such as degree of feathering and pig- mentation also aid in embryo age determination. DISCUSSION Data in this report provide the basic information required to age embryos of Ring-billed Gulls to within 3 days. Of interest are the statistically insignifi- cant differences between equivalent age embryos from the center and periph- ery of the colony. The general lack of differences in development character- istics and size of newly hatched chicks from the 2 areas suggests the young have approximately the same chance of survival. Dexheimer and Southern Ryder and Somppi • GULL EMBRYO DEVELOPMENT 251 (1974) found no significant difference in fledging success between central and peripheral Ring-billed Gull chicks on an island similar to our study area where flooding posed no problem. They did find significant differences in fledging success on an island where peripheral chicks were exposed to wave action and excessive wetting causing death. SUMMARY Characteristics for aging Ring-billed Gull embryos to within 3 days are given. Body weight, total length, back length, and culmen length from 10 days after laying to hatch- ing were the most accurate aging parameters. No overlap at 1 standard deviation occurred in any of the above characters in 3-day intervals. ACKNOWLEDGMENTS Financial support for this and related studies of Ring-billed Gulls was obtained from the National Research Council (A6520 to J. P. R.). We thank Mr. R. Trowbridge for allowing us to base operations at Bonavista. We appreciate the field assistance of Mr. T. Carroll and thank Prof. B. Spencely for producing Fig. 2 and Mrs. B. Salo for typing the manuscript. Helpful suggestions on an earlier draft of this paper were received from J. Burger and R. W. Schreiber. LITERATURE CITED Beer, C. G. 1965. Clutch size and incubation behaviour in Black-billed Gulls {Larus bulleri) . Auk 82:1-18. Dawkins, B. A., W. H. Behle, and W. W. Newby. 1965. The embryonic development of the California Gull {Larus calif ornicus) . Univ. Utah Biol. Ser. 13:1-26. Dexheimer, M. and W. E. Southern. 1974. Breeding success relative to nest location and density in Ring-billed Gull colonies. Wilson Bull. 85:288-290. Drent, R. H. 1970. Functional aspects of incubation in the Herring Gull. Behaviour (Suppl.) 17:1-132. Gilbertson, M. and R. Hale. 1974. Early embryonic mortality in a Herring Gull colony in Lake Ontario. Can. Field-Nat. 88:354-356. Harris, M. P. 1964. Aspects of the breeding biology of the gulls Larus argentatus, L. fuscus and L. marinus. Ibis 106:432-456. Maunder, J. E. and W. Threlfall. 1972. The breeding biology of the Black-legged Kittiwake in Newfoundland. Auk 89:789-816. Paludan, K. 1951. Contributions to the breeding biology of Larus argentatus and L. fuscus. Vidensk. Medd. fra Dansk. Naturh. foren. 114:1-128. Parsons, J. 1972. Egg size, laying date and incubation period in the Herring Gull. Ibis 114:536-541. Rugh, R. 1962. Experimental embryology. Burgess Publ. Co., Minneapolis, Minn. Ryder, J. P. 1975. Egg-laying, egg size, and success in relation to immature-mature plumage of Ring-billed Gulls. Wilson Bull. 87:534-542. Schreiber, R. W. 1970. Breeding biology of Western Gulls {Larus occidentalis) on San Nicolas Island, California, 1968. Condor 72:133-140. Vermeer, K. 1969. Egg measurements of California and Ring-billed gull eggs at Miquelon Lake, Alberta, in 1965. Wilson Bull. 81:102-103. 252 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 . 1970. Breeding biology of California and Ring-billed gulls. Can. Wildl. Serv. Rep. Ser. 12:1-52. Westerkov, K. 1950. Methods for determining the age of game bird eggs. J. Wildl. Manage. 14:56-67. DEPT. OF BIOLOGY, LAKEHEAD UNIV., THUNDER BAY, ONTARIO, CANADA P7B 5E1. ACCEPTED 25 MAR. 1976. HABITAT PARTITIONING IN A COMMUNITY OF PASSERINE BIRDS Robert C. Whitmore Since the concept of the niche was brought into the forefront by G. E. Hutchinson (1957), many ecologists have sought to analyze niche relation- ships in natural communities. The quest for quantification has led to numer- ous field studies from which huge quantities of ecological data have been amassed, much of it concerned with birds (e.g. MacArthur 1958, Hespenheide 1971, Willson 1974) . One problem in such studies is to visually synthesize relationships from complex data matrices. Several techniques have been developed to address this problem. One such technique is ordination. Briefly stated, ordination is an arrangement of units in a uni- or multi-dimensional order as opposed to a classification in which units are arranged in discrete classes (Bray and Curtis 1957). Classically, ordinations have been restricted to plant complexes but ecologically meaningful ordinations can be con- structed of animal data as well (e.g. James 1971, Whitmore 1975a). Additional operational problems are listed by Green (1971) : (1) there is a practical limit to the number of environmental parameters which can be measured, and (2) many of the parameters measured are likely to be highly correlated (redundant), and some may be relatively invariate or irrelevant. The use of multivariate statistical analyses, especially those techniques which reduce the number of variables to a more easily visualized set, can help pro- vide answers to these last 2 problems. Combining ordination with multi- variate statistics can give insight into all of the above problems (e.g. James 1971, Whitmore 1975a) . Once the position of the birds along environmental gradients has been established, generalizations can be made about their rela- tionships with each other and other species. The purpose of this study was to quantify the relationships of a community of passerine birds in an attempt to ascertain which variables are important in habitat selection, to develop an ordination along environmental gradients, and to determine the range of habitat use. STUDY AREA AND METHODS The Virgin River Valley is located in the southwest corner of Utah, northwest corner of Arizona, and southeast tip of Nevada at about 37°N 113°W. Lower Sonoran desert sur- rounds the valley on 3 sides and the Pine Valley Mountains border on the north. The Valley is an isolated oasis from the rather harsh surrounding environment and the den- sity of birds in it is quite high. I collected data in the streamside vegetation along Santa Clara Creek to the Virgin River and along the Virgin River to Zion National Park. The area is characterized by stands of mature Fremont cottonwood {Populus fremontii) , large 253 254 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 clumps of tamarix {Tamarix pentandra) , and isolated patches of sand bar willow (Salix exigua) . Much of the river valley is heavily planted with alfalfa {Medicago sativa) and numerous species of fruit, nut, and ornamental trees. The rivers and streams of the area usually flow year round, though excessive removal of water for irrigation or unusually low rainfall will sometimes cause drought in August and September. Avifaunal investigations in the valley have been primarily restricted to species accounts, most notable those of Behle (1943), Wauer and Russel (1967), and Wauer (1969). Much collecting, under the direction of W. H. Behle, has been done in the valley and sur- rounding areas. Vegetational data were collected between 1 May and 30 June 1973 using a modification of the range finder circle method described by James and Shugart (1970). I measured 10 vegetational variables in a 0.04 ha circular plot around each singing, territorial male bird encountered while walking along the river. Habitats for 421 individuals of 24 spe- cies of passerine birds were measured. Table 1 lists the passerine species I encountered. In order to determine which variables were important in species’ separation, the data were subjected to stepwise discriminant analysis (Dixon 1970), a multivariate statistical technique. The underlying theory for the use of the discriminant function in ecology is discussed elsewhere (Green 1971, 1974, James 1971, Whitmore 1975a) and will not be considered here. The stepwise adaptation of the discriminant analysis allows for insertion of each variable in a stepwise manner based on its ability to achieve discrimination be- tween species. The order of insertion determines the order of importance in group separation. When dealing with an n-dimensional data matrix, interpretations may be more meaning- ful if the number of axes can be reduced to a number which can be easily visualized. Principal component analysis (PCA) can be used to accomplish such a task. PGA pro- duces linear combinations of the original variables in such a manner as to explain progressively smaller portions of the total variance in the data. This total variance is the sum of the variances for each of the variables. The first axis is constructed so that linear combination of variables represents the greatest amount of response variance. The second axis, which is orthogonal to the first, represents the second greatest amount of variance. The third represents the third greatest amount and so on. The sum of the variance of components is the total variance. The data were subjected to PCA in order to produce an ordination along vegetational gradients. Vegetational resource use was calculated by dividing each of the measured variables into discrete units. For example, canopy cover was divided into 10, 10% classes and the number of individuals in each species present in each class was determined. After the development of resource matrices for each of the 4 most important variables as deter- mined by stepwise discriminant analysis, I calculated vegetational resource use values using the procedures deseribed by Colwell and Futuyma (1971) for expanded matrices. RESULTS Discriminant analysis. — Results of the stepwise discriminant analysis have been presented elsewhere (Whitmore 1975a). Prior to calculating the discriminant functions the stepwise discriminant analysis program performs a multivariate analysis of variance (MANOVA) among the species based on the measured variables. The advantage of such an analysis is that it not only accounts for the variate but also the covariate Whitmore • PASSERINE HABITAT PARTITIONING 255 Table 1 List of Passerine Species Encountered in the Virgin River Valley from 1 May TO 30 June 1973 Giving Symbols used in Future Tables and Figures WK Western Kingbird (Tyrannus vertical is) AF Ash-throated Flycatcher {Myiarchus cinerascens) * Black Phoebe (Sayornis nigricans) WF Willow Flycatcher {Ernpidonax traillii) ❖ Western Flycatcher (Empidonax difficilis) ❖ Western Wood Pewee (Contopus sordidulus) HW House Wren {Troglodytes aedon) BW Bewick’s Wren {Thryomanes bewickii) * Rock Wren iSalpinctes obsoletus) BGN Blue-gray Gnatcatcher {Polioptila caerulea) * Phainopepla (Phainopepla nitens) WV Warbling Vireo (Vireo gilvus) Orange-crowned Warbler (V ermivora celata) LW Lucy’s Warbler (V erinivora luciae) YW Yellow Warbler (Dendroica petechia) AW Audubon’s Warbler (Dendroica coronata auduboni) YT Common Yellowthroat (Geothlypis trichas) YBC Yellow-breasted Chat (Icteria virens) MW MacGillivray’s Warbler (Oporornis tolmiei) WW Wilson’s Warbler ( Wilsonia pusilla) ❖ Yellow-headed Blackbird (Xanthocephalus xanthocephalus) RWB Red-winged Blackbird (Agelaius phoeniceus) BHC Brown-headed Cowbird (Molothrus ater) * Hooded Oriole (Icterus cucullatus) BO Bullock’s Oriole (Icterus galbula bullockii) * Summer Tanager (Piranga rubra) BHG Black-headed Grosbeak (Pheucticus melanocephalus) BG Blue Grosbeak (Guiraca caerulea) IB Indigo Bunting (Passerina cyanea) LB Lazuli Bunting (Passerina amoena) * House Finch (Carpodacus mexicaniis) LG Lesser Goldfinch (Spinus psaltria) * Rufous-sided Towhee (Pipilo erythrophthalmus) AT Abert’s Towhee (Pipilo aberti) SS Song Sparrow (Melospiza melodia) * Not included in analyses since the sample was less than 5. i, j , relationships among habitat variables (Anderson and Shugart 1974). The { regular assumptions required for statistical tests are needed here and are considered to be met based on the multivariate central limit theorem (Morri- son 1967). A total of 276 species comparisons can be made from 24 different j species; of these all but 4 were significantly different (p < .01). These 4 <1 256 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 2 Correlation Matrix (r) for 10 Vegetational Variables N = 24 SPTI T-7 T-15 T-22 T-30 T-38 % SD % CC CHM SPT T-7 0.65* * T-15 0.74* 0.83* T-22 0.78* 0.85* T-30 0.69* 0.40* T-38 0.63* 0.21 %SD -0.09 0.28 %cc 0.78* 0.64* CHM 0.85* 0.39* %GC -0.32 -0.18 0.90* 0.59* 0.68* 0.34* 0.47* 0.83* 0.01 0.01 -0.22 0.62* 0.66* 0.61* 0.44* 0.55 0.66* -0.30 -0.25 -0.39* -0.30 0.60* -0.10 0.74* -0.29 -0.81* -0.33 -0.28 -0.17 -0.20 1 See key to abbreviations of variables in Table 3. * Significant at p < 0.05. will be discussed later. It is remarkable that 10 measured variables can sepa- rate 272 of the possible 276 species pairs, especially since the environment is restricted and low in plant diversity. Whitmore (1975a) presented a 2-di- mensional ordination along the first 2 discriminant function axes and com- pared those results with those of James (1971). Principal component analysis. — Since PCA is usually based on a correla- tion matrix, it is of value to examine the correlations among the vegetational variables. As can he seen in Table 2 many of the variables are highly cor- related. Values greater than 0.34 are significant (p < .05). James (1971) found many similar vegetational correlations in a study in Arkansas. Per- cent ground cover is negatively correlated with all other variables. This cor- responds to going from the high biomass forested areas, cottonwood and tamarix stands, to the low biomass open areas, alfalfa and open field. The highest correlations are found among the 3 classes of middle and small sized trees, possibly corresponding to the isolated willow and tamarix stands. The 2 classes of large trees are positively correlated due to the presence of several sizes of large cottonwoods. Canopy cover and canopy height are also posi- tively correlated. Other positive correlations occur between the number of species of trees and the number of trees in each of the size classes. Thus, tree species number per unit area is positively correlated with vegetational diver- sity. Shrub density is not strongly correlated with any of the other variables, indicating rather uniform distribution throughout the study area. It is, how- ever, correlated positively with small trees, again corresponding to the clumps of willow^ and tamarix, and correlated negatively with the large tree variables and ground cover. Whitmore • PASSERINE HARITAT PARTITIONING 257 Table 3 Summary of the Results of the Principal Component Analysis of each of 10 Vegetational Variables for 24 Species of Passerine Birds Correlations With Original Variables Component I II Number of Species of Trees (SPT)^ -0.65 -0.14 Number of Trees 7.6-15.2 cm DBH (T-7) -0.54 0.42 Number of Trees 15.2-22.9 cm DBH (T-15) -0.65 0.23 Number of Trees 22.9-30.4 cm DBH (T-22) -0.71 0.10 Number of Trees 30.4-38.1 cm DBH (T-30) -0.64 -0.36 Number of Trees > 38.1 cm DBH (T-38) -0.53 -0.58 Percent Shrub Density (% SD) 0.01 0.74 Percent Canopy Cover (% CC) -0.49 -0.22 Canopy Height in Meters (CHM) -0.43 -0.48 Percent Ground Cover (%GC) Percentage of Total Variance Accounted 0.75 -0.16 for Cumulative Percentage of Total Variance 56.56 16.87 Accounted for 56.56 73.43 1 Abbreviations for variables used in Tables 2 and 4. The results of the PC A are summarized in Table 3. The first component accounts for 56.6% of the variance in the original data. Percent ground cover shows a high positive correlation with the first axis. Species of birds having high values on this axis occur where there is high ground cover. The first axis also shows negative correlations with the measured variables for trees. Therefore, this axis represents a gradient starting with the forested areas with low ground cover and preceding to open areas with high ground cover, i.e. going from cottonwood stands to alfalfa fields. The second axis, which accounts for an additional 16.9% of the variance, is correlated positively with shrub density and small trees and negatively with large trees and canopy height. This corresponds to a gradient going from areas of tall trees, if trees are present at all, with low shrub density to areas of high shrub density and no large trees. A 2-dimensional plot is presented in Fig. 1. Species in the lower left of this ordination, e.g. Bullock’s Oriole, Yellow Warbler, Audubon’s Warbler, and Black-headed Grosbeak, are those associ- ated with high canopy cover and many trees. Ground cover and shrub den- sity are low in this area of the ordination. A species such as the Warbling Vireo would be expected to be found in areas with the same amount of canopy cover and ground cover as the Bullock’s Oriole, but with increased shrub density. A group of 5 species; Yellow-breasted Chat, Red-winged Blackbird, 258 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Fig, 1. Two-dimensional ordination of bird species along the first and second principal component axes. The first, horizontal, axis accounts for 56.6% of the response variance and second, vertical, axis accounts for an additional 16.9%. Increasing values on the first axis correspond to an increase in ground cover and a decrease in the numbers of trees. Increasing values on the second axis represent an increase in shrub density. Spe- cies’ names are abbreviated as in Table 1. Yellowthroat, Song Sparrow, and Willow Flycatcher, is located on the upper end of both axes. These are high shruh density species with little canopy cover and moderate ground cover in their territories. Six species cluster in the lower right corner of Fig. 1. These species are found in either open country or the alfalfa fields as evidenced hy their positions on axis I. If these 6 species are considered a separate habitat guild, then a regression line (r = .77 1 can he drawn through the remaining 18. This line can be viewed as a third gradient and corresponds to going from densely forested areas with low ground cover (lower left), to dense shrub areas with moderate ground cover (upper middle). Therefore, 3 separate gradients are apparent on this 2-dimensional ordination, thus increasing its value. In this study the addition of a third PGA axis adds little new information in that only 11.2% more variance is accounted for and there are no strong correlations with the original variables. The 4 most important variables, as determined by the stepwise discriminant program, are all accounted for by the first and second principal component axes. Therefore, discussion of Whitmore • PASSERINE HABITAT PARTITIONING 259 Species Habitat Use Values Table 4 OF THE 4 Most Important Variables^ Species % CC2 %SD T-7 %GC BO^ .208 .511 .274 .563 WK .383 .376 .256 .393 AF .513 .344 .303 .464 BHG .328 .678 .626 .403 LB .399 .310 .256 .457 BG .245 .288 .276 .356 YW .342 .255 .332 .310 AW .263 .300 .379 .362 WW .193 .440 .430 .560 IB .317 .376 .274 .390 YT .201 .385 .428 .399 YBC .437 .546 .358 .378 MW .351 .356 .364 .325 LW .253 .413 .422 .379 AT .416 .443 .327 .361 BHG .372 .658 .463 .257 GHN .378 .593 .497 .605 RWB .392 .365 .309 .354 HW .410 .341 .309 .331 BW .408 .667 .523 .475 WF .179 .271 .311 .334 LG .391 .219 .435 .309 WV .393 .310 .470 .436 ss .362 .223 .435 .285 1 Calculated by formulae in Colwell and Futuyma ( 1971 ). 2 See key to abbreviations of variables in Table 3. 3 See key to abbreviations of species’ names in Table 1. species’ distributions along gradients constructed using PCA will be confined to the first 2 axes. Habitat resource use. — Values for resource matrices constructed from the 4 most important variables are found in Table 4. Most of these values are less than 0.5, indicating that the species are restricted in habitat use. Those species that show consistently low values and with a low mean value for the 4 resource use determinations, may be termed habitat specialists. Included in this designation are the Blue Grosbeak, Yellow Warbler, Willow Flycatcher, Lesser Goldfinch, and Song Sparrow. Species with high resource use values, are relative habitat generalists. They include the Black-headed Grosbeak, Yellow-breasted Chat, Brown-headed Cowbird, Blue-gray Gnatcatcher, and Bewick’s Wren. Some species, e.g. Bullock’s Oriole, Lazuli Bunting, and Wilson’s Warbler, are high in 1 variable and low in others. 260 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 DISCUSSION Certain aspects of the ecological distribution of species in the Utah study area lend themselves to comparisons with previously published data. For this reason the following species or groups of species will be examined in more detail. Indigo Bunting and Blue Grosbeak. — One of the 4 species’ pairs that were found to be not significantly different by the MANOVA was the Indigo Bunt- ing and the Blue Grosbeak. In a table presented by Shugart and James (1973) only moderate overlap between these species was recorded, the Blue Gros- beak being found solely in clonal persimmon field plots while the Indigo Bunting was scattered throughout several habitat types, most notably forest edges. Stewart and Kantrud (1972) found the Blue Grosbeak in the Coteau Slope of North Dakota while the Indigo Bunting was found in the Coteau Slope and Northeastern Drift Plain. In 2 types of ordinations, James (1971) found moderate separation between the species. Using the same techniques, discriminant analysis and principal component analysis, I found little sepa- ration between the species. In my study the Blue Grosbeak was one of the most restricted in habitat use (Table 4). Therefore, in the Virgin River study area their habitat use was almost indistinguishable. Even though spe- cies specific habitat differences were not detected it is reasonable to assume that the 2 might take different sized food items, based on bill size alone. In other southwestern studies (Dixon 1959, Raitt and Maze 1968, Austin 1970, Carothers et al. 1974j Indigo Buntings were not observed. Whitmore (1975b) suggests that the Indigo Bunting is new in Utah, coming from the southeast approximately 30 to 40 years ago. Perhaps as a result of inter- specific competition with its congener, the Lazuli Bunting, it may be forced into a suboptimal habitat, therefore causing overlap with the Blue Grosbeak. Bewick" s Wren and Song Sparrow. — Although these species seem to require river lowlands with dense vegetation and cover (Behle 1943), as in the previ- ous pair of species, effective partitioning may be carried out by means of different food preferences and feeding behavior. Two other pairs of birds indistinguishable in habitat preference. House Wren and Western Kingbird, and Abert’s Towhee and Ash-throated Flycatcher, also differ behaviorally and in food selection. House Wren and Bewick" s Wren. — Inasmuch as Kroodsma (1973) recorded instances of competition between the House and Bewick’s wrens in Oregon one might expect similar activity in the Virgin River Valley. I observed 16 House and 20 Bewick’s wren territories in my study area and recorded no instances of interspecific territoriality. Behle (1943) states that the House Wren only winters in the lowlands while breeding in the mountains in Utah. This is not consistent with my observations nor those of Wauer and Carter W^hitmore • PASSERINE HABITAT PARTITIONING 261 (1965) who stated that there are several records of the House Wren in the riparian woodland during the breeding season. Habitat use was, however, similar to that reported by Kroodsma (1973) in that the Bewick’s Wren was confined to the dense thickets and House Wren occurred where shrub density decreased and grassy substrate increased. This latter point can be noted also by the positions of the species in Fig. 1. As noted in Table 4, the Bewick’s Wren has one of the broadest habitat ranges of all the species measured and therefore species’ overlap with it is to be expected. Parulidae. — All 7 species of the family Parulidae observed in the Virgin River Valley fall on or very close to the regression line drawn through the 2- dimensional principal component ordination. With the exception of the chat and yellowthroat, 2 species of different size, the species seem to be evenly distributed along the gradient going from the forested areas to dense shrub zones. Warblers are not found on the gradient going to the open country. A species that Cody (1974) found to be a generalist, the Yellow Warbler, is found here to be one of the most restricted species. This could be due to the high number of warbler species in such a restricted habitat. Carothers et al. ( 1974) found only 3 warbler species in their study in the riparian habitat of Arizona. As evidenced by the uniform distribution of warblers along the forest-shrub gradient one might think that competition is severe. The presence of large numbers of Audubon’s Warblers, the closest warbler on the ordina- tion, could exert a competitive influence on the Yellow Warbler, but I have no direct evidence that they are affecting one another. Behle (1943) and Wauer and Carter (1965) stated that the Audubon’s Warbler is an abundant migrant through the Virgin River Valley during April and May but breeds only in the conifers found at higher elevations. I observed them actively countersinging and defending territories through June. Therefore, even if these birds are non-breeding in the area, they probably affect community structure. I do not know if Audubon’s Warblers ever left during the breeding season and if they did leave the effect of competitive release on the other warbler species. One explanation for the appearance of many Audubon’s Warblers late in the season is that the severe winter of 1972-1973 could have delayed northward migration. The 2 warblers most closely associated in habitat use are the Common Yel- lowthroat and Yellow-breasted Chat. As noted from their positions on the ordination, these species inhabit areas of dense shrubs. Both species are rather broad in their habitat use (Cody 1974; Table 4) but differ in body size and foraging behavior. The chat, the larger, feeds primarily by gleaning insects from the foliage of shrubs whereas the yellowthroat often hawks insects from exposed perches or flies to the ground to pick prey out of the grass. In spite 262 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 of the closeness of the 7 warbler species in habitat use, I observed few in- stances of interspecific aggression. Brown-headed Cow bird. — With the exception of the Ash-throated Flycatcher all of the 24 passerine species analyzed in this paper are known hosts of the Brown-headed Cowbird (Friedmann 1963, 1971). Since the Ash-throated’s eastern congener, the Great-crested Flycatcher {Myiarchus crinitis) is a known host, I assume that the Ash-throated Flycatcher is also parasitized. To be effective at nest parasitism it should be advantageous for the cowbird to be broad in its range of habitat choices, thus allowing it better access to more nests. Examination of Fig. 1 shows that the Brown-headed Cowbird is cen- trally located in the ordination and almost equidistant from the ends of the regression gradient. Its mean habitat use value of .438 is one of the highest, indicating broad use of the 4 variables tested. James (1971) states that the Brown-headed Cowbird shows remarkable latitude in habitat use. One might conclude, therefore, that the cowbird, in order to take advantage of as many hosts as possible, is not as restricted as other species in habitat use. Willow Flycatcher. — Of all of the species in this analysis, the Willow Fly- catcher has the lowest habitat use value. This species is confined to areas of shrub density ranging from 70% to 100%. Trees of any size or species seldom occur in its defended territories. Behle (1943) listed this species as a common breeder in the streamside willows throughout the valley and cited numerous specimens collected along the Virgin River and Santa Clara Creek ( including 6 collected from one site 3 km southwest of St. George, Utah). This species is now uncommon to rare. A possible explanation for this is habitat change. Christensen (1962) documented the introduction and spread of the shrub tamarix (Tamarix pentandra) in Utah. This colonization has taken place at the expense of the willows. The stands of streamside willow discussed by Behle are almost totally gone. In fact, one is hard pressed to find any sub- stantial willow stands along the Santa Clara Creek. Probably tamarix does not provide a suitable nest site for the Willow Flycatcher and as a result the bird has been forced to go elsewhere. Possible evidence for this exists in that Wauer and Carter (1965) observed the species in the remaining willows at the Springdale Ponds area near the mouth of Zion Canyon. Habitat changes such as this are probably responsible for many of the differences in recent observations as compared to the older published data of Behle (1943) and V oodbury et al. (1949). Blue-gray Gnat catcher. — Also found near the center of the ordination is the Blue-gray Gnatcatcher. James (1971), Whitmore (1975a), and Kimberly Smith ( pers. comm.) observed and discussed the wide range of habitats se- lected by this species. In a more definitive study. Root (1967) lists many habitats in which the Blue-gray Gnatcatcher may be found and remarks on Whitmore • PASSERINE HABITAT PARTITIONING 263 the variability of selected sites in various areas of its geographic range. Based on the 4 most important variables, gnatcatchers in the Virgin River Valley had the broadest habitat use of any of the species in the community. My ob- servations of the foraging behavior of this species are consistent with those of Root in that foraging was primarily confined to the foliaged portions of the available habitat, most notably the outer most foliage of large mature trees. Hawking for insects was observed, but it was mostly confined to times when gnatcatchers were flying between trees. Occasional sorties low over the alfalfa were also observed, possibly, as mentioned by Root, to catch grasshoppers. The use of indices such as niche breadth and overlap has been discussed by Cody (1974) but one point should be re-emphasized, that of weighting. The aspect of the Colwell and Futuyma (1971) calculations that make them so valuable is that the species themselves determine the weights of each of the resource states by their positions in the habitat, i.e. their use of each state. This type of calculation effectively eliminates the misconception that each of the subdivisions of the resource matrix is equally important to the species and, therefore, provides a more meaningful interpretation of the data. How- ever, these calculations, to date, have been confined to one resource matrix at a time. Vdiat is needed, and is currently being worked on, is an n-dimen- sional habitat use matrix, i.e. one that will allow simultaneous analysis of several variables. It is possible to combine several variables into one by the construction of an index, such as an importance value, but in this type of statistic much information is lost. It is frequently reported in the literature (Wiens 1969, James 1971, Ander- son and Shugart 1974) that bird species select certain parts of the habitat based on specific search images. But care should be taken in emphasizing habitat keys. Perhaps the perceptual world (niche-gestalt) described by James (1971) is an artifact of the observer, i.e. the ecologist may be recog- nizing distinct habitats or positions along environmental gradients, but the bird species present may not be capable of the same distinctions or their dis- tinctions may not be equivalent to those of the observer (Vandermeer 1972 ). Vandermeer argues that the ecologist will never be able to view the niche through the eyes of a bird, even though the sensory systems are similar. How- ever, if the goal of the ecologist is, as Bronowski (1973) stated, to have the ability to visualize the future and to foresee what may happen, i.e. gain an index of predictability, then placing the species along environmental gradients offers useful insights. Acknowledging the conceptual problems involved, it is still useful to derive axes that allow the scientist to predict the behavior of a species, particularly in respect to the concept of environmental change and its impact on the community. The validity of ordination work can be tested by subsequent field observation. 264 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 SUMMARY Habitat relationships within a community of passerine birds were examined using multivariate statistical techniques and one index of niche breadth here termed “habitat use.” Four species pairs were not significantly distinct when analyzed using multivariate analysis of variance. A 2-dimensional ordination along known vegetational gradients was constructed using principal component analysis. A regression line was drawn through this ordination providing a third gradient. Habitat use analysis defined several habitat generalists, including the Blue-gray Gnatcatcher and Brown-headed Cowbird, and several habitat specialists, including the Willow Flycatcher and Blue Grosbeak. Reasons for taking care in discussing avian habitat selection were presented. ACKNOWLEDGMENTS I wish to acknowledge the help and guidance of Clayton M. White during the course of this study. Kimball Harper, Joseph Murphy, and Alvin Rencher read drafts of this manuscript and their critical comments are appreciated. William Evenson wrote programs for calculation of the habitat use values. I wish to thank James Karr and Frances James for their many helpful suggestions concerning the final draft of the manuscript. Financial assistance was obtained through a D. Elden Beck Scholarship from Brigham Young Uni- versity, the Frank M. Chapman Fund, and the National Audubon Society. LITERATURE CITED Anderson, S. H. and H. H. Siiugart. 1974. Habitat selection of breeding birds in an east Tennessee deciduous forest. Ecology 55:828-837. Austin, G. T. 1970. Breeding birds of desert riparian habitat in southern Nevada. Condor 72:431-436. Behle, W. H. 1943. Birds of the Pine Valley Mountain region. Southwestern Utah. Univ. Utah Biol. Ser. 7:1-85. Bray, J. R. and J. T. Curtis. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol. Monogr. 27:325-349. Bronowski, j. 1973. The ascent of man. Little, Brown and Co., Boston, Toronto. Carotiiers, S. W., R. R. Johnson, and S. W. Aitciiison. 1974. Population structure and social organization of southwestern riparian birds. Am. Zool. 14:97-108. Christensen, E. M. 1962. The rate of naturalization of tamarix in Utah. Am. Midi. Nat. 68:51-57. Cody, M. L. 1974. Competition and the structure of bird communities. Monogr. Popul. Biol. No. 7. Colwell, R. K. and 1). J. Futuyma. 1971. On the measurement of niche breadth and overlap. Ecology 52:567-576. Dixon, K. L. 1959. Ecological and distributional relations of desert scrub birds of western Texas. Condor 61:397-409. Dixon, W. J. 1970. Biomedical computer programs. Univ. of Calif. Press, Los Angeles. Friedmann, H. 1963. Host relations of the parasitic cowbirds. U. S. Natl. Mus., Bull. 233. . 1971. Further information on the host relations of the parasitic cowbird. Auk 88:239-255. Green, R. H. 1971. A multivariate statistical approach to the Hutchinsonian niche: Bivalve molluscs of Central Canada. Ecology 52:543-556. Whiltnore • PASSERINE HABITAT PARTITIONING 26S . 1974. Multivariate niche analysis with temporally varying environmental factors. Ecology 55:73-83. Hespenheide, H. 1971. Food preference and the extent of overlap in some insectivorous birds with special reference to the Tyrannidae. Ibis 113:59-72. Hutchinson, G. E. 1957. Concluding remarks. Cold Spring Harbor Symp. Quant. Biol. 22:415-427. James, F. C. 1971. Ordinations of habitat relationships among breeding birds. Wilson Bull. 83:215-236. AND H. H. SiiUGART. 1970. A quantitative method of habitat description. Audu- bon Field Notes 24:727-736. Kroodsma, D. E. 1973. Coexistence of Bewick’s Wrens and House Wrens in Oregon. Auk 90:341-352. MacArthur, R. H. 1958. Population ecology of some warblers of northeastern conifer- ous forests. Ecology 39:599-619. Morrison, D. F. 1967. Multivariate statistical methods. McGraw-Hill Book Co., N.Y. Raitt, R. J. and R. W. Maze. 1968. Densities and species composition of breeding birds of a creosote bush community in southern New Mexico. Condor 70:193-205. Root, R. B. 1967. The niche exploitation pattern of the Blue-gray Gnatcatcher. Ecol. Monogr. 37:317-350. 5hugart, H. H. and D. James. 1973. Ecological succession of breeding bird populations in northwestern Arkansas. Auk 90:62-77. Stewart, R. E. and H. A. Kantrud. 1972. Population estimates of breeding birds in North Dakota. Auk 89:766-788. Vandermeer, J. H. 1972. Niche theory. Annu. Rev. Ecol. Syst. 3:107-132. Wauer, R. H. 1969. Recent bird records from the Virgin River Valley of Utah, Arizona, Nevada. Condor 71:331-335. AND D. L. Carter. 1965. Birds of Zion National Park and vicinity. Zion Nat. Hist. Assoc. AND R. C. Russel. 1967. New and additional records of birds in the Virgin River Valley. Condor 69:420-423. Whitmore, R. C. 1975a. Habitat ordination of the passerine birds of the Virgin River Valley, southwestern Utah. Wilson Bull. 87:65-74. . 1975b. Indigo Buntings in Utah with special reference to interspecific competi- tion with Lazuli Buntings. Condor 77:509-510. Wiens, J. A. 1969. An approach to the study of ecological relationships among grass- land birds. Ornithol. Monogr. No. 8. Willson, M. F. 1974. Avian community organization and habitat structure. Ecology 55:1017-1029. Woodbury, A. M., C. Cottam, and J. W. Sugden. 1949. Annotated checklist of the birds of Utah. Univ. Utah Bull. 39:1-40. DEPT. OF ZOOLOGY, BRIGHAM YOUNG UNIV., PROVO, UTAH 84602 (PRESENT AD- DRESS: DIVISION OF FORESTRY, WEST VIRGINIA UNIV., MORGANTOWN 26506.) ACCEPTED 22 JAN. 1976. BREEDING DISPLAYS OF THE LOUISIANA HERON James A. Rodgers, Jr. Information on the breeding behavior of herons is for the most part not sufficiently detailed to permit comparative analysis. However, as a conse- quence of modern ethological research methods and theory (see, for example, Lorenz 1950, Tinbergen 1952, Hinde 1970), the behavioral patterns of nu- merous species of ardeids are now better understood. Meyerriecks (1960), while concentrating on the breeding behavior of the Green Heron (Butorides virescens ) , also made a preliminary study on the Great Blue Heron ( Ardea herodias) , Snowy Egret {Egretta thula) , and Reddish Egret iDichromanassa rufescens) . Later, Meyerriecks (1962) synthesized what was known at that time of the breeding behavior of all North American ardeids. Blaker (1969) recently completed a detailed monograph on the breeding behavior of the Cattle Egret (Ardeola ibis). The objective of this paper is to present detailed information on the breed- ing displays of the Louisiana Heron ( Hydranassa tricolor). Each display is discussed with respect to its form, function, and possible evolutionary origin. METHODS AND STUDY AREA Most field observations on the Louisiana Heron were made on Grand Island, Barataria Bay, Louisiana, during 1972 and 1973. The predominant vegetation growing on Grand Island is black mangrove {Avicennia nitida) and saltmarsh cordgrass (Spartina alterni- flora). I studied Louisiana Heron behavior for over 900 h from 2 observation blinds. Observations were made with 7X binoculars and were documented by tape recordings and photography (35 mm still and 8 mm movie). The line drawings and diagrams in the text were made from field notes or photographs. DISPLAYS While the displays of the Louisiana Heron have been treated in summary form by various authors (see Noble in Bent 1926, Meyerriecks 1962), a de- tailed description of the individual displays and the role they play in the overall reproductive cycle has not been presented. In furnishing such a de- scription, I have adhered as closely as possible to the terminology of Meyer- riecks (1960, 1962). Agonistic Behavior Scott (1956) defines agonistic behavior as any behavior associated with conflict or fighting. The behaviors treated in this section are grouped together because they exhibit the most obvious external signs of agonistic behavior. 266 Rodgers • LOUISIANA HERON DISPLAYS 267 Alert Posture. — In assuming an Alert Posture, a Louisiana Heron extends the head upward, holds the wings tight to the body, the neck and legs become very straight, and the body axis is oriented toward the vertical. The feathers all over the body, including the white crest plumes, are sleeked ; and the head is oriented in various directions as the bird attempts to locate the disturbance. Frequently the heron utters a “scaa/i” vocalization. I did not observe tail- flipping as described for the Green Heron (Meyerriecks 1960) as a part of the alert behavior of the Louisiana Heron. Territorial males are quick to rise to the Alert Posture at the slightest dis- turbance. Neighboring birds readily react to nearby herons and also assume the Alert Posture. A Louisiana Heron that has been startled while incubating or brooding exhibits this behavior and the ^^scaaK’’’ call when observing the surroundings or when returning to its nest or territory after having been driven away. The Alert Posture, with sleeking of the feathers, holding the wings to the body, and extension of the head upwards exhibits preparatory components of taking-off and may be derived from the intention movements of flight. Meyerriecks (1960) believes the amount of neck extension reflects the degree of escape tendencies; the greater the escape tendency, the greater the exten- sion of the neck. Often associated with the Alert Posture is repeated sleeking and partial erection of the crest and upper neck feathers. Just as the sleeking component may indicate escape tendencies (Meyerriecks 1960), the erection component may indicate aggressive tendencies. Aggressive behaviors such as the Upright display are characterized by partial erection of the crest and upper neck feathers by Louisiana Herons. Upright display. — This aggressive display is characterized by the extension of the neck to an almost vertical position above the body, orientation of the bill horizontally, and a moderate amount of feather erection of the crest and the upper and lower neck. The Upright may be accompanied by a nasal “aaa/i” call which is probably indicative of a more aggressive behavior, since this sound also is a component of other aggressive displays (described below). The Upright is generally performed at the approach of an intruding heron or some other disturbance. Most often, a trespassing heron is intimidated by the Upright and flees or at least halts its approach. The Upright is observed in other situations. For example, a Louisiana Heron returning to the heronry exhibits the Upright when preparing to feed its nearly fledged juveniles as they approach the parent. Forward display. — The Forward display is characterized by extreme erection of the crest, the feathers of the upper and lower neck, and the scapular ai- grettes. The white crest plumes point upward almost perpendicular to the 268 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 head. From a front or rear view of the head, the erected white crest plumes take on a multipronged appearance that calls attention to the head and the bill, the heron’s weapon. The head and neck are extended fully upwards, with the bill held horizontal and the bends of the slightly drooped wings exposed. The Forward display usually intimidates an intruding heron and thus ends the confrontation without further interaction. If the intruding heron con- tinues to approach and violate the territory, the resident heron then usually attacks. In what Meyerriecks (1962) calls the Full Forward display, the mandibles are open, and the wings are fanned out to the sides with the dorsal surface facing the opponent. The extension of the wings out to the sides and the extreme feather erection greatly increases the apparent size of the heron and probably heightens the threat or intimidation effect of the display. From a crouched posture, with the neck partially retracted, the heron lunges at its opponent while emitting an “aaa/i” vocalization. At the fullest extension of the lunge, the mandibles are closed producing a snap which is audible at 3 to 5 m. If the attacking heron is on the ground, it frequently runs toward its opponent with the wings held out to the sides while giving several “aaa/i” calls. Often the attacking heron chases after the fleeing bird as far as 15 to 20 m away from the nest. In flight, the legs dangle beneath the pursuing heron, the head is held up, and the neck is coiled in an S-shaped position. Rapid forward extensions of the head occur as the heron lunges at the escaping bird. Aerial fighting. — Highly aggressive interactions occasionally include 2 Louisi- ana Herons flying up from the ground or top of a bush, attaining heights of up to 15 m. During the encounter the fighting herons often give loud “aua/i” calls as they repeatedly lunge their bill and thrash their feet at one another. The wings beat rapidly, as each heron faces its opponent. Aerial fighting be- tween males contesting territorial boundaries is common during the courtship period. Twig Shaking. — Twig Shaking is characteristic of unpaired male Louisiana Herons during the courtship period, and is also performed by paired males and females in conflict situations. While Twig Shaking, the heron leans for- ward, extends the head out and down and grasps a twig in the nest or a nearby branch. During this action the feathers of the crest and upper and lower neck regions are moderately erected. The twig is then shaken from side to side. The intensity of Twig Shaking varies with the circumstances surrounding its performance. During more intense Twig Shaking, the nest or branch sways from the force exerted by the displaying heron. For example, an un- paired male often performs vigorous Twig Shaking during the approach of a Rodgers • LOUISIANA HERON DISPLAYS 269 female or a neighboring male. In less intense forms of Twig Shaking, the twig or branch is merely grasped or only slightly shaken. Twig Shaking may be a low intensity Snap, without a prominent down- ward pumping motion. These displays include similar feather erection and body posture. Arguments against Twig Shaking being a Snap are as follows: (1) In the Snap a twig is only sometimes grasped. (2) The downward pump observed in the Snap is absent in Twig Shaking. (3) Twig Shaking appears to exhibit more aggressive tendencies than the Snap. Territorial males fre- quently Twig Shake before or during a border confrontation with a neigh- boring male, but they do not perform the Snap in such a situation. (4) Only unpaired male Louisiana Herons perform the Snap, while both males and females perform Twig Shaking. The side-to-side movements that accompany Twig Shaking resemble the twig placement and rearrangement that are part of nest building behavior. This may indicate that Twig Shaking is derived from some component of nest building, though twig arrangement lacks feather erection and vigorous twig manipulation. However, Twig Shaking in the Louisiana Heron appears to be a redirected agonistic behavior vented toward an inanimate object such as a twig, or to be in the process of evolving into a “full” display. The erection of the feathers appears to indicate aggressive tendencies. Blaker (1969) speculated that Twig Shaking in the Cattle Egret is aggressive in origin and function. I observed no obvious signal function of Twig Shaking as it exists in the Louisiana Heron, and designate it as a display at this time only for comparative purposes. Tail- flipping. — Tail-flipping, as described for the Green Heron by Meyerriecks (1960), was not observed in the Louisiana Heron. Withdrawn Crouch. — A Louisiana Heron assumes the Withdrawn Crouch in the presence of another Louisiana Heron, especially when the first heron is being threatened or attacked by the second. In the Withdrawn Crouch, the feathers are sleeked, the legs bent, and the head and neck are tucked back onto the body so that the horizontal posture presents a relatively low profile. The bill is held either horizontal or downward, never in an upward direction or toward the opponent. This is also the case in the Green Heron (Meyer- riecks 1960) and the Cattle Egret (Blaker 1969). The Withdrawn Crouch apparently serves to reduce the aggressive tenden- cies of the attacker. Female Louisiana Herons usually assume the Withdrawn Crouch during their approach toward a displaying male. Darwin (1872) and Morris (1956) have pointed out that a submissive posture in overall form is often the opposite of aggressive postures in a species. Such is indeed the case with the Louisiana Heron. In the Withdrawn Crouch the heron shows inten- 270 THE WILSON BULLETIN • Vol. 89, No. 2, Jane 1977 tion movements ( terminology of Daanje 1950) of takeoff such as feather sleeking, crouching, and tucking in the head to the body, thus indicating possible escape tendencies. Wing Preen. — When a female Louisiana Heron approaches an unpaired male during the courtship period, the male begins rapid and intense preening. This preening differs markedly from the slow, smooth motion of normal preening. In preening performed between courtship displays, the heron droops 1 wing slightly and then runs the bill down through the primaries 1 or more times in succession (once 64.9%, twice 30.9%, 3 times 4.0%, 4 times 0.2%; N == 126 performances). Blaker (1969) describes a similar behavior of Wing Touching in the Cattle Egret. While Wing Preening is very stereotyped in form in the Louisiana Heron, I did not observe any indication of its signal function. Agonistic display discussion. — The form and vocalizations of the agonistic displays of the Louisiana Heron are similar to other North American ardeids, especially the Reddish Egret (Meyerriecks 1960). Only twice did I observe a display corresponding to Meyerriecks’ (1962) description of the Aggressive Upright Threat for the Louisiana Heron. Both instances were brief encoun- ters and an accurate description was not made. Snap and Stretch Displays The Louisiana Heron appears to be unusual among North American ardeids in that it tends to perform the Snap and Stretch displays in one sequence, and for purposes of description they will be treated as a unit. Snap and Stretch displays are performed only by the male Louisiana Heron, either from the nest or less frequently, from some other site on the territory. Both displays cease after pair formation. The form and sequence of movements are as follows: (1) In one smooth motion the male moderately erects the scapular aigrettes and the feathers of the crest, upper and lower neck, leans slightly for- ward, and extends the head fully outwards (Fig. lA) and downward until the bill is perpendicular and near or below the level of the feet (Fig. IB). (2) The legs bend almost completely and the wings droop slightly as the body drops quickly downward. The head is thus lowered well below the nest. In its extreme downward position, the male usually grasps a branch of the mangrove and shakes it with twisting side-to-side mo- tions of the bill (Fig. 1C). This concludes the Snap portion of the sequence. Rodgers • LOUISIANA HERON DISPLAYS 271 (3) In one smooth motion, the head is raised and laid on the mantle re- gion, with the bill pointed toward the zenith. At this point the body axis is oriented about 60° above the horizontal, the wings are held tight against the body, and the legs are partially straightened (Figs. ID, 2). Moderate feather erection is still present. (4) The legs straighten lifting the body upward, and the wings begin to open and droop slightly as the heron approaches the maximum height. At the pinnacle of the upward motion, the neck is partially uncoiled and the head is lifted clear of the mantle region (Fig. IE). Mean- while, the eyes bulge and the mandibles begin opening. 272 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Fig. 2. The Stretch display performed by a male Louisiana Heron from its nest. (5) As the head lowers onto the back the legs bend, causing the body to descend toward the nest, and the wings are further opened and par- tially extended out to the sides (Fig. IF). During this downward pumping motion, the mandibles are brought together to produce a mechanical “snap” sound. Rodgers • LOUISIANA HERON DISPLAYS 273 Fig. 3. Swelling of the throat region (indicated by stippling) in preparation for the '^unh” call of the Stretch display (above) and deflation during the call (below). (6) Steps 4 and 5 are usually repeated once or twice (Fig. IG-H). After the last downward pump, the heron rises and assumes the posture described in step 3 and now usually gives one or more low-pitched, resonant “wAi/t” calls. Each call is accompanied by swelling and de- 274 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 flating of the throat and upper region of the neck as the bill is slightly opened then closed (Fig. 3). During this calling the head often relaxes slightly off the vertical. Ending the display, the male lowers the head and turns it from side to side, looking around the heronry, possibly to locate females in the vicinity. Excluding the calls, a normal Snap-Stretch requires 3 to 5 sec depending on the duration of the twig-grasping component. High wind velocities can either prolong individual performances or result in a decrease in frequency. A great deal of variation occurs in the Snap-Stretch of the Louisiana Heron. The male may (88.4%, N = 96 displays) or may not grasp a twig in the downward stroke of the Snap. In 7 displays I observed, the male grasped and released one twig, then seized a second before performing the Stretch. Some- times the shaking is so slow and deliberate that no movement of the mangrove tree results, but often such force is exerted that the nest or mangrove moves. The downward pumping in both the Snap and the Stretch varies from smooth, unhurried movements, to rapid and forceful ones. The presence of a female near the displaying male seems to increase the probability of forceful per- formances of both displays. In some Stretch displays, during the upward lift of the body and head, the bill may leave the vertical (sideways or backwards) by as much as 15 to 20°. Though the Stretch is most often immediately preceded by the Snap, male Louisiana Herons sometimes perform the Stretch alone. These were most often associated with the sudden appearance of a female (14 of 17 observa- tions) and may indicate a high degree of excitation on the part of the male. The Snap too may occur by itself. Meyerriecks (1962) applies the term “low-intensity” to these isolated Snaps. However, I observed Louisiana Herons performing Snaps characterized by vigorous twig seizing and forceful down- ward pumping motions, without adding a Stretch. Isolated vigorous Snaps were observed at all stages of the courtship phase right up to copulation. The frequency of Snaps that occur outside the Snap-Stretch association varies among individuals at similar stages of reproductive activity and tends to increase in each individual as its courtship period advances. For example, male A1 in a period of 60 min performed 49 Snap-Stretch sequences and 15 isolated Snaps. Male B3 in a period of 38 min exhibited 44 Snap-Stretch sequences and 22 isolated Snaps. In these instances, any isolated Snap that appeared to be an interrupted Snap-Stretch sequence was discounted, and A1 and B3 both had been displaying continuously since the early morning about 6 hours earlier. Since all other ardeids thus far investigated perform the Snap and Stretch separately, the linkage of the Snap and Stretch in the Louisiana Heron ap- pears to represent a derived condition. I suspect that the Louisiana Heron is Rodgers • LOUISIANA HERON DISPLAYS 275 Table 1 Variation in Number of Pumps and Calls PER Stretch Display Character Range Mode Mean! male Al^ pumps 1-4 2 2.0 (N = 136) calls 1-6 4 2.4 male B3" pumps 1-3 2 1.8 (N = 117) calls 1-10 5 3.1 total displays® pumps 1-4 2 1.9 (N = 438) calls 1-10 4 2.6 1 Mean values include displays that possessed no calls. 2 Data for males A1 and B3 were collected during the 1973 breeding season. 3 Data for total displays are for the 1972 and 1973 breeding seasons. Still in the process of evolving the Snap-Stretch association, since the Snap and Stretch are at present also performed separately. The number of upward and downward pumping motions (1 to 4 per dis- play) and the number of calls (0 to 10 per display) vary among males and among performances by a single male. In Table 1, the 2 males showed indi- vidual variations in the number of components of the Stretch, but differed greatly only in the number of calls per display. Twenty-one % of 438 Stretch displays included no call. Blaker (1969) reported that 37% of 27 Stretch displays performed by unmated male Cattle Egrets possessed no call. Meyer- riecks (1962) stated the number of pumps performed by Louisiana Herons at Lake Alice, Florida, varied from 1 to 4 with 3 being most typical. A variation observed only 9 times was the alternation in the sequence in which pumps and calls were performed in the Stretch. Following a Snap, a male pumped twice, gave one call, then two more pumps and three calls with- out performing another Snap. Another male performed the Snap, did 2 pumps, gave 1 call, followed by another pump and 3 calls. A third male performed the Snap, did 1 pump, gave 1 call, followed by another 1 pump and 1 call. Variations were also noticed in the calls between individuals. While no recordings were analyzed, I could distinguish differences between some displaying males. The vocalization varied from soft and low-pitched to harsh and loud. The large variability in the performances of the Snap and Stretch requires some comment. Those courtship displays with a greater number of each com- ponent may make the displaying male more obvious and convey more infor- mation to a potential mate and possibly neighboring males. Morris (1957) has pointed out that display postures that exhibit differences from one per- formance to another indicate that only a “typical intensity” for the display 276 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 has developed. With a high or low motivational state on the part of the per- forming animal, minute variation in the form of even the most “fixed” display can be detected. In the Louisiana Heron, the form and vigor of the Snap and Stretch varies with the amount of stimulus (e.g., the presence of a female and sexual state of the male). Finally, since both morphological and behavioral characters are in large measure genetically determined among birds, and since anatomical features are known to vary from individual to individual, variations on either side of the central tendency for the phenotypic expression of a fixed action pattern should not cause surprise. Various anatomical features of the Louisiana Heron are accentuated by the Stretch. The magenta iris, black bill tip, turquoise-cobalt orbital skin, and white throat are made more conspicuous by the vertical motions of the dis- playing male. The white of the throat, belly and partially opened wings, are particularly obvious in a frontal view from a distance of more than 30 m away as the male moves up and down during the Stretch. These anatomical and behavioral features may be functioning as individual releasers or all acting together in heterogeneous summation (Seitz, in Tinbergen 1951) to render the Stretch more obvious. Both the bill-snapping and the call of the Stretch also focus attention on the displaying male. With wind velocities of less than 10 km per hour away from the observer, both the call and the bill-snapping are audible from as far away as 10 m. These sounds alone may serve to notify females wan- dering through the heronry of the presence of the displaying male, even one that is displaying from a site in dense vegetation and not visible from more than 3 to 4 m away. Because greater wind velocities interfere with the per- ception of the sounds, the visual clues are probably the primary factors attracting females to the displaying male. Discussion. — The erection of the crest, upper and lower neck, and scapular aigrettes during the Snap-Stretch may be indicative of agonistic components in the Louisiana Heron. There is an increase in the degree of feather erection by the male in the presence of a female. Still, the Snap-Stretch sequence is primarily sexual in function. The Snap often possesses actual twig grasping in the Louisiana Heron, whereas the same display of the Green Heron and the Great Blue Heron rarely exhibit twig grasping (Meyerriecks 1960, 1962). In addition, the Snap of the Louisiana Heron is not of the bowing type as performed by the Reddish Egret ( Meyerriecks 1960). The Snap of the Louisiana Heron appears to be unusual in 2 respects: it is often associated with the Stretch to form a single behavioral sequence, and the Snap possesses a downward pumping motion. Douglas Mock (pers. comm.) has informed me that the Great Egret iCasmero- dius albus) also possesses a downward pump in its Snap. Rodgers • LOUISIANA HERON DISPLAYS 277 I I The Stretch of the Louisiana Heron exhibits 3 basic movements: head laid on back with the bill oriented skyward, upward jump off, and downward pumping motion. Daanje (1950 ) has suggested that the second phase of the Stretch (the “jump off”) in A. cinerea is derived from the intention move- ments of preparing to takeoff in flight. The Stretch of the Louisiana Heron exhibits numerous components similar to the movements of taking off in flight, and may resemble the more primitive condition by retaining an actual “jump off” motion. Tomlinson (1974) reports a downward pumping motion and bill-snap in the Stretch of the Purple Heron (A. purpurea). Circle Flight Display The Circle Flight of the Louisiana Heron is performed exclusively by the male, and consists of the following movements: (1) The heron bends its legs slightly and leans forward, head and neck are tucked against the body, and the wings held to the body in prepara- tion for taking off. No feather erection is visible. (2) The legs straighten, providing the thrust for jumping off, as the head and neck are fully outstretched. The bill axis is oriented slightly up- ward from the plane of the body, tail is spread, and the wings beat in slow, very deep strokes creating a loud “whomp-whomp-whomp” sound. The legs temporarily dangle below the heron during the initial takeoff, then are brought up and held straight behind, where they move up and down with each wing beat. (3) The heron continues to gain altitude and distance from the takeoff site, wings still beating in slow, deep arcs and head fully extended, for as far as 10 to 12 m. (4) The midsection of the neck begins showing formation of a downward bend, and soon the head is tucked back onto the mantle region. (5) As the deep wing flapping gives way to normal wing beating, the heron assumes the normal flying posture. The heron turns from a straight flight line, flies in a roughly circular path going past the territorial bush, executes another sweeping turn, and approaches the territory in basically a straight flight path. (6) When 10 to 25 m from the nest, the bird fully extends its head. The legs begin moving up and down again, then finally drop downward. The tail is spread fully, and the wings are alternately flapped rapidly and held motionless for gliding. (7) The heron begins giving a long series of ^^culh-cuW^ vocalizations while moving the bill in up and down motions. The number of these THE WILSON BULLETIN • Vol. 89, \o. 2, June 1977 278 calls ranges from 5 to 12 per display. There usually is erection of the crest and upper and lower neck feathers, and aigrettes as the heron begins descending toward the nest. The vocalizations continue, with the gliding giving way to rapid wing-flapping, as the heron slows its flight. (8) The legs are lowered as the heron lands on its territory, usually on or near the nest. Landing is usually followed by a continuation of head nodding motions with additional calls (Fig. 4). \^lien these nodding motions cease, so do feather erection and vocalizations. In general, the male Louisiana Heron takes off into the wind when per- forming the Circle Flight display. Initial Circle Flights at the beginning of the courtship period may cover a path as large as 50 to 75 m in diameter. This circuit begins shrinking as the courtship period continues, and dwindles to its shortest dimensions just before pair formation, as small as 20 to 25 m in diameter. On windy days, the takeoff distance is greatly reduced, and the head is tucked in sooner. Lnder these circumstances, the male usually flies a greater distance past the nest bush and an exceptionally long approach distance results. The Circle Flight is performed less frequently than the Snap-Stretch or even the isolated Snap. The length of a performance of the Circle Flight ranges from 15 to 20 sec, depending on the size of the arc the male flies. Its rate of occurrence increases when a female Louisiana Heron is nearby, but her pres- ence is not essential. It also becomes more frequent with the passage of time after the male first establishes a territory. Most variations in the performance of the Circle Flight are probably due to external conditions. The effect of high wind velocities has been discussed. If a male has a nest deep in dense bushes that prevents an immediate takeoff, he must climb to an open site for takeoff. During this climb the wings are drooped and partially extended out to the sides for balance. Often the male does not return to the takeoff site, but lands on another part of his territory. Sometimes he even drops onto the territory of a neighboring male and is driven off. Occasionally a returning male diverges from the straight-line return to attack another Louisiana Heron in the vicinity of his territory, or one that has flown onto his territory during the Circle Flight. X^Tiile females apparently do not perform this display, frequently as many as 3 female Louisiana Herons were observed following a single male during a Circle Flight. The flight of the females on these occasions resembled nor- mal flight with the following exceptions: shortly before landing, the females extended their heads forward and often gave several ''culh-culh'' calls; after landing they often engaged in head nodding and calling. On such occasions Rodgers • LOUISIANA HERON DISPLAYS 279 the females made fewer calls than the male. Meyerriecks ( 1960 ) reported mutual performance of the Circle Flight by male and female Green Herons, Reddish Egrets, and Snowy Egrets. The posture assumed by the male and vocalizations he gives in his approach to the nest are the same as in the Greeting display, engaged in by both the male and female Louisiana Heron after pairing. In many instances, the female was observed to jump onto the nest bush as the male returned and to join him in head nodding and calling. However, the male usually chased the fe- male away. As is the case with the Snap and Stretch displays, the Circle Flight is no longer performed after pair formation. Discussion. — I agree with Meyerriecks (I960) that the Circle Flight is more Fig. 4. Male engaged in the “Greeting display” phase of the Circle Flight after landing on the nest bush. Compare this posture with that in Fig, 5. 280 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 indicative of sexual than hostile behavior. The Circle Flight display prob- ably functions to attract potential mates. The Circle Flights of the Louisiana Heron and Reddish Egret are similar. The Reddish Egret performs head-tossing movements and ^‘‘crog-crog^^ vocali- zations during the return approach to the nest (Allen 1954, 1955). Meyer- riecks (1960) calls the aerial displaying the “Aerial Stretch display” and the component performed after landing the “Stationary Stretch display.” Further discussion of the return flight of the Circle Flight is found in the section describing the Greeting display. Bill-nibbling Bill-nibbling is performed by both male and female Louisiana Herons. It is common during precopulatory and postcopulatory behavior and as a part of the greeting ceremony after pair formation. A series of sounds resembling gentle “rattling” is made by rapidly opening and closing the mandibles. The hill is usually oriented downward towards the nest, but either sex often ex- tends the nibbling mandibles over and around to the sides of its mate. Louisi- ana Herons were also observed to move the nibbling mandibles in side-to-side shaking motions in front of the mate. The mandibles are sometimes laid on or into the feathers of its mate, but I have not observed feather grasping such as reported for the species by Huxley (in Bent 1926) and Meyerriecks (1962). No feather erection was noted during Bill-nibbling. Bill-nibbling functions in appeasement and reducing hostile tendencies be- tween 2 herons. During precopulatory encounters, the female performs Bill- nibbling in association with the Withdrawn Crouch and seems to reduce the male’s aggressiveness towards her. In this instance, the bill is oriented down- ward or away from the male and is not waved over and around him. Discussion. — Laying the nibbling mandibles on the feathers and waving them over and around the mate probably represents ritualized allopreening (Hud- son 1965). Hudson (1965) and Blaker (1969) summarize the presence of Bill-nibbling in the Ciconiiformes. Greeting Ceremony This is not a display, but rather a relatively rigid display sequence. The greeting ceremony is composed of the Greeting display. Bill-nibbling, and often twig passing. The male Louisiana Heron constructs the foundation of the nest by himself during the courtship period, but after pair formation the labor is divided. Both engage in a greeting ceremony which is characterized by the male bringing nest twigs and passing them to the female who works them into the nest and finishes the structure. This results not only in the nest being completed, but also in reinforcing the new pair bond. Rodgers • LOUISIANA HERON DISPLAYS 281 Fig. 5, Greeting display performed by a new pair of Louisiana Herons. The male (left) holds the twig skyward in his bill as he passes it to the female. Note the amount of feather erection by both herons. A male performing the Greeting display returns to the nest with a twig in his mandibles and extends the head fully upward and exhibits extreme feather erection of the crest, upper neck, to a lesser extent the lower neck, and the aigrettes. Simultaneously, the wings are held out to the sides, and the bill, with the twig in the tip, is pointed repeatedly toward the zenith, then down- ward toward the nest, while the male gives several “cw/A-ca//i” calls (Fig. 5j. The female on the nest engages in the same behavior as she reaches out and takes the twig in her mandibles from the male. Both herons head-nod several more times, giving additional calls, and the pair usually engages in Bill- nibbling. Feather erection ceases soon after the female places the twig into the nest as the male looks on intently. The male then leaves, and the procedure is repeated again until the nest is completed. If the nest is erected deep inside a dense mangrove bush, the male extends his head down into the bush and the female reaches up from within the foliage to take the twig. 282 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 As the ceremony is repeated again and again, there is a noticeable decrease in the degree of feather erection, amount of head nodding, and number of calls. The crest, feathers of the upper neck, and aigrettes are not erected as much as during the period immediately after pair formation, when twig passing be- havior began. Erection also lasts a shorter time in the later stages. The male may nod his head only once or twice and give just a few calls, and the female in response may simply reach out and take the twig from the male after only a single head nod toward the zenith. In some instances, the female engages in the ceremony and works the twig into the nest without rising from the prone position. This variation is especially frequent during the incubating and brooding phases. After nest construction is completed and the eggs are laid, one heron remains away from the nest for long periods of time while its mate assumes incubating or brooding duties. Upon return of the mate, both herons engage in the ceremony during nest relief. As the returning mate begins an approach glide of 15 to 20 m toward the nest it extends its head fully outward, much as in the approach phase of the Circle Flight. Feather erection, head nodding, and vocalizations are identical to those of the Greeting display. The returning bird never lands directly on the nest, but on top of the nest bush or one nearby, maintaining the fully extended head. The wings are held away from the sides, and head nodding and calling continue. The heron then begins moving toward the nest while continuing this posturing and calling. Its mate then begins reciprocating, and the pair engages in the Greeting display while face to face or standing side by side. Bill-nibbling usually is performed by both herons during this time. The relieved heron then flies away, leaving the nest duties to the returned male. If the incubating or brooding heron fails to rise from the nest when its mate returns, the newly arrived heron continues performing the greeting ceremony, with intense Bill-nibhling until its mate rises off the nest. Often the mate that has been relieved flies off a short distance, finds a twig, then returns and presents the twig to its mate while performing the Greeting display. This is done repeatedly with as many as 9 twigs presented in a period of 5 min, though 2 to 3 twigs in 2 min is more common. After the twig is passed, it is then added to the nest. Twig passing is common during the inculcating and early brooding phases, especially if the relieved bird has been on the nest for a long period, and may persist until late in the breeding cycle when the nestlings are up to 2 weeks old. The nest at this time is still in good condition and probably does not require additional twigs for repair. I believe the twig passing behavior and greeting ceremony under these cir- cumstances serves primarily to reinforce the pair bond during a period when the partners are separated for great lengths of time. The role of the greeting Rodgers • LOUISIANA HERON DISPLAYS 283 ceremony in pair bond maintenance for several North American ardeids is discussed by Meyerriecks (1962). Although sex identification after pair for- mation is often impossible because of fading soft part colors, I believe both male and female Louisiana Herons engage in this behavior to about the same degree. This conclusion is supported by observations on many pairs that performed several nest reliefs during the day, in which, each member of the pair returned with twigs after being first relieved by its mate. The number of calls uttered by a heron varies from 2 to 17 during the Greeting display. It is my impression that more calls are given when the pair has been separated for a period of time. Performances of elements of the Greeting display were observed under different circumstances from those noted above. Between long sessions of incubating or brooding, a lone parent often leaves the nest and moves a short distance away to preen or sunbathe. Upon returning to the nest, it often per- forms the Greeting display alone before resettling on the eggs or nestlings. This shows how spontaneous the performance of the display is when an adult Louisiana Heron returns to the nest. Under similar circumstances, but with older nestlings (1 to 2 weeks old), the parent is often met upon its return by aggressive, lunging snaps from its young. The adult may then perform the Greeting display which usually ends the aggressive behavior of the nestlings. When a territorial male moves towards an intruding female to attack and drive her away, the female Louisiana Heron occasionally performs the greeting ceremony (Greeting display with Bill-nibbling). The male sometimes returns the display but usually drives the female off during the early part of court- ship. In late courtship, the greeting ceremony probably functions to reduce the male’s aggressiveness towards the female until he allows her to stay on the territory. Fledged Louisiana Herons perform a similar greeting cere- mony to one another. The display is enacted when 2 siblings approach one another after a time of separation. By beginning the displays at a distance from the nest the incoming heron apparently gives its mate time to recognize the returning partner by sight or sound. This may safeguard against aggressive incidents that might occur between the pair members if the returning heron were to land suddenly. Fighting on the nest would be selected against if it upsets the nest itself or endangers the eggs or nestlings in it. The display occurs in several different contexts and seems to function in mate recognition, reduction of aggressive tendencies between mates, appeasement, and reinforcement of the pair bond in the Louisiana Heron. The last phase of the Circle Flight and the Greeting display are very similar, but in the return flight of the Circle Flight, no twig is grasped and waved up and down in the mandibles. 284 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 SUMMARY The Upright display involves slight feather erection of the crest, upper neck, while the heron raises the head upward. In the Forward, the heron exhibits extreme erection of the crest, upper and lower neck feathers, and aigrettes while the head is extended upward. Both aggressive displays are typically accompanied by the “oaa^” call. While Twig Shak- ing, the heron moderately erects the crest, upper and lower neck feathers, and aigrettes as it extends the head out to grasp and shake a branch. The Louisiana Heron appears to be unusual in that it tends to combine the Snap and Stretch displays into 1 sequence, though each is observed being performed separately. As the male moderately erects the crest, upper and lower neck feathers, and aigrettes, he extends the head out and down with the bill oriented downward. Bending of the legs results in a lowering of the head and body. At the farthest downward extension, a branch is usually grasped. Thereupon, the Snap ends. Next, the heron rises, lays the head on the mantle, and orients the bill toward the zenith. In an upward movement, the head is partially lifted off the mantle, the wings begin to open and droop, and the bill is slightly opened. As the heron drops by bending the legs, the opened mandibles are snapped closed, and the wings are partially fanned out to the sides. The up and down motions are often repeated one or more times before the male returns to the head-on- mantle posture and gives the “unA” call. The intensity and number of each component in the Snap and Stretch vary greatly. In the Circle Flight, the male jumps into the air, the head is extended fully outward and held there for an exaggerated amount of time as the wings beat in deep arcs. The head is then tucked back as normal flight is assumed in a circular path. As the male returns to the nest, the legs drop, the crest, feathers of the upper and lower neck, and aigrettes are erected, and the head is fully extended. Several “culh-culh’’ calls are given during this time. Both male and female Louisiana Herons engage in Bill-nibbling, in which they open and gently close the mandibles and thereby create a “rattling” sound. This behavior is thought to function in appeasement and in reducing the agonistic behavior between herons. In the Greeting display, the crest, feathers of the upper and lower neck, and aigrettes are erected, the head is extended up, and the bill usually holding a twig is repeatedly oriented to the zenith and then waved downward, as the heron gives several calls. Bill-nibbling usually accompanies the Greeting display. This behavior is performed by both members of the pair during twig passing associated with nest building and the nest relief of an incubating or brooding mate. The greeting ceremony probably func- tions in appeasement, reducing agonistic behavior, and mate recognition. ACKNOWLEDGMENTS This paper represents part of a thesis submitted in partial fulfillment of the require- ments for the M.S. degree in the Department of Zoology and Physiology, Louisiana State University, Baton Rouge, Louisiana. I would like to thank George H. Lowery Jr. of the Museum of Zoology for providing both assistance and direction during my graduate re- search. I am also grateful to numerous graduate students of the museum who provided both suggestions and stimulating discussion, especially Donald Buden and Robert S. Kennedy. I am indebted to the Louisiana WildLife and Fisheries for furnishing me with equipment and use of the Marine Biological Laboratory on Grand Terre. The personnel of the Marine Laboratory were very helpful, particularly the supervisor Ralph Latapie. The Louisiana Ornithological Society awarded me a grant that helped defray the cost of Rodgers • LOUISIANA HERON DISPLAYS 2R5 research during 1973. I am grateful to Douglas Mock for reading the manuscript and making many helpful suggestions. Finally, I would like to thank my wife Linda for her understanding, encouragement, and assistance throughout the entire investigation. LITERATURE CITED Allen, R. P. 1954. The Reddish Egret: bird of colors and contrasts. Audubon Mag. 56:252-255. . 1955. The Reddish Egret. Audubon Mag. 57:24^27. Bent, A. C. 1926. Life histories of North American marsh birds. U.S. Natl. Mus. Bull. 135:1-490. Blaker, D. 1969. Behaviour of the Cattle Egret Ardeola ibis. Ostrich 40:75-129. Daanje, a. 1950. On locomotory movements in birds and the intention movements de- rived from them. Behaviour 3:48-98. Darwin, C. 1872. The expression of the emotions in man and animals. John Murry, London. 1948. Watts and Co., London. Revised and abridged by C. M. Beadnell. Hinde, R. a. 1970. Animal behaviour, a synthesis of ethology and comparative psy- chology. McGraw-Hill Book Co., New York. Hudson, M. J. 1965. Bill clappering display in the Common Heron Ardea cinerea. Ibis 107:460-465. Lorenz, K. 1950. The comparative method in studying innate behaviour patterns. Symp. Soc. Exp. Biol. 4:221-268. Meyerriecks, a. J. 1960. Comparative breeding behavior of four species of North American herons. Publ. Nuttall Ornithol. Club No. 2. . 1962. Contributions in Handbook of North American birds, vol. 1 (R. S. Palmer, ed.). Yale University Press, New Haven, Conn. Morris, D. 1956. The function and causation of courtship ceremonies. Fondation Singer-Polignac, Colloque Internat. sur LTnstinct (Paris, 1954) :261-266. . 1957. “Typical intensity” and its relationship to the problem of ritualisation. Behaviour 11:1-12. Scott, J. P. 1956. The analysis of social organization in animals. Ecology 37:213-221. Tinbergen, N. 1951. The study of instinct. Oxford Univ. Press, London. . 1952. “Derived” activities; their causation, biological significance, origin, and emancipation during evolution. Q. Rev. Biol. 27:1-32. Tomlinson, D. N. S. 1974. Studies of the Purple Heron, Part 2: behaviour patterns. Ostrich 45:209-223. MUSEUM OF ZOOLOGY, LOUISIANA STATE UNIV., BATON ROUGE 70803. PRESENT ADDRESS: DEPT. OF BIOLOGY, UNIV. OF SOUTH FLORIDA, TAMPA 33620. AC- CEPTED 16 APRIL 1976. BREEDING BIOLOGY OF CLIFF SWALLOWS IN VIRGINIA Gilbert S. Grant and Thomas L. Quay The Cliff Swallow (Petrochelidon pyrrhonota) has recently extended its breeding range southeastward into the Piedmont regions of Virginia, North Carolina, South Carolina, Georgia (Tedards 1965, 1966, Scott 1966, Dopson and Peake 1967, Parnell 1967, Cohrs and Cohrs 1972), and Florida (Sykes 1976) . This paper summarizes our studies of the breeding biology of one of these new populations of Cliff Swallows near the John H. Kerr Reservoir on the Roanoke River in the Piedmont of Virginia and North Carolina. MATERIALS AND METHODS Systematic examination of Cliff Swallow nests began on 12 June and ended on 21 August 1969. We gained access to nests by using a boat and a ladder. The nest contents were examined with a flashlight and a dental mirror. Of the 36 bridges within the perimeter of the reservoir, only 7 had significant numbers of nesting swallows. We found 354 nests under these 7 bridges, but 84 nests were inac- cessible, 90 were either always empty or partially destroyed old nests, and 39 were occu- pied by House Sparrows {Passer domesticus) . Nest success data from 72 of the remaining 141 nests were used. The 69 nests excluded were either examined too infrequently or young were found on first examination of nest contents and clutch size could not be deter- mined. The nests at 2 colonies, Goodall’s Landing and Occoneechee, were examined at weekly intervals and the largest colony, Bluestone Landing, was visited twice a week. All of the 72 nests were in Virginia, on the upper reaches of the reservoir near Clarksville. RESULTS AND DISCUSSION Nest lining. — The lining of 119 active Cliff Swallow nests was recorded. Of these, 102 were lined solely with straw, 4 with straw and chicken feathers, 5 with straw and Cliff Swallow feathers, 4 with straw and unidentified feath- ers, 1 with straw and a material resembling cigarette filters, 1 with hair, and 2 had no lining at all. Nests have been found lined with straw, wool, and feathers ( Davie 1898) ; grass and feathers (Forbush and May 1939) ; grass stems, feathers, and other materials ( Bent 1942) ; straw or hay (Samuel 1971) ; and fine grasses with occasionally a few sticks, hairs, and feathers, but with many nests nearly devoid of any material (Emlen 1954). Clutch size. — The mean clutch size of 60 nests that had only one brood for the nesting season was 3.32 ± 0.72 (Table 1) (range 2 to 5 eggs). Twelve nests had 2 complete broods. We were unable to determine the clutch size of the first brood in these, but the mean clutch size of the second brood was 3.00 ± 0.85 ( Table 1), also with a range of 2 to 5 eggs. Mayhew (1958) felt that 3 to 4 eggs per clutch was the normal condition in California and Erskine 286 Grant and Quay • CLIFF SWALLOWS IN VIRGINIA 287 Nesting Success of Table 1 Cliff Swallows at Kerr Reservoir, Virginia Eggs Young % Total Nesting Success Clutch Size No. Of Clutches No. Laid No. Hatched % Hatched No. Fledged % Fledged First Clutches 2 6 12 11 91.7 11 100.0 91.7 3 32 96 79 82.3 74 93.7 77.1 4 19 76 61 80.3 51 83.6 67.1 5 3 15 11 73.3 9 81.8 60.0 Total 60 199 162 145 Mean 3.32 2.70 81.4 2.42 89.5 72.9 S.D. 0.72 0.99 1.04 Second Clutches 2 3 6 4 66.7 4 100.0 66.7 3 7 21 15 71.4 15 100.0 71.4 4 1 4 3 75.0 3 100.0 75.0 5 1 5 5 100.0 4 80.0 80.0 Total 12 36 27 26 Mean 3.00 2.25 75.0 2.17 96.3 72.2 S.D. 0.85 1.20 1.04 and Teeple (1970) reported a mean clutch size of 3.74 in New Brunswick. Myres (1957) found the mean clutch size to range from 3.6 to 3.9 eggs ( 54 nests) in a British Columbia colony and Samuel (1971) found an average first clutch of 3.31 eggs (35 nests) and an average second clutch of 2.89 eggs (9 nests) in West Virginia. Incubation period. — The length of time from the laying of the last egg to the hatching of the last young averaged 13.5 days (20 nests) in this study, ranging from 11 days (1 nest) to 16 days (2 nests). Using this same mea- sure, Mayhew (1958) found the incubation period to be 16 days (2 nests) in California and Samuel (1971) found 15 days (7 nests) to be the mean in West Virginia. Myres (1957) took the incubation period to be 14 days, but stated 13 days may be more accurate and Burns (1915) gives 12-14 days. Cliff Swallows begin incubation the day before the last egg is laid (Mayhew 1958, Samuel 1971). Brood size and egg mortality. — The mean brood size (number of eggs hatched per nest) was 2.70 ± 0.99 (Table 1) in the 60 first broods. Thirty- seven of 199 (18.6%) first-clutch eggs failed and there were 2 nest failures (no eggs hatched) in this sample. In the 12 nests with second broods, the 288 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 mean brood size was 2.25 ± 1.20 (Table 1). Nine of 36 (25.0%) second- clutch eggs failed to hatch and there was only one nest failure in this group. Hatching failure may be due to infertility. Samuel (1971) found that 35.2% of all Cliff Swallow eggs laid (first and second broods) did not hatch. Nestling period. — Samuel (1971) found that the period of time between hatching and leaving the nest averaged 23.6 days in 6 broods. Mayhew (1958) found first-flying young 23 days after hatching and Burns (1921) found young left the nest 16 to 24 days after hatching. One young made its first flight at 22 days at Kerr Reservoir. Cliff Swallows often remained in the nest several days after they were able to fly, making an exact determination of the duration of the nestling period difficult to compute. Nestling mortality and total mortality. — The mean number of young fledged per nest was 2.42 ± 1.04 (Table 1) with a nestling mortality of 10.5% in the first broods. The total first-brood nesting mortality was 27.1%. Second clutches fledged 2.17 ± 1.04 young (Table 1) for a 3.7% nestling loss and a total nesting mortality of 27.8%. More eggs hatched, more young fledged, and there was an overall higher nesting success in first-brood nests with smaller clutches (Table 1). Although the sample size was small, nesting success in second broods tended to increase with larger clutches. However, by multiplying percent total nesting success against clutch size it can be seen that the larger clutches produced more young despite a lower percent of fledging and hatching success in first broods. Greater success with larger clutches in second broods versus those of first broods was possibly correlated with insect abundance, but quantitative mea- surements were not made. Both weather and the experience or skill of the adults might have been better during second broods. Samuel (1971) found a total nesting mortality (both clutches combined) of 41.5% in West Virginia, while the nesting mortality for all clutches at Kerr Reservoir was 27.2%. In Samuel’s (1969) study, the high mortality was attributed to nest abandonment caused by barn alterations, House Sparrow interactions, fallen nests, and unknown reasons. We noted little influence of House Sparrows on the reproductive success of Cliff Swallows except under bridges w ith small numbers of swallows. No nests fell down during our study at Kerr Reservoir. Swallows nesting under bridges were obviously not sub- jected to the same set of losses incurred by those nesting inside of barns. Samuel 1 1969, 1971) believed that the unknown-reasons group may have included abandonment by birds joining migrating flocks. Foster (1968) also reported instances of Cliff Swallows departing as a colony and leaving eggs and nestlings behind in the nests. Five nests (3 with eggs and 2 with young) were abandoned for unknown reasons at Kerr Reservoir. The different colony GTant and Quay • CLIFF SWALLOWS IN VIRGINIA 289 sites involved and the asynchronous timing precluded the departing-as-a- colony phenomenon reported by Foster (1968) and Samuel (1969, 1971). Stewart (1972) reported heavy nestling losses in Cliff and Tree swallows {Iridoprocne bicolor) when adults abandoned young as a result of 24 hours or more of continuous rain. Although most birds had departed by 31 July, adults with young generally remained at these Virginia colonies until the last young left the nest (21 August). Second broods. — Cliff Swallows typically raise one brood, although 2 broods have been suggested by Bent (1942) and McCann (1936). Samuel (1971) studied 9 pairs that raised 2 broods in West Virginia. The 12 second broods at Kerr Reservoir were all found in nests that had contained first broods earlier in the nesting season. It seems that 2 broods could occur commonly under optimum conditions at Kerr Reservoir. Grinnell (1937) found that 48 days elapsed between the day of first arrival and first-flying young and Samuel (1971) found this period to range from 38 to 48 days. Swallows were first noted on 12 April 1969 (our first visit; they probably arrived a few days earlier) and the last young left the nest on 21 August 1969. Thus, 132 days is more than ample time to accommodate 2 broods, but only 16.7% of the swallows completed 2 broods in 1969. Cliff Swallows arrive at nesting sites in successive waves of migrants (Mayhew 1958) and it seems probable that favorable spring weather could initiate early synchronous first clutches and thereby facilitate the completion of second broods. SUMMARY We studied the breeding biology of the Cliff Swallow in Virginia where it has recently become established as a new breeding bird. Data are presented on nest lining, clutch size, incubation period, hatching success, brood size, egg mortality, nestling period and mor- tality, and second broods. Little difference in breeding biology was found between the Virginia colonies and those in the remainder of their range. ACKNOWLEDGMENTS The personnel at Occoneechee State Park, Virginia provided material assistance with the field research, especially J. A. Brown (Superintendent) and J. E. Thomas (Chief Ranger). Sandy Grant typed and proofed several drafts of the manuscript. Carl L. John- son was our field associate, on an NSF-Undergraduate Research Participation award (Grant No. GY-5963) at North Carolina State Lfniversity under the direction of C. H. Bostian and T. L. Quay. Paul W. Sykes, Jr. made original field data taken in 1966 and 1967 available and also gave advice. Support was provided to the senior author by the Department of Zoology and the Office of Financial Aid at North Carolina State Univer- sity. D. E. Samuel and T. R. Howell read a draft of the manuscript and offered helpful suggestions. This is Paper No. 4756 of the Journal Series of the N. C. Agricultural Experiment Station. 290 THE WILSON BULLETIN • Vol. 89, \o. 2, June 1977 LITERATURE CITED Bent, A. C. 1942. Life histories of North American flycatchers, larks, swallows, and their allies. U. S. Natl. Mus. Bull. 179. Burns, F. L. 1915. Comparative periods of deposition and incubation of some North American birds. Wilson Bull. 27:275-286. . 1921. Comparative periods of nesting life of some North American Nidicolae. Wilson Bull. 33:177-182. CoHRS, D. AND D. CoHRS. 1972. Cliff Swallow nest found in Atlanta. Oriole 37:19. Davie, 0. 1898. Nests and eggs of North American birds, 5th ed. Landon Press, Colum- bus, Ohio. Dopson, C. W., Jr. and R. H. Peake. 1967. The Cliff Swallow; A new breeding bird for Georgia. Oriole 32:29-31. Emlen', J. T. 1954. Territory, nest building, and pair formation in the Cliff Swallow. Auk 71:16-35. Erskine, a. J. and S. M. Teepee. 1970. Nesting activities in a Cliff Swallow colony. Can. Field-Nat. 84:385-387. Forbush, E. H. and j. B. May. 1939. A natural histor>- of American birds of eastern and central North America. Houghton Mifflin Co., Boston. Foster. W. A. 1968. Total brood mortality in late-nesting Cliff Swallows. Condor 70: 275. Grinnell. j. 1937. The swallows at the Life Sciences Building. Condor 39:206-210. Mayhew, W. 1958. The biology of the Cliff Swallow in California. Condor 60:7-37. McCann, H. D. 1936. A colony of Cliff Swallows in Chester County, Pa. Auk 53:84—85. Myres, M. T. 1957. Clutch size and laying dates in Cliff Swallows. Condor 59:311-316. Parnell. J. F., ed. 1967. Southern Atlantic Coast Region. Audubon Field Notes 21:557. Samuel, D. E. 1969. Nest al)andonment and possible management of the Cliff Swallow in \^'est Virginia. Trans, of the Northeast Fish and Wildlife Conference 26:181-185. . 1971. The breeding biology of Barn and Cliff swallows in West Virginia. Wil- son Bull. 83:284-301. Scott. F. R., ed. 1966. Middle Atlantic Coast Region. Audubon Field Notes 20:558. Stewart. R. M. 1972. Nestling mortality in swallows due to inclement weather. Calif. Birds 3:69-70. Sykes, P. 1976. Cliff Swallow breeding in south-central Florida. Wilson Bull. 88:671- 672. Tedards. a. M. 1%5. Cliff Swallows nesting in South Carolina. Chat 29:95-97. . 1966. Continued nesting of Cliff Swallows at Lake Hartwell in western South Carolina. Chat 30:110. DEPT. OF BIOLOGY, UMV. OF CALIFORNIA, LOS ANGELES 90024 AND DEPT. OF ZOOLOGY, NORTH CAROLINA STATE UNTV., R.ALEIGH 27607. ACCEPTED 16 JAN. 1976. WEATHER INFLUENCES ON NOCTURNAL BIRD MORTALITY AT A NORTH DAKOTA TOWER Michael Avery, Paul F. Springer, and J. Frank Cassel Most studies of bird losses at towers have dealt with weather conditions in a general manner (e.g. Tordoff and Mengel 1956, Kemper 1959, Taylor and Anderson 1973) because losses usually were not monitored on a daily basis throughout the entire migration season. Thus, weather conditions prevailing on nights of large, spectacular kills have received the most attention. Such nights are usually characterized by overcast skies, often with precipitation, winds favorable for migration, and in the fall the passage of cold fronts (e.g. Brewer and Ellis 1958). In the course of a study of bird migration and mortality at the U.S. Coast Guard’s Omega Navigation Station, located approximately 3 km west of La- Moure, North Dakota in the James River Valley (Avery et al. 1975), it was apparent that while occasional large kills occurred on overcast nights, con- siderable losses took place throughout the migration seasons under non-over- cast skies, particularly in the spring. Since mortality was monitored daily and accurate weather data were available from nearby, it was possible to analyze the losses with respect to cloud cover and wind conditions during 4 entire migration seasons. METHODS The 366-m Omega tower is supported by 3 sets of 5 guy wires (34.9-60.3 mm diameter) spaced 120° apart. The guy wires are attached at heights of 53, 109, 167, 228, and 293 m. The lower 2 guys are anchored 122 m from the tower, the next 2 at a distance of 213 m, and the last at 297 m. In addition, 16 evenly spaced transmitting cables (50.8 mm diam.) extend from the top of the tower to a circular perimeter road 732 m away. Searches for tower casualties were made every morning at daybreak (except for 7 days) during the study periods: 30 March-4 June and 8 August-15 November 1972 and 2 April-2 June and 12 August-3 November 1973. Because the size of the tower site and the dense vegetation on it made it difficult to find all bird casualties, the area under the guy wires (approximately 168 ha) was divided into 4 concentric strata (Avery et al. 1975) : A, 0-46 m (0.66 ha) from the tower; B, 47-92 m (1.97 ha) ; C, 93-183 m (7.88 ha) ; and D, 184-732 m (157.61 ha). During the daily searches, stratum A was checked completely. The approximate areas searched in the other strata were: B — 0.37 ha (18.8%), C — 0.50 ha (6.3%), and D — 1.51 ha (1.0%). The location and condition of each bird were recorded as it was collected, and only specimens judged to have died during the previous night were included in the analyses presented here, unless stated otherwise. Because official weather data are not available from LaMoure, hourly weather reports were obtained from the Federal Aviation Administration Flight Service Station at James- town, 72 km north-northwest of LaMoure. A few of the records were discarded inad- 291 292 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 vertently by the station prior to analysis; thus, cloud cover and wind data are not avail- able for these nights. Each of the 225 nights for which we have records of cloud cover was characterized as overcast or non-overcast. Four classes of cloud cover are recognized in the official weather reports — < 0.1 sky cover, 0.1 to < 0.6, 0.6 to 0.9, and > 0.9. These were assigned the numbers 0, 1, 2, and 3, respectively, and the 13 hourly figures from 1800-0600 GST were summed. The mean was calculated and a night was designated over- cast if the mean was > 2.5. All other nights were called non-overcast. This distinction is somewhat arbitrary* and, conceivably, if different criteria were used, slightly different interpretations of the data would result. Wind direction and estimates of cloud cover made at the Omega tower corresponded well with the official weather reports from James- town. The mean, nightly, surface wind direction was calculated from the hourly records; and the mean directions were grouped into four 90° sectors. Nocturnal bird migration in this region is primarily along a northwest-southeast axis (Richardson and Gunn 1971, Avery et al. 1976), and in this paper winds are referred to as favorable if from the 106°-195° quadrant on spring nights and 286°-015° on fall nights. Losses in the 3 most frequently killed fcimilies, Rallidae, Parulidae, and Fringillidae, were examined during their respective periods of peak migration as indicated by field surveys conducted several times weekly near the tower site. Only the losses on nights within each of these peak migration periods w*ere used in these analyses. Chi-square goodness-of-fit tests were used to determine if, within each family, the losses occurring during the entire peak periods in each cloud-wind category were in the same proportions as the number of nights in those categories. The G-test (Sokal and Rohlf 1969) was used to ascertain independence between cloud cover and distance of kill from the tower (Table 4) and between cloud cover and season (Table 5). Prior to analysis, the data used in Tables 4 and 5 were corrected for differences in the area searched in each stratum. In all tests p < 0.05 was accepted as statistically significant. RESULTS AND DISCUSSION The 5, largest, single-night kills and the accompanying weather conditions are listed in Table 1. All occurred under an overcast sky with the exception of the night of 14-15 May 1972 which was moonless and clear. Weather data revealed no conditions of overcast or poor visibility anywhere in the region that night. The behavior of birds at the tower during the kills on overcast nights was generally similar to that described by previous authors (e.g. Coch- ran and Graber 1958) and is treated in more detail in another paper (Avery et al. 1976). Although the largest collections of dead and injured birds were made fol- lowing overcast nights, mortality occurred consistently on clear nights as well. Table 2 shows that during spring migration the percent of losses of rails and fringillids occurring on overcast nights was about the same as the rate of occurrence of those nights; however, spring mortality in warblers and fall mortality among all 3 families occurred on overcast nights in greater-than- expected percentages (p<0.05). During their peak periods of spring migration, rails and fringillids were Avery et al. • MORTALITY AT A NORTH DAKOTA TOWER 293 Table 1 The 5 Largest Single-night Losses at the Omega Tower in 1972 and 1973 Night of kill Birds found Weather conditions during night 25-26 Sept. 1973 69 overcast, light rain, light ENE wind 4- 5 Oct. 1972 48 overcast, NE wind 5-15 k 14-15 May 1972 27 clear, light S wind 10-11 May 1972 25 overcast, drizzle, light S wind 21-22 Aug. 1972 23 overcast, NW wind 10-15 k killed in significantly greater numbers on non-overcast nights with south- easterly winds than on nights with other conditions (Table 3). Conversely, warblers were killed in significantly greater numbers on overcast nights. In the fall, losses at night during peak migration periods under the various con- ditions of cloud cover and wind direction were distributed in about the same frequency as the occurrence of nights with these conditions except for warblers and fringillids which were killed in significantly greater numbers on overcast nights with northeasterly winds. The high proportion of fall losses on overcast nights within 12 h after the passage of a cold front is consistent with other published reports (Brewer and Ellis 1958, Tordoff and Mengel 1956, Laskey 1960, Taylor and Anderson 1973). The fall losses presented in Tables 2 and 3 were due primarily to the few, large, single-night kills (Table 1), each of which was preceded by a cold front through the LaMoure area. Table 2 Percent of Losses Occurring on Overcast Nights at the Omega Tower IN 1972 AND 1973 Family or group Spring! (%) Number of birds Fall (%) Number of birds Rallidae 35 34 41* 22 Other non-passerines 36 11 50 10 Parulidae 64* 44 81* 135 Fringillidae 32 104 64* 100 Other passerines 48 57 70 46 Total 42 250 70 313 Percent of overcast nights 32 22 1 * Indicates statistical significance between % of loss and % of overcast nights. 294 THE WILSON BULLETIN • Vol 89, No. 2, June 1977 Table 3 Percent Losses in Relation to Cloud Cover and Wind Direction at the Omega Tower in the Peak Migration Periods in 1972 and 1973 F amily Cloud coveri NW 286-015° Surface wind quadrant^ NE SE 016-105° 106-195° SW 196-285° No. of birds and (nights) Spring Rallidae o 20(27)" 40(39) 40(23) 0(12) 5(26) n 29(42) 12(16) 53(24)* 6(18) 17(55) Parulidae 0 32(15) 0(23) 59(54)-* 9( 8) 22(13) n 20(47) 0(14) 50(25) 30(14) 10(36) Fringillidae 0 21(15) 21(35) 57(40) 0(10) 28(20) n 3(43) 16(18) 68(27)* 13(12) 62(49) Fall Rallidae 0 33(38) 67(31) 0(13) 0(19) 3(16) n 11(24) 11(20) 33(26) 44(30) 9(66) Parulidae 0 23(13) 70(50)* 3(13) 4(25) 69 ( 8) n 37(23) 21(27) 16(17) 26(33) 19(30) Fringillidae o 2(24) 91(24)* 2(29) 6(24) 54(17) n 20(24) 5(10) 10(27) 65(39) 20(41) 1 o = overcast, n = non-overcast. 2 * Indicates statistical significance between % of loss and % of nights with indicated weather conditions. 3 % of nights in each wind category are in parentheses. Spring losses were not characterized by large kills but were smaller and more evenly distributed throughout the season. There was no direct associa- tion of spring losses with frontal movements; the bulk of the losses occurred on nights with favorable ( i.e. southeasterly) winds. Ceilometer observations made at the tower revealed that the hulk of spring migration took place on nights w ith southeasterly winds. The percent of losses of birds recovered within various distances of the tower varied with cloud cover (Table 4). In each family or group the per- cent killed in stratum A on overcast nights was similar to that on non-over- cast nights. Among rails and other non-passerines, the losses on non-overcast nights were distributed approximately evenly among the 4 strata. Losses to passerines on non-overcast nights consistently exceeded those on overcast nights in strata C and D. In each family or group, losses on non-overcast nights in stratum D were 3 or 4 times those on overcast nights. Non-passerines suffered substantially greater losses in the outermost stratum than did pas- serines, particularly on non-overcast nights. Overall, losses on overcast nights were concentrated near the tower in strata A and B, whereas losses on non- overcast nights were more evenly distributed, 9% occurring at least 184 m ^very etal. • MORTALITY AT A NORTH DAKOTA TOWER 295 Table 4 Percent of Losses by Stratum at the Omega Tower on Overcast and Non-overcast Nights in the 1972 and 1973 Migration Seasons Percent by stratum Family or groups Cloud cover- A B C D Number of birds Rallidae o 23 36 36 5 21 n 22 28 28 22 35 Other non-passerines o 43 21 29 7 9 n 31 19 25 25 12 Parulidae* o 44 43 13 1 137 n 50 20 26 4 42 Fringillidae* () 29 46 23 2 97 n 30 35 28 8 107 Other passerines o 34 38 27 2 59 n 25 41 28 6 44 All birds* 0 36 42 20 2 323 n 31 32 27 9 240 1 * Indicates statistical significance between overcast and non-overcast nights. 2 o overcast, n = non-overcast. from the tower. Within warblers, finches, and total birds, the distribution of kill by strata on overcast nights differed significantly from that on non-over- cast nights. Table 5 shows how the distance of kills from the tower varied with cloud cover and season. In both spring and fall, greater percentages of the seasonal losses were generally found in the 2 innermost strata under overcast condi- tions than under non-overcast. Conversely, in strata C and D, relatively more birds were found dead following non-overcast nights in both spring and fall than following overcast nights. When mortality between seasons is compared, spring losses were generally less than fall losses in strata A and B but ex- ceeded the fall losses in strata C and D on both overcast and non-overcast nights. In both spring and fall, the differences in mortality between overcast and non-overcast nights within the strata were statistically significant and indicate that the distance of losses from the tower was influenced by cloud cover. The differences in location of tower casualties in spring and fall is de- picted in Fig. 1. This graph includes all of the tower casualties found in 1972 and 1973 and consists of raw data uncorrected for differences in areas searched. It shows that in each year the percent of fall losses exceeded those of spring within 92 m of the tower. Beyond 92 m the situation was reversed, 296 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 5 Percent of Losses by Stratum at the Omega Tower on Overcast and Non-overcast Nights in 1972 and 1973 Season’^ Cloud cover2 Percent by stratum Number of birds A B C D Spring* 0 34 34 28 4 104 n 26 32 30 12 146 Fall* o 37 45 17 1 219 n 40 31 23 6 94 1 * Indicates statistical significance in % of losses by strata between overcast and non-overcast nights. 2 o = overcast, n = non-overcast. except for the 184-229 m interval in 1972. These results, although not sta- tistically significant in 1973, show that, except for this one exception, larger spring losses consistently occurred at greater distances from the tower than did fall losses. Cloud conditions seem to have a considerable effect on the manner in which bird mortality actually occurs at the Omega tower. From the results obtained it appears that most fall mortality takes place when large numbers of birds are aloft on overcast nights. Such nights are usually closely associated with the passage of a cold front. On overcast nights, migrants congregate around the tower (Avery et al. 1976) and are killed near the structure by colliding with it, the guy wires and transmitting cables, or other birds. On the other hand, spring migrants are apparently aloft when winds are favorable, regardless of cloud cover (Table 3), and thus much mortality occurs on non-overcast nights when migrants are not congregated at the tower. On such nights, mi- grants actually seem to avoid the structure (Avery et al. 1976). Consequently, in the spring, sizable losses occur on non-overcast nights far from the central structure through collisions with outlying guy wires and the transmitting cables. The regular occurrence of substantial bird losses on non-overcast nights is perhaps peculiar to the Omega tower with its widespread system of cables. Losses do occur on non-overcast nights at other towers with less extensive cable arrays (e.g. Stoddard and Norris 1967), but apparently they are not as great as at the Omega tower. Birds deviating from their flight path to avoid most towers may remove themselves from the danger of the supporting guy wires. The 16 transmitting cables extending from the top of the Omega tower, however, pose additional problems; and birds avoiding the tower, and hence the innermost supporting guy wires, are still liable to collide with the outer transmitting cables. Avery et al. • MORTALITY AT A NORTH DAKOTA TOWER 297 Fig. 1. The % of total seasonal losses collected at 46-m intervals from the Omega tower in 1972 and 1973. Some of the differences in mortality among groups of migrants may be due to interspecific (or interfamilial) behavioral differences. For instance, at the Omega tower, warblers were prone to be killed close to the central structure (Table 4). Possibly warblers are influenced by red tower lights more so than are other groups, or perhaps warblers are less able to change direction to avoid inner guy wires than are other migrants. The sizable pro- portions of some kinds of non-passerines killed away from the tower, espe- cially on non-overcast nights (Table 4), suggests behavioral differences that may be even more basic than family or group-level differences. Overing (1936, 1937) also noted differences in the responses of various passerines to tall, lighted structures. On 20 October 1935, hundreds of Field Sparrows, Spizella pusilla, perched on benches at the base of the lighted Washington Monument; “None of these sparrows struck the monument that night, nor did they seem confused by the lights nor fly against the shaft, as the vireos and warblers were doing.” The following fall, there was a similar occurrence. Of the 523 birds collected by Overing in the falls of 1935 and 1936, only 7 were fringillids. Further differences are suggested by Stoddard 298 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 and Norris (1967) who noticed that during nights of heavy rainfall, fringil- lids tended to persist in their migratory flight while warblers, vireos, and thrushes sought ground cover. No experimental evidence exists detailing differences among various taxa of nocturnal migrants in their response to tall, lighted structures. This area warrants more attention because conceivably such an investigation could lead to methods whereby losses of some species at towers can be reduced. SUMMARY An examination of the cloud cover and wind conditions that accompanied bird losses at a 366-m tower in southeastern North Dakota revealed that most fall losses occurred under overcast skies associated with the passage of cold fronts. In the spring, 58% of the mortality took place on non-overcast nights, generally with southeasterly winds. Rails were killed in relatively equal proportions on overcast and non-overcast nights in both spring and fall. Warblers were killed in significantly greater numbers on overcast nights in both seasons, as were fringillids in the fall. Losses on non-overcast nights tended to be distributed farther from the tower than were those on overcast nights. Fall losses were concentrated closer to the tower than were spring losses because fall losses occurred mostly under overcast skies as migrants milled about the tower. Spring losses seemed to occur primarily on non-overcast nights through collisions with outlying guy wires and the transmitting cables. Behavioral differences among species or families of migrants may be involved in migrant mortality at towers. ACKNOWLEDGMENTS This study was sponsored by the Northern Prairie Wildlife Research Center, James- town, North Dakota in cooperation with the Department of Zoology, North Dakota State University. The statistical advice provided by Robert Carlson and Douglas H. Johnson and the editorial review by Harrison B. Tordoff and W. Wilson Baker is greatly ap- preciated. Willing cooperation was provided by the U.S. Coast Guard’s Omega Naviga- tion Station. This paper is a portion of the master’s thesis of the senior author. LITERATURE CITED Avery, M. L., P. F. Springer, and J. F. Cassel, 1975. Progress report on bird losses at the Omega tower, southeastern North Dakota. Proc. N.D. Acad. Sci. 27 :40-49. , , AND . 1976. The effects of a tall tower on nocturnal bird migra- tion— a portable ceilometer study. Auk 93:281-291. Brewer, R. and J. A. Ellis. 1958. An analysis of migrating birds killed at a television tower in east-central Illinois, September 1955-May 1957. Auk 75:400-414. Cochran, W. W. and R. R. Graber. 1958. An attraction of nocturnal migrants by lights on a television tower. Wilson Bull. 70:378-380. Kemper, C. A. 1959. More TV tower destruction... Passenger Pigeon 21:135-142. Laskey, A. R. 1960. Bird migration casualties and weather conditions, autumns 1958- 1959-1960. Migrant 31:61-65. OvERiNC, R. 1936. The 1935 fall migration at the Washington Monument. Wilson Bull. 48:222-224. Avery etal. • MORTALITY AT A NORTH DAKOTA TOWER 299 . 1937. The 1936 fall migration at the Washington Monument. Wilson Bull. 49: 118-119. Richardson, W. J. and W. W. H. Gunn. 1971. Radar observations of bird movements in east-central Alberta. In Studies of bird hazards to aircraft. Can. Wildl. Serv. Rept. Ser. 14:35-68. SoKAL, R. R. AND F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Fran- cisco. Stoddard, H. L. and R. A. Norris. 1967. Bird casualties at a Leon County, Florida TV tower: an eleven-year study. Tall Timbers Res. Sta. Bull. 8. Taylor, W. K. and B. H. Anderson. 1973. Nocturnal migrants killed at a central Florida TV tower; autumns 1969-1971. Wilson Bull. 85:42-51. Tordoff, H. B. and R. M. Mengel. 1956. Studies of birds killed in nocturnal migra- tion. Univ. Kansas Publ. Mus. Nat. Hist. 10:1-44. DEPT. OF ZOOLOGY, NORTH DAKOTA STATE UNIV., FARGO 58102 (MA AND JFC) AND NORTHERN PRAIRIE WILDLIFE RESEARCH CENTER, JAMESTOWN, ND 58401 (present address ma: 231 gibson road, annapolis, md 21401; PRESENT ADDRESS PFS: U. S. FISH AND WILDLIFE SERVICE, C/O HUMBOLDT STATE UNIV., ARCATA, CA 95521). ACCEPTED 3 NOV. 1976. BREEDING BIOLOGY OF HOUSE SPARROWS IN NORTH MISSISSIPPI James N. Sapplngtox Detailed studies on various aspects of breeding biology of the House Spar- row {Passer domesticus) have been made in Europe (e.g., Summers-Smith 1963; Seel 1968, 1970; Mackowicz et al. 1970 j and in the Lnited States (e.g., Weaver 1943, Mitchell et al. 1973, North 1973, Will 1973). This study was conducted in Oktibbeha County, ^lississippi, during the breeding seasons of 1972-1974 ( Sappington 1975) and is the first to be made below 34° lati- tude. It includes data on the activity of breeding birds and helpers at the nest with corresponding analyses of nest-building, egg-laying, incubation, hatching success, nestling feeding rate, nesting efforts, fledging age, and fledging suc- cess. MATERIALS AND METHODS This study was conducted on the main campus and adjoining areas of Mississippi State University, Oktibbeha Co., Mississippi. 33° 28' north latitude and 88° 48' west longitude. It continued for 3 years, 1972 through 1974, with emphasis on the breeding season which normally lasted from Februarv' to August. Four nesting areas were studied: (1) 4 trees {Magnolia grandiflora) espaliered on the walls of Lee Hall in the central part of the cam- pus, designated as ‘'Tree”; (2) a large barn where sheep and horses were fed and kept overnight, designated as “Horse Barn”; (3) an equipment shed and orchard, designated as “Shed”; and (4) a cluster of buildings associated with a pig feed lot, designated as “Pig Farm.” Only Tree colony was studied in 1972. All colonies were studied during 1973, but only Pig Farm and Shed colonies were studied in 1974. Individual House Sparrows from each area were captured by (1) baited traps, (2) mist nets at roost sites (Sappington and Jackson 1973), (3) hand nets at nests, and (4) hand-lifting young from nest before fledging. Each captured bird was banded with a U.S. Fish and ildlife Service metal leg band and/or color-marked with a coded combi- nation of colored plastic leg bands and released or put back into the nest. Only records of marked birds that were readily identifiable were used for analysis. The mated pair at each of 280 nests was identified by the arrangement of the metal and colored leg bands. Nest-building activities and feeding of nestlings were observed through 7 X 35 binoculars. The birds at Tree, Pig Farm, and Horse Bam colonies were accus- tomed to people passing near their nests at all hours of the day. No observation blinds were necessary in observing breeding activities at these locations. At Shed colony which occupied a single, remotely-located building, the birds were war\- of observers. However, with an automobile as a blind, I was able to make observations within 6 m of the nest without disturbing the birds. Construction of 221 nests at 100 sites was observed. I plotted the location of each nest on a sketch map and recorded the identifying markings of the mated pair and presence of any helper. Individual nests or blocks of nests located in the same area were observed for AS or full-day periods beginning at 07:00 and lasting through 18:00 CST. 300 Sappington • MISSISSIPPI HOUSE SPARROWS 301 A total of 285 nest-days was spent in observing nest-building. A nest-day is defined as a day spent by one person observing a single nest. At 229 nests I was able to establish the date of the appearance of the first egg. These nests were checked daily between 07:00 and 08:00 until the last egg was laid. The date of the laying of the first egg and appearance of the last egg in each clutch were recorded. For those nests in which incubation occurred, clutch size was determined to be the num- ber of eggs at the beginning of incubation. The incubation period was taken as the time between the laying of the last egg and date of hatching of the last young when all eggs hatched. One week after each clutch was completed, each nest was checked for hatching 3 times daily — between 07:00 and 08:00, 12:00 and 13:00, and 18:00 and 19:00. The dates of the hatching of the first young and the hatching of the last young were recorded. Hatch- ing success (%) was calculated from the total number of eggs laid. Feeding of nestlings, ranging in age from 1 to 20 days, was observed at 254 nests be- tween 05:00 and 19:00 for 177 days. My observations include 145 days on which I made continuous observations from 05:00 to 19:00 and 32 days on which I observed nests for shorter intervals. Because of the placement of nests I was generally able to observe from 1 to 7 nests at one time, giving a total of 673 nest-days of observation. The number of visits to the nest with food was used as a measure of feeding activity. Royama (1966) stated that feeding frequencies for another hole-nesting species {Pams major) are far too variable to be used as a true index of food consumption per nestling, but according to Pettingill (1970), no matter how food is supplied, the individual nest- lings receive an equal amount during the course of the day due to automatic apportion- ment. Records were kept of the number of visits to the nest by each parent and helper (a bird other than the parents). Observations were made during all weather conditions ex- cept heavy rain. According to Kendeigh (1952), cloudiness, fog, or wind do not affect feeding rates; only heavy rain affects them and then only temporarily. For calculations of total daily visits to the nest and visits per nestling, only those observation periods which lasted the entire day were used. A nesting effort was arbitrarily considered to be an incubated clutch. Fledging was considered to be the time when a young bird took flight from the nest the first time. Age at fledging, or nestling period, was calculated from the day of hatching until the day of fledging. Fledging success (%) was calculated from the total number of eggs laid and number of eggs hatched. Statistical analyses were performed using the UNIVAC 1106 computer at Mississippi State University. The analysis of variance, as well as basic statistics, including means, standard deviation, and standard error were obtained from these data by the first option of UNIVAR (1973 version), a basic statistics program written by D. M. Power. Basic statistics for samples of more than 2000 cases were obtained by using the program, BMD OlV, ANOVA for One-Way Design of the Biomedical Computer Programs f Dixon 1974). When only 2 numbers were compared for significant difference, the Chi-square test was used. Yates’ correction (Chase 1967) was applied when the expected frequency was fewer than 5 cases. When 2 percentages were tested for significant differences, as in percent success, a computer program written by Jerome A. Jackson of Mississippi State University was used. The program calculates a t-value which may be compared with a tabular t. The method of this test is based on the arcsine transformation as suggested by Sokal and Rohlf (1969) . I used a probability level of 0.05 as the criterion for significance in all statistical analyses. 302 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 RESULTS AND DISCUSSION During the 3 breeding seasons of 1972-1974, 987 House Sparrows were color-banded at the 4 locations; 311 were adults, and 676 were juveniles. The male-female ratio of total marked adult birds was 1:1.03 (153 males and 158 females). Other studies have shown that the sex ratios for House Spar- rows have not differed greatly from 1:1 (Summers-Smith 1963, Will 1973, North 1973) . I also used observational data from roosting and feeding studies to get an indication of the male-female ratio for the study areas. Among roosting birds, males constituted 52.1% of the total, and among feeding birds, they constituted 50.1% of the total. These figures do not differ significantly from my findings of 49.2% males in the breeding situation. Of the marked adults, 82.6% were breeders (78.4% males; 86.7% females). Summers-Smith (1958) stated that the mated male and female House Sparrow remain faithful to each other and to their nest site for life. He does, however, cite exceptions involving bigamy, desertions, and the holding of more than one nest site by a single male. In my study of 100 nest sites, the mated pair remained faithful to each other for only 60.5% of the cases during a particular breeding season. However, the number of males having only 1 mate (69) was significantly higher than those having more than 1 mate (45) during a breeding season. There was no evidence that pairs remained to- gether for 2 consecutive breeding seasons. Although 39.5% of the males had more than one mate during a single season, no cases of simultaneous polygamy were observed. Of 156 breeders of known origin, 125 (80.1%) nested at the location where they were handed. The difference between numbers of sedentary breeders and transient breeders was highly significant. Generally, breeding House Sparrows returned to the area where they were handed but not necessarily to the same nest site of previous years. Attachment to the nest site appeared to be strongest in the male. Of the total nest sites, 86.0% were retained by the male for the en- tire breeding season as opposed to 45.0% by the female. There was a highly significant difference between number of sites occupied by a single male (86) and number occupied by more than one male (14 ) during the breeding season hut no significant difference between number of sites occupied by a single female (45) and number occupied by more than one female (55). Individual males showed very little attachment to their nest site after the breeding duties were over. In subsequent breeding seasons only 10% (6 of 60 cases) of the sedentary breeding males returned to their previous nest site. After the breed- ing season all birds used communal roosts instead of their nest sites for the fall and winter months. The old nests were torn out by people, or weathering deteriorated them. These factors may have been reasons why so few spar- rows returned to their old nest sites in subsequent years. Sappington • MISSISSIPPI HOUSE SPARROWS 303 During the 3 seasons of 1972—1974, 584 nests were built at 296 sites at the 4 locations. This activity embraced a time span of approximately 5 months each year. Typically, it began in early February and lasted until near the end of July. During the 3-year period the earliest nest was started 10 Febru- ary, and the latest was started 21 July. Bent (1958j stated that nesthuilding occurs in various places of the United States during every month of the year. Nest-building was observed at 221 nests of which 5 ( 2.26%) had helpers. Although cooperative nest-building is common among some weaver finches (Crook 1960, MacLean 1973), I have seen no previous reference to this prac- tice by House Sparrows. A rhythmic pattern was manifested in both nest- building and egg-laying. Peak periods of nest-building were generally fol- lowed within a week by intensive egg-laying. Others have reported similar patterns (Mitchell et al. 1973, Summers-Smith 1963, Weaver 1939). Of 584 nests built, 532 contained eggs. Egg-laying during each of the 3 years embraced a span of approximately 5 months beginning in late February and continuing through July. The first egg was laid 24 February, and the last egg was laid July 28 for the 3-year period. I did not quantify the amount of time spent incubating by each sex of the mated pair. Weaver (1943) stated that only the female incubates since the male was never observed to sit on the eggs. On the other hand, Daanje (1941) stated that both sexes incubate the eggs. Summers-Smith (1963) found that both sexes spent spells of time on the eggs during incubation, but since the male does not develop a brood patch, it cannot be truly said that he incubates. I observed that the male relieved the female at the nest 5 or 6 times per day for periods up to 20 min in length. Presumably the female was feeding at this time. At night only the female sat on the eggs. Although communal be- havior in egg-laying and incubation has been exhibited by other social species, Mexican Jay {Aphelocoma ultramarina) (Brown 1970) and Smooth-billed Ani (Crotophaga ani) (Davis 1940), there was no evidence of such behavior in the House Sparrow. For 229 nests the mean incubation period was 12.2 days (S.E. = 0.12) ( Range = 10 to 17 days). The 17-day period occurred in the Horse Barn colony in 1973 and may be attributed to the sudden occupancy of the area by a Barn Owl {Tyto alba) which caused such stress that both incubation and feeding were often halted for several hours at a time. My data do not differ significantly from that found by others (12 days. Weaver 1943; 11.2 days. Seel 1968; 11.3 days, Mitchell and Hayes 1973). Weaver (1943) and Sum- mers-Smith (1963) stated that the hatching period may be spread over 2 or 3 days. I found a much shorter time span for hatching completion. Hatch- ing began in the early morning and did not last beyond 18:00 of the same day. There was only 1 exception in which 1 nest in 1973 required 2 days for 304 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 1 Comparison of Mean Daily Visits to Nests With Helpers (W) and Nests Without Helpers (WO) Parameter! Mean Standard Error Visits by parents W 222.7** ±2.0 WO 206.6 ±2.9 Visits by helpers 31.4 ±0.5 Total visits W 254.1** ±2.8 WO 206.6 ±2.9 Visits per nestling W 71.5** ± 0.4 WO 58.1 ±0.7 Number of nestlings W 3.6 ± 0.04 WO 3.6 ± 0.04 1 Based on 321 nest-days of observation at nests with helpers and 352 nest-days of observation at nests without helpers. ** (p^.Ol). hatching completion. For the entire study period the earliest hatching date was 11 March and the latest was 16 August. Feeding of nestlings was observed at 254 nests, of which 161 (63.4%) had multiple-feeders (helpers). A chi-square test indicated there were signifi- cantly more nests with helpers than without helpers. In nests with helpers mean daily total visits amounted to 254.1 of which 31.4 (12.4%) were from helpers (Table 1). Nests without helpers received a mean of 206.6 daily visits, significantly fewer (p < 0.01) than at nests with helpers. For all nests observed the mean daily feeding frequency was 229.3 with 64.5 per nestling. The mean hourly feeding rate was 16.5 (S.E. = 0.09) with a mean of 3.6 nestlings per nest. Kendeigh (1952) reported a feeding rate of 20 times per hour for nestling House Sparrows with 4 young per nest. Comparison of our data sets using chi-square indicates that they do not differ significantly. Temporal patterns were present in the hourly feeding rate (Sappington 1975). Three peak periods occurred daily, late morning, mid afternoon, and late afternoon. Feeding was minimal between 05:00 and 06:00 and between 18:00 and 19:00. A great deal of variation occurred in both total visits to the nest and visits per nestling depending on ages and number of young per nest. However, Sappington • MISSISSIPPI HOUSE SPARROWS 305 these parameters varied very little (no significant difference) after the 8th day of age in nests with both 3 and 4 nestlings (Sappington 1975). It ap- peared that the feeding rate changed very little after the first half of nestling life. Summers-Smith (1963) found that the frequency of feeding nestling House Sparrows increased until the 14th day, alter which there was a de- crease. My results were similar to those of Seel (1969) who discovered a common pattern of feeding in broods of nestling House Sparrows of all sizes which was a rising phase up to nestling day 8^/4 to IP/^, followed by a levelling off phase. Increased feeding in my study was halted at the approximate time that homeothermy is supposedly accomplished in the nestling House Sparrow (Pettingill 1970) . One might surmise that the larger the brood the more visits to the nest. I found a direct relationship between brood size and daily visits to the nest, but there was an inverse relationship between brood size and visits per nest- ling (Sappington 1975). Seel (1969) found the same relationship for broods of 1 to 3 nestling House Sparrows, and Moreau and Moreau (1940) found that the smaller the brood the greater the number of feedings that each re- ceived. These findings agree with the results of von Haartman (1953). He found that it is not the number of nestlings but their reactions that stimulated parents to bring food. Kendeigh (1952) found that nests with fewer birds fledged earlier. My study generally showed this trend but the differences are very slight and are non-significant for nests which fledged 2, 3, 4, or 5 birds with or without helpers. Moreau and Moreau (1940) also found that the smaller brood does not fledge earlier. I observed that a marked change in the behavior of both the nestlings and helpers occurred 2 or 3 days prior to fledging. Nestlings become exceedingly quiet, lying crouched in the nest. Helpers no longer fed them. On the day of fledging parents rarely fed the young until they left the nest. Fledging generally occurred in the early morning and seldom did all young leave the nest at once. Summers-Smith (1963) and Weaver (1943) found that the fledging period of a single nest may be spread over 2 or 3 days. From 180 nests in my study all young fledged by 12:00 on the same day. Never were there more than 4 h between the fledging of the first and last nestling from the same brood. This synchronized pattern of fledging should not be considered unusual, but could be expected because of the small age differential in nestlings within the same brood as experienced in this study. My study of 180 nests showed that time spent in the nest varied from 14 to 23 days with an overall mean of 17.1 days (S.E. = 0.15). The 23-day period which occurred in Horse Barn colony may be attributed to the presence 306 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 of the Barn Owl. Summers-Smith (1963) gave a nestling period of 11 to 19 clays (mode of 18) for Great Britain. Kendeigh (1952) cited the nestling period to be 14 to 16 days for Illinois. Weaver (1942) found that the time spent in the nest varied from 12 to 16 days with a mean of 14.4 days for New York. Although my study showed a higher nestling period than did the Il- linois or New York study, there was no significant difference. Also, there was no significant difference in mean number fledged from nests with helpers (2.8) and nests without helpers (2.9). Each nest site was used an average of 2.84 times (S.E. = 0.10) during a single breeding season. This figure is somewhat higher than those of 2.1 (Summers-Smith 1963), 2.0 (Will 1973), or 1.68 (Weaver 1943) but ap- proaches 3.0 found at a site studied by Mitchell et al. (1973). Use was high- est in 1974 when each nest site was occupied an average of 3.2 times. In 1974 there was a scarcity of sites as compared to the 2 previous years. Nest sites were limited to only the Pig Farm and Shed locations during 1974, and at the Pig Farm 5 of the 7 buildings normally used were torn down, thus limiting the number of available sites. The number of incubated clutches per pair (N = 142) ranged from 1 to 4 with a mean of 1.84 (S.E. = 0.07) which compares favorably with 2.1 of Graggs (1967) and Summers-Smith (1963). Percent of pairs having 1, 2, 3, or 4 nesting efforts were 39.4, 40.9, 15.5, and 4.2 respectively. The number of eggs laid per female per season (N = 596) ranged from 3 to 16 with a mean of 7.46 (S.E. = 0.29). This figure is not significantly different from the 7.95, 8.94, or 8.61 eggs per female per season reported by Weaver ( 1943), Will (1973), and Summers-Smith (1963) respectively. The number of eggs hatched per pair ( N = 495 ) ranged from 2 to 16 with a mean of 6.37 ( S.E. = 0.26) as compared with 5.88 (Will 1973) and 6.11 (calculated from data from Summers-Smith 1963). Breeding success was based on results from 224 nests. Glutch size ranged from 2 to 6 eggs with a mean of 4.2 (S.E. = 0.06). Nests containing 4 eggs each accounted for 60.3% of the clutches. McAtee ( 1940) found that clutch size in Maryland ranged from 2 to 6 with a mode of 5. Bent ( 1958) gave a range of 3 to 7 with a mode of 5 throughout the United States. My average is well within these ranges. Others report similar means from Great Britain — 3.9 Seel (1968) and Graggs (1967), 4.1 Summers-Smith (1963) — and from the United States — 1.3 Mitchell et al. (1973), 4.4 Will (1973), 4.7 Weaver ( 1943 ) . My average is not significantly different from these previous studies. The number hatching per nest was 3.4 (S.E. = 0.08) in nests in which at least 1 hatched. The number fledging per nest was 2.8 (S.E. = 0.09) in nests in which at least 1 fledged. Hatching success, based on all eggs laid (Table 2), was 83.2% which is Sappington • MISSISSIPPI HOUSE SPARROWS 307 Table 2 Overall Breeding Success of House Sparrows at Mississippi State, Mississippi Category Value Total clutches 224 Clutches lost 19 Eggs lost 72 % of eggs surviving 92.1% Number of eggs hatched 758 % of total eggs hatched 83.2% % of surviving eggs hatched 90.4% Number of young fledged 584 % of total eggs 64.1% % of incubated eggs 69.6% % of eggs hatched 77.0% significantly higher (p<.01) than 71.0% (Summers-Smith 1963), 61.0% (Mitchell et al. 1973), or 65.8% (Will 1973). It is also somewhat higher than the average of 77.0% which Nice (1957j attributed to altricial hole- nesting species, but is not significantly different from 85.4% calculated for House Sparrows from data reported by Seel (1968). Fledging success, based on total eggs laid, was 64.1% for the entire period (Table 2). This figure compares with 66.0% given by Nice (1957) for hole-nesting altricial birds, but is significantly higher (p < .01) than 50.0% (Summers-Smith 1963), 41.0% (Mitchell et al. 1973), or 35.1% (Will 1973) reported for House Sparrows in other studies. Although some previous studies were quite de- tailed, specific mention of cooperative breeding activities is almost lacking and at best fragmentary. One might assume that my high breeding success could be attributed to the activity of helpers. However, there was no signifi- cant difference in fledging success from nests with helpers (68.5%) and nests Avithout helpers (72.2%). The higher breeding success in my study does not appear to be the result of assistance by helpers, but may be attributed to the small percentage of eggs lost to breakage and predation (7.9%), high hatching success of surviving eggs (90.4%), and rather small percentage of nestlings ( 23.0%) which died or fell to predators. SUMMARY Breeding biology of the House Sparrow (Passer domesticus) was studied in 4 colonies of marked birds during the breeding seasons of 1972-1974 in Oktibbeha Co., Mississippi. The male-female ratio of breeding birds was 1:1. The male generally remained faith- ful to his nest site (86.0%) but not so faithful to his mate (60..5%) for the entire breed- ing season. 308 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Each nest site was used an average of 2.84 times per season, but the number of in- cubated clutches per pair was 1.84. There was a trace of cooperative nest-building, but there was no evidence of communal egg-laying or incubation. Rhythmic patterns were manifested in both nest-building and egg-laying with peak periods of nest-building fol- lowed within a week by intensive egg-laying. Mean clutch size was 4.2 eggs. The mean incubation period was 12.2 days. A mean of 3.4 eggs hatched, and 2.8 young fledged per nest. Young fledged at 17.1 days of age. Nestlings were fed at a mean rate of 16.5 times per hour. Feeding of nestlings by sparrows other than the parents was observed in 161 of 254 nests in which feeding was significantly higher than in those nests without helpers. Helpers accounted for 12.4% of the feeding in those nests which they visited. Hatching success (83.2%) and fledging success (64.1%) were significantly higher than those reported by other researchers but do not appear to be the result of assistance by helpers; they may be attributed to the small percentage of eggs lost to predation and breakage, high hatching success of surviving eggs, and low mortality of nestlings. ACKNOWLEDGMENTS I wish to thank Dr. Jerome A. Jackson who served as chairman of my doctoral dis- sertation committee and for his criticism of this manuscript. Credit is also given to Ken- neth Bicker, Robert Kirkland, Gordon McWilliams, and Robert Stewart for their valu- able assistance in banding and behavioral observations. Constructive criticism of the manuscript by Drs. Richard Johnston and S. Charles Kendeigh is greatly appreciated. I gratefully acknowledge the Frank M. Chapman Memorial Fund of the American Mu- seum of Natural History for part of my financial support. LITERATURE CITED Bent, A. C. 1958. Life histories of North American blackbirds, orioles, tanagers, and allies. U.S. Natl. Mus. Bull. 211. Brown, J. L. 1970. Cooperative breeding and altruistic behavior in the Mexican Jay {Aphelocoma ultramarina) . Anim. Behav. 18:366-378. Chase, C. I. 1967. Elementary statistical procedures. McGraw-Hill Book Company, N.Y. Craggs, J. I). 1967. Population studies of an isolated colony of House Sparrows iPasser doniesticus) . Bird Study 14:53-60. Crook, J. H. 1960. Studies on the social behavior of Quelea q. quelea (Linn.) in French West Africa. Behavior 16:1-55. Daanje, a. 1941. Uber das Verhalten des Haussperlings (Passer d. domesticus). Ardea 30:1-42. Dams, I). E. 1940. Social nesting habits of the Smooth-billed Ani. Auk 57:179-218. Dixon, W. J. (ed.). 1974. Biomedical computer programs. Univ. Calif. Press, Berke- ley. Kendeigh, S. C. 1952. Parental care and its evolution in birds. Illinois Biol. Monogr. 22:i-x, 1-356. Mackowicz, R., j. Pinowski, and M. Wieloch. 1970. Biomass production by House Sparrow {Passer d. domesticus L.) and Tree Sparrow {Passer m. montanus L.) populations in Poland. Ekol. Polska 18( 23) :465-501. McAtee, W'. L. 1940. An experiment in songbird management. Auk 57:333-348. Sappington • MISSISSIPPI HOUSE SPARROWS 309 MacLean, G. L. 1973. The Sociable Weaver, Part 2: Nest architecture and social or- ganization. Ostrich 44:191-218. Mitchell, C. J. and R. 0. Hayes. 1973. Breeding House Sparrows, Passer domesticus in captivity. Ornithol. Monogr. 14:39-48. , R. 0. Hayes, P. Holden, and T. B. Hughes, Jr. 1973. Nesting activity of the House Sparrow in Hale County, Texas, during 1968. Ornithol. Monogr. 14:49-59. Moreau, R. E. and W. M. Moreau. 1940. Incubation and fledgling periods of African birds. Auk 57:313-325. Nice, M. M. 1953. The question of ten-day incubation periods. Wilson Bull. 65:81-93. . 1957. Nesting success in altricial birds. Auk 74:305-321. North, C. A. 1973. Movement patterns of the House Sparrow in Oklahoma. Ornithol. Monogr. 14:79-91. Royama, T. 1966. Factors governing feeding rate, food requirement and brood size of nestling Great Tits Parus major. Ibis 108:313-347. Sappington, J. N. 1975. Cooperative breeding in the House Sparrow (Passer domesti- cus). Ph.D. thesis, Mississippi State Univ., Mississippi State, Mississippi. AND J. A. Jackson. 1973. A technique for mist netting birds at communal roosts. Inland Bird Banding News 45:102-106. Seel, D. C. 1968. Clutcb size, incubation, and hatching success in the House Sparrow and Tree Sparrow Passer spp. at Oxford. Ibis 110:270-282. . 1969. Food, feeding rates, and body temperature in the nestling House Spar- row Passer domesticus at Oxford. Ibis 111:36-47. . 1970. Nestling survival and nestling weights in the House Sparrow and the Tree Sparrow Passer spp. at Oxford. Ibis 112:1-14. SoKAL, R. S. AND F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Fran- cisco. Summers-Smith, D. 1958. Nest-site selection, pair formation, and territory in the House Sparrow Passer domesticus. Ibis 100:190-203. . 1963. The House Sparrow. Collins, London. VON Haartman, L. 1953. Was reizt den Trauerfliegenschnapper [Muscicapa hypoleuca) zu futtern? Vogelwarte 16:157-164. Weaver, R. L. 1939. Winter observations and a study of the nesting of English Spar- rows. Bird-Banding 10:73-79. . 1942. Growth and reproduction of English Sparrows. Wilson Bull. 54:183-191. . 1943. Reproduction in English Sparrows. Auk 60:62-74. Will, R. L. 1973. Breeding success, numbers, and movements of House Sparrows at McLeansboro, Illinois. Ornithol. Monogr. 14:60-78. DEPT. OF ZOOLOGY, MISSISSIPPI STATE UNIV., MISSISSIPPI STATE 39762 (PRESENT ADDRESS: DEPT. OF BIOLOGY, WILLIAM CAREY COLLEGE, HATTIESBURG, MS 39401). ACCEPTED 24 MAR. 1976. NESTING BEHAVIOR OF YELLOW-BELLIED SAPSUCKERS Lawrence Kilham This communication presents aspects of the nesting of Yellow-bellied Sap- suckers {Sphyrapicus varius) either not mentioned in previous accounts (Bent 1939; Johnson 1947; Howell 1952; Kilham 1962a, b; Lawrence 1967) or, if mentioned, open to amplification and new or other interpretation. Activities covered extend from the start of excavating through egg-laying and incubation to fledging. They do not include agonistic and courtship behavior which are being described separately. METHODS With 5 pairs studied in special detail and 4 others for a more limited time, I made observations from % to 2 hours a day every day, with a few missed from late April or early May through fledging in July, either between 08:00-10:00 or 15:00-17:00, these times having been found to be equivalent in terms of heights of activity. Percentages of time that the male or the female of a breeding pair spent in such activities as incubation or brooding refer, as shown in Tables 1 to 4, to total observation times, a method also used by Lawrence (1967). They are given for convenience of description and are not intended to imply total coverage that would have demanded dawn to dusk observations 7 days a week throughout the nesting period. In addition to counted hours 1281) I spent many uncounted ones in partial observations on 12 other nesting pairs in Lyme, New Hampshire. In regard to attentiveness I have not used methods employed, among others, by Stickel (1965), Lawrence (1967), and Skutch (1969). Stickel, who gives the most detail, designates “attentiveness as the time adults spent excavating a cavity, sitting beside it, guarding and incubating eggs, and once the birds bad hatched, as that time the parents remained at the nest cavity.” What Stickel refers to as sitting by a nest, guarding it, etc., I have considered under the general term of “loitering.” Only those times, therefore, that tlie sapsuckers actually spent in the work of excavating, sitting on eggs, or brooding young within the nest, have been considered in making calculations. A session at the nest, in contrast, has been regarded as the total time that one of a pair spends in or by the nest until relieved by its partner. Descriptions of the vocalizations, drummings, tappings, and displays of 5. varius mentioned in this report are given elsewhere (Kilham 1962a). EXCAVATION Share done by males and females. — The amount of work done by either sex depends on circumstances. At 5 nests that were first excavations of the breeding season, I found that the males did nearly all of the work. When they excavated, they worked continuously from 15 to 30 min at a stretch. When females changed places with them they spent much of their time in preening and resting as Lawrence (1967) has described. Their excavating was often token in character, with little sawdust removed, particularly after completion 310 Kilham • SAPSUCKER NESTING BEHAVIOR 311 of the entrance corridor. But females can work under special circumstances. When Pair A abandoned a first excavation in 1974 and began a second one, Female A (FA) did 68% of the excavating observed in the first 4 days. This was excavating of the hardest type, for FA dug through 6.5 cm of the living, outer shell of an aspen {Populus tremuloides) . The aspen had heart rot due to infection with Fames igniarius (Kilham 1971). When the easier digging at the center was reached, the male took over and did 79% of the excavating. The cavity was completed in the next 5 days. FA worked so continuously in her first stretch that she took no time for rest- ing and preening during periods of observation and on a number of occasions refused to leave when her mate came to change places. Why should she have worked in this manner? A possibility is that females are a reserve in regard to excavating. By not working under usual circumstances they conserve energy for forming and laying eggs. When a first excavation fails, a female that had been becoming ready physiologically for egg-laying, suddenly finds herself with no place to do so. As a result her drive to excavate may become even greater than that of her mate. Tight fit of nest entrances. — A feature of holes in Fome5-aspens is that entrance corridors carved by males, when they do most of the excavating, are a snug fit for their bodies. Females do not usually enter nest cavities until the time of egg-laying when, if their body size is larger than that of their mates, they may have a hard time getting in. The first time I saw female C 1 FC) enter her nest cavity was on 20 May. She had to wriggle to force her way. After remaining inside for 5 min she had difficulties getting out. She pushed her head outside 5 times, moving it violently up and down as she struggled, without success until the 5th time, to force herself through. Even later on, when feeding young, she pumped her head up and down in struggles to emerge. I have observed the same phenomenon for a female Downy Woodpecker (Picoides puhescens) that had an even more difficult time emerging at the start of egg-laying. Female A, who had carved her own corridor, had no such difficulties. Attempt to reuse nest hole of a preceding year. — Sapsuckers carve fresh nest holes each year. On 15 April 1975 Male D made the start of an excavation in a Fome5-aspen 8 days before I saw the female. The excavation was completed in May and used successfully. A pair of sapsuckers (Pair H) returned to the same aspen in 1976 but instead of starting a new excavation, for which there appeared to be no suitable sites, they attempted to reuse the old hole of the year before. This was evidenced by tapping, by the male entering and giving breeding calls from the entrance, and by courtship flights. On 29 April Male H entered twice, coming out with black fecal matter that appeared to be distasteful, for he wiped his bill many times on a pine limb. 312 THE WILSON BULLETIN • VoL 89, No. 2, June 1977 Pair H abandoned the hole in early May and excavated a fresh one in a new Fomes aspen. It seemed from this experience that accumulation of fecal mat- ter, which I had measured as being 6 cm deep at the end of the 1975 nesting season, had possibly acted as a deterrent to reuse of the old hole. EGG-LAYING FA was in Nest A from 05:35 until 06:35 on 2 June. This was longer than any single stretch that she spent during the incubation period. When MA came to the entrance at 05:57 as if to change places, she remained out of sight and when he returned at 06:21 she struck out sharply, driving him away. Similar behavior was observed in Female G on 24 May when she had 4 eggs in the nest. This dominance of the female at time of egg-laying may be a general phenomenon, for I have also noted it in the Common Flicker (Colaptes auratus; Kilham 1959). In 1976 Pairs H and I finished their excavations by the middle of May. Egg-laying and the start of incubation that should have followed appeared to be delayed by 9 days of cold, wet weather. Although I did not find when first eggs were laid in either nest. Nest H contained 1 egg on 23 and 4 on 28 May, the first day of incubation. Of these eggs, only 2 hatched. The history of Nest I was more complicated. There were 3 eggs on 23 and 4 on 24 May. Neither the male nor the female appeared much disturbed when I put a ladder up to their nest. I was thus unable to account for destruction of 3 eggs that I watched Male I carry from the nest on 25 May. On 29 and 30 May a single egg remained. Four more were laid between 1 and 4 June. Of this total of 5 eggs, 3 hatched. There were thus high rates of failure (50% in H and 40% in I > in both nests. These failures to hatch may have related to exposure during the prolonged spell of cold weather. The percentages of egg failure appear high. Ricklefs (1969) found that only 8.1% of 3226 eggs of 6 passerine species failed to hatch. All 5 of the unhatched sapsucker eggs were removed by the parent sapsuckers within a day. INCUBATION Although most pairs settled down to incubating promptly, the females of Pairs C and D frequently loitered outside of their nests following changeovers with their mates during the first 5 days of incubation. In the last 6 or 7 days of incubation, they became more attentive than the males, FC doing 60% and FD 86% of the total incubating. Experience of a following year with Pairs H and I is shown in Table 1. Here the 2 females incubated from the beginning of the period and were as attentive or nearly so as the males in their sessions at the nest. As their ses- Kilham • SAPSIJCKER NESTING BEHAVIOR 313 Table 1 Attentiveness of Males (M) and Females (F) of 2 Pairs of Sapsuckers During Incubation and Brooding; Both Females were Black Polymorphs Activity Pair and Sex Total time of sessions ( min ) Attentiveness HM 610 94% HF 293 94% Incubation IM 239 97% IF 189 93% HM 375 80% HF 236 83% Brooding IM 107* 85% IF 184* 58% * Regular brooding only lasted 4 days in contrast to 9 days for Pair H. sions were fewer than those of the males, the latter did the greater part of the incubating in both pairs. I have found sapsuckers very restless during incubation periods, often looking or coming out of their nest holes, regardless of how near or far away I sat watching. These periods of abandoning the eggs added up to considerable amounts of time, especially in the case of the females. Thus in nearly 31 hours of watching at Nests A, C, and D (Table 2) I found that eggs were left uncovered for a total of 5 hours or close to 16% of the time. Can weather affect incubating? This appeared to be the case on only one day for the 3 pairs of sapsuckers shown in Table 2. The hottest day of the spring of 1974 was on 10 June which was the 8th day of incubation for Pair A. The temperature was 36° C in the shade and presumably hotter in the nest hole that was exposed to the mid-afternoon sun. Although the members of this pair had been incubating close to 100% of their time for the previous 3 days, they became very restless on 10 June leaving their eggs unguarded for 34 out of the 60 min that 1 watched them. PERIOD OF FEEDING YOUNG Brooding. — I have considered brooding as the days when each member of the pair remained on the nest for a high percentage of its time until relieved by its mate. Tables 1 and 3 give ranges of these percentages that, as shown in Table 1, were lower than those noted during the period of incubation. As nest J were brooded 82% of the time I watched, the 3 young of nest I were 314 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Table 2 Amounts of Time Eggs were Left Uncovered During Periods of Orservatton IN Incubation Periods of 3 Pairs of Sapsuckers Pair A Pair C Pair D Duration of observations (min) 520 782 550 % of observation times eggs left uncovered 6% 18% 23% shown in Table 3 brooding was of 8 to 10 days duration. I did not consider as brooding the periods of 2^ min that sapsuckers spent in nests on subsequent days. On some of these I could hear excavating. This suggested that the males (and less often the females) were producing sawdust used in relation to removal of fecal matter from the nest (Kilham 1962b). There were 2 exceptions to stoppage of brooding on a definite day. One was with Pair A that, after ceasing to brood on 24 June, brooded for 42 of 60 min on 26 June which was rainy and exceptionally cold. The other was with Pair I (Table 1) where the brooding was regular only through day 4, then irregular and infrequent until day 7 when it ended. As shown in Table 4 and described in the following section, I had nests with 1, 2, 3, and 4 young. A point of interest was that whereas the 2 young of Nest J were brooded 82% of the time I w atched, the 3 young of Nest I were brooded 67 and the 4 young of Nest A 73%. While the differences were not great, the nest with the fewest young received the most brooding. Royama Table 3 Results of Observations on the Number of Days 3 Pairs of Sapsuckers Brooded Their Newly Hatched Young and the Ways Duties were Shared Between Males (M) and Females (F) Pair A Pair C Pair D Parameters M F M F M F Amounts of time sexes lirooded (min) 251 220 163 164 94 120 Percentages of total time 53% 47% 50% 50% 44% 56% Average duration of periods M and F brooded (min) 19 16 13 Duration of brooding (days) 9 10 8 Observation time (min) 644 360 321 Percent of observation time that sapsuckers brooded 73% 91% 67% Kilham • SAPSIJCKER NESTING BEHAVIOR 315 Table 4 Effects of Number of Young in the Nest (1, 2, 3, or 4) on Feeding Rates of Pairs of Sapsuckers and one Lone Male* Pairs of Sapsuckers J = Lone Male H I A Period No. of young Rate/h No. of young Rate/h No. of young Rate/h No. of young Rate/h Brooding 2 7.3 3 9.3 4 10.6 End of brooding thru day 15 after hatching (Nest not under observation) 2 13.4 3 15.4 4 17.1 Last 9 days before nest leaving Day of nest leaving 1 6.2 10.1 26 ( End of nest; predation by weasel? ) 4 28 20.8 * Observation time, 89.4 h. ** 1 died; cause unknown. (1966) writing of the Great Tit [Pams major) emphasizes that there was greater heat loss in small broods, where young were less efficient at keeping each other warm. They would, therefore, need more brooding and/or more feeding. Feeding young. — A feature of feeding young noted at all nests was that the sexes shared the tasks almost equally, with the males making 6 to 10% more visits than their partners (Fig. 1). Figure 2 gives the combined male plus female feeding rates of Pair A. I found by inspection that Nest A contained 4 young and I visited it for an hour or more every day from hatching on 15 June to fledging 28 days later. After a low average of 10.5 when parents were spending much of their time brooding, the rate jumped to 24 visits per hour with its cessation. This high average continued to within 4 days of fledging when it fell to 16. Findings with Pair D, recorded in a similar manner, followed an almost identical curve. A feature of Lawrence’s paper is the all-day feeding rate. My nests were visited either in the mid-morning or mid-afternoon, at which times I found no differences in rates as would seem to be the experience of Lawrence ( 1967:116) as well. As stated by Lawrence (1967:113) “it is difficult to arrive at a valid analysis of the feeding rate without knowledge of the exact number of young.” I was fortunate in 1976 to have 3 nests that, being within 3 to 5 m of the ground were easily accessible by ladder. These nests contained 1, 2, and 3 316 THE WILSON BULLETIN • Vol 89, No. 2, June 1977 K ^ / V / ACTIVITY FEEDING YOUNG REMOVAL OF FECES NO. OBSERVATIONS 527 213 151 81 39 29 Fig. 1. Observations on 3 pairs of nesting sapsuckers showing that while the task of feeding the young was shared by both sexes to a nearly equal extent in all pairs, that of nest sanitation varied considerably. Observation times, here combined, were between 08:00-10:00 and 15:00-17:00. young, while Nest A in 1974 contained 4. As shown in Table 4 the feeding rates declined in stepwise fashion from Nest A with 4 young to Nest H with only 2. But the decline was not proportionate to the number of young, for the fewer the nestlings, the more each one received from its parents. These extra feedings may have hastened the time of nest leaving; the single nestling in Nest H, for example, having left on day 26 as contrasted with the 4 nestlings of Nest A that left on day 28. Pair H provided an exception to the finding that males and females, in general, fed young almost equally. Although the members of the pair had shared the feedings almost equally when they had 2 young, the male did almost all of the feeding (75%) in the last 9 days when only a single nestling remained. Vocalizations of nestlings. — The vocalizations of nestling sapsuckers are described by Lawrence (1967:125) as well as by Kilham Q962a). They carry for considerable distances and can be of aid in locating nest trees. They might, therefore, also serve to attract predators. If they are a hazard they must, it would seem, provide compensating advantages. It is conceivable that the harsh “check-check-checks” of the young stimulate parents to keep the feeding at a high rate and to thus raise more young. A line of evidence suggesting that the vocalizations do stimulate adult sapsuckers involves intrusions by lone adult male and female sapsuckers. In sapsucker nests that I followed every day through the nestling period I seldom saw' these intruders until the last week or two of the nestling period when they began coming to nest trees, sometimes repeatedly, to look into the nest hole in spite of being driven away. Kilham • SAPSUCKER NESTING BEHAVIOR 317 Fig. 2. Curve of combined male plus female feeding rates of a pair of nesting sap- suckers, showing sharp increase of rate at termination of brooding. Observation times, here combined, were between 08:00-10:00 and 15:00-17:00. They thus behaved as if something, such as the vocalizations that carried for a distance, were attracting them. A parent sapsucker, on the other hand, may return repeatedly to an empty and silent nest. The young of Pair A left early on the morning of 13 July. Male A came to the hole later, bowed in 3 times, then left to add sap to his 318 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 bill-full of insects before returning. He made 5 such visits before flying off to feed the insects plus sap to a juvenile. The behavior of Male I was even more striking. After his young had been killed by a predator, he came to the hole 10 times in 17 min with a load of insects that became steadily larger as he continued to catch insects between visits. On his 10th visit, MI swallowed the insects and left. The return of MA and MI to their silent nests was possibly due to habituation over previous days. Skutch (1976) noted a Golden-naped Woodpecker { Melanerpes chrysauchen) that continued to bring food lor 6 days after a nest was desolated. The range of vocalizations of nestlings is considerable and probably conveys a variety of messages ranging from hunger to alarm. A peculiar episode took place at Nest D the day before nest-leaving when an intruder, a female, as- cended the nest aspen cautiously, then poked in at the nestlings. They immedi- ately set up a wailing noise that I had not heard previously and continued it for 30 sec after the intruder had left. The reaction suggested that the young sapsuckers were able to recognize the intruder as being a foreigner and not their own mother. JNest sanitation. — Both Johnson (1947) and Lawrence (1967) noted that males do most of the nest cleaning, but I have found more variation in this task than in feeding the young (Fig. 1). Among 7 pairs of sapsuckers followed in the same manner, the amount of the work done by the females ranged from 0% in 3, to 2, 22, 30, and 56% respectively in 5 other nests. In one out of 7 therefore, the female did more than the male. Another parameter showing variation is the day when parents cease to remove feces. In 3 nests, those of Pairs F, H, and A, this was 4, 5, and 8 days before the young left the nest. The amount of black, tarry fecal matter that accumulated in Nest H was 5 cm and in Nest A, 6 cm deep as measured after- ward. Male G, in contrast to males of other pairs, removed 7 large bill-fulls of feces in 30 min on 22 July, the day before fledging, and Lawrence (1967:120) saw a male remove feces after 1 fledgling had left. Emergence. — At 07:10 on 30 June, about 30 sec after it had been fed, a juvenile flew from its nest hole on a circular, downward flight that carried it 6 m to a stub, where it rested silently. The young one had been looking about with its head out of the hole for the previous hour. Neither at this nor any other nest have I seen parent sapsuckers make special efforts to induce young to leave. OTHER ASPECTS Lone parents. — I have encountered 5 nests where young were tended by widowed parents, of which 3 were females and 2 males. Behavior differed between the sexes. Whereas the females fed their young at exaggerated rates Kilham • SAPSUCKEK NESTING BEHAVIOR 319 of up to 20 to 28 times per hour, bringing little prey with each visit, the males fed at a slow rate, closer to what they would have used had their partners re- mained alive. Thus lone Male J (Table 4) fed its single young at a rate of 6.2/h in the last 9 days of the nestling period. Male H, who also had a single young but had a mate, fed at a rate of 7.7/h. Of the 5 nests with lone parents, 4 failed, 1 due to predation and the others, I believed, to starvation, for the vocalizations of the young became very feeble. Male J raised one young successfully and Lawrence (1967:117) describes a lone male that raised and fledged 2 young. It would seem from these experiences that males are more apt to succeed in raising young alone than are females. As described elsewhere (Kilham, in press) one of the lone females I watched succeeded in attracting a new male that, after 2 days, started to feed her young. The nest, however, was destroyed by predation on the following night. Predation by weasel ( ? ) . — The aspen of Pair I was 1 m from a stone wall. I noticed a weasel {Mustela sp.) running along the wall on 11 June when Female I made ‘‘^quare’’^ notes in alarm. The weasel stood up to look at me at close range. On 30 June I found remnants of a nestling, with wing feathers still in sheaths, below the hole and similar remains at the bottom of the nest. The predator had seemingly been able to enter. There was no ring of tooth marks around the entrance, such as I have noticed when raccoons attack a sapsucker nest (Kilham 1971) . On examining a rough place on the bark at the foot of the aspen, I collected over 30 whitish hairs a centimeter or slightly more in length; hairs that might have come from the belly of a weasel. Although a snake might have entered, there are no tree climbing snakes in woodlands of central New Hampshire, to my knowledge, and a snake would have swallowed its prey whole. A presumption, therefore, was that a weasel was the predator. Johnson (1947) described attacks of a weasel on a sapsucker nest. Temperament. — How close should one sit when in the open and without a blind? When a sapsucker is disturbed by one’s being too close, it makes repeated “tcaan.” notes, raises its crest, alights on the opposite side of the nest tree from the observer, and may bow into the hole repeatedly before entering. These signs of shyness are generally present at the start of a breeding season, but largely disappear as nesting progresses. The members of a pair may then appear remarkably tame. When I set a step ladder by Nest H, the parents fed their nestlings without hesitation when I was less than 3 m away. An occasional individual is more shy. Female I was unusually nervous, but this was mainly on her first visit to the nest after I had arrived. By a second or third visit she entered the nest with little hesitation. I never felt that my sitting close (at 7-8 m) ever kept her from an intended visit 320 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 to eggs or young. The curious thing was that at Nest H where the members of the pair were both tame, MH made 59% and his mate 41% of the feeding visits. At Nest I, at the same height (4 m) above the ground, MI made 105 and FI, the shy female, 104 of the feeding visits made as I watched close by; these figures supporting an impression that FI was not unduly disturbed. The reason for sitting close to the nests was that I wanted to see, as clearly as possible, the types of prey parents were feeding nestlings. Insects fed to nestlings. — Sapsuckers are versatile insect catchers, moving rapidly up trunks and limbs to glean from bark; flying against clusters of leaves and, on warmer days, catching prey in midair. They occasionally go to the ground, possibly more on wet days. They thus appear to catch insects of a wide variety in New Hampshire and I have never noted that they pick up ants any more than casually. Only a very small fraction of their prey, furthermore, is caught near their sap holes (Kilham 1964). The size of prey may vary considerably. On 7 June 1968 I watched a male fly to the ground to catch a luna-sized moth which it carried to an “anvil” of rough bark. He pulled off the wings, then fed the body to a nestling that was 4 days from fledging. The male stayed by to poke in at the nestling to assist it in managing its large meal. On 17 July of another year I watched a male struggle with a willow sawfly (Cimbex) 3.4 cm in length. After battling for some minutes, the sapsucker gave up and I recovered the crippled but unsubdued insect from the ground. Although sapsuckers pick up very small prey such as ants, it has seemed, from observations made close to nests, that most insects fed to young are large and more soft-bodied. Thus, to cite Pair C as an example, I noted between days 5 and 18 after hatching, that the male in 32% of 59 visits and the female in 34% of 73 visits, had legs and gauzy wings of insects projecting a centimeter or more from their bills. On a number of occasions I have watched foraging sapsuckers pause to compact their load of insects, bringing into their bills all projecting appendages; then after more foraging, fly to feed their young with no sign of larger prey being visible. This has made it seem that what one actually sees in bills at times parents arrive at the nest may be only a crude index of how much larger prey they are actually catching. Beetles, further- more, of a centimeter or more in size, are usually so beaten and dismembered as to be difficult to recognize. In June in New Hampshire I have found large crane flies (Tipula sp.) to be common in woods where I have watched sapsuckers. These may have accounted for many of the legs and wings projecting from parents’ bills. On one morning after a rain, FH flew to the ground and picked up a smooth larva, 2.5 cm long, possibly of a crane fly. In two successive years the male at one nest aspen carried feces to a place Kilham • SAPSUCKEK NESTFNC; HEHAVIOR 321 3 m above ground on a low tree, discarding his load against the hark. I cleared the ground below and was thus able to make daily collections. A sample of these, as kindly examined by Dr. G. Thomas Fisher, Dept, of Entomology, University of New Hampshire, contained for the most part remnants of major and minor workers of carpenter ants iCamponotus pennsylvariicus ) . These findings coupled with those of Beal 1 1911 ) have made me feel that the remains of ants pass through digestive tracts particularly well. As guides to what sapsuckers actually feed nestlings, however, they may he deceptive. Other views and experiences on the subject of sapsucker foraging are given by Lawrence (1967j and Tate (1973 ). \^Tiile sapsuckers bring sap as well as insects to nestlings, I have never seen them bringing fruit. This may he because trees fruiting in June and early July were scarce in woods where 1 did my watching. Sapsuckers are not unique in bringing sap to feed young, for Thonen ( 1966 ) has noted the same habit in the Three-toed Woodpecker i Picoides tridactylus) of Europe. Black polyniorph females. — In the course of studying sapsuckers over 25 years and finding 69 nests, 1 have encountered 12 females that were “black polymorphs” having black or nearly black crowns. Attempts to find consistent differences in their breeding behavior have been unsuccessful. DISCUSSION Males were the more domestic-minded of pairs of sapsuckers studied in New Hampshire. While females sometimes equaled them in attentiveness to incubat- ing, brooding, and, for periods, feeding nestlings, males generally performed the larger share of these activities as well as doing most of the excavating and nest cleaning. The females might be regarded as a reserve, exerting them- selves to the full when a nest contains a full brood of 4 or more. In a nest with only 1 nestling, on the other hand, as was the case with Nest H, the female left almost all of the care to the male. This greater role of males may explain differences of behavior in lone males as compared to lone or widowed females. Of 3 lone males, to combine one described by Lawrence (1967) with 2 of mine, 2 succeeded in raising their young, feeding them in normal fashion. Of 3 females that I observed, all fed their young in an inefficient, exaggerated fashion. The nestlings of 2 died, seemingly of starvation. The third lone female succeeded in attracting a new mate who started to feed the nestlings 2 days later ( Kilham, in press ) . A point of comparison was that while one of my lone males and one of Lawrence’s attracted new females, these new females, although they came to the territory, showed no interest in the nests. This is perhaps what one might predict, fe- males being on the whole the less domestic-minded. The successful remating, in terms of care of the young, involved a new male. 322 THE WILSON BULLETIN • Vol. 89, No. 2, June 1977 Lawrence’s account (1967) of the nesting of sapsuckers in Ontario differs from mine in a number of respects. One is where (pp. 95-96) she states that of 4 woodpeckers (of which one was the sapsucker) that she studied, none left “their eggs uncovered more than a minute or two at a time or, to be exact, a total of 27 min in 90 hours of observation.” Were my observations of much more time (Table 2) unusual? It would not seem so from what other observers have recorded for other picines. Thus, Skutch (1969:469), during 5 hours of observing Red-crowned Woodpeckers (Melanerpes ruhri- capillus) , noted that they left their eggs uncovered 86 min or 22% of the time. The restlessness that he describes for this and for the Golden-naped Woodpeckers in the incubation period is almost exactly what I have noted for 5. varius. Although he only gives times of actually incubating in his table fp. 486) his figures for 4 pairs of M. chrysauchen show that they left eggs uncovered for 0, 10, 11, and 24% of the time respectively. Skutch’s experiences with Acorn Woodpeckers (M. formicivorus) were similar for in one watch of 11.5 hours, he found eggs were left uncovered for 141 min. One can say that all of these species were melanerpine and observed in the subtropics. It is of interest, therefore, that Pynonnen (1939:114), in all-day watching at 2 nests of Greater Spotted Woodpeckers (Picoides major) in Finland, found eggs left uncovered for periods totaling 5-6 h for one pair and 4 h for the other. dTie most restless woodpeckers I have observed were a pair of Hairy Wood- peckers iP. villosus) in Lyme. Both the male and female left eggs uncovered up to 30 and 40% of tours on duty on the nest and this was throughout the incubation period, as judged by periods of watching limited to 40-90 min at a time that I made on scattered days, at a distance that was far enough away not to frighten them. To cite Skutch’s account (1969) again there are parallels in nest sanitation between S. varius and at least 3 species of Melanerpes woodpeckers. He states that Golden-napes “often allow nestling’s droppings to accumulate in the hole and then remove them in a spell of concentrated house cleaning.” Golden-napes use their nest cavity for roosting after fledging and keep it clean the whole time. On the other hand Golden-fronted iM. aurifrons) and Red- crowned woodpeckers, that do not use theirs after fledging, “appear not to clean the nest at all after the young birds can take their meals through the doorway.” Sapsuckers appear to be ambivalent in this regard, some pairs ceasing well before and others continuing to carry out feces until the time of nest leaving. The functions of the loud, persistent vocalizations of nestlings, that could attract predators, is a subject that needs further study. As pointed out by Skutch ( 1976) they are possibly more important in hole-nesters that cannot he guided by the speed, strength, and color of gaping reactions when feeding Kilham • SAPSUCKER NESTING BEHAVIOR 823 young in a darknened nest cavity. The best experiments are seemingly those of von Haartman (1953). He was able to show, by dividing a nest box of the Pied Flycatcher {Muscicapa hypoleuca) , that the rate at which parents fed their young was not guided by their number but by the clamor raised by the hungriest of them. If birds of temperate zones raise the largest broods possible, then begging cries would seem a device that might insure efficient and constant feeding. Quite opposite to these views, however, are those of Lawrence (1967:125) who states that she “found no evidence of the wood- pecker nestlings “chatter notes” having any direct stimulating effect upon parents’ feeding rates.” Differences of opinion are, or should be, a stimulus to further studies. If the natural history of sapsuckers or other woodpeckers is to be known with any completeness, it would seem that far more studies are needed by different observers studying them and their nesting habits in different parts of their total range. SUMMARY Activities covered in this report extend from excavation through time of fledging of Yellow-bellied Sapsuckers. Males did nearly all of the excavating. When a first excavation failed, one female excavated harder and longer than her mate in starting a new one. Two females were dominant at the nest hole at time of egg-laying. Males started incubating more promptly than some females, but after 4 to 5 days delay, females of 2 nests incubated more than their mates in daylight hours. Eggs were left uncovered 16% of the time. The tasks of brooding and feeding the young were shared by both sexes almost equally. When free of brooding, which lasted for 8-10 days, the combined feeding rate of the sexes doubled almost immediately in 2 pairs. Males did the most of the nest cleaning. Comparisons are made of feeding rates of parents caring for 1, 2, 3, and 4 young and of the behavior of 2 lone or widowed males as compared to 3 lone females. Both situations brought out the greater role of males. The loud vocalizations of young sapsuckers are considered as having selective value in stimulating parents to a high rate of foraging and feeding of nestlings. LITERATURE CITED Beal, F. E. L. 1911. Food of the woodpeckers of the United States. U.S. Dept. Agr. Biol. Surv. Bull. 37. Bent, A. C. 1939. Life histories of North American woodpeckers. U.S. Natl. Mus. Bull. 174. Haartman, L. \on. 1953. Was reizt den Trauerfliegenschnapper (Muscicapa hypo- leuca) zu futtern? Vogelwarte 16:157-164. Howell, T. R. 1952. Natural history and differentiation in the Yellow-bellied Sapsucker. Condor 54:237-282. Johnson, R. A. 1947. Role of male Yellow-bellied Sapsucker in the care of the young. Auk 64:621-623. Kilham, L. 1959. Early reproductive behavior of flickers. Wilson Bull. 71:323-336. . 1962a. Breeding behavior of Yellow-bellied Sapsuckers. Auk 79:31-43. 324 THE WILSON BULLETIN • Vol. 89, No. 2, June vtgc . 19621). Nest sanitation of Yellow-bellied Sapsucker, Wilson Bull. 74:96-97. . 1964. The relations of breeding Yellow-bellied Sapsuckers to wounded birches and other trees. Auk 81:520-527. . 1971. Reproductive behavior of Yellow-bellied Sapsuckers. I. Preference for nesting in Eowes-infected aspens and nest hole interrelations with flying squirrels, raccoons and other animals. Wilson Bull. 83:159-171. . Altruism in nesting Yellow-bellied Sapsucker. Auk (in press). Lawrence, L. de K. 1967. A comparative life-histor\- study of four species of wood- peckers. Ornithol. Monogr. No. 5. Pynnonen, a. 1939. Beitrage zur kenntnis der biologie finnischer spechte. Ann. Soc. Zool-Bot. Fennica Vanamo 7:1-166. Ricklefs, R. E. 1969. An analysis of nesting mortality in birds. Smithson. Contrib. Zool. No. 9:1-48. Royama, T. 1966. Factors governing feeding rate, food requirement and brood size of nestling Great Tits Parus major. Ibis 108:313-347. Skutch, a. F. 1969. Life histories of Central American Birds III. Pacific Coast Avi- fauna No. 35. . 1976. Parent birds and their young. IJniv. Texas Press. Austin. Stickel. 1). W. 1965. Territorial and breeding habits of Red-bellied Woodpeckers. Am. Midi. Nat. 74:110-118. Tate. J. Jr. 1973. Methods and annual sequence of foraging by the sapsucker. Auk 90:840-856. Tii(")Nen, W. H. 1966. Dreizebenspecht futtert nestjunge mit baunsaft. Ornithol- beobachter 63:21-23. DEI’T. OF MICUOBIOLOGY, DAKTMOUTIl MEDICAL SCHOOL, HANOVER, NH 03755. ACCEPTED 1 MAR. 1976. GENERAL NOTES Sex differences in alarm responses of wintering; Evening Grosbeaks. — Flocked birds probably attract the attention of predators to a greater degree than do solitary indi- viduals. Conspicuous colors, movements, or vocalizations tend to promote flock cohesion while increasing predation risks; thus flocked birds must effect a compromise between the need to communicate among themselves and the need to evade predators (Wiley, Auk 88:881-892, 1971; and referenees cited therein) . A speeies illustrating such a compromise is the Evening Grosbeak ( Hesperiphorm ves- pertina) , which occurs in sizable flocks during the winter in several parts of the U.S. The flocks are conspicuous, both visually (due particularly to the boldly patterned plumage of males) and auditorily (due to the grosbeaks’ frequent use of loud calls). During a study of winter social behavior in this species, I observed the alarm responses of males and of females in grosbeak flocks. This paper describes some sexual differences in alarm responses which might function to reduce predation in this species. Methods. — Wintering Evening Grosbeaks were observed from February to May 1976 at Logan, Utah on a 200-m semi-urban wooded stretch bordering the Logan River. Three hundred grosbeaks were banded of an estimated total of at least 700-800 individuals that frequented the study area (Ralph and Ralph, Rird-Randing 47:340-344, 1976). Alarm responses within feeding groups of grosbeaks were documented in April and early May at a 0.86 X 0.86 m elevated platform provisioned with sunflower seeds and situated 1.5 m from the nearest cover. The birds were watched through 1-way glass from a blind positioned 0.5 m from the platform. Counts were made at 15- to 30-sec intervals of the number of males and females present. If, during a given period between counts, the feeding group showed a generalized alarm response in whieh some individuals adopted a “freezing” stanee at the platform and others escaped, a count immediately was made of the number of males and females re- maining at the platform. One hundred sueh instances were recorded. Additional observations were made of the incidence of distress calls or “squeals” among 189 individuals during trapping and banding operations. Squealing was recorded as pres- ent if a bird produeed the call one or more times when handled. Alarm responses within feeding groups. — Evening Grosbeaks feeding in groups at the platform sporadically exhibited fright responses. Such responses appeared sometimes to be caused by the flight overhead of a hawk or shrike or by human activity, although in most instances I was unable to discern an obvious stimulus for the behavior. Typically, the birds suddenly became alert and ceased to eat; then, a fraction of a second later, some or all members of the group flew away, often to nearby trees. Sometimes a bird at the platform gave the keer call, which in most contexts functioned as a contact vocal- ization, just prior to the initial moment of alertness. Escaping birds either were silent or, more often, gave the bree call ordinarily associated with agonistic interactions. Occasion- ally, an escaping bird gave a chirping sound. Individuals that remained at the platform during a fright adopted a crouched posture and remained motionless and silent for several seeonds or longer. Rirds freezing in this manner appeared extremely alert and tended to fly away at the slightest fear-producing stimulus. If they did not become further alarmed, they eventually resumed feeding at the platform. The mean number of males counted at the platform just prior to a fright was 12.8 (n = 100, SDr=6.16), and of females, 17.7 (n = 100, SDi=6.76). The pre-fright sex ratio thus was 1.38:1.00 in favor of females. This ratio did not differ significantly from 325 326 THE WILSON BULLETIN • Vol 89, No. 2 Table 1 Occurrence of Distress Calls Among Male and Female Evening Grosbeaks at Successive Captures Capture Previous Behavior Males Captured (n) Calling Females Captured (n) Calling 1 — 87 67 102 93 2 Called at 1 7 3 11 10 Did not call at 1 2 0 1 0 3 Called at 1 and 2 — — 1 1 Called at 1 but not 2 2 0 — — the sex ratio of 254 grosbeaks trapped in the same part of the winter (1,31 females to LOO males! (x“ 0.18, df = 1, P > 0.5) . The mean number of males observed to freeze at the platform during a fright was 1.6 (n = 100, SI) = 1.63), whereas that of females was 6.8 (n iz: 100, SD 4.36), producing a post-fright sex ratio of 4.30:1.00 in favor of females. In 92 of 100 cases, a greater proportion of females than males exhibited the freezing response; this result differs sig- nificantly from chance expectation ( x" = 70,56, df =: 1, P< 0.001). If one compares male to female numbers at the platform before and after the onset of frights, a highly significant difference is again obtained (x"= 149.34, df 1, P< 0.001), Thus, fe- males showed a greater tendency than males to freeze at the platform when alarmed, and males exhibited a greater tendency than females to flee. This difference of behavior appeared also to hold for groslieaks that became alarmed while resting or feeding on buds in leafless trees, although I did not make (juantitative observations of responses in this situation. Distress behavior of restrained birds. — Evening Grosbeaks that I trapped, handed, and measured often produced squeals when handled. Of 189 individuals for which I re- corded the presence or absence of this vocalization at initial capture, 77% of males s(|uealed, whereas 91% of females gave the call (Table 1). A significantly greater pro- portion of females than males squealed when handled for the first time ( x" = 7.25, df = 1, P<0.01). In general agreement with this finding is J. Ogden’s (quoted in Nor- ris and Stamm. Bird-Banding 36:83-88, 1965) observation that, of 10 Evening Grosbeaks captured in mist nets, 5 females gave distress calls during removal from the nets, whereas 5 males gave no calls. Parks’ (Bird-Banding 16:32-36, 1945; Bird-Banding 18:57-76, 1947) descriptions also suggest that females of this species squealed more often than males when handled. Twenty-one of the 189 individuals were recaptured one or more times 1-10 days after they were handed. Such birds either repeated the initial response when handled and measured on later occasions or, alternatively, called on the first hut not on subsequent occasions (Table 1). Although my sample sizes are too small to permit a meaningful statistical analysis, the fact that several birds squealed at initial capture but not on later occasions, whereas none showed the reverse pattern, suggests that habituation to han- dling was a factor affecting response frecjuencies. In all but one case, individuals that called at initial capture hut not thereafter were males. It thus seems possible that Eve- ning Grosbeaks differ sexually not only in the propensity to give distress calls at first June 1977 • GENERAL NOTES 327 I capture, but also in the tendency to cease responding when handled for a second or tiiird I time. Squealing by hand-held grosbeaks often was accompanied by a visual display in which , the bill was opened and the tail spread. Some birds also raised the crest; and occa- ! sional individuals spread the wing, maintaining it for several seconds in a horizontally extended position, when I released my hold on one wing. The latter behavior appeared i to be more common among vocal females than among vocal males. Discussion. — Cryptically colored animals may avoid detection by predators by remain- ing motionless against an appropriate background (reviewed by Alcock, Animal Behavior: an Evolutionary Approach, Sinauer Associates, Inc., Sunderland, Mass., 1975). It seems j possible that sex-associated differences in the alarm responses of wintering Evening Grosbeaks are adaptively related to the species’ sexual dimorphism in plumage coloration. To me, the plumage of females appears much less conspicuous than that of males against a variety of backgrounds in winter and spring. The sexual difference in visual con- spicuousness seems more marked when the birds are stationary than when they are in flight. If the difference is similarly evident to animals that prey upon this species, then female Evening Grosbeaks should benefit more than males by freezing in relatively open i situations. Conversely, males should increase their chances for survival by fleeing im- I mediately to cover when alarmed. Males, by fleeing, might secondarily reduce the prob- 1 ability of a predator’s detecting exposed, motionless females. One might expect to find , similar differences of behavior in males and females of some other bird species showing : comparable dimorphism in plumage coloration. j The sexual differences I observed in the distress responses of restrained Evening Gros- i, beaks seem problematical. One could speculate that these differences bear some relation- ship to differential alarm responses in free-ranging flocks. Since females apparently rely i more extensively than males upon concealment in open situations, they should be more vulnerable than males in the event of detection by a predator. A second line of defense (i.e. squealing) thus might be more important to females than to males. However, an ; examination of Norris and Stamm’s (1965) findings on interspecific differences in the occurrence of distress calls among restrained birds reveals no obvious relationship be- ! tween cryptic coloration and the tendency to squeal when handled. I In many species, squealing by a captured individual stimulates birds of the same and j sometimes of other species to mob the predator, potentially providing the victim with an I opportunity to escape (reviewed by Thorpe, Cambridge Monogr. Exp. Biol. 12:1-143, 1961). Free-ranging Evening Grosbeaks did not mob me when I removed squealing con- specifics from traps, although the possibility remains that the birds might mob a predator 1 under different circumstances. In any event, it seems possible that squealing — with its I concomitant visual display, which contains several postural elements used intraspecifically in threat — might function in this species to startle or repel a predator and thereby to 1 increase a victim’s chances for escape. I Summary. — Observations were made of alarm responses among Evening Grosbeaks wintering at Logan, Utah. When a feeding flock became alarmed, females exhibited a significantly greater tendency than males to freeze rather than to fly to cover. Females also showed a significantly greater tendency than males to give distress squeals when I handled. It is hypothesized that sexual differences of response to danger stimuli may be 1 adaptively related to sexual dimorphism in plumage coloration in this species.- Martha ! Hatch Balph, Dept, of Wildlife Science, Utah State Univ., Logan 84322. Accepted 7 Dec. I 1976. 328 THE WILSON BULLETIN • Vol 89, No. 2 Nesting of Turkey and Black vultures in Panama. — On 23 March 1976 I found a Turkey Vulture {Cathartes aura) nest with 1 young, and a Black Vulture (Coragyps atratus) nest -with 2 young on a 0.5 ha island in Gatun Lake, Panama Canal Zone. The island is 525 m from the nearest mainland and 275 m offshore Barro Colorado Island, the Smithsonian Tropical Research Institute's biological reserve. The vegetation is young secondarv' forest about 20 m high with a relatively open canopy and a predominance of palms. The undergrowth is not dense and consists of tree seedlings, vines, and grasses. One-third of the island has been cleared for an experiment and there has been almost daily human activity there since Januajy 1976. Other than vultures, the largest animals regularly encountered on the island are basilisk lizards i Basiliscus basiliscus) and marine toads (Bufo marinus) . One armadillo was seen on the island for a short time. The scarcity of animals which might prey on the eggs or nestlings may make the island a particularly favorable nesting site. The Turkey Vulture nest encompassed an area of about 1 nr under a tangle of grass and brush at one end of the island. Leaves and debris under this cave-like shelter had not been arranged in any way. It appeared that the 24-day-old nestling moved about in the entire area. The nestling was removed from the nest 8 April 1976 and hand-reared. From photo- graphs and descriptions of the development of a Turkey Vulture (Bent. U.S. Natl. Mus. Bull. 167:18, 1937) I estimated its age on 8 April to he about 24 days. Egg-laying, there- fore, was the first week of February for incubation is “almost 40 days" (Coles. Studies in the Life History of the Turkey Vulture, Ph.l). thesis, Cornell Univ., Ithaca. N.Y., 1938) and hatching occurred the week of 14 March. The bird fledged when 70 days old. on 21 -May. which age corresponds to that provided by Coles ( 1938) of 70 to 80 days. The nest site was periodically examined for 4 months after the young had been removed and there was no evidence of a second nesting attempt. A Black Vulture nest was 35 m from the Turkey Vulture nest. The nest was located at the base of a clump of palms in a small tunnel-like shelter formed by dead palm leaves among the trunks. No nest material appeared to have been brought in or arranged by the adults. Both the Turkey and Black vulture nesting sites were similar in being somewhat concealed from view, providing some shelter from rain, and encompassing about the same area. The Black Vulture young had apparently left the nest at least 17 days before fledging: no young had been seen after 23 March until 7 April when one flightless young observed near the site was easily captured. By 9 April, however, 2 young and 2 adults were seen on the shore of the island. I believe that the yt>ung were fully capable of flight by this time, and that the 4 were the 2 adults and young from the nest on the island, although none had been banded. Subseciuent observations reinforced that assumption. Four Black \ ultures were (M)inmonly observed on the island for 15 weeks after the young had fledged; a tall tree at the edge of the cleared area was a favorite roost. Tlie 4 Black Vultures formed a cohesive group, flying in together to perch in the tree, and always flying off together. The composition of the group was always 2 young and 2 adults; it was un- usual to see vultures other than these 4 roosting on the island. Incubation for Black Vultures bas been reported as 38 days, and the young fledge in 80 days from hatching (Stewart, Auk 91:595-600, 1974). Thus, the Black Vultures which had fledged by 9 April probably hatched the first week of Februarv ; egg-laving then would have occurred the last week of December 1975. For at least 5 weeks after fledging the 2 Black Vulture young would beg for and re- ceive food from the adults. The young would vigorously flap their wings, bob their heads. June 1977 • (;ENERAL NOTES 329 and emit a low-pitched hissing sound. On 26 .July, more than 15 weeks after fledging, the young were still begging food but were not fed on the one occasion I observed them. Long post-fledging dependence is also indicated for the Turkey Vulture by the hand-reared young’s recognition of and behavior towards me. For 7 weeks after fledging the young Turkey Vulture would, upon my appearance, assume a posture with wings slightly spread and arched downwards, the head lowered, and the tail inclined upwards with the feathers separated. There was much head-bobbing and wing-flapping when food was presented, but no vocalizations. The Turkey Vulture did not behave in this manner if approached or fed by other individuals. The nesting of Turkey and Black vultures I observed corresponds to the dr>- season in Panama which is from January’ to April. Wetmore’s (Smithsonian Misc. Coll., 150 d ) : 160-161, 1965) incidental reports of Black Vulture nesting activity, nestlings, fledglings, and an egg collection also indicate that nesting occurs during the dry season in Panama for this species. His only information on Cathartes sp. nesting in Panama is for the Yellow- headed Vulture, C. burrovianus. for which he reported 2 young “only recently able to fly” on 14 May 1953. The Turkey Vulture from the nest on the island fledged 21 May. Vulture nesting activity correspondence with the dry season may be due to several fac- tors. Carrion resources for vultures may he greater during the dry season. However, for frugivorous animal populations at least, highest mortality might he expected at the end of the wet season, during November, December, and January, when fruit abundance is lowest (Smythe, Am. Nat. 104:25-35, 1970). Alternatively, the dry season weather may he more favorable for raising young. The dry season months are windier and drier than the rest of the year and foraging time is likely maximal then. The wet season’s rains and calmer weather may make soaring difficult and thus make hunting less energetically ef- ficient. Also, the higher relative humidity and rainfall then might be detrimental to the health and development of the terrestrial young. I thank Eugene S. Morton for commenting on the manuscript. — Lauhie A. McHargue, Smithsonian Tropical Research Institute, Box 2072, Balboa, Canal Zone. Accepted 11 Nov. 1976. Fulvous histling Duck populations in Texas and I^ouisiana.-— Nund)ers of Fulvous Whistling Ducks { Dendrocygna bicolor) declined rapidly in the 1960’s apparently from exposure to pesticides applied to rice (Flickinger and King. J. Wildl. Manage. 36: 706-727, 1972). The species was listed as endangered by the Texas Organization for En- dangered Species since 1972. However, numbers have increased since 1970 when many rice growers began to voluntarily discontinue the use of aldrin-treated rice seed in Texas and to substitute drill planting for aerial seeding of treated seed in Louisiana. Aldrin treatment of rice seed was suspended by the U. S. Environmental Protection Agency in 1974. Ground or aerial surveys of Fulvous Whistling Ducks were made at all their traditional concentration areas in 14 southeast Texas counties or 4 Parishes of southwest Louisiana along the Gulf Coast. Counts and estimates were alternated between Texas and Louisiana between 1968 and 1974 when the first late summer estimates were made for both states. Spring and later summer censusing was done in Texas for the first time in 1975. Most spring counts were made from the ground from mid-April through May with the excep- tion of the count in Texas in 1975 (made 15 to 30 April). Late summer birds in both states were estimated from aircraft during mid-September except in Texas in 1975 when a ground count was taken. In the spring of 1968 only 1123 Fulvous Whistling Ducks were counted in both states 330 THE WILSON BULLETIN • Vol. 89. No. 2 Fulvous Whistling Table Ducks Observed in 1 Texas and Louisiana, 196(3-1975 Year Texas Louisiana Spring® 1966 b 192 1967 519 672 1968 379*= 744 (1,000) 1969 978 — 1970 — 1,441 (4,000) 1971 — 1,555 (4,000) 1973 ( 2,000) — 1975 1,650" Late Summer 1973 — (6,000- 8,000) 1974 ( 15,000) " (8,000-10,000) 1975 6,700 (8,000-12,000) “ Counts and estimates of total numbers in Louisiana from J. Lynch, U, S. Fish and Wildlife Service (pers. comm.); ( — ) = No observations made; 'Ground and aerial counts made in the rice belt; Estimate for the Texas Rice Belt (from C. Stutzenbaker) in Bellrose (The Ducks, Geese and Swans of North America, Stackpole Books, Harrisburg, PA, 1976:75; ® Estimate of ChambcTs and Jefferson counties from J. Dunks, Texas Parks and Wildlife Department (pers. comm. ). (Tal)l(^ I) when tlie use of aldrin-treated rice seed was at a peak. Following a decline in aldrin use, numbers increased in the spring in Louisiana (1441 in 1970) and in Texas (about 2000 in 1973). There were also fewer Fulvous Whistling Ducks found dead in rice fields in 1970 and 1971 (Flickinger and King, J, Wild. Manage. 36:706-727, 1972) than in earlier years. During the s])ring of 1975 largest eoneentrations in Texas oeeurred in Chambers and Wharton counties with the peak number of 650 birds in Wharton County on 26 April. The Texas counties of Chambers and .lefferson had concentrations in both spring and late summer. In late summer there was an eastward movemen't of Fulvous Whistling Ducks into the I’ort Arthur area from the western part of the rice belt (Flickinger et ah, J. Wildl. Manage. 37:171-175, 1973), This may explain the increase in numbers in Cham- bers and .lefferson counties and their absence in w^estern rice belt counties at this time of year. In late summer of 1974, 15,000 birds were estimated from aircraft during a general waterfowl survey of Chambers and Jefferson counties i}. Dunks, pers. comm.). In 1975, 6700 birds were recorded by ground counts made in late summer. The highest count of 5700 birds was made in Chambers County on 14 September. However, most of the rice eountrv' in Chambers and Jefferson counties was not surveyed by aircraft in the late summer of 1975. Therefore, the decrease in late summer numbers in Texas from 1974 to 1975 is probably more apparent than real and does not indicate a sudden decline in the population. From 8000 to 10,000 Fulvous Whistling Ducks were estimated for southern Louisiana in the late summer of 1974 and from 8000 to 12,000 in 1975. The largest number observed (5000-7000 birds) was in Acadia Parish on 15 September. A reasonable estimate would be that about 17,000 Fulvous Whistling Ducks were in southeast Texas and southwest Louisiana in the late summer of 1975 (7000 in Texas June 1977 • GENERAL NOTES 331 and 10,000 in Louisiana). This should provide a hase-line index for future late suinnu'r censuses. — Edward L. Flickinger, U.S. Fish and Wildlife Service, Victoria, TX 77901 ; David S. Lobpries, Texas Parks and Wildlife Dept., Port Arthur, TX 77640; Hugh A. Bateman, Louisiana Wild Life and Fisheries Commission, Baton Rouge, LA 70804. Ac- cepted 14 Jan. 1977. Slipper shells, a major food item for White-winged Scoters. — In the winters of 1973 and 1974, large feeding flocks of White-winged Scoters {Melanitta deglandi) moved into the New Bedford Harbor region of Buzzards Bay, Plymouth Co., Massachu- setts. I collected 28 White-wings and examined their upper digestive tracts. Six indi- viduals did not have sufficient food material in the gullet for analysis. The contents of the 22 individuals analyzed differed from those in the literature and from my own previous observations. Slipper shells iCrepidula fornicata) comprised 88% of the hulk organic matter by volume. The remainder of the stomach contents was mostly oyster spat and soft shelled clams. These did not exceed 25% of the bulk in any individual. Cottam (U.S.D.A. Tech. Bull. 643, 1939) reports from an examination of 819 adult White-wings, that % of their food was mollusks, of which bivalves comprised 63% and less than 2% were slipper shells. Scott and Olson ( Ecol. 54:996-1007, 1973) found in New Hampshire that 89% of the total volume of food of White-winged Scoters was bivalves and Siliqua costata was the dominant food; they recorded no slipper shells. Trawl and dredge samples from the feeding area revealed a high accumulation of shell. These shell deposits are from shucking operations of local sea and bay scallop industries. Three species of Crepidula were attached to the shell deposits; C. fornicata was the domi- nant species. Hoff fSci. Teach. 38:1, 1971) reported a lu'avy organic load in the sur- rounding waters, the primary source of which is untreated sewage from a nearby munici- pal sewer outfall. It is apparent that the combination of bigb concentration of organic nutrients and shell substrate have provided an ideal habitat for slipper sbells. These in turn have provided a different food budget for Wbite-winged Scoters in southeastern Massachusetts. — James G. Hoff, Southeastern Massachusetts JJniv., Dartmouth 02747. Accepted 16 Jan. 1976. Egg movement by a female Gadw^all between nest bowls. — Gadwalls (Anas strepera) nest commonly on the Woodworth study area located 4.8 km east of Woodworth, North Dakota on the Missouri coteau IKirsch and Higgins, Wildl. Soe. Bull. 4:16-20). This 1231 ha area is a research station of the Northern Prairie Wildlife Research Center. On 13 June 1975 a Gadwall nest containing 10 eggs was found at the station head- quarters inside an open-topped enclosure measuring 6 m by 6 m and fenced with 5 cm by 5 cm chain link wire mesh. Vegetation at the nest site consisted of smooth bromegrass (Bromus inermis) and absinth (Artemisia absinthium) . This nest was in a corner of the enclosure with the rim of the nest touching the fence. The clutch had been incubated approximately 2 days. We revisited the nest on 5 July and found that 8 of the 10 eggs had been moved into a new nest bowl on the other side of the fence adjacent to the original one. The other 2 eggs were in the original nest bowl and were cold. Incubation of the 8 eggs in the new nest was about 22 days. We moved the other 2 eggs into the new nest with the remainder of the clutch. Vegetation at this site was comparable to that at the first nest site. On 7 July all 10 eggs had been moved back through the fence into the original nest and some of the eggs had hatched. Eight ducklings had hatched and left the original nest by 8 July and 2 dead embryos in partially pipped eggs remained in the nest. 332 THE WILSON BULLETIN • Vol. 89, No. 2 A possible explanation of how this egg movement was accomplished has been dis- cussed by Oring (Auk 81:88-89, 1964) who reported that some Pintails (A. acuta) and Mallards (A. platyrhynchos) , but none of 15 trapped Gadwalls, moved their eggs into new nests after a trap was placed over the original nest. The ducks used the ventral sur- face of their bills to pull the eggs through the trap. This Gadwall may have moved her eggs in the same manner when she returned to her nest from opposite sides of the fence. The authors thank Kenneth F. Higgins and Harold F. Duebbert for reviewing the manuscript. — Robert F. Johnson Jr., Dept, of Forestry, Michigan Technological Univ., Houghton 49931 and Leo M. Kirsch, Northern Prairie Wildlife Research Center, U.S. Fish and W ildlife Service, Jamestown, ND 58401. Accepted 3 Feb. 1976. Foods of western Clapper Rails. — The Colorado River population of the Clapper Rail (Rallus longirostris yumanensis) is presently listed as endangered (U.S, Fish and Wildlife Service, United States List of Endangered Fauna, p, 11, 1974). Its presence in fresh-water habitat during the 5-month breeding season and probable migration to Mexican coastal salt water swamps (Tomlinson and Todd, Condor 75:177-183, 1973) are considered unusual for Clapper Rails in general. In June 1971, Tomlinson and R. L. Todd collected 35 Clapper Rail specimens in the southwestern United States and western Mexico to determine if a racial distinction oc- curred among the geographically isolated populations. Three separate races (/?. 1. yu- manensis, R. 1. rhizophorae, and R. 1. nayaritensis) were confirmed from the collection (Banks and Tomlinson, Wilson Bull. 86:325-335, 1974). Because the food habits of these rails were unknown, 32 stomachs ( proventriculus and gizzard) were preserved for later food habits analysis. This analysis provides the first insight into the freshwater food habits of R. 1. yumanensis. and should be useful in future preservation and manage- ment considerations. The birds were collected in fresh-water marshes along the Colorado River from Needles, California south to the Delta in Sonora, Mexico and in mangrove ( Avicennia germinans and Rhizophora mangle) swamps from Guaymas, Sonora, to San Bias, Nayarit, Mexico (Table 1), Airline distance from the northernmost to southernmost l)oints is approximately 1800 km. The specimens, including specific data relating to their collection have been deposited in the U.S. National Museum, Washington. D.C. Each Clapper Rail stomach was wrappt'd in cheesecloth in the field and preserved in a 10% formalin solution. Analysis was conducted at Arizona State University following procedures described by McAtee (Auk 20:449-464, 1912). The contents of each stomach were separated into ingesta types in a gridded petri disli and examined under a dissecting microscope. Each food type was visually estimated as a percentage of the total content in a particular stomach. The major foods of Yuma Clapper Rails were invertebrates; little vegetative material was ])resent (Table 2). Crayfish ( Procambarus and Orcopectes are the common genera) were the dominant food in 9 of the 10 stomachs from Topock Marsh on the lower Colorado River south to Imperial Reservoir in Arizona and California; the other stomach was empty. Of 2 specimens collected at the confluence of the Gila and Colorado rivers, one contained primarily an introduced fresh-water clam ( Corbicula sp.) (98%). and the other primarily isopods (97%). Colorado River Delta specimens in Mexico contained a greater variety of food organisms, but the major components w'ere water beetles and fish. Of the 16 R. 1. yumanensis stomachs, 9 had crayfish. 11 contained insect fragments, 4 had water beetles. 4 had fish, and 3 contained clams. In addition to the water beetles, other insect matter consisted of small amounts of weevils (3 stomachs), damselfly nymphs (2 stomachs), dragonfly nymphs, grasshoppers, and insect eggs. Spiders, leeches, prawns. June 1977 • GENERA I. NOTES 333 Table 1 Clapper Kail Collection Sites Number of Race Specimens Location Coordinates Yuma Clapper Rail {R. 1. yumanensis) 4 Topock Marsh and Gorge 34°44'N, 114°30'W 1 Bill Williams Delta 34°17'N, 114°04'W 2 Cibola Lake 33°14'N, 114°10'W 1 Below Cibola Lake in Colorado River 33°1LN, 114°10'W 1 Martinez Lake 32°58'N, 114°29'W 1 Imperial Reservoir 32°52'N, 114°28'W 2 Confluence of Gila and Colorado rivers 32°43'N, 114°33'W 4 Colorado Delta in Sonora, Mexico 32°02'N, 115°06'W 16 Sonora Clapper Rail (R. /. rhizophorae) 2 Laguna Del Soldado, Guaymas, Sonora 27°57'N, 110°59'W 2 Bahia Tobari, Sonora 27°05'N, 109° 56' W 2 Topolobampo, Sinaloa 25°38'N, 109° 03' W 6 San Bias Clapper Rail {R. 1. nayaritensis) 2 Altata, Sinaloa 24°38'N, 107°56'W 3 Mazatlan, Sinaloa 23°13'N, 106°22'W 5 San Bias, Nayarit 21°34'N, 105°17'W 10 Grand Total 32 and a small mammalian bone also were found. Vegetative matter consisted of twigs (10% in 1 stomach), 2 legume seeds (1 stomach), and 18 unidentified black seeds (3 stomachs) . All 16 of the R. 1. nayaritensis and R. 1. rhizophorae stomachs contained crabs (Table 3) . Of the 6 R. 1. rhizophorae samples, 5 contained 99% or more crab material and one contained 75% pulmonate snail plus 23% crab. The stomach contents of each of the 10 R. 1. nayaritensis specimens consisted of 89% or more of crab parts. Six morphologically distinct types of crab were differentiated by exoskeleton and chelae in the sample but further identification was not accomplished. Other identifiable items were insect eggs in 2 stomachs and insect fragments in 3. Vegetative matter was slight, consisting of 2 unidentified seeds, each found in a separate stomach. One stomach contained 2 white stones. Esophageal contents from 1 bird (possibly part of a regurgitated casting) con- sisted of 6 crab legs. 334 THE WILSON BULLETIN • Vol 89, No. 2 Table 2 Stomach Contents of R. L. YUMANENSIS FrOM THE LoWER COLORADO RiVER AND DeLTA^ General Locations of Collection Topock Marsh to Confluence of Gila Imperial Lake and Colorado Rivers Colorado Delta Total Birds Contain- ing Food Items % Composition % Composition % Composition FOOD ITEMS Average Range Average Range Average Range Crustacea Astacidae (Crayfish) 94.67 80-100 9 Palaemonidae (Shrimp) .25 0- 1 1 Isopoda 48.50 0-97 1 Insecta Hydrophilidae (Wa- ter beetles) Carabidae (Ground 56.50 1-80 4 beetle) 0.11 0- 1 1 ITnidentified Coleoptera 0.56 1- 5 1 Curculionidae (Weevils) Anisoptera ( Dragonfly 2.78 5- 10 3 nymphs) 0.50 0- 2 1 Zygoptera < Damselfly nymphs) Orthoptera 0.11 0- 1 2.00 0- 8 2 ( Crrasshoppers) 0.11 0- 1 1 Insect eggs 0.11 0- 1 1 Unidentified parts 0.78 1- 6 1.50 1- 2 4 Arachnida < Spider) 0.56 0- 5 1 Ilirudinea (Leech) 3.75 0-15 1 Mollusca {Corhicula) 0.06 0- 0.5 50.00 2-98 3 Vertehrata Ihiidentified fish Uitidentified mam- 31.75 9-98 4 mal hone 0.06 0- 1 1 Plant matter Seeds 0.11 0- 1 2.75 1- 5 2 Twigs 100.02 100.00 2.50 0-10 100.00 1 Total Birds Examined 9 2 4 15" 1 Date presented as estimated percentages of total volume of stomach food content. - One additional bird taken at Bill ^^■illiams Delta had no food in its digestive tract. June 1977 • GENERAL NOTES 335 Table 3 Stomach Contents of R. l. rhizophorae and R. l. nayaritensis FROM Mexico^ FOOD ITEMS R. 1. rhizophorae % Composition R. 1 nayaritensis % Composition Total Birds Containing Food Items Average Range Average Range Crustacea Brachyura ( Crabs ) 86.83 23-100 98.40 89-100 16 Mollusca Pulmonata (Snail) 12.50 0- 75 1 Insecta Eggs 0.17 0- 1 1.00 0- 10 2 Alisc. Parts 0.17 0- 1 0.20 0- 2 3 Plant Brown seed 0.17 0- 1 .10 0- 1 2 Miscellaneous White Stones .10 0- 1 1 Feathers 0.17 0- 1 .10 0- 1 2 Unidentified black fragments .10 0- 1 1 100.01 100.00 1 Data are presented as estimated percentages of total volume of stomach food content. Samples included 6 R. 1. rhizophorae and 10 R. 1. nmjaritensis. Food selections by birds in our sample were similar to those of other Clapper Rail populations. The major food item of Georgia Clapper Rails iR. /. ivaynei and R. I. crepitans) was crabs, supplemented by other invertebrates and the seeds of cordgrass (Spartina sp.) fOney, J. Wildl. Manage. 15:106-107, 1951). Pellet castings by R. /, crepitans in Delaware revealed crab exoskeleton and clam shell fragments IMeanley. Auk 79:113, 1962). Foods of western races of Clapper Rails, other than those reported herein, also consisted mainly of invertebrates with a minor amount of plant material. Mof- fitt (Condor 43:270-273, 1941) reported that the California race (R. I. obsoletus) ingested horse mussel (Modiolus demissus) as its main food item, with other invertebrates and seeds of cordgrass as supplements. Test and Test (Condor 44:228, 1942) found repre- sentatives of amphipods (Amphipoda) in one specimen of R. I. obsoletus, and Williams (Condor 31:52-56, 1929) observed this race to feed on clams (Macoma sp.). Applegarth (M.A. thesis, Stanford Univ., Standford, Calif; 1938) also listed a variety of invertebrates in the diet of the California birds and stated that this race lives almost entirely upon invertebrates. It was therefore not surprising that the 3 populations sampled in our study also relied heavily on invertebrates, particularly crabs and crayfish. However, we noted several interesting observations relative to food selection. First, despite a great abundance and variety of invertebrate food species in the mangrove swamps of Mexico, the birds sampled there apparently selected crabs in preference to other available foods. In the Colorado Delta (which generally contains brackish water and no crabs), the birds were adaptable and consumed a greater variety of foods. Up stream in the freshwater 336 THE WILSON BULLETIN • Vol. 89, No. 2 marshes of the Colorado River (which are relatively limited in invertebrate species and 'i numbers; Grinnell, Univ. Calif. Publ. Zool. 12:15-294, 1914), the rails’ principal food was crayfish. Thus, within the limits of this investigation. Clapper Rails were selective, opportunistic, or limited in the variety of foods eaten depending upon habitat type. On the basis of the available literature ( Ortmann, Proc. Am. Phil. Soc. 41Q71):267- i 400, 1902) it is interesting to note that crayfish were absent on the lower Colorado River prior to 1900. In recent years, crayfish have become relatively common through introduc- tion and/or natural expansion. The increase of a major food item, combined with cre- ation of stable marsh habitat behind dams during the same period (Ohmart, et ah, Trans. i 40th N.Am. Wildl. and Nat. Res. Conf,, 240-254, 1975) strongly support a hypothesis sug- gested by Tomlinson and Todd (Condor 75:177-183, 1973) and supported by Ohmart and Smith (USER contract no. 14-06-300-2409, Boulder City, Nev., 1973) that R. 1. yu- 1 manensis has since 1904 increased its distribution from the Colorado Delta northward along j the Colorado River to approximately Needles, California. Further documentation of early river development and Clapper Rail distribution can be found in Dickey (Auk 40:90-94, 1923), Phillips et al. (Tlie Birds of Arizona, Univ. of Ariz. Press, 1964), and Welsh 1 Audubon Field Notes 20:590, 1966). We are grateful to R. L. Todd (Arizona Department of Game and Fish) for his help j in securing specimens. Jill B. Leigh and Nancy Stamp aided in the sorting and identifi- cation of stomach contents. — R. 1). Ohmart, Dept, of Zoology, Arizona State Univ., Tempe, , 85201 and R. E. Tomlinson, Patuxent Wildl. Res. Center, U.S. Fish & Wildl. Serv., Laurel, MD 20811. (Present address RET: U.S. Fish & Wildl. Serv., P.O. Box 1306, Albuquerque, NM 87103.) Accepted 19 Jan. 1976. Aggression in foraging migrant Seini|>alniatt*d Sandpipers.- The comparative study of foraging in young and older birds is a current interest in ornithology (e.g. Orians, Anim. Behav. 17:315-319, 1969), l)ut few accounts assess the specific components that affect foraging efficiency, for example age-related differences in mechanical abilities or diffeiT'nces in social factors (e.g. aggression) related to foraging. We describe here some social and mechanical aspects of foraging in juvenile and adult Semipalmated Sandpipers (Calidris pusilla) which we observed at Plymouth. Massachu- setts on 29 and 30 August 1973. Juveniles were easily identified hy their juvenal plumage (see Bent. U.S. Natl. Mus. Bull. 142:248, 1927). The observations were made during an es])eeially higli tide when prey items, mostly amphipods. were unusually visible, even to us. Semijialmated Sandpipers in Plymouth usually rest during high tides and, except for brief periods during falling tides, they normally locate prey tactually. Our observations on 29 August were made to compare the frequency of aggression among aliout 20 adult and 5 juvenile sandpipers. Chasers were usually in a “Tail-up” posture quite similar to what Drur>' (Fig. 5 in Auk 78:17(>-219. 1961) likens to Sharp- tailed (irouse ( Pedioecetes phasianellus) dance postures. Dominant birds in virtually all chases we saw were the individuals that initiated a particular chase. The results (Table 1) are assessed by the same method Hailman (Bird-Banding 46:236-240. 1975) used in his analysis of sparrow aggression and show (1) that juvenile sandpipers were more frequently aggressive than adults (x“= 19.88, P< 0.001). but (2) that they were no more ag- gressive towards adults than towards other juveniles. Our oliservations on 30 August were made under conditions similar to those of the 29th. but were directed more toward tallying rates of feeding attempts rather than toward determining social interactions between adults and juveniles. About 45 juveniles and 45 adults were present in the observation area, more than on the previous day. We chose a Jane 1977 • GENERAL NOTES 337 Table 1 Frequencies of Chases Among Adult and Juvenile Semipalm ated Sandpipers Feeding IN Inundated Tidal Wrack Bird chased Chaser AdiUt J uvenile Total Adult 34^ 5 39 (42.88)- ( 10.72) (53.60) Juvenile 25 3 28 (10.72) (2.68) (13.40) Totals 59 8 67 (53.60) (13.40) 1 Observed Frequency. - Expected Frequency. single bird, either an adult or a juvenile, in either a normal or an aggressive Tail-up pos- ture, and with a stop-watch timed its activities including the number of feeding attempts and aggressive encounters, for 30-120 sec. The time intervals varied because we often lost track of individuals in the melee of other birds. The summarized results (Table 2) show clearly that birds in Tail-up postures initiated chases more often than birds in normal postures, and that birds in normal postures were the victims of chases more often than birds in Tail-up postures. These results were regardless of age. This relationship may explain why the rates of feeding attempts were similar in all 4 possible age/posture groups (Table 2). Because we did not record the ages of birds being chased on the 30th, we can not state quantitatively whether or not there was any change in dominance relation- ships among juveniles from the previous day. Our impression was that there was little change. Our intent is to show that in one circumstance, juvenile Semipahnated Sandpipers were more frequently aggressive than conspecific adults and that conseiiuently they dominated adults proportionately more than they were dominated by adults. Thus all young birds are not necessarily submissive to adults while foraging, something which is often assumed. Our efforts to quantify whether or not this aggression resulted in their obtaining more food were inconclusive because we could rarely discriminate between successful and un- Table 2 Feeding Attempt Rates and Frequency of Chasing by Adult and Juvenile Semi- palmated Sandpipers Feeding in Inundated Tidal Wrack Age Posture No. sec observed Mean no. of attenipts/sec No. of times chaser No. of times chased Adult Normal 1068 0.50 3 28 Adult Tail-up 880 0.45 68 8 Juvenile Normal 960 0.44 4 15 Juvenile Tail-up 855 0.50 77 8 338 THE WILSON BULLETIN • Vol. 89, No. 2 successful feeding attempts. We noted, however, that aggressive and non-aggressive sandpipers had similar feeding attempt rates. According to Recher and Recher (Wilson Bull. 81: 140-154, 1969) a point is reached when the frequency and intensity of aggression among sandpipers declines as they be- come more concentrated in an area of abundant food. The adult sandpipers we watched may have reached this point but the juveniles may not have — possibly because they were less efficient than adults (see Recher, Ecology 47: 393-403, 1966) in catching prey and therefore had a higher threshhold for lowering aggression. We thank D. G. Ainley, J. P. Mailman, and M. A. Howe for their helpful comments. This report is part of the results we have obtained in studies of migratory shorebirds funded by the Migratory Bird and Habitat Research Station, U.S. Fish and Wildlife Service, Contract No. 14-16-0008-687. — Brian A. Harrington and Sarah Groves, Mano- met Bird Observatory, Manomet, MA 02345. (Present Address SG: Dept, of Zoology, Univ. of British Columbia, Vancouver, B.C.). Accepted 9 Apr. 1976. Herring Gull eating bayberry. — Several studies of the Herring Gull i Lotus argen- tatus) (Harris, Ibis 107:43-53, 1965; Threlfall, Can. Field-Nat. 82:176-180, 1968; Tin- bergen, The Herring GulTs ff arid, 1960) have demonstrated the omnivorous and opportunistic qualities of its diet. In addition to the well known animal and garbage com- ponents, Herring Gulls consume grasses, grain, and blueberries ( Vaccineum angustif olium) when available (Threlfall, Nature in Wales 11:67-73, 1968; Davis, Br. Birds 49:400M'04, 1956; Haycock and Threlfall, Auk 92:678-697, 1975). This note describes a previously unrecorded vegetable food source. On 30 August 1975 I observed an adult Herring Gull feeding on the fruit of bayberry {Myrica pennsylvanica) at Great Gull Island, Suffolk County, New York. The bird flew to the bush from downwind, lowered its feet and spread them in the upper twigs of the bush, and kept its wings spread so that it was supported by the wind. While in this posi- tion the bird bent its head several times and picked berries off the upper twigs. The gull fed in this manner for ai)proximately 2 min and then flew off upwind. Pellets of either Herring Gulls or Great Black-backed Gulls - by Bond (1972). Dr. Marcus Buchanan ( pers. comm.). Director of the Virgin Islands Ecological Re- search Station on St. John saw 2 male bullfinches flying near Centerline Road above Coral Bay on 18 October 1971. On 16 November, Truesdell located at least 10 individuals (Bond 1972) in the Nanny Point subdivision near where the first female was observed and a Mrs. Learner on 25 November found 2 birds 1.1 km west-northwest of the head of Kiddel Bay Trail. From 16 January to 5 February 1972 Daniel Roby observed bullfinches on 9 days with a maximum of 15 being observed in a single day. He estimated having seen 40 L. noctis, but due to the inaccessibility of much suitable habitat adjacent to that where sightings were made we think that 100 birds is probably a more accurate estimate of the number of bullfinches on St. John at the time. Roby searched for L. noctis throughout the island, including the remote east end, but only found the species in the southeast corner. The greatest concentrations of the Lesser Antillean Bullfinch were on the western side of Ram Hill, on the hill behind Nanny Point, and at the head of Salt Pond Trail. With the exceptions of the sightings by Buchanan and Learner and a pair of females seen by Roby on the east side of Europa Bay Salt Pond (3.2 km westnorthwest of Salt Pond), all bullfinch observations were within 1.5 km of Salt Pond. Roby noted that the underparts of all closely observed adult male bullfinches were smokey gray. On 7 December 1972 Raffaele collected a female L. noctis at the foot of Ram Hill on St. John. This specimen was sent to the Bird and Mammal Laboratories at the U.S. National Museum and was identified by Richard Banks as L. n. ridgwayi. The specimen will be deposited in the University of Puerto Rico, Rio Piedras campus col- lection. Habitat. — The area around Salt Pond is among the driest on St. John and the vege- tation type is referable to the cactus woodland of Robertson (Auk 79:44-76, 1962). This association is characterized by a predominance of columnar cactus [Cephalocereus roy- enii) , and the century plant (Agave americana) ; woody plants occur only as low scrub. Except for 2 females feeding by mangroves at the edge of Europa Bay Salt Pond, all perched L. noctis were in cactus woodland a short distance from the sea. Feeding. — Roby observed L. noctis feeding on fruits on 6 plant species. Five individuals were recorded feeding on dildo cactus iC. royenii) . Three fed on barrel cactus (Cactus intortus) , 2 on common sage (Lantana involucrata) and manchineel iHippomane manci- nella) (a plant poisonous to man), one bullfinch ate fruits of gumbolimbo the female entered the nest and brooded the young for a short time. Both birds were ^ (juite tame and were easily photographed from a distance of 4 m. There were several indications that some bullfinches had fledged their young at the time of these observations in late January and early February, while others were just be- ginning to nest. Roby observed a female with a fully-fledged immature bird following her and gaping for food. On Nanny Point Hill he found a nest placed high in a tall C. royenii that was very similar in construction to the 2 active nests. On the west side of Ram Hill, 5 partially completed nests were found in O. mbescens. Roby heard male bullfinches singing at Kiddel Bay, the head of the trail to Salt Pond, and 200 m south of the second active nest. Method of introduction. — Marcus Buchanan observed 2 schoolgirls from Barbados ar- rive on Tortola with 2 caged male Lesser Antillean Bullfinches on 1 July 1971. Possibly then, the establishment of L. noctis on St. John might be the result of an introduction. James Bond (Nineteenth supplement to the check-list of the birds of the West Indies (1956), Acad. Nat. Sci., Phila., 1974) suggests another feasible mechanism for the transjrortation of L. noctis to St. John is via cruise ship en route to Charlotte Amalie fronr St. Martin or Antigua. While such a mechanism may be excellent for explaining the arrival of a flocking species sirch as the House Sparrow (Passer domesticus) , it has its drawbacks when one considers a territorial species like L. noctis that demands trees or hriisli and is not likely to be found near docks or flying offshore. There are several factors that support natural colonization by this bullfinch: (1) LoxifiiUa n. ridgwayi occurs on the Lesser Antillean islands of Anguilla, St. Martin, St. Bartheleniy, Barbuda and Antigua that lie adjacent to the Virgin Islands. Therefore this race is the most likely of the 9 Lesser Antillean Bullfinch races to have invaded the Virgins. Anguilla and St. Martin, the closest islands in the range of L. n. ridgwayi to the Virgin Islands are also the closest of all Lesser Antillean islands (supporting bull- finches) to the Virgins. Saba, the closest source to the Virgins of another race of Lesser Antillean Bullfinch (L. n. coryi), is not only farther from these islands (with the excep- tion of St. Croix which is not known to have bullfinches) than Anguilla and St. Martin, hut Loxigilla is rare on Saha while it is common on Anguilla and St. Martin. l2 l The only islands directly in the path of expansion of L. n. ridgwayi from its native islands to St. John are Norman and Peter islands. Both of these have now been reported to support Lesser Antillean Bullfinch populations. Though no bullfinches have been found on St. Thomas nor St. Croix ( Murray, Birds of the Virgin Islands, Dukane Press Inc., Hollywood, Fla., 1969; G. A. Seaman, pers. comm.), Anegada (La Bastille and Richmond, June 1977 • GENERAL NOTES 341 Carib. J. Sci. 13:91-110, 1973), Virgin Gorda and Beef Island ( Raffaele, pers. observ.), nor on Tortola (Raffaele, pers. observ.; A. Wetmore, pers. comm.), none of these islands is in the most direct potential route of expansion of L. n. ridgwayi from Anguilla or other islands inhabited by this bullfinch. Norman Island lies 8 km due east of Salt Pond, the population center of L. noctis on St. John, and is thus directly in the path of any ex- pansion from Anguilla. Peter Island, because of its location 2.2 km northeast of Norman Island is probably less important than that island to any immigration. In May of 1972 Marcus Buchanan received reports of bullfinches inhabiting both Norman and Peter islands. (3) Hurricane Donna provides a plausible mechanism for the transport of the bullfinch to the Virgin Islands. On 5 September I960 the eye of hurricane Donna passed directly over Anguilla. At this time the San Juan, Puerto Rico Weather Bureau reported the highest winds above 135 kph extending 135 km in a northeast semicircle and 72 km in a southwest semicircle. Outside these areas were gale force winds ranging from 61-133 kph extending 35 km northeast and 180 km southwest of the hurricane’s center. Later that morning Donna passed a short distance north of St. Thomas in the Virgin Islands. Wind gusts up to 108 kph were reported there even before the hurricane. This hurricane which passed directly over Anguilla and through the Virgin Islands could have been responsible for transporting L. noctis to the Virgin Islands. Since 1957, when the thorough study of Robertson (1962) indicated that there were no bullfinches on St. John (Robertson, pers. comm., spent 5 days on the coast from Lames- hur to John’s Folly and Ram Head), the only other major storm that passed between Anguilla and the Virgin Islands was hurricane Faith in August 1966. This hurricane, however, was weaker than Donna and passed farther to the north of the islands under consideration, striking them with lesser winds (61-133 kph) (Herbert, Weatherwise 20:17-23, 1967). We doubt that Faith’s winds would have been strong enough to dis- lodge bullfinches from Anguilla, and as this hurricane passed relatively far to the north of the Virgin Islands we could expect the Lesser Antillean Bullfinch’s invasion to have involved more northerly islands in the group. (4) The pattern of L. noctis distribution in the Virgin Islands when compared to that of human settlement argues for natural expansion rather than human introduction as the means of bullfinch dispersal. If the species had been brought in as a cage bird and had escaped, the probability of such an incident occurring would be greatest near human population centers. The bird would then establish itself in the nearest suitable habitat to these centers. This has been the case in Puerto Rico where of the II Ploceidae and Fringillidae found to be recently established on that island the majority have their population centers in or near large cities and appear to be spreading outward from them while only I species has done all its colonizing completely removed from a heavily populated metropolitan area. Biogeography .—KaXhex than representing a haphazard invasion into an area, the Lesser Antillean Bullfinch’s expansion to the Virgin Islands appears to fit a trend if looked at as the expansion of a Lesser Antillean element. Of 13 endemic West Indian genera known from the Lesser Antilles at least 3 have relatively recently expanded to Puerto Rico or the Virgin Islands. Among these are the 2 hummingbirds {Sericotes and Orthorhyncus) , of South American origin (Robertson 1962), and now Loxigilla. Also new to these is- lands is the Caribbean Elaenia (Elaenia martinica) , an endemic West Indian species (see Robertson 1962). Margarops, a fourth West Indian genus that appears to be in- creasing in numbers on Puerto Rico, is not considered here to be a recent arrival to that island because of other evidence suggesting long residency there (Bond, Eighteenth sup- 342 THE WILSON BULLETIN • VoL 89, No. 2 plement to the check-list of the birds of the West Indies (1956), Acad. Nat. Sci., Phila., 1973 j. In contrast to Puerto Rico and the Virgin Islands only a single West Indian genus has undergone a range expansion in recent times anywhere in the Lesser Antilles. I That is the case of Eulampis, another hummingbird, which has been found in Grenada | and Barbados < Bond. Eleventh supplement to the check-list of the birds of the West j Indies <1956), Acad. Nat. Sci., Phila., 1966). The expansion of L. noctis into the Puerto Rico- Virgin Islands region strengthens the hypothesis suggested by Robertson (1962) i that the species arriving there may be part of a contingent that moved through the Lesser Antilles more or less at the same time. Certainly the 3 endemic West Indian genera ex- tending their ranges through the Virgin Islands to Puerto Rico as compared to the 1 for all of the Lesser Antillean islands combined suggests an unusual circumstance that needs an explanation, particularly with respect to the source area and time of initiation of such a dispersal. Loxigilla noctis is surviving well on St. John and habitat similar to that which the species inhabits there abounds on other nearby islands. We might therefore expect the dispersal of L. noctis through the Virgin Islands to Puerto Rico where it may compete with its congener L. portoricensis. Should the ranges of L. noctis and L. portoricensis come to overlap, the interaction of the species should be carefully observed as this might shed light on the extinction of L. p. grandis on St. Kitts which at one time coexisted with L. noctis there. Acknowledgments. — W'e thank William Truesdell, Park Naturalist of the Virgin Is- lands National Park during the course of our investigation on St. John, for his assistance in numerous ways. We also thank Cameron B. Kepler for his advice and critical reading j of the manuscript, William B. Robertson, Jr. for his suggestions and reviewing of the j manuscript and Richard C. Banks for his specimen identification. Bedford Brown of the j U.S. Weather Bureau gave invaluable assistance with respect to the hurricane data. — j Herbert A. Raffaele, Dept, of Naturcd Resources, Box 5887, Puerto de Tierra, PR ] 00906 and Daniel Roby, 228 W' . W'ashington Lane, Phila., PA 19144. Accepted 13 Jan. \ 1976. I Foraging behavior of the White Ibis. — The foraging behavior of many ciconiiforms I is fairly well known. There is a particularly extensive literature on herons (Kushlan, | Auk 93:86-94, 1976) and storks ( Kahl, Behaviour 27:76-106, 1966; J. Ornithol. 112: 21-35, 1971; Ibis 114:15-29, 1972; Condor 75:17-27, 1973). However, little is known about the feeding behavior of ibises. Most accounts note merely that they probe in the water or on land. Bent (U.S. Natl. Mus. Bull. 135, 1926) reported Audubon’s claim that the American White Ibis < Euducimus albus) can force crayfish from burrows by placing mud in them, and Vestjens lEmu 73:21-22, 1973) reported that the Australian White Ibis (Threskiornis molucca) breaks mussels on stones. The purpose of this paper is to document the various behaviors used by the American White Ibis and to note some of ! the circumstances in which they are used. I hope that this will provide a foundation for ' future study of this generally neglected group. Observations reported here were made both in the field and under various experimental conditions on captive birds. The White Ibis is primarily a non-visual, tactile forager, and most techniques involve placing the partially opened bill in the water or bottom sediment and closing the tip on encountered prey. Ibises often swallow items by thrusting the head downward. Prey can also be worked upwards to the gullet by closing the bill tip since there is a gap between ^ the mandibles midway up the biU when the tips are closed. This may permit backward propulsion of a food item when the bill tips are brought together. The gap between | June 1977 • GENERAL NOTES 343 the ibis' bill is similar but not as extensive as that of the Limpkin (Aramus guarauna) (Snyder and Snyder, Living Bird 8:117-223, 1969) or the 2 openbill storks { Anastomus spp.) (Kahl, J. Ornithol. 112:21-35, 1971). Although it has been hypothesized that this feature is an adaptation for mollusc predation in other birds, particularly catching and extracting snails, the ibis does not extract snails. It is possible that the primary function of the bill gap in all 4 species is to effect a tweezer-like apposition of the bill tips for better grasping of prey as has also been suggested by Wetmore ( How'ell, Florida Bird Life, Coward-McCann, N. Y., 1932) . The probe is the most characteristic and most commonly used feeding behavior. Several types of probing can be distinguished by the depth and rapidity of the stab and the ex- tent of accompanying locomotion. Shallow probing is directed to the top or less than 2 cm into the sediment. It may consist of multiple tactile nibbles at the sediment or ground surface. Deep probing is the insertion of the slightly open mandible deep into the sediment, under plant roots, or under rocks. A deep probe may consist of multiple ex- ploratory- thrusts in the same hole. Several types of movement accompany probing. In stationary probing, ibises remain in one place. In step-probing, ibises generally alternate a single shallow probe with 2 or more steps. In multi-probing, ibises take several steps followed by several shallow or deep probes. Obviously intermediate behavior occurs. Other feeding behaviors are used less frequently. Pecking is the picking up of sighted objects without inserting the bill into the substrate. This is usually used on land. In water, even when prey items are visible, ibises generally use vision only to choose a par- ticular area and then probe non-visually to locate prey. This was demonstrated repeatedly by captive birds in a 10 cm deep pool. Groping is holding a widely gaping bill in the water while moving the tip along the bottom. This is similar to the behavior of Wood Storks (Mycteria americana) . Head swinging is moving the partially submerged and gaping bill from side to side in the water. At the termination of each swing, the ibis’ bill and head face to one side, with the plane of the dorsal surface of the bill perpendicular to the water. This behavior is similar to that used by spoonbills (Platalea spp.) except that spoonbills swung their head and neck from side to side while the dorsal surface of the bill remains at an angle of about 45° to the surface of the water. I have also seen head swinging in the Scarlet Ibis {Eudocimus ruber) and Glossy Ibis iPlegadis talcinel- lus) , suggesting the behavior is widespread in ibises. Since spoonbills also probe, it is probable that both probing and head swinging are homologous behaviors in the 2 sub- families and each group typifies a line of adaptation leading to the perfection of one of the feeding techniques. Although White Ibis foraging behavior is labile and almost any technique may be used in any situation, certain behavior patterns are characteristic of particular habitat conditions. The more usual behavior sequences observed under particular foraging cir- cumstances are shown in Fig. 1. Birds feeding with flocks in shallow open marshes are generally restricted to probing-while-walking behaviors (Fig. la). Often the entire flock moves as a loose unit through a feeding location. When movement is fast, as in a tight flock or on land (Fig. lb), probing is generally shallow. Stationary multi-probing is used especially along the roots of plants (Fig. Ic) and around and under other objects in the water. Deep probing is characteristic of feeding in locations with soft, drying mud and little surface water (Fig. Id). Birds feeding alone in deeper water (Fig. le, f) use a succession of techniques such as deep probing and groping. Figure Ig is a particularly varied sequence of a lone bird feeding in deep water and around a rock. Exploration around and under the rock occupied much of the birds’ time. Head swinging was re- stricted to deep, open water (Fig. le, g) and was often performed after observing an- 344 THE WILSON BULLETIN • Vol. 89, No. 2 Time in seconds 0 20 In flock, marsh prairie, **ater z: 10 cm. 4^ 40 5- 60 80 100 120 b. In flock, dry pasture, water 0 cm. 4" .sv clump, water 10 cm. move to another clump d. In flock, drying pool of mud. Z /X#' water 4 cm. e. Alone.open marsh edge of ^ /L canal, water 20 cm. f. Alone, edge of mangrove stream, water 20 cm. 4 4" 4^ ZZ g. Alone, edge of large rock and in deep open water, 20 cm. cj.o'* 5.'> fly to top of rock I fly to open water Tn water at edge of roclT walk to rock at edge of rock Fig. 1. Ethogram of feeding behavior of \^'hite Ibis in various circumstances. Abbrevi- ations: Sp-SPr = step-shallow probing, St-SPr = stationary-shallow probing, Sp-DPr = step-deep probing. St-Dpr = stationary -deep probing, St-MlPr = stationary-multiple deep probing, Pk = pecking, Gp = groping. Hs = head swinging. other bird head swinging. In the sequence of Fig. Ig, the ibis flew to within 2 m of a foraging Roseate Spoonbill ( Ajaia ajaja) before beginning to head swing. These sequences illustrate the nature of ibis foraging behavior. Foraging is generally restricted simply to probing, groping for and picking up objects, usually without specific visual cues. Variability in behavior is primarily a matter of speed of movement and water depth. Yet within the narrow constraints imposed by morphology, a number of subtle behavioral variations are formed by combinations of probing and locomotor movements that can be used to explore various microhabitats for prey. Thus the feeding behavior of the White Ibis is characterized by use of relatively few primary techniques but a number of subtle variations that permit tactile foraging in any location shallow enough to allow walking or standing. The adaptable repertoire available suggests the White Ibis could obtain a wide variety of prey but specializes on those types easily caught by its non-visual techniques (Kushlan and Kushlan, Florida Field Nat. 3:31-38, 1975). Ibis feed- ing behavior, as presently understood, is less diverse than that of many herons and storks, lacking aerial components and making no special use of feet or wings. Future studies of the White Ibis and comparative work on other species may reveal additional June 1977 • GENERAL NOTES 345 components and complexity. — James A. Kusiilan, Dept, of Biology, Univ. of Miami, Coral Gables, FL (Present address: V.S. National Park Service, Everglades National Park, Homestead, FL 33030) . Accepted 14 Mar. 1976. Birds of five families feeding from spider webs. — Burtt et al. (Wilson Bull., 88: 157-158, 1976) observed a Cedar Waxwing {Bombycilla cedrorum) removing specks from 2 spider webs in the top of a dead tree. They suggested that the waxwing was removing insect prey that had become entangled in a possibly abandoned web. Since reports of web-feeding are scarce in the literature, Burtt et al. (1976) suggested that opportunities for web-feeding might be rare. Because of their hovering abilities, hum- mingbirds appear to be pre-adapted for web-feeding, and, indeed, the only literature re- ports we have found come from the family Trochilidae (Wolf, Condor 72:1-14, 1970; Bent, U.S. Natl. Mus. Bull. 176:377, 1940; Bullock, 1825, in Bent, op. cit., 431). There- fore, we thought it important to report our observations on web-feeding in species of 5 avian families, and in 1 species to compare web-feeding with gathering of web material, possibly for use in a nest. The first 4 species mentioned were observed by R.B.W. in Mexico, the last by J.P.H. in Madison, Wisconsin. On 26 March 1973, a nesting Fawn-breasted Hummingbird {Amazilia yucatanensis: Trochilidae) pecked repeatedly at a vertically-oriented spider web in an area of dry deciduous forest 16 km south of Xpujil, Campeche. The bird hovered in front of the web and darted forward several times, touching the web with its bill on each occasion. Whether or not the bird removed insects from the web could not be determined because of poor light. The bird did not appear to be grasping web materials, nor did it begin nest-building after it left the spider web. On 16 March 1974, a Blue Bunting (Cyanocompsa pareZ/ina :Fringillidae) pecked at an orb-weaver (Family Araneidae) web located about 50 cm off the ground. The bird flew up from a perch 10 cm above the ground and hovered near the spider web. The bird pecked several times at the web and then returned to its perch near the ground. The bird repeated its actions 3 times, the third time directing its pecks toward a second web ad- jacent to the first. Whether the bird was taking insect prey or small spiders could not be determined. This sequence occurred about 200 m from the first observation. In December 1974, at the Chicanna Archaeological Zone, 8 km west of Xpujil, Cam- peche, another Blue Bunting was observed dismantling a spider web about 3 m from the ground. The bird approached the web along a twig, grasped a strand of the web and pulled. In pulling, the bird assumed an upright posture with the long axis of the body perpendicular to the perch and the head held perpendicular to the body axis. The bird struggled with the web for 15 sec and finally broke off the strand and flew away. The action of web-gathering appeared substantially different from the pecking motions de- scribed above. Another instance of web-feeding occurred at Chicanna on 17 July 1975 while a White- bellied Wren {TJropsila /eucogas/m :Troglodytidae) was foraging in a tree 4 m above the ground. The bird was moving rapidly from twig to twig, actively foraging by pecking at twigs and hawking insects. At one point, the bird pecked twice at a twig, turned, and delivered 2 pecks to a spider web, and then moved away. The web which the bird pecked appeared abandoned and had a large amount of vegetable or animal matter entangled in it. A Yellow-green Vireo {Vireo flavotdridis ’.Nireomdae) also pecked at a spider web while foraging at Xpujil on 5 July 1974. The bird was foraging 10 m up in a 13-m tree at the edge of a clearing. The bird grasped a speck in a spider web with its bill, pulled force- 346 THE WILSON BULLETIN • VoL 89, No. 2 fully for a moment, and then swallowed. The bird then wiped its bill twice and moved away. On 11 May 1975, a Yellow-rumped Warbler i Dendroica coronata :Parulidae) hovered in front of a vertically-placed spider web attached to the tips of branches high in a tree, plucked one prey item from the web, and flew off. There were no perches convenient to the web as in tbe case of the Cedar Waxwing reported by Burtt et al. The web bad 13 remaining prey items visible in it and despite many warblers of several species foraging in this tree and others nearby, no other bird fed from or even inspected the spider web in the 10 min after the feeding instance. Our observations suggest that web-feeding is more widespread than the paucity of liter- ature on the subject suggests. Perhaps observers have overlooked it or assumed that it was already so well known that it was not important to mention it. It may be significant that 3 of the 4 tropical species were observed web-feeding during the breeding season, when a need for higher protein intake may occur. Hummingbirds in particular are not well adapted for capturing insects, but their hovering abilities make it possible for them to secure protein by cleptoparasitism from spider webs. — Robert B. Waide and Jack P. Hailman, Dept, of Zoology, Univ. of Wisconsin, Madison 53706. Accepted 1 May 1976. Winter nest microclimate of Monk Parakeets. — Monk Parakeets ( Myiopsitta rnonachus) have a broad distribution in South America where their range extends from tropical Bolivia and Brazil well into the temperate regions of Argentina (Bull, Wilson Bull. 85:501-505, 1973; Olrog, Las Aves Sudamericanas, Universidad Nacional de Tucu- man, Argentina, 1968) . While the species normally encounters a wide range of local climates, in North America it survives winters which are more severe than those of its native range (Bump, U.S. Fish and Wildl. Serv., Bureau of Sport Fisheries and Wildlife, Wildlife Leaflet No. 496, 1971). Among psittacids Monk Parakeets are unique in build- ing large enclosed nests composed of interwoven twigs. (Forshaw, Parrots of the World, Doubleday and Co., Inc., New York, 1973). Unlike most birds Monk Parakeets occupy their nests throughout the year. The role of enclosed nests in contributing to the main- tenance of a favorable microclimate has been demonstrated for several species (Ricklefs, in Avian Energetics, R. Paynter, ed., Publ. Nuttall Ornithol. Club 15:152-297, 1974). Tolerance of low winter temperatures in this species may be improved by the use of these stick nests for nighttime roosts. In this paper we investigate the possibility that during the winter the nest of the Monk Parakeet contributes to energy savings by creating a favorable microclimate. Methods. — Measurements of air temperature (Ta) and nest temperature (Tn) were made at a nest which had been constructed by a breeding pair in the upper corner of a large (4x3x2 m) outdoor flight cage. The nest was situated approximately 10 cm below the cage roof, but otherwise exposed on all sides. The nest (Fig. 1) was 0.5 m long, 0.3 m deep and 0.3 m wide. At the time of study it was occupied by a mated pair. Temperatures were measured with 20-gauge copper-constantan thermocouples and were recorded at intervals of 2 min with a Honeywell recording potentiometer (model 112). The uncertainty of measurement did not exceed 0.2°C. Ta was recorded from a thermo- couple placed 5 cm from the back of the nest. Tn was recorded from thermocouples im- planted at several locations within the nest, but concentrated around the inner nest cham- ber (Fig. 1). An additional thermocouple was positioned on the floor of the inner nest chamher such that it was in contact with the birds when they occupied the nest. This probe signaled when the birds entered or left the nest. Measurements were made during 10 days in January and February usually from 16:00 June 1977 ■ GENERAL NOTES 347 10 cm. Fig. 1. Representation of the Monk Parakeet nest seen in longitudinal section (left) and on end (right). Points marked a and b correspond to the position of the thermo- couples represented in Fig. 2 by the dots and solid line respectively. until 09:00 of the following day. Wind velocity was obtained from records of the univer- sity meteorological station located approximately 100 m from the nest. Results. — The birds entered the nest at various times during the day for short periods, especially during inclement weather. They also entered the nest around 17:30 (Mj h after sunset) and remained in the nest until approximately 06:30 {Vi h before sunrise) the following day. While the birds were in the nest, temperatures within the inner chamber generally exceeded Ta; temperatues within the tunnel never exceeded Ta. This reflected the birds’ position within the nest. Fig. 2 illustrates a typical record of night time Ta and Tn. Early in the night Tn’s were only slightly above Ta. However, with time the difference between Tn and Ta in- creased. The increase in Tn over Ta during the early morning hours corresponded with a decrease in wind velocity as shown in Table 1. The maximum temperature difference (Tn-Ta) observed on any night was 4.6°C. The mean Ta calculated during the time the birds were in the nest was -5.65° C while the mean nest temperature (calculated as Wind Velocity and the Table 1 Gradient Between Nest Temperature AND Air Temperature Time of day Wind velocity!, m/sec T2, »c hours maximum minimum maximum minimum 1700-1950 8.5-11.6 1.8-3.1 1.0 -0.1 2000-2250 8.0- 8.9 1.8-2.7 1.4 0.0 2300-0150 5.4- 8.0 1.8 2.0 0.1 0200-0450 3.6- 5.8 0.4^0.9 3.3 0.1 0500-0750 3.1- 4.5 0-0.9 3.0 0.8 1 Ranges for values determined over 10 min intervals. 2 Range of values for the difference between nest temperature and air temperature in Fig. 2 while birds occupied the nest. 348 THE WILSON BULLETIN • VoL 89, No. 2 1700 1800 1900 2000 2100 2200 CJ o CD 3 O CD CL E CD Time of day Fig. 2. Relation of air temperature (circles) and nest temperatures (line and dots) to time of day. Arrows indicate time when birds entered and left the nest. the mean of the 2 thermocouple temperatures recorded within the inner nest chamber) was .1.89°C. Thus on the average Tn exceeded Ta by 1.76°C. Discussion. — The nest examined in this study was relatively small. As Monk Parakeets nest gregariously, nesting assemblages may contain up to 20 inner chambers, weigh 200 kg, and measure 2m or more across (Forshaw 1973; Roscoe, et al., N.Y. State Fish and Game J. 20:170-173, 1973). Despite this fact, temperatures near the inner chamber of the nest were as much as 4.6°C above Ta, and on a typical night exceeded Ta by an average of 1.76°C. It seems probable that temperatures would be even more moderate in larger nests which should be better insulated. Additionally, since winter nests may be occupied by both adults as well as the juveniles of the previous breeding season fpers. obs.), even higher temperatures might be anticipated. Although nest temperature was higher than air temperature, a major advantage of us- ing the nest as a winter roost might be in reducing radiant heat loss, since heat loss to the cold night sky may represent an appreciable fraction of the total heat loss (^Schmidt- Nielsen, Oxford Univ. Press, New York, 1964; Webster, J. Appl. Physiol. 30:684^690, June 1977 • GENERAL NOTES 349 1971). By ameliorating the effects of wind, particularly in large nests, convective losses also may be reduced. The minimum energy savings resulting from elevated nest temperatures can he esti- mated from the relation of oxygen consumption (VO:>) to Ta of Monk Parakeets (Weathers and Caccamise, Oecologia 18:343-358, 1975). Below 25°C VO2 of Monk Para- keets increases linearly with decreasing Ta at a rate of 0.099 ml O2 g"^ hr~^ °C'^. The energy savings attributable to the nest is the product of the excess of Tn over Ta and this factor. Taking the average Tn - Ta to be 1.76°C then Energy Savings r= (1.76°C) X (0.099 ml O2 g'' hr'^ °C-9 = 0.174 ml O2 g'' hr~\ The resting metabolic rate of Monk Parakeets measured at night and in the thermal neutral zone is 1.17 ml O2 g"^ hr"’^ (Weathers and Caccamise, 1975) . Thus the mean energy savings for the night depicted in Fig. 2, based on the difference between Tn and Ta, represents 15% of the standard metabolic rate. Expressed as the difference in predicted metabolic rate at -5.65°C versus -3.89°C the energy savings is 3.7%. This value does not take into account the effect of the nest in reducing radiant heat loss and therefore should be considered a minimal estimate of the energetic advantage of using the nest as a winter night-time roost. We thank Helen Mozdic and Peter Alexander for technical assistance, and Mr. Charles Wagg of the Division of Plant Industry, New Jersey Department of Agriculture who was instrumental in obtaining the birds. This report is a paper of the Journal Series, New Jersey Agricultural Experiment Sta- tion, New Brunswick, New Jersey. — Donald F. Caccamise and Wesley W. Weathers, Dept, of Entomology and Economic Zoology and Dept, of Environmental Physiology, Rutgers, The State Univ. of New Jersey, New Brunswick 08903 (Present address WWW: Dept, of Avian Sciences, Univ. of California, Davis 95616). Accepted 17 Feb. 1976. Snake predation on Bell’s Vireo nestlings. — Bent (U.S. Natl. Mus. Bull. 197: 260, 1960) found that cats and cowbirds were the Bell’s Vireo’s {Vireo bellii) worst ene- mies; he makes no mention of snakes. However, Mumford (Wilson Bull. 64:231, 1952) and Barlow (Univ. Kansas Puhl. Mus. Nat. Hist. 12:291, 1962) both suggested snake predation as a cause of nest losses. Nolan (Condor 62:241, 1960) implicated the black rat snake {Elaphe obsoleta) and black racer {Coluber constrictor) as nest predators based on their abundance in the scrub habitat he studied. This note documents 3 observations of snake predation on nestling Bell’s Vireos in Douglas County, Kansas. On 14 June 1974, about 08:45, I heard the scolding notes of a pair of Bell’s Vireos and several Dickcissels {Spiza arnericana) and Northern Yellowthroats {Geothlypis trichas) near a dogwood {Cornus drummondii) where I had discovered a vireo nest 3 days earlier. I reached the nest in time to see a black rat snake {E. o. obsoleta) with a vireo nestling in its mouth climbing quickly off the nest limb to the ground. The nest, which previously held 4 nestlings, was empty. This pair of vireos built a new nest about 10 m from the old site and laid a clutch of 3 eggs. On 25 June 1975, about 10:00, I heard scolding notes of a pair of Bell’s Vireos in the vicinity of a small dogwood. I located the nest on the southeast corner about 38 cm from the ground and observed a red-sided garter snake (Thamnophis sirtalis parietalis) on the supporting limb of the nest with its head in the nest engulfing the back half of a very young nestling. It dropped to the ground as I pulled back a branch for a closer look, and as I attempted to grab the snake, it dropped the live but bleeding nestling and escaped. I returned the nestling to the nest, but when I checked the nest again some 10 min later 350 THE WILSON BULLETIN • VoL 89, No. 2 the nestling was missing. Best (Auk 91:169, 1974) found that snakes would return for the remainder of the nest contents when disturbed. This pair of vireos renested about 15 m from the old site and successfully fledged 3 young. On 27 June 1975, I heard mobbing calls from several species of birds near a Bell’s Vireo nest. As I approached the dogwood where the nest was located I saw a black rat snake slipping away through the grass. I killed the snake and found that it had eaten the 3 Bell’s Vireo nestlings from the nest. Bell’s Vireos nest very close to the ground and it is not surprising that snakes are im- portant predators on them. Black rat snakes are good climbers, regularly take birds as prey (e.g., Jackson, Wilson Bull. 82:329-330, 1970), and can often be located by mob- bing calls of birds (Fitch, Copeia 1963:649-658, 19631. Although red-sided garter snakes are not regarded as climbers they occasionally take nestling birds as prey (Fitch, Univ. Kansas Publ. Mus. Nat. Hist. 15:493-564, 1965). My thanks to Ivan Boyd of Baker University, Baldwin City, Kansas for permission to use the Baker Wetlands Research Area and to Richard F. Johnston for his comments on the manuscript. — Calvin L. Cink, Museum of Natural History and Dept, of Systemat- ics and Ecology, Univ. of Kansas, Lawrence 66045. Accepted 12 Jan. 1976. Crow predation on Black-crowned Night Heron eggs. — Corvids prey on the eggs of many species of birds. They feed on some eggs at nest sites and fly off and cache others at a distance. Descriptive and experimental data on crow predation has come from work in gull colonies located on flat ground with high visibility because the vege- tation is sparse and low (e.g. Tinbergen et al.. Behaviour 28:207-321, 1967). However, Fish {Corvus ossifragus) and Common crows ( C. brachyrhynchos) also prey on the eggs of herons, egrets, and ibises that nest in dense tree colonies where nests are not as visible from the air (Milstein et ah, Ardea 58:171-255, 1970; Meanley, Wilson Bull. 67:84-99, 1955; Uusi and Uusi, Wilson Bull. 80:458-466, 1968). No estimation of actual preda- tion rates or descriptions of crow-heron interactions from heron and egret colonies are available. In this note we report on interactions between Black-crowned Night Herons ( Nycticorax nycticorax) and Common and Fish crows observed in 5 heronries in southern New Jersey from 1973 to 1975. Data were collected in 3 Cheriy' {Prunus sp.) and Poison Ivy (Rhus toxicodendron) tree colonies of Black-crowned Night Herons, Snowy Egrets iEgretta thula) , and Glossy Iliis (Plegadis falcinellus) on Little Beach Island, Brigantine National Wildlife Refuge; and in 2 Phragmites (with scattered Iva and Juniperus bushes) colonies of Black-crowned Night Herons, Snowy Egrets, Common Egrets (Casmerodius albus). Cattle Egrets ( Bubulcus ibis) and Glossy Ibis on Big Heron and Islajo islands near Atlantic City (see Adams and Miller, EBBA News 38:103-107, 1975 for de- scription) . All 5 heronries examined had at least 1 active crow nest. Two of the 3 heronries on Little Beach Island had cache nests ( used only to store food items) , located near the active nests. These contained heron eggshells, the unbroken eggs of night heron. Clapper Rail (Rallus longirostris) and Glossy Ibis, diamondback turtle (Malaclemys terrapin) eggs and hatchlings, and a dead Glossy Ibis chick. We often observed crows eating the numerous bird and turtle eggshells that were scattered on Spartina mats in the grassy areas near the heronries. We systematically searched the tree areas on the northern end of Little Beach Island and found 2 inactive heronries each with an inactive crow nest, but no other crow' nests. Similarly, crow nests w'ere only located in the heronries fin Juniperus) on Islajo and Big Heron islands. June 1977 • GENERAL NOTES 351 Heron egg success was determined in 1 heronry on Little Beacli Island hy marking all eggs and checking nests every other day. In 1974 the heronry contained 22 Hlack-crowned Night Heron nests, 6 Snowy Egret nests, and 1 crow nest. Erom a blind we observed crows taking all of the eggs in 5 and some of the eggs in 2 night heron nests. Egg loss at 5 other nests was attributed to crows when shells with peck holes were found. No other aerial predators were observed in this colony. We also observed crows carrying night heron eggs. We observed no egg loss in Snowy Egret nests. In 1975 this colony contained 21 night heron nests, 6 Glossy Ibis nests, 2 Snowy Egret nests, and 1 Fish Crow nest. In 1975 only 10% of the eggs were eaten compared to 36% in 1974. In both years there was an active crow nest, hut in 1975 the nest was deserted halfway through the night heron incubation period. This crow nest was only 1 m above a night heron nest, and the heron pair might have caused the crow’s desertion. Corvids were the only aerial predators we observed in any of the heronries. Common Crows were seen most often. We observed 1 Blue Jay i Cyanocitta cristata) take and eat a Black-crowned Night Heron egg. We observed Black-crowned Night Herons from blinds for 50 hours each year. Night herons react to crows in several ways; (1) an incubating bird remains silent on the nest when a crow flies over, (2) a night heron standing near its uncovered eggs (less than 3.5 m) walks quickly down to its nest and resettles when a crow approaches, (3) a night heron standing more than 3.5 m from its uncovered eggs flies to the nest and settles when a crow approaches and (4) a night heron whose mate is incubating either ignores a crow% flies out of the heronry squawking, or aggressively moves toward a crow while vocalizing. Crows were successful in taking eggs only when night herons failed to return quickly to their nests, when an incubating bird left before its mate returned, and before incuba- tion began. Usually night herons did not actively chase or attack crows although night herons did supplant crows sitting near night heron nests 20% (n — 25) of the time. On 2 occasions we saw night herons chase and mob crows flying over the heronry located in Phragmites. The night herons did not dive-bomb or attack the crow, but flew very close to it. Crows were observed flying slowly over areas known to contain heron nests, and swiftly over other areas. Once crows descended into the trees within the heronry’, the night herons squawked until the crow left. Although nestling Black-crowned Night Herons actively and pugnaciously defend their nests (this study; Teal, Wilson Bull. 77:257-263, 1965), adults do not overtly attack crows or other predators near their nests. Other herons such as Snowy and Cattle egrets seemed far more aggressive. Thus one advantage of nesting in mixed species colonies may be the anti-predator behavior of other species. For example, the senior author, work- ing at Agassiz National Wildlife Refuge in northwestern Minnesota (1969-1971) ob- served Black-crowned Night Herons nesting in cattails with Franklin’s Gull (Larus pipixcan) . Whenever the gulls changed the location of their colony, the night herons also deserted the old location and nested in the new colony. When predators (weasel, skunk) approached the colony, the night herons remained on their nests and the gulls mobbed the predator. This research was funded by grants from Sigma Xi fto D.C.H.) and the Research Council of Rutgers University (to J.B.) and MNIH grant 16727 (to Colin Beer). — Jo- anna Burger, Dept, of Biology, Livingston College, Rutgers Univ., New Brunswick, NJ 08903, and D. Caldwell Hahn, Institute of Animal Behavior, Rutgers Univ., Newark, NJ 07102. Accepted 8 Apr. 1976. ORNITHOLOGICAL LITERATURE Evolution and the Diversity of Life: Selected Essays. By Ernst Mayr. Belknap Press of Hanard Universit>- Press. Cambridge, Mass., 1976: ix + 721 pp. S20.00. — In this book Ernst Mayr presents 47 essays on various topics in evolutionar\- biology originally published in many books and journals between 1940 and 1974, though most are from the past 2 decades. Some of the essays have been shortened, have had the references up- dated. or have been adapted by the addition of new comments clearly distinguished from the original version. Two have been translated into English. The essays are di\ided into 9 sections, their titles and the number of essays in each being: (!• Evolution (9 essays); < II) Speciation (7); iIII) Histon of Biology (6); (IV > Philosophy of Biology <4i; (V) Theoiy of Systematics (4); (VI) The Species (5); (VII) Man (1) ; 'VIII) Biogeography (9); and (IX) Behavior (2). For each section Mayr has written a brief introduction placing the topic in historical context and assessing his own con- tribution to it. Many of the essays deal directly with birds and will be of special interest to ornithologists; these include “Bird speciation in the tropics.” “History- of the North American bird fauna.” “Inferences concerning the Tertiar>- North American bird faunas,” “The origin and history of the Polynesian bird fauna.” “Fragments of a Papuan Or- nithogeography," “The omithogeography of the Hawaiian Islands.” and "The nature of the colonization of Birds.” Most of the essays do not deal directly or principally with birds, but with general aspects of evolution and systematics that are applicable to ornithol- ogy as well as other branches of evolutionar\- biology. This is a valuable collection of essays, many of which are otheiAvise available only in specialized libraries. For anyone interested in the evolutionary aspects of avian biology, this book is indispensible. — Robert J. Raikow. Ecology and Evolution of an Andean Hummingbird iOreotrochilus estella), by F. Lynn Carpenter. University of California Press, Berkeley, California, 1976: 106 pp., 8 black-and-white plates, 14 text figs., 11 tables. $2.50. — Probably more than any other avian group, hummingbirds are presently serving as vehicles for testing ecological- evolutionary theory. The attention is well deserved. The ease with which they can be observed and captured, their territorial tendencies and dependency of spatially and temporally restricted nectar sources, and the great degree of species’ sympatrv, are all at- tributes that make them a convenient group to study. Surprisingly, however, there are only 2 monographs on individual species; one by F, Gary Stiles on the behavioral ecology of Calypte anna, and the recent one considered here. Carpenter continues the trend set by Stiles — dependence on long-term systematic study, thorough documentation, and in- sistence that the way to understand the evolution of a species is through its extant ecologv. Carpenter follows the activity of Oreotrochilus estella through 2 wet seasons and 1 dry- season in southern Peru and northern Chile. 0. estella is one of only a handful of resident hummingbird species living above 4000 m, though species literally swarm on the slopes and in the lowlands below. Carpenter set out to identify the morphological, behavioral, and ecological adaptations that are integrated into the life history strategy of 0. estella and which allow it to exploit this climatically rigorous and vegetatively sparse habitat. Hummingbirds seem to be more specialized and less variant in morphology, physiology-, and behavior than most other avian groups, thus Carpenter has selected this particular species as the one most likely to be “aberrant" and show striking differences in these categories from “typical” tropical or temperate species. 352 June 1977 • ORNITHOLOGICAL LITERATURE 353 The monograph is organized into 5 major sections: (1) morphological adaptations, 12) behavioral adaptations (foraging and roosting), (3) reproductive adaptations, <4) ter- ritorial adaptations, and (5) energetics. In the first section adaptive value of large size, dull coloration of both sexes, and large foot size, characteristics unusual in hum- mingbirds, are discussed. Coloration is seen as an antipredator device peculiar to a grass- land habitat totally devoid of trees, and the large size of both sexes ( x = 8 g » is be- lieved related to increasing energy use efficiency. Large foot size is explained as an adaptation to vertical roosting on canyon and cave walls. In the second section behaviors associated with these unique roosting and foraging habits are discussed in the context of the climatic and biotic factors that have determined the evolution of morphological adaptations. Carpenter suggests that cave and rock roost- ing is an adaptation for avoiding nocturnal predation and for protection against ambient winter night temperatures which sometimes approach 0° C. Foraging technique at flowers is shown to involve significantly less hovering than in other hummingbirds, but much more perch feeding, an adaptation Carpenter sees as an energy-saving response to high altitudes where thin air produces a large energy demand for hovering. Pollen feeding by O. estella on the only winter blooming hummingbird-flowered plant, Chuquiraza spinosa, is discussed as a probable coevolutionar>- pollination relationship in which the plant is more reliably pollinated by O. estella at this time of year than by more externally temperature-influenced insects. Section 3 deals with the selection of nest sites, construction of nests, the phenolog>- of nesting, and nesting success. Most nests are constructed under rock overhangs, are well insulated, and occur near the richest nectar sources. Carpenter argues that these features protect the nestlings from severe weather and nest predators, and account for the high nesting success compared to other hummingbirds. Because such sites are limited, compe- tition for nest sites was intense and was believed to account for the evolution of an ag- gressive female territoriality associated with the nest site in which both male and female conspecifics are usually vigorously repulsed (Section 4). The aterritorial males appeared to be subdominant to females, and males visit females on territories and display and at- tempt to mate with them, the reverse of what occurs in all other hummingbird systems described to date. Carpenter believes this atypical arrangement is forced by the need for females to occupy gorges and construct nests there. Because, incidentally, the richest nec- tar sources are also found in these areas, males are forced out onto open hillsides where food is dispersed and scarce and therefore not dependable. The larger size of males is also seen as related to food supply: larger body size is believed to result in more efficient linear flight between distant food sources. In section 5 an energetics model of feeding territoriality is generated for 0. estella by measuring resting and torpid metabolism, estimating cost of linear and hovering flight, then computing time and energy budgets for both sexes in summer and winter. Winter is seen as the critical time of year when most mortality occurs despite a ver>' physiologi- cally efficient torpidity system finely tuned to the daily winter ambient temperature re- gime. An integrative concluding discussion section recounts the morphological, physiological, and behavioral adaptations that allow 0. estella to exploit the high Andes, discusses the possibility of a taxon cycle in 0. estella and other high Andean species, and concludes with a discussion of evolution and colonization at high altitudes. Topics covered include numbers of trochilid species in high altitude vs. low altitude environments, the relation between competition and speciation in high altitude hummingbird species, and the mag- nitude and speed of evolutionary change to life at high altitude. 354 THE WILSON BULLETIN • Vol 89, No. 2 The monograph is well written in a crisp, easily read style. Jargon is kept to a mini- mum, so the interested layman will be able to glean much from each section. Figures < and tables are kept simple and are effective in helping to grasp textural explanation. i Statistical analysis is straightforward. Most major contentions are well argued. There is effective integration of field observa- tions and laboratory experimentation, particularly regarding torpidity. The choice of this “aberrant” species for investigation was a good one and Carpenter generally makes a strong case for the major hypotheses that the peculiar adaptations of O. estella have been forced by the climatic and physical factors characteristic of the altiplano and, secondly, that hummingbirds, usually regarded as a highly specialized and adaptively limited group, can evolve adaptations which allow them to exploit unusual and climatically i rigorous environments. 1 The work is not without weaknesses. In my opinion there is excessive speculation * throughout, particularly in the first section (morphological adaptations). Arguments regarding body size differences between O. estella and other hummingbirds and ex- planations for sexual dimorphism in O. estella are protracted and overextended. Occa- sionally, major arguments are based on very small sample sizes. Conceptual confusion also sometimes occurs. Early on, with little evidence, winter mortality is identified as the most important population limiting factor and implicitly regarded as ultimately limiting. Later, the introduced winter blooming Eucalyptus is be- lived to have decreased winter mortality, but this is not regarded as important because at this point Carpenter tells us that ultimate limitation occurs through nest site scarcity. In another section, he says that “the nesting stage is still probably the most vulnerable of the whole life cycle,” but no data are offered in support. Still later in the energetics sec- tion the species’ winter energetics strategy is pictured as extremely efficient and finely tuned. If so, one wonders why winter mortality occurs to the extent that it supposedly does. The reader is left guessing what Carpenter really does believe ultimately limits the population. Arguments in the concluding discussion also are flawed. The relationship of trochilid species diversity at high vs. low altitudes is simplistic and arguments marshalled to explain the reason for low diversity at high altitude are confused, e.g., “rigorous” en- vironment is poorly defined and seemingly confused with “stressful,” and it is not clear whether it is the “rigor” of the environment or the lack of evolutionary- time that is the main explanation for the low diversity. Further, though the effect of speciation rate on diversity is considered important, the importance of extinction rates of species in high elevation environments is not even mentioned, an unexpected oversight for one who argues earlier for the presence of a taxon cycle in 0. estella and related high-altitude congeners. In general, however, this study is a valuable addition to our increasing knowledge of hummingbird systems and is one of the few to date that provides an overview of the ecol- ogy and evolution of a single species. Serious students of hummingbird ecology will find it very useful. — David L. Lyon. Crows of the World. By Derek Goodwin, illus. by Robert Gillmor. Cornell Univer- sity Press, Ithaca, New York, 1976: vi + 354 pp., 3 color plates, many line drawings and distribution maps. $28.50. — This is a reference guide to the family Corvidae. Most of the book (286 pages) is devoted to individual accounts of each of the 116 species of corvids. In each account there is a detailed description of the species, followed by June 1977 • ORNITHOLOGICAL LITERATURE 355 discussions of field characters, distribution, feeding and general habits, nesting, voice, and display and social behavior. For each species there is also a range map. Those familiar with the range descriptions given in most check-lists (e.g. Khorasan, tlie Dasht i Lut desert and Persian Baluchistan) will appreciate this visual aid. For most species a line drawing is also provided. Although some of these illustrations are rather poorly reproduced, they add a new dimension to a work of this kind. Readers are usually forced to choose between books that provide many color photos and plates but do not give much information on the birds depicted, or those that provide detailed information with little visual reinforcement. This book provides an excellent compromise between these formats. The species-by-species analysis of the Corvidae is one of the most comprehensive studies of the vast amount of information that has been recorded concerning this family. In his introduction, Goodwin states that, “one of the book’s purposes is to indicate what is not known” about various members of this family. By reading the description and synopsis of the behavior and biology of each species, and then reviewing the references that are listed immediately following the discussion of each bird, the reader becomes aware of the gaps in the information concerning that form. It is probable that these analyses may stimulate readers who are in a position to record observations on some forms to do so, and not disregard the birds with a paraphrase of Aesop’s “It’s only a crow and that signifies nothing!” The first chapter discusses taxonomic nomenclature and mechanisms of speciation. For those familiar with these concepts there is little new information, but the chapter should prove helpful to readers not well-versed in these areas. Fig. D.l. is a dendrogram showing the “presumed relationships of the corvine genera.” The author does not give data or references that could enable the reader to quickly interpret the meaning of this “phylogeny” or determine the basis by which it was constructed. The next 2 chapters deal with the adaptive radiation and adaptive characters, and the plumage and coloration of the Corvidae. They provide information that will help the reader to better understand the descriptions given in the species accounts. Chapter 4, on behavior, is longer and more detailed than the preceding chapters. This is under- standable considering the vast amount of information that has been recorded on the complex behavioral patterns of the Corvidae. Goodwin has ably summarized much of this information and presents it in a well-organized, easy to read form. The author states, “In the species section forms believed to be most closely related to each other are placed together in so far as this is possible within the confines of linear arrangement.” Although most readers will find the author’s arrangement for the most part acceptable, I feel that he should have included a list of the references or observations that were used in constructing this sequence. The book concludes with indexes of com- mon and scientific names. Crows of the World is a well-written and informative book that provides a well- organized mass of information on all 116 species of Corvidae. It would be a useful addi- tion to any ornithological library. — Stephen R. Borecky. Collected Papers in Avian Paleontology Honoring the 90th Birthday of Alex- ander Wetmore. Edited by Storrs L. Olson. Smithsonian Contributions to Paleobiology, No. 27, Smithsonian Institution Press, Washington D.C., 1976: xxvi 211 pp. — Alex- ander Wetmore has done more to revitalize avian paleontology during the past half century than any other person, and this volume is a remarkable testament to the vigor and excitement of the field today. Festscrift volumes have not been popular recently, at 356 THE WILSON BULLETIN • Vol. 89, No. 2 least in this country, which may not be a bad thing because they are often collections of major papers by minor authors, or minor papers by major authors; assemblages of journal rejects rescued from the bottom drawers of file cabinets. Such is decidedly not the case with this volume. An outstanding group of workers has presented a collection of always interesting and often exciting reports on a variety of topics. Though the book opens with brief “appreciations” of Wetmore’s work and influence on the field by S. Dillon Ripley and Jean Delacour, and a scientific biography of Wetmore’s work by Storrs L. Olson, it is the papers themselves that offer the finest testament to the role that Wetmore has played in stimulating many workers to enter and explore the field. This is further demonstrated in the acknowledgements of Wetmore’s influence by various authors, and by the number of new taxa that they have named in his honor. More than just a collection of fine research reports, this volume is also a summary statement of the present state of avian paleontology'. For this reason, rather than try' to review all of the papers separately, I will take this opportunity to assess the strengths and weaknesses of the field, and its current role in avian biology, at least as I percieve it from the viewpoint of an interested nonparticipant in the field. Paleontology in recent years has shifted from an emphasis on the description of new forms, as important as that continues to be, to an analysis of the significance of its findings in terms of evolutionary' theory'. Thus it has become more closely associated with biology and less with geology than was formerly the case, a trend that is recog- nized in the increasing use of the term paleobiology. In this context its greatest potential contributions to ornithology would seem to lie in three general areas: Q) the origin of birds and of their peculiar adaptations, especially for flight, (2) analysis of the phylogeny of the higher categories of birds, and (3) clarification of the role of birds in the evolu- tion of ecosystems through time. I will consider the field as exemplified by the present collection of papers in relation to these three areas. The problem of the origin of birds has been reexamined by several workers in recent years, following several decades during which Heilman’s thecodont theory went un- challenged. Of the several ideas recently suggested for avian origins, John Ostrom’s theory of an origin from the coelurosaurian dinosaurs has been the most convincing. In this volume Ostrom uses his intimate knowledge of the anatomy of Archeopteryx to speculate on some of the anatomical changes that must have occurred between Archeop- teryx and modern birds so as to make powered flight possible. The main interest centers on the coracoid and its changes in form that were apparently associated with a conver- sion of the action of the supracoracoideus muscle, converting it from a depressor of the wing to an elevator, thus making possible the recovery stroke of powered flight. The toothed birds of the Cretaceous are reexamined in two papers. Philip D. Gingerich assesses the significance of these birds to avian phylogeny. He argues that Hesperornis did indeed have teeth, and also a paleognathous palate. This leads him into a consider- ation of the ratite problem, and he argues against Cracraft’s theory that the ratites are monophyletic because (among other reasons) the paleognathous palate is considered to be a primitive condition among birds rather than a derived state. Larry D. Martin and James Tate Jr. provide a detailed description and analysis of Baptornis, a diving bird closely related to the better-known Hesperornis. In order to understand both the phylogeny of birds and the evolution of faunas, it is obviously important to know at what periods in time the various groups occurred, and especially when they first appeared. Fossils are the source of this information, and many useful data have been provided by avian paleontologists, including several important contributions in this volume. Nevertheless, the overall picture is subject to June 1977 • ORNITHOLOGICAL LITERATURE 3S7 frequent reevaluation and must be regarded with a certain caution owing to the frequency with which specific fossil forms are reallocated to different higher taxa. In the present volume, for example, Alan Fedducia reassigns the genus Neanis, previously considered to be the oldest known passeriform, to the order Piciformes. Feduccia and Larry D. Martin discuss Uintornis, which was originally placed in the Picidae, then the Buc- conidae, then shifted to the Cuculiformes. They return it to the Piciformes, but in a new family Primobucconidae. Storrs L. Olson shifts Protornis from the Alcedinidae to the Momotidae, thus giving this presently New World family an Old World history. Charles T. Collins moves the extinct family Aegialornithidae from the Apodiformes to the Caprimulgiformes. Reallocations of this sort are not uncommon in the avian paleontological literature. It is unclear whether this is because the often fragmentary remains are ambiguous, or because earlier workers were less rigorous in their studies than more recent investigators. However, in some cases, including those in this volume, it seems that new interpretations follow upon the analysis of larger and better samples than those available to earlier workers. This is an important problem because the whole picture of avian phylogeny, to the extent that paleontology contributes to it, depends on the accurate recording of the occurrence of different groups in time and space, and unexpected discoveries may have a profound effect on our ideas of the history of particular groups. This is clearly illustrated by Storrs L. Olson, who presents us with Motmots in Europe and Todies in Wyoming. Another important problem is how to interpret fossil forms that are intermediate in characteristics between living taxa. “Intermediate” in this context means that the fossil forms are mosaics of characters, some shared with one taxon and some with another. Pierce Brodkorb describes a new form, Alexornis, that shares some char- acteristics with the Bucconidae (Piciformes) and some with the Momotidae (Coracii- formes) . Pat Vickers Rich and David J. Bohaska describe the oldest known owl, Ogygoptynx, as intermediate between the Tytonidae and Strigidae. What can such forms represent phylogenetically? The usual solution is to suggest that the intermediate is some sort of “link” between the later taxa, perhaps ancestral to both of them. The problem of interpretation arises in part at least, from a failure to analyze the meaning of the individual characters, rather than just listing and totalling them. Rather, they should be interpreted in terms of the directions of evolution of evolving characters within the larger taxa; some “similarities” may then be found to be primitive character states, and some derived. If phylogenetic affinities are hypothesized only on the basis of the latter, much of the confusion from conflicting evidence could be eliminated. Of course there is the problem that many of the characters used in avian paleontology are minor variations in bony knobs, projections, grooves, etc., easily subject to convergence and difficult to analyze cladistically with much confidence. Still, it might be worthwhile to try. Several papers provide descriptons or reinterpretations of fossil birds and early avifaunas, and contribute to an increased understanding of the ecological roles of birds at various times and places in the past. Among the most significant are studies by Alan Feduccia and Larry D. Martin on Eocene Piciformes, and by Storrs L. Olson on Oligo- cene Coraciiformes. It is suggested that these groups were the dominant small land birds in these epochs, and that the ascendency of the Passeriformes to their current dominance may not have occurred until Miocene time. Among the analyses of more restricted areas, Oscar Arredondo’s review of the Pleistocene predatory birds of Cuba presents a remarkable picture of giant eagles, owds, and vultures analyzed in terms of their ecological relationships with the contemporary mammals. Joel Cracraft analyzes 358 THE WILSON BULLETIN • VoL 89, No. 2 the Moas of New Zealand and demonstrates how an adequate fossil record can be subjected to detailed mathematical analysis. He presents a story of adaptive radiation in these giant flightless birds, and also provides a new classification in which many fewer species are recognized than were admitted by earlier workers. Other noteworthy studies in this volume include reviews of the Lower Miocene swifts by Charles T. Collins, of the Pleistocene pied-billed grebes by Robert W. Storer, of a Pleistocene avifauna of Ecuador by Kenneth E. Campbell, Jr., and of the Paleogene birds of Asia by E. N. Kurochkin. Descriptions of a new Miocene osprey by Stewart L. Warter and of a new Miocene flightless auk by Hildegarde Howard are also included. The final paper, by G. Victor Morejohn, brings us to Recent times with the discovery in California Indian middens of bones from the flightless duck Chendytes lawi, previously known from Pleistocene remains. Altogether, this volume demonstrates that avian paleontology is alive and well, and making important contributions to several areas of systematic ornithology, though it is also troubled by some of the philosophical and methodological problems faced by sys- tematics generally. Students of avian biology owe a debt of gratitude to Storrs L. Olson for his work in organizing this outstanding collection of reports in avian paleontology, — Robert J. Raikow. A Guide to Eastern Hawk Watching. By Donald S. Heintzelman. Pennsylvania State University Press, University Park & London, 1976: 99 pp., maps, charts, and black- and-white photos. Cloth S8.95, paper S5.95. — Following brief written descriptions of each species, there are sections on field equipment, migration seasons, mechanics of hawk flights, and hawk lookouts. For each state or province there is information on hawk- watching localities, giving the quality of spring and fall observations, a description of the area, and directions for reaching it. Areas covered include the Great Lakes region and Eastern Canada, New England, the Middle Atlantic States, and the Southern Appalachian States. The book ends with appendices listing raptor conservation organizations, sample field data forms, as well as a short bibliography and index. The heart of the book con- sists of some 70 plates of raptors in flight; most of these are photographs but a few are line drawings. These will aid the observer in identifying flying birds. There are no illustrations in which different species are shown together for rapid comparison, but such are available in other books such as Peterson's field guides, to which the present volume should prove a handy supplement. — Robert J. Raikow, The Birds of the Ligonier Valley. By Robert C. Leberman. Special Publication No. 3, Carnegie Museum of Natural History, Pittsburgh, PA, 1976: 67 pp., many line drawings, 6 color plates, numerous photographs. $5.00. — This is an unusually handsome regional list. The Ligonier Valley lies in Southwestern Pennsylvania between the western- most ridges of the Allegheny Mountains; the area covered is about 100 square miles. Following brief introductions to the area and its ornithological history, the bulk of the book is devoted to species accounts giving the abundance and seasonal occurrences of all birds reported from the area. The line drawings by Carol H. Rudy are attractive and lifelike, illustrating many of the species discussed. H. Jon Janosik has contributed 6 outstanding paintings of locally observed forms, including a seldom-pictured immature Kirtland's Warbler, Unfortunately some of the brilliance of the original plates has been June 1977 • ORNITHOLOGICAL LITERATURE 359 subdued by the printer. The Ligonier Valley is an exceptional birding area, and anyone interested in exploring it will find this book an attractive and useful aid. — Robert J. Rai- KOW. Bird Flight Photography. By Roger F. Cram. Creative Arts Photography, P.O. Box 642, Hiram OH 44234: 35 pp., line drawings. $3.95. — Uncomplicated instructions for amateur bird photography using only simple and inexpensive equipment. Order directly by mail.- — R.J.R. Analysis of Vertebrate Populations. By Graeme Coughley. John Wiley & Sons, New York, 1977: ix + 234 pp. — This is a detailed guide to a study of the mechanics of population biology. Chapters are devoted to The Population; Age, Abundance; Rate of Increase; Dispersal; Fecundity; Mortality; Relationship between Parameters (calculation of r, birth rates, death rates, etc.) ; Mark-Recapture; and Population Analysis in Management. The first 3 chapters are so cursory as to be of little value, but from Chapter 4 (Abundance) onward this text becomes an interesting “how to” book for a study of basic population parameters. Many statistics, formulae and indices are given, and the level of the math is algebra or less. Great pains were taken to avoid even a hint of calculus (e.g. the fundamental growth equation, dN/dt = rN is only given in its algebraic form, and the differential is not even mentioned). Nevertheless, the mathematics are presented logically and lucidly. This work is valuable in that it pulls together methodological techniques and criticisms for most population parameters. By- avoiding complicated mathematics the volume has value to beginning students, specialists and field workers in conservation and game areas. This useful book should form a part of the library of most population biologists, particularly those interested in the analysis of field data and in various techniques of obtaining the data. — Michael A. Mares. My Recipes Are for the Birds. By Irene Cosgrove & Ed Cosgrove, Doubleday & Co., Inc., New York, 1976: 31 pages, paper cover. $2.95. — A handy collection of recipes for bird feeders designed to attract various species, with helpful advice on the design and placement of feeders. — R.J.R. Annotated Checklist of the Birds of Ontario. By R. D. James, P. L. McLaren, and J. C. Barlow. Life Sci. Misc. Pub., Royal Ontario Museum, 100 Queen’s Park, Toronto, Canada, 1976: 75 pp., paper covers, 2 maps. $2.50. — Status, breeding status, distribution, frequency, dates of occurrence, and egg dates for 427 species. — R.J.R. A Guide to Bird-Watching in Mallorca. By Eddie Watkinson. AB Grafisk Form- givning, Stockholm; Available from M. Philbrick, PO box 83, Vashon, WA 98070; no date given: 56 pp., paper covers, many maps. $3.90. — Not a guide to tbe birds, but to finding them. Detailed maps and instructions on many local areas, with information on how to get there and what birds to expect at any season, along with helpful advice on local customs, laws, transport, and so forth. — R.J.R. CONSERVATION COMMITTEE REPORT Falconry: Effects on Raptor Populations and Management IN North America Falconry can be defined as the sport of hunting with trained raptors. Historically, falconry referred to the training and use of falcons in hunting but now the term is used to describe the use of all raptors trained to take prey. Because of recent declines in population levels of some species of raptors such as the American Peregrine Falcon iFalco peregrinus anatum) , (Hickey 1969, Cade and Fyfe 1970), Cooper’s Hawks (Ac- cipiter cooperii) (Snyder et al. 1973), and use of falcons in entertainment (i.e, Atlanta Falcons Football Club and Air Force Academy Falcons Sports), concern has been ex- pressed since the early 1960’s about both the future of falconry as a sport and its impact on raptor populations. With listing of the American Peregrine Falcon as an endangered species on 8 March 1969 by the U.S. Fish and Wildlife Service, interest in the welfare of most species of raptors increased. Stringent controls and regulations on taking raptors and on the sport of falconry were instigated in many states in the early 1970's. The Sup- plementing Agreement to the Convention for the Protection of Migratory Birds and Game Mammals with Mexico in 1972, and subsequent drafts of proposed federal falconry per- mit regulations created controversy and conflict among falconers, environmental groups, state conservation agencies, and the U.S. Fish and Wildlife Service. The Conservation Committee of the Wilson Ornithological Society undertook this review of the sport of falconry, impacts of removing raptors from the wild, and the final form of federal fal- conry permit regulations (Federal Register 41 (10) :2237-2240, 41(37) :8053) issued in February 1976. FALCONRY The earliest records of falconry are from Mongolia and Egypt (about 2000 B.C.) where it probably originated not as an art or sport but as a method to secure food. The use of trained falcons for hunting evolved into an art and sport which reache4 its peak popu- larity in the Middle Ages. From the 8th to the 17th century, falconry flourished in Europe where the type of raptor permitted correlated with social class. After the Middle Ages, interest in falconry declined, perhaps because of firearms development and changes in social order (Nye 1966). Falconry developed slowly in North America, probably because it was not typical of immigrant peoples or appropriate to pioneer life. By the early 1900’s, interest slowly increased, stimulated by R. L. Meredith and an article in the National Geographic Maga- zine in 1920 by Louis Agassiz Fuertes (Nye 1966). As the result of efforts by Meredith and others to stimulate interest in falcons and falconry, the art and sport grew such that, by the late 1930’s, over 100 falconers were active in the United States. Growth of falconry was slow in the 1940’s and 1950’s but greatly increased in the 1960’s and early 1970's when there were an estimated 1500 active falconers. This can be partially at- tributed to increased publicity and legalization of the sport. In 1964 the U.S. Fish and Wildlife Service permitted the use of trained raptors in the taking of migratory game birds. A survey of states and Canadian provinces and territories by the Conservation Committee revealed that 6 of 11 provinces and territories responding and 34 of 47 states responding allow falconr>\ Supported by the federal regulations in 1976, the tradition of taking raptors for falconry and the “sport” have been solidly established by law. 360 June 1977 • CONSERVATION COMMITTEE REPORT 361 The Environmental Assessment prepared by the U.S. Fish and Wildlife Service (1976) on the proposed federal falconry regulations states that about 2769 falconry permits had been issued by 29 states in 1974 and projects that 59(X) falconers will be licensed in 1980 if all 50 states were to allow falconry. This estimated 2-fold increase may be realized because of previously unlicensed falconers applying for licenses, better law enforcement, and increased awareness of the sport. The Conservation Committee doubts the projected increase will occur. Falconry re- quires proper facilities, hard-won expertise, long hours, and dedication. We estimate conservatively that fewer than half of the about 2700 permittees presently licensed are active falconers. The remainder keep raptors, perhaps due to the novelty or supposed glamour of the activity. Administration of federal and state regulations on training, ex- perience, facilities, testing, record keeping and restrictions on the species of raptor al- lowed is such that license applications will not rapidly increase. Indeed, we estimate that the increase may approximate 40% by 1980 or about 5% per year. STATUS OF RAPTOR POPULATIONS The Environmental Assessment of the proposed federal falconry regulations lists 18 species of raptors of “importance” to falconry in North America, but only 6 species, in- cluding the Red-tailed Hawk iButeo jamaicensis) , Goshawk {Accipiter gentilis), Harris Hawk (Parabuteo unicinctus) , American Kestrel (Falco sparverius) , Prairie Falcon (F. mexicanus) and Peregrine Falcon represent most used at the present. Red-tailed Hawks are certainly the most used in North America. The status of any wildlife population may be evaluated in 3 general ways: (1) com- plete annual counts, (2) development of an abundance index in some stage of the life cycle, or (3) a life-equation approach which provides an indirect evaluation of the popu- lation’s condition. These approaches are not new and fall within the general outline pre- sented by Leopold (1933:139) . A complete count of a raptor population is difficult or impossible unless the area is small and few reliable long-term data are available. Indices of abundance based on counts of fall migrants are available but are confounded by weather, numbers of ob- servers, changes in counting sites, etc. We have relied primarily upon the life-equation approach to evaluate the status of selected raptor populations. A recruitment standard was estimated for most species discussed based on models. Knowledge of some parameters used in the models may be excellent but others are poorly understood. Nevertheless, recruitment standards, the long-term average production rate per breeding-age female required to maintain a stable population, have been estimated (Henny 1972; Henny and Wight 1972). The number of young fledged per successful nest, based primarily on banding records, provides a long-term index of production but does not reveal, of course, population decline since young are banded only at existing nests. Finally, production rates based on intensive short-term nesting studies were compared with the long-term recruitment standards. The importance of pesticides as a factor in population declines of raptors used in falconry was kept in mind in the review of available data on raptor popu- lations. Considerable progress has been made in the last decade demonstrating the relationship between population declines of several species of raptors useful for falconry' and pesti- cide contaminants. Soon after Ratcliffe’s (1967) initial report of eggshell thinning in British Peregrines and Sparrowhawks (Accipiter nisus) , similar thinning was discovered in North American Peregrines (Hickey and Anderson 1968), here associated with high levels of DDT and its metabolites, especially DDE. Measurements of the eggs of 362 THE WILSON BULLETIN • Vol. 89, No. 2 many other species of raptors were soon made, and reductions in eggshell thickness were found in several, especially Sharp-shinned Hawks {Accipiter striatus) , Cooper’s Hawks and Prairie Falcons (Anderson and Hickey 1972). Other species important in falconry, such as the Goshawk and Red-tailed Hawk, which feed primarily on mammals, were shown to be much less involved in the shell-thinning problem than species which feed heavily on birds. Peregrine and Prairie Falcons. — The correlation of DDE in egg contents and thin egg- shells has been clearly established for Peregrines (Cade et al. 1971) and Prairie Falcons (Fyfe et al. 1969; Enderson and Wrege 1973). Enderson and Berger (1970) showed that fledging success is inversely correlated with thin-shelled eggs in Prairie Falcons. Despite obvious correlations among DDE residues, eggshell thinning, and reproductive failure in some raptors, it is difficult to establish that these events are in fact the prime cause of population decline of such species as Peregrines (Hickey 1969). A major argu- ment has been that DDT and its metabolites were not abundant enough to account for declines apparent by 1950. Peakall (1974) cleverly demonstrated that DDE was present in amounts accounting for pronounced shell-thinning in Peregrine eggs in California as early as 1948 by extracting residues from shell membranes of museum eggs and com- paring the levels with more recent eggs for which the egg contents had also been analyzed. A major circumstance to be remembered is that pesticide contamination or population declines of Peregrines and Prairie Falcons have not been uniform throughout the range of these species in North America. Prairie Falcons declined 34% in occupancy of ter- ritories in western Canada with the reduction being concentrated in 4 of 6 areas studied (Fyfe et al. 1969), hut one report (Enderson 1969) and several unpublished accounts from the Rocky Mountain region reveal they are normally abundant and reproducing well. Peregrines disappeared as a breeding bird in the eastern United States, but are not declining and are not significantly contaminated with DDE in the Aleutian Islands (White et al. 1973) . Peregrines in North America, except for those in the Aleutian Islands, are experiencing reproductive difficulties due to DDE, and the threat is probably greater to this species than any other. Even in the Queen Charlotte Islands, British Columbia, Peregrine eggs contain about 3 times the DDE residues found in Aleutian Peregrine eggs (White et al. 1973). Fewer pairs nested in the Queen Charlottes in 1970 than in 1960 (Nelson 1970). Elsewhere in North America, populations are generally declining or steady with some reproductive difficulty apparent. In the Rocky Mountain region from northern New Mexico to Montana, 14 pairs fledged only 3 young in 1973 (Enderson and Craig 1974). In 1974 only 10 of those pairs could he found, hut they reproduced well despite thin- shelled eggs. In 1975 only 7 of the original 14 pairs could he found and fledging suc- cess was poor. A few pairs are known to persist in western Texas ( G. Hunt, pers. comm. » and in California. It is probably true that only between 20 and 30 pairs of Peregrines were known to nest in the contiguous United States in 1975. American Kestrel. — The status of the American Kestrel in North America was reviewed by Henny ( 1972) . The number of young banded per successful nest in the Northeast provided a crude index to production. The mean number handed per nest was 3.92 young for 1925-45. 3.92 for 1946-59, and 3.63 for 1960-68. The data for the latter period suggest a 7% decrease in young fledged per successful nest. Hackman and Henny (1971) also suggested a slight population decrease in the East. Lincer and Sherburne (1974) found relatively high levels of DDE in the eggs of Kestrels breeding near Ithaca, New York, and felt pesticides were being obtained in the wintering grounds in the southeastern United States. June 1977 • CONSERVATION COMMITTEE REPORT 363 The Environmental Assessment of the proposed falconry regulations mentions a recent increase in eastern populations, perhaps due to DDT bans. Kestrel eggshells were thinned 10% by one application of DDT at .84 kg per ha on 172,800 ha in the 1974 DDT-Tussock Moth Spray Program in the Pacific Northwest ( Henny et al. 1976, unpubl. report). This was accompanied by a 5-fold increase in residue levels of DDT and its metabolites in the eggs over those from controls 30-50 km from the spray area. Red-tailed Hawk. — The Red-tailed Hawk nests throughout much of North America and few reports of Red-tailed Hawk population declines appear in the literature. Further- more, Henny and Wight (1972:246) report no significant change in average production per successful nest during the DDT era in North America. They concluded that repro- duction is balancing mortality rates in the populations. Unlike other raptors, counts of migrant Red-tailed Hawks at White Marsh, Maryland during the 1950’s and early 1960‘s also remained stable (Hackman and Henny 1971). Anderson and Hickey (1972) found eggshell thinning exceeding 9% in 5 Montana eggs, but other eggshells from the United States were near normal. In recent years a number of short-term nesting studies of Red-tailed Hawks have been conducted. The observed production rates have approximated the long-term production standard estimated by Henny and Wight ( 1972). A six-year study ( 1966-71) in Saskatche- wan showed production slightly in excess of that believed required (Harris 1971), while a 3-year study (1967-69) in Alberta showed a population declining slightly (Luttich et al. 1971). Orians and Kuhlman (1956) reported slightly above normal production in Wisconsin in 1953-55, while Gates (1972) reported production slightly below' normal in east-central Wisconsin in 1962-64. Similarly, Seidensticker and Reynolds (1971) re- ported below-normal productivity and 10.9% shell thinning during a 2-year study , (1966-67) in south-central Montana. A more recent study (1971-72) in southwestern I Montana, however, shows normal production (Johnson 1975). Production in southern , California in 1973 appeared to be excellent in an area where man was not interfering with nesting (Wiley 1975). Annual production rates are variable, depending upon local con- ditions, but the results of the short-term studies in the pesticide era reveal the species is reproducing normally. Coopefs Hawk. — Accipiters experienced eggshell changes about twice as great as the buteos (Anderson and Hickey 1972) paralleling the reported declines of Cooper’s Hawks I and Sharp-shinned Hawks in eastern North America (Spofford 1969, Hackman and Henny 1971, Snyder et al. 1973) . Young banded per successful nest in the Northeast was taken as a crude index to production ( Henny and Wight 1972). The mean number banded was 3.53 young for 1929-45, 3.08 for 1946-48, and 2.67 for 1949-67. More recently, calcu- lations for 1968-74 show an improvement to 3.36 young (Henny, unpublished data). These findings accompany the contention of the Environmental Assessment that Cooper's Hawks have increased in the East during recent years. PROBLEMS ASSOCIATED WITH MANAGEMENT OF RAPTOR POPULATIONS Major problems associated with management of raptor populations include contamina- tion of food chains by pesticides, loss of habitats necessary for foraging and nesting, and man-caused mortality (White 1974). Unlike the role of contaminants in declines of some species of raptors, losses of nesting habitat, suitable foraging areas, and prey species are less well documented (White 1974, Braun et al. 1975) but may be important locally. Disturbance of nesting sites by photographers, banders, and the general public can re- duce hatching and fledging success iFyfe and Olendorff 1976). Alteration of nesting sites by nuclear tests has had similar results ( Stahlecker and Alldredge 1976). Shooting 364 THE WILSON BULLETIN • Vol. 89, No. 2 increases mortality of raptors despite nearly complete legal protection from this source. With increased awareness of values of raptors and enforcement of regulations, losses of all birds of prey through shooting appear to be low compared to earlier periods. Man- caused mortality by vehicles, electrocution, trapping, poisoning, and egg collecting ap- pears to be low and stable at present although data are lacking for definitive statements. These causes undoubtedly are responsible for some mortality but actual impacts on over- all populations are unknown. The Environmental Assessment of the proposed federal falconry regulations docu- mented little mortality of raptors trapped and transported for purposes of falconry. In a survey of British falconers, Kenward (1974) estimated that from 11 to 53% of raptors taken as nestlings, depending upon species, died in the first year of captivity. The sur- vey further indicated that between 50 and 93% of the trained raptors were eventually lost or released. Similar data are not available for North America but it can be safely assumed that deaths of birds taken as nestlings, especially by inexperienced people, are quite high and may exceed those values reported by Kenward (1974) . Cade (1968) estimated that from 25 to 50% of the annual production of Gyrfalcons (Falco rusticolus) in Iceland was exported under strict regulations for falconry each year for several centuries with no noticeable impact upon the breeding population. Blood (1968) reported that removal of about 12 nestling Peregrines annually from 1952 to 1967 from the Queen Charlotte Islands did not appear to adversely affect breeding Pere- grine populations. Exploitation rates of raptors for falconry appearing in the Environ- mental Assessment suggest that less than 0.5% of any species population is taken an- nually. If first-year mortality rates of most raptors in the wild approximate 40 to 60% ( Shor 1970, Luttich et al. 1971) and most raptors taken for falconry are first year birds, it is obvious that removal of raptors from the wild for falconry at even double the present levels could not be responsible for regionwide population declines. Recently, interest in management of raptors has increased, primarily the result of con- cern about local populations. Olendorff and Stoddart (1974) give an excellent state- ment of management ])ossibilities for grassland raptors, including the likely possibility of increasing nesting sites by providing man-made structures. Perhaps the largest and most successful attempt in providing artificial nesting sites was conducted in Michigan in 1967 where 43 nest platforms were constructed for Ospreys {Pandion haliaetus) . The majority of the platforms became occupied in later years and pairs using them experienced lower nestling mortality than pairs using the dwindling supply of natural sites (Postu- palsky and Stackpole 1974). In an area in California, the total number of Osprey nests producing young was increased by 37% over the previous 3 years after artificial nest platforms were erected (Garber et al. 1974). R. Fyfe (pers. comm.) has shown that Prairie Falcons nest readily in holes dug in dirt banks where none existed before. The most massive effort to restore reduced or extirpated populations is now underway for the Peregrine. There are several options available and all are difficult. Population re- covery of Peregrines could lie effected 1)> 3 means: (1) reduce DDE contamination through limiting or prohibiting the use of DDT, (2) artificially increase the production of young, and (3) reduce mortality rates. Reducing levels of long-lived pesticides in the environ- ment. especially outside of the United States, will l)e difficult and will necessitate inter- national cooperation. Artificial increase of fledging rates could be effected by manipulating eggs in nests in the wild or by manipulating young which have been bred in cap- tivity. For example, first clutches from nests in the wild could be removed and artificially incubated and second clutches will be produced. Further, eggs or young can be placed into the nests of wild Peregrines. This was done successfully in Colorado in 1973 and June 1977 • CONSERVATION COMMITTEE REPORT 365 is perhaps the best technique as long as wild nesting pairs can be found. Peregrine eggs or young can be put in the nests of other species. It has been shown that hawks can successfully fledge Prairie Falcon young (R. Fyfe, pers. comm.). In some regions the Prairie Falcon would presumably be a suitable foster species, but possible adverse effeets of “imprinting” by young Peregrines on their adult foster parents have not been determined. Young Peregrines can be released to the wild in the absence of adults by allowing them to fledge from a protected place and by supplying food until they are in- dependent. This procedure has already been used successfully by H. Meng and by Cornell University in reintroducing captive-produced Peregrines in the eastern United States, avoiding the possibility of “imprinting,” and can be used where no adult Pere- grine pairs are present. The third means involves the techniques of falconry. Qualified falconers could take wild Peregrine nestlings, train them to hunt and release them after the first or second winter. Good falconers keep a higher percentage of young birds alive than oceurs in the wild and perhaps twice as many eould be released as would be ex- pected to naturally survive (Cade 1974). Certainly there are many falconers capable of handling Peregrines successfully, but the scheme would require careful organization and so far there is little information relating to survival rates of trained falcons. There is little doubt that rehabilitation of Peregrine populations south of Canada will consist mainly of the release of captive-bred birds produced by a few large breeding projects. Smaller projects have thus far served to provide reservoirs of the appropriate subspecies and have been useful in developing some of the necessary techniques. People proficient in falconry have clearly been crucial in the progress made in breeding falcons in captivity and are certain to be indispensable to further population rehabilitation efforts. A major concern relates to the intentional or accidental release of exotic Peregrines in North America. One school contends only the form originally occurring in the area should be flown or released there. The other extreme contends that Peregrines are all substantially similar and that natural selection will determine the genetic attributes of any wild population, regardless of its origin. The argument is academic for the eastern United States since no captive stock of that former population exists and any Peregrines introduced there forming a viable population may well be better than no Peregrines at all. In the West, adequate captive stocks originating in the wild are being bred so that enzootic birds may be re-stocked. In any event. Peregrine populations in temperate North America will be to a major degree artificial in that they must be managed as long as DDE remains the dominant adversary. Recent success of the Cornell program for captive breeding of Peregrines indicates that production of 250 young Peregrines per year from captive sources is possible by 1980. While some young produced will be retained for eventual captive breeding, many will be available for reintroduction. It is also likely that captive bred falcons will be eventually available to falconers. Since most of the present stock of breeding Peregrines in captivity was obtained on loan from falconers, it is logical for them to expect to receive some of the progeny. Regulation of the disbursement and final disposition of captively bred raptors will be a major problem until firm understandings supported by appropriate legislation are reached. PERSPECTIVE Interest in falconry is expected to increase at a slow rate because of more stringent regu- lations at the federal, state, and provincial levels and because of the difficulty in main- taining and training captive birds. No appreciable impact on wild populations of any species is anticipated because species or subspecies listed as endangered cannot be legally 366 THE WILSON BULLETIN • Vol. 89, No. 2 used for falconry. Falconry will continue with the more common species such as Red- tailed Hawks and American Kestrels because other species are more difficult to acquire. Present official attitudes of most states and provinces surveyed reflect little apprehension that falconry is unmanageable or a drain on raptor populations. Some states and prov- inces believe that regulation of falconry is inordinately expensive in time and money because of demands of the few practitioners. SUMMARY The art of falconry in North America, practiced by a few individuals for many years, attracted little attention until the 1960’s. Presently about 2800 falconers are licensed in the United States with less than one half considered to be active. While interest in this art is expected to increase, we believe growth will be slow, probably 5 to 10% per year, due to rigorous demands on time and equipment required and restrictive regulations. Many different species of raptors have been used in falconry. Presently 6 species are commonly used, especially the Red-tailed Hawk and American Kestrel. Present evidence suggests that only 2 races of the Peregrine Falcon are threatened in North America, and declines may have occurred in local populations of other species. Declines in populations of Peregrines are attributed to pesticide contamination of food chains. Apparent de- clines in other populations of raptors are also attributed to pesticides and locally to changes in land use and possibly indiscriminate shooting. Removal of raptors from wild populations for falconry has not had documentable adverse effects except possibly at local nesting sites. Continuation of the art of falconry under the framework of the recent federal regulations is not expected to have measurable impacts on region-wide populations. Management of raptors is poorly developed and relatively unexplored. Captive breeding of raptors holds much promise for production of birds both for re-establishment and as a source of birds for falconry. Falconers have contributed much to the continued improve- ment of the Cornell University Peregrine program in terms of breeding stocks and tech- ni(jue development. RECOMMENDATIONS 1. Additional data are urgently needed for the monitoring of changes in raptor popula- tions over large areas. This may entail the development of new census techniques and establishment of national or continental surveys. 2. Falconry is a legitimate art and has a place in wildlife management and conserva- tion. The art should not be popularized. 3. The federal falconry regulations should be adopted immediately by all states as they represent the initial step in uniform regulation of falconry and the taking of raptors. The Conservation Committee does not see the need for placing Great Horned Owls (Bubo virginianus) under the falconry statutes. 4. It is crucial that a practical banding or tattooing system for permanently identi- fying individual captive raptors be immediately established since enforcement of falconry regulations will be exceedingly difficult without it. 5. Any species of raptor bred in captivity, including properly accounted for and marked endangered species, should be allowed for falconIy^ The effect will be to encourage captive breeding and will reduce dependence on wild populations. 6. Properly accounted for and marked birds used in falconry should be allowed free interstate transport provided proper state and federal permits have been obtained. June 1977 • CONSERVATION COMMITTEE REPORT 367 ACKNOWLEDGMENTS We acknowledge the contribution of the personnel of state and provincial Conserva- tion Agencies who responded to the questionnaire survey. Several individuals greatly facilitated the Committee by making data available which improved our knowledge of falconry and raptors. Specifically, our appreciation and thanks are extended to J. L. Ruos of the U.S. Fish and Wildlife Service and R. M. Stabler of Colorado College. LITERATURE CITED Anderson, D. W. and J. J. Hickey. 1972. Eggshell changes in certain North American birds. Proc. Int. Ornithol. Congr. 15:514-540. Blood, D. A. 1968. Population status of Peregrine Falcons in the Queen Charlotte Is- lands, British Columbia. Can. Field-Nat. 82:169-176. Braun, C. E., F. Hamerstrom, T. Ray, and C. M. White. 1975. Conservation com- mittee report on status of eagles. Wilson Bull. 87:140-143. Cade, T. J. 1968. The Gyrfalcon and falconry. Living Bird 7:237-240. . 1974. Plans for managing the survival of the Peregrine Falcon. Pp. 89-104, in Management of Raptors, Raptor Res. Found., Inc. Raptor Res. Rept. No. 2. AND R. Fyfe. 1970. The North American Peregrine survey, 1970. Can. Field- Nat. 83:191-200. , J. L. Linger, C. M. White, D. G. Roseneau, and L. G. Swartz. 1971. DDE residues and eggshell changes in Alaskan falcons and hawks. Science 172:955-957. Enderson, j. H. 1969. Prairie Falcon nesting success in Colorado in 1965. Pp. 362-363, in Peregrine Falcon populations: their biology and decline, (J. J. Hickey, ed.) Univ. Wisconsin Press, Madison. AND D. D. Berger. 1970. Pesticides: eggshell thinning and lowered production of young in Prairie Falcons. BioScience 20:355-356. AND G. Craig. 1974. Status of the Peregrine in the Rocky Mountains in 1973. Auk 91:727-736. AND P. H. Wrege. 1973. DDE residues and eggshell thickness in Prairie Fal- cons. J. Wildl. Manage. 37:476-478. Fyfe, R. W. and R. R. Olendorff. 1976. Minimizing the dangers of nesting studies to raptors and other sensitive species. Can. Wildl. Serv., Occas. Pap. No. 23. , J. Campbell, B. Hayson, and K. Hodson. 1969. Regional population declines and organochlorine insecticides in Canadian Prairie Falcons. Can. Field-Nat. 83: 191-200. Garber, P. G., J. R. Koplin, and J. R. Kahl. 1974. Osprey management on the Lassen National Forest, California. Pp. 119-122, in Management of Raptors, Raptor Res. Found., Inc., Raptor Res. Rept. No. 2. Gates, J. M. 1972. Red-tailed Hawk populations and ecology in east-central Wisconsin. Wilson Bull. 84:421-433. Hackman, C. D. and C. J. Henny. 1971. Hawk migration over White Marsh, Maryland. Chesapeake Sci. 12:137-141. Harris, W. C. 1971. Red-tailed Hawk nesting success, 1971. Blue Jay 29:203. Henny, C. J. 1972. An analysis of the population dynamics of selected avian species. U.S. Fish and Wildl. Serv., Res. Rep. 1. AND H. M. Wight. 1972. Population ecology and environmental pollution: Red- tailed and Cooper’s Hawks. U.S. Fish and Wildl. Serv., Res. Rep. 2, pp. 229-250. 368 THE WILSON BULLETIN • Vol. 89, No. 2 , M. W. Nelson, and S. R Gray. 1976. Impact of 1974 DDT spraying for tussock moth control on American Kestrels. Unpublished Report submitted to U.S. Forest Service, Portland, OR (mimeo.) 23 pp. (A copy has been deposited in the van Tyne library-.) Hickey, J. J. (Editor). 1969. Peregrine Falcon populations; their biology and de- cline. Univ. Wisconsin Press, Madison. AND D. W. Anderson. 1968. Chlorinated hydrocarbons and eggshell changes in raptorial and fish-eating birds. Science 162:271-273. Johnson, S. J. 1975. Productivity of the Red-tailed Hawk in southwestern Montana. Auk 92:732-736. Kenward, R. E. 1974. Mortality and fate of trained birds of prey. J. Wildl. Manage. 38:751-756. Leopold, A. 1933. Game management. Charles Scribner’s Sons, New York. Linger, J. L. and J. A. Sherburne. 1974. Organochlorines in Kestrel prey: a north- south dichotomy. J. Wildl. Manage. 38:427-434. Luttich, S. N., L. B. Keith, and J. D. Stephenson. 1971. Population dynamics of the Red-tailed Hawk (Buteo jamaicensis) at Rochester, Alberta. Auk 88:75-87. Nelscjn, R. W. 1970. Langara Island, Queen Charlotte Islands. In the North Ameri- can Peregrine survey, 1970, (T. Cade and R. Fyfe). Can. Field-Nat. 84:244-245. Nye, a. C., Jr. 1966. Falconry. Pp. 164-173, in Birds in Our Lives, ( A. Stefferud and A. L. Nelson, eds.) U.S. Dept. Interior, Bur. Sport Fish, and Wildl., Fish and Wildl. Serv. Olendorff, R. R. and J. W. Stoddart. 1974. The potential for management of raptor populations in western grasslands. Pp. 47-88, in Management of Raptors. Raptor Res. Found., Inc., Raptor Res. Rept. No. 2. Orians, G. and F. Kuhlman. 1956. Red-tailed Hawk and horned owl populations in Wisconsin. Condor 58:371-385. Peak ALL, D. B. 1974. DDE: its presence in Peregrine eggs in 1948. Science 183:673- 674. PosTUPALSKY, S. AND S. M. Stackpole. 1974. Artificial nesting platforms for Os- preys in Michigan. Pp. 105-117, in Management of Raptors. Raptor Res. Found., Inc. Raptor Res. Rept. No. 2. Ratcliffe, D. a. 1967. Decrease in eggshell weight in certain birds of prey. Nature 215:208-210. Seidensticker, j. C., IV and H. V. Reynolds, HI. 1971. The nesting, reproductive performance, and chlorinated hydrocarbon residues in the Red-tailed Hawk and Great Horned Owl in south-central Montana. Wilson Bull. 83:408-418. Shor, W. 1970. Peregrine Falcon population dynamics deduced from band recovery data. Raptor Res. News 4:49-59. Snyder, N. F., H. A. Snyder, J. L. Linger, and R. T. Reynolds. 1973. Organochlorines, heavy metals, and the biology of North American Accipiters. BioScience 23:300-305. Spofford, W. R. 1969. Hawk Mountain counts as population indices in northeastern America. Pp. 323-331, in Peregrine Falcon populations: their biology and decline, IJ. J. Hickey, ed.) Univ. Wisconsin Press, Madison. Stahlegker, D. W. and A. W. Alldredge. 1976. The impact of an underground nuclear fracturing experiment on cliff-nesting raptors. Wilson Bull. 88:151-154. U.S. Department of the Interior. 1976. Environmental assessment: proposed fal- conry regulations. Fish and Wildlife Serv., Washington, D.C. June 1977 • CONSERVATION COMMITTEE REPORT 369 White, C. M. 1974. Current problems and techniques in raptor management and con- servation. Trans. N. Am. Wildl. Nat. Resour. Conf. 39:301-312. , W. B. Emison, and F. S. L. Williamson. 1973. DDE in a resident Aleutian Island Peregrine population. Condor 75:306-311. ViLEY, J. W. 1975. The nesting and reproductive success of Red-tailed Hawks and Red-shouldered Hawks in Orange County, California, 1973. Condor 77:133-139. Conservation Committee Clait E. Braun, Chairman James H. Enderson Charles J. HexNny Heinz Meng Alva G. Nye, Jr. ORNITHOLOGICAL NEWS HAWK MOUNTAIN RESEARCH AWARD The Board of Directors of Hawk Mountain Sanctuar>- takes pleasure in announcing an annual award of S250 for support of raptor research. The Hawk Mountain Research Award will be granted annually to a student engaged in research on raptors ( Falconi- formes) . To apply, students should submit a description of their research program, a curricu- lum vitae, and 2 letters of recommendation by 31 October 1977 to: Mr, Alex Nagy, Hawk Mountain Sanctuary- Association, Route 2, Kempton, PA 19529. A final decision will be made by the Board of Directors in Februar\- 1978. Only students enrolled in a degree granting institution are eligible. Both under- graduate and graduate students are invited to apply. Projects will be judged com- petitively on the basis of their potential contribution to improved understanding of raptor biology and their ultimate relevance to conservation of North American hawk populations. COLONIAL W ATERBIRD GROUP MEETING The Colonial Waterbird Group, organized during the Wading Bird Conference at Charleston, South Carolina last October 1976, will hold its first annual meeting on 21-23 October 1977, at Northern Illinois University, in DeKalb. The conference will include paper sessions, subgroup meetings (surveys, conservation, etc.) and an important busi- ness session. Any person wishing to present a paper on an aspect of research or man- agement of pelicans, cormorants, herons, ibises, gulls, terns, alcids or other colonial waterbirds should submit a single page abstract no later than 15 August 1977 to the National Audubon Research Department, 115 Indian Mound Trail, Tavernier, FL 33070. Additional information on the conference will appear in the mid-summer CWG news- letter, or may be obtained by writing the above address. REQUESTS FOR ASSISTANCE Shorebird color-marking. — In 1977, the Canadian Wildlife Service will again be carry- ing out extensive banding and color-marking of shorebirds in James Bay. Last year, over 12,400 shorebirds were captured during July and August resulting in over 580 reports of color-marked birds in eastern North America and South America. Much valuable in- formation on migration routes is being obtained and observers are again asked to look out for and report any color-dyed or color-banded shorebirds that they may see. Reports should include details of species (with age if possible), place, date, color-marks, and if possible, notes on the numbers of other shorebirds present. For color-dyed birds, please record the color and area of the bird that was dyed. For color bands and standard metal leg bands, please record which leg the bands were on, whether they were above or below the “knee,” the colors involved, and the relative position of the bands if more than one was on a leg (e.g., right leg. blue over metal etc.). All reports will be acknowledged and should be sent to Dr. R.I.G. Morrison, Canadian Wildlife Service, 2721 Highway 31, Ottawa, Ontario, Canada KIA 0E7. International Shorebird Surveys, 1977-78. — A cooperative International Shorebird Survey scheme was started in 1975 to obtain information on shorebird migration and to identify and document areas of major importance. This scheme has been highly suc- 370 June 1977 • ORNITHOLOGICAL NEWS 371 cessful, with much very valuable information on shorebird distribution and migration coming from contributors throughout eastern Canada and the L.S.A., the Caribbean Islands, and Central and South America. Information from the scheme will be valuable in assessing requirements for the future protection and conservation of the birds and their habitat. In 1977 we are anxious to continue and extend the scheme in as many areas as possible. Any observer who may be able to participate in regular survey counts of shorebirds during spring and autumn migration periods, as well as during the winter in shorebird wintering areas, are asked to contact one of the undersigned. Occasional counts from observers visiting shorebird areas on an irregular basis would also be most welcome. For areas in Canada: Dr. R.I.G. Morrison, Canadian Wildlife Service, 2721 Highway 31, Ottawa, Ontario, Canada KIA 0E7. For areas in the U.S.A., Caribbean Islands, Central and South America: Brian A. Harrington. Manomet Bird Observatory, Manomet. MA 02345, U.S.A. Wing-tagged Laughing Gulls. — Juvenile Laughing Gulls have been wing-tagged with green or orange and green wing tags (and with metal leg bands) in Barnegat Bay. New Jersey to study behavior, migration, habitat selection, and survival rates. The tags are round (5 cm in diameter) and numbered. Please report all sightings to Bird Band- ing Laboratory, Office of Migratory Bird Management, Laurel. MD 20811. Please in- clude date, time, location, color of tag. and number of the tag if possible. Information may also be sent to Joanna Burger, Dept, of Biology. Livingston College. Rutgers Uni- versity, New Brunswick. NJ 08903. Bird-strip mine and bio-indicator literature ivanted. — A bibliography on birds found on strip mines and the use of birds as bio-indicators of the quality of the environment is being compiled for the Institute of Mining and Minerals Research in Kentucky. Any- one wishing to submit published or unpublished articles, reports, theses, etc., to be in- cluded in the bibliography should send them to: Pierre N. Allaire, Dept, of Science and Mathematics, Lees Junior College. Jackson, KY 41339. Needed: Egg date records for Purple Martins (Progne subis). — Please send date on which each martin pair at colony laid its 1st egg. Age of each pair (adult or subadult* if known and yearly total of martin young raised at colony also would be helpful. If dates represent 2nd nesting attempts or 2nd broods, please indicate. Records for as many years as possible are needed. Each contribution will be acknowledged. — Charles R. Brown, 2601 Turtle Creek Drive, Sherman, Texas 75090. This issue of The Wilson Bulletin was published on 27 June 1977. ■p: ■ 'M ' ■ -1'- 2j’i*r- jfilW ^r- Sr'j ' *r^ ‘*^rr '.'•Vf/HfJi'* .>Kti t'.i 'V *.i T Sit-l ’ l \ri 'v** i ' ’.u nlttlU9><^ * ' li'*>»”.-*ifl4 11? ' * ■ • •'■ ■wn^t-T(..i;“v .' V». V*. ■ j,’. *|f -vf> *, Ul'' . , >-s r ■? '■ ,-;y i ’ t- i>^ ^v- ■•' -i.r -ifl J»*.- 1 A**! .'t. '’st ’ x.mm » V i.' *5^ ‘V* ; ■; ♦• . -jk or *3 • ■ ^;- ' -v>| ■> '^■'•^4i ' ' ^ y i ^ei ■ '■ ■“ii , ‘ j* ''r?*‘ L--‘ ■■* ■*’ .^•^ ‘ r* ' ‘ ’■■ ■ -r' Iff ■ . ; ' - ;?. M *' ’•■ - ■..■•i*f ‘ » . . . S' _ . i V-* ■>. 4# ^>v hy^ .VLm‘V / 'I t. K' I"# , r ■ ,c- - rtf. • — ■' ■ .. * ' • « ' • f iTS i - • : J.ll The Wilson Bulletin Editor Jerome A. Jackson Department of Zoology P.O. Drawer Z Mississippi State University Mississippi State, MS 39762 Editorial Assistants Bette J. Schardien Patriqa Ramey Bonnie J. Turner Martha Hays Wayne C. Weber Review Editor Robert Raikow Color Plate Editor William A. Lunk Department of Life Sciences 865 North Wagner Road University of Pittsburgh Ann Arbor, MI 48103 Pittsburgh, PA 15213 Suggestions to Authors See Wilson Bulletin, 87:144, 1975 for more detailed “Suggestions to Authors.” Manuscripts intended for publication in The Wilson Bulletin should be submitted in dupli- cate, neatly typewritten, double-spaced, with at least 3 cm margins, and on one side only of good quality white paper. Do not submit xerographic copies that are made on slick, heavy paper. Tables should be typed on separate sheets, and should be narrow and deep rather than wide and shallow. Follow the AOU Check-list (Fifth Edition, 1957) and the 32nd Supplement (Auk, 90:411-419, 1973), insofar as scientific names of U.S. and Canadian birds are concerned. Summaries of major papers should be brief but quotable. Where fewer than 5 papers are cited, the citations may be included in the text. All citations in “General Notes” should be included in the text. Follow carefully the style used in this issue in listing the literature cited; otherwise, follow the “CBE Style Manual” (1972, AIBS). Photographs for illustrations should have good contrast and be on gloss paper. Submit prints unmounted and attach to each a brief but adequate legend. Do not write heavily on the backs of photographs. Diagrams and line drawings should be in black ink and their lettering large enough to permit reduction. Original figures or photographs submitted must be smaller than 22 X 28 cm. Alterations in copy after the type has been set must be charged to the author. Notice of Change of Address If your address changes, notify the Society immediately. Send your complete new address to the Treasurer, Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. He will notify the printer. The permanent mailing address of the Wilson Ornithological Society is: c/o The Museum of Zoology, The University of Michigan, Ann Arbor, Michigan 48104. Persons having business with any of the officers may address them at their various addresses given on the back of the front cover, and all matters pertaining to the Bulletin should be sent directly to the Editor. THE BREEDING BIOLOGY OF THE OAHU ‘eLEPAIO Sheila Conant 193 LIFE HISTORY FACTORS AFFECTING THE INTRINSIC RATE OF NATURAL INCREASE OF BIRDS OF THE DECIDUOUS FOREST BiOME Richard Brewer and Lynda Swander 211 NESTING HABITAT OF COMMON RAVENS TN VIRGINIA Robert G. Hooper 233 GROWTH AND DEVELOPMENT OF KNOWN-AGE RING-BILLED GULL EMBRYOS John P. Ryder and Lynn Somppi 243 HABITAT PARTITIONING IN A COMMUNITY OF PASSERINE BIRDS Robert C. Whitmore 253 || BREEDING DISPLAYS OF THE LOUISIANA HERON Jamen A. Rodgers, Jr. 266 ' ■ BREEDING BIOLOGY OF CLIFF SWALLOWS IN VIRGINIA — . Gilbert S. Grant and Thomas L. Quay 286 WEATHER INFLUENCES ON NOCTURNAL BIRD MORTALITY AT A NORTH DAKOTA TOWER Michael Avery, Paul F. Springer, and J. Frank Cassel 291 BREEDING BIOLOGY OF HOUSE SPARROWS IN NORTH MISSISSIPPI James N. Sappington 300 NESTING BEHAVIOR OF YELLOW-BELLIED sAPSUCKERs Lawrence KUham 310 GENERAL NOTES SEX DIFFERENCES IN ALARM RESPONSES OF WINTERING EVENING GROSBEAKS Martha Hatch Ralph NESTING OF TURKEY AND BLACK VULTURES IN PANAMA Lauric A. McHargUe 325 328 4 FULVOUS WHISTLING DUCK POPULATIONS IN TEXAS AND LOUISIANA Edward L. Flickinger, David S. Lobpries, and Hugh A. Bateman 329 SLIPPER SHELLS, A MAJOR FOOD ITEM FOR WHITE-WINGED SCOTERS James G. Hoff 331 1 EGG MOVEMENT BY A FEMALE GADWALL BETWEEN NEST BOWLS Robert F. Johnson, Jr. and Leo M. Kirsch 331 FOODS OF WESTERN CLAPPER RAILS R. D. Ohmart and R. E. Tomlinson 332 AGGRESSION IN FORAGING MIGRANT SEMIPALMATED SANDPIPERS Brian A. Harrington and Sarah Groves 1 336 HERRING GULL EATING BAYBERRY Roger F. Pasquier 338 THE LESSER ANTILLEAN BULLFINCH IN THE VIRGIN ISLANDS Herbert A. Raffaele and Daniel Roby 338 FORAGING BEHAVIOR OF THE WHITE IBIS — - James A. Kushlan 342 BIRDS OF FIVE FAMILIES FEEDING FROM SPIDER WEBS Robert B. W aide and Jack P. Mailman 345 1 WINTER NEST MICROCLIMATE OF MONK PARAKEETS Donald F. Caccamise and Wesley W. Weathers SNAKE PREDATION ON bell’s vireo NESTLINGS - - Calvin L. Gink t 1 346 I 349 CROW PREDATION ON BLACK-CROWNED NIGHT HERON EGGS Joanna Burger and D. Caldwell Hahn 350 ORNITHOLOGICAL LITERATURE - - - — 352 CONSERVATION COMMITTEE REPORT (Falconry: Effects on Raptor Populations and Management in North America) i 360 ORNITHOLOGICAL NEWS — 370 TIieWlsonBulletin PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY VOL. 89, NO. 3 SEPTEMBER 1977 PAGES 373-520 COMP. ZOOl:, library OCT 1 7 1977 harvard UHIVKRSITY The Wilson Ornithological Society Founded December 3, 1888 Named after ALEXANDER WILSON, the first American Ornithologist. President — Douglas A. James, Department of Zoology, University of Arkansas, Fayetteville, Arkansas 72703. First Vice-President — George A. Hall, Department of Chemistry, West Virginia Univer- sity, Morgantown, W. Va. 26506. Second Vice-President — Abbot S. Gaunt, Department of Zoology, Ohio State University, Columbus, Ohio 43210. Editor — Jerome A. Jackson, Department of Zoology, P.O. Drawer Z, Mississippi State Uni- versity, Mississippi State, Mississippi 39762 Secretary — James Tate, Jr., P.O. Box 2043, Denver, Colorado 80201. Treasurer — Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. Elected Council Members — Sidney A. Gauthreaux, Jr. (term expires 1978) ; James R. Karr (term expires 1979); Clait E. Braun (term expires 1980). Membership dues per calendar- year are: Active, $10.00; Sustaining, $15.00; Life memberships, $200 (payable in four installments). The Wilson Bulletin is sent to all members not in arrears for dues. The Josselyn Van Tyne Memorial Library The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of Michigan Museum of Zoology, was established in concurrence with the University of Michigan in 1930. Until 1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines from members and friends of the Society. Now two members have generously established a fund for the purchase of new books; members and friends are invited to maintain the fund by regular contribution, thus making available to all Society members the more important new books on ornithology and related subjects. The fund will be administered by the Library Committee, which will be happy to receive suggestions on the choice of new books to be added to the Library. William A. Lunk, University Museums, University of Michi- gan, is Chairman of the Committee. The Library currently receives 104 periodicals as gifts and in exchange for The Wilson Bulletin. With the usual exception of rare books, any item in the Library may be borrowed by members of the Society and will be sent prepaid (by the University of Michigan) to any address in the United States, its possessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers, as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to “The Josselyn Van Tyne Memorial Library, University of Michigan Museum of Zoology, Ann Arbor, Michigan.” Contributions to the New Book Fund should be sent to the Treasurer (small sums in stamps are acceptable). A complete index of the Library's holdings was printed in the September 1952 issue of The Wilson Bulletin and newly acquired books are listed periodically. The Wilson Bulletin The official organ of the Wilson Ornithological Society, published quarterly, in March, June, September, and December. The subscription price, both in the United States and elsewhere, is S15.00 per year. Single copies, S4.00. Subscriptions, changes of address and claims for undelivered copies should be sent to the Treasurer. Most back issues of the Bulletin are available and may be ordered from the Treasurer. Special prices will be quoted for quantity orders. All articles and communications for publications, books and publications for reviews should be addressed to the Editor. Exchanges should be addressed to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, Michigan. Known office of publication: Department of Zoology, Mississippi State University, Mississippi State, Mississippi 37962. Second class postage paid at Mississippi State, Mississippi and at additional mailing office. Allen Press, Inc., Lawrence, Kansas 66044 THE WILSON BULLETIN A QUARTERLY MAGAZINE OF ORNITHOLOGY Published by the Wilson Ornithological Society VoL. 89, No. 3 September 1977 Pages 373-520 NESTING HABITAT OF BACHMAN’S WARBLER^A REVIEW Robert G. Hooper and Paul B. Hamel Bachman’s Warbler {V ermivora bachmanii) is on the verge of extinction; no populations are known. Recent systematic searches in widely separated parts of its range failed to locate nesting birds (Stevenson 1972, pers. comm., Hamel et al. 1976). The 2 most recently published sightings were apparently of transient birds as each was seen only once (Imhof 1973, American Ornithologists’ Union 1975). Most observations of the bird were made in the late 1800’s and early 1900’s when little attention was given to habitat. Often habitat descriptions were ambiguous and thus, misleading when con- sidered alone. By comparing the varied descriptions collectively, a clearer conception of the warbler’s habitat evolves. Much of the information on the nesting habitat of Bachman’s Warbler is unpublished and generally not known to exist. We reviewed published and unpublished descriptions of habitat in order to better identify the most likely breeding areas. REPORTS OF nests We know of 40 nests of Bachman’s Warbler that have been found. Some mention of habitat was made in the account of 32. Eleven of these descriptions were published. The remaining 21 habitat descriptions were found in the field notes of A. T. Wayne which are on file at the Charleston Natural History Museum. We examined Wayne’s notes from 1901, when he rediscovered the bird in South Carolina (Wayne 1901), to 1928 when he apparently stopped taking notes. We believe the set of field notes is complete (A. E. Sanders pers. comm.) ; but, Wayne did not record 3 of the nests in his notes. Observations on the same nest were often made on different days; how- ever, Wayne apparently collected all Bachman’s Warbler nests he found and recorded the names of persons to whom he sent them. This information along with radically different habitat descriptions allowed accurate determination of the number of nests located. Wayne found at least 35 nests in TOn Swamp, near Charleston, South Carolina, between 1906 and 1919. Bachman’s Warbler (V ermivora bachmanii), male above and female below. Watercolor and acrylic painting by Sidney A. Gauthreaux, Jr. 374 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 Widmann (1897, 1899) reported 2 nests from Missouri. One nest was reported from Kentucky (Embody 1907) and 2 from Alabama (Holt 1920, Stevenson 1938). A nest from Georgia originally reported as that of Bachman’s Warbler (Arnow 1908) was later attributed to Swainson’s Warbler {Limn othly pis swainsonii) by Wayne (1912). Burleigh (1958) credited the Georgia nest to Bachman’s Warbler but we accepted Wayne’s judgment. With the exception of records of Barnes (1954), Dawn (1958), and Chamber- lain (1958), we disregarded sightings of the bird when a nest was not found. In these cases a male was observed throughout much of the nesting season and was apparently on territory even though presence of a nest or female was not confirmed. RESULTS AND DISCUSSION The swamp habitat. — All Bachman’s Warbler nests were found in low, wet forested areas. The 35 nests Wayne found in South Carolina were in the headwater swamp and adjacent wet flats of I’On Swamp. The other 5 nests were in bottomland forests along coastal plain rivers. Twelve of the descrip- tions of nest sites mentioned water, but as water levels fluctuate rapidly in both headwater swamps and bottomlands, the lack of water at other nest sites is of unknown significance. Also, it is possible that Wayne failed to record in his brief notes the proximity of water to some nests. Descriptions of the overstory vegetation near Bachman’s Warbler nests were poor. Wayne (1907) said of the area where his first 6 nests were found, “The trees are chiefly of a deciduous character, such as the cypress, black gum, sweet gum, tupelo, hickory, dogwood, and red oak. In the higher parts of the swamp short-leaf pines [probably P. taeda]., water oaks, live oaks, and magnolias abound.” At least some of the other nests Wayne found were in different parts of I’On Swamp up to 2.4 km from the site of the first nests. However, no further reference was made to the overstory trees. Widmann (1897) described the forests of the St. Francis River basin but failed to associate the site of his nest discovery with any particular forest community. Embody (1907) found, “The tulip tree, sweet and black gums, sycamore, elm and various oaks occur in abundance.” He, too, was talking of the bottomland not the actual nest site. Holt ( 1920) said, “The burn (where the nest was located ) was surrounded by the virgin swamp growth of Pinus taeda. Magnolia virginiana, Pieris nitida, Ilex coriacea, Persea and other hydrophytic vegetation.” Stevenson ( 1938 ) mentioned that elm ( Ulmus alata) and cherry {Primus serotina) were near the nest. rOn Swamp, the headwater swamp where Wayne found 35 nests, is a complex of forest communities. Non-alluvial headwater swamps occur in the interstream areas of the lower coastal plain where poorly developed drainage patterns and depressions impound rainwater and seepage. Wet flats are the better drained areas bordering the swamps ( Stubbs 1966). The deep water Hooper and Hamel • BACHMAN’S WARBLER HABITAT 375 areas in the headwater swamps are forested with swamp tupelo [Nyssa sylvatica var. bijlora) and bald cypress {Taxodium distichum) with almost no understory. Shallower areas support pond cypress [T. ascendens) and swamp tupelo with an impoverished understory, again offering little potential for nesting habitat. The next drier zone is forested by the sweet bay {Mag- nolia virginiana) — swamp tupelo — red maple {Acer rubrum) association that often has a moderate to dense understory. Slightly drier sites, the wet flats, support sweet gum {Liquidambar styraciflua) , willow oak (Quercus phellos) , numerous other hardwoods and scattered loblolly pine (P. taeda) . Drier sites support other hardwood associations that grade into loblolly pine sites. The driest areas are forested in longleaf pine (P. palustris). All of these conditions occur in FOn Swamp. Stands of the sweet bay — swamp tupelo — red maple association and the wet flats seem the most probable areas where Wayne found nests. Current flood plains, also called first bottoms, are dissected by former stream channels that create sloughs, which are generally the last part of the bottom to dry. Sloughs are forested by bald cypress and water tupelo (A^. aquatica) stands that, because of the inhibiting effect of long standing water, have poorly developed shrub layers and thus, are probably poor habitat for the warbler. The remainder of the first bottoms and the slightly higher former flood plains or terraces are forested by some 70 species of trees (Putnam et al. 1960) including those listed by Widmann (1897), Embody (1907), Holt (1920), Stevenson (1938), and Barnes (1954). The drier portions of the first bottoms and wetter areas of the terraces seem to best fit the descriptions given by those workers. It appears that Bachman’s Warbler used the portions of the river bottoms and headwater swamps that were inundated for relatively short periods compared to the lowest and wettest areas. The territory in a longleaf — loblolly pine stand (Dawn 1958, Chamberlain 1958) was the most radical deviation from the use of such areas; however, nesting was not determined. In view of all nests being found in swamps and bottomlands, there is little reason to believe that pine stands were of much importance to the bird as nesting habitat. Serai condition of the over story. — Only twice was reference made to virgin or primeval condition of the forest (Wayne 1907, Holt 1920). Because the area was surrounded by virgin forest, it is important that the site of the nest Holt (1920) found had undergone disturbance: “The nest was ... in a small burned-over area covered with a thin, new growth of blackberry briers.” Wayne spoke of the primeval swamp in which he worked as having an understory that was “chiefly cane, aquatic bushes, and swamp palmetto. 376 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 while patches of blackberry brambles and thorny vines are met with at almost every step.” The patches of blackberry (Rubus spp.) are evidence that disturbance had occurred in the overstory: these are relatively shade intolerant plants that appear to grow best in full sunlight. In hardwood stands, cane (Arundinaria tecta) develops into thick patches following disturbance to the overstory (Hughes 1957). In the case of virgin forests, the disturbance to the overstory would have been due to natural mortality of the trees. The description by Widmann (1897) of the area in which he found a nest also indicated disturbance: “...two acres of blackberry brambles... medley of half-decayed and lately felled tree tops . . . steaming under a broiling sun. . . .” Stevenson (1938) found a nest “. . . in a thicket between two branches of an unused logging road . . .” indicating an opening in the overstory. There is no evidence that the swamp where Wayne (1907) found his nests was virgin forest. A study by D. B. Urbston and D. R. Mudge (pers. comm.) of former land use patterns in TOn Swamp indicated only 6% of the swamp hardwood forest was over 30 years old in 1900-1920. The swamp had been cultivated extensively for rice, the practice of which declined rapidly after 1865. Outlines of the former fields show clearly on aerial photographs and dikes are prominent landscape features. There is evidence that Wayne found some of the nests in abandoned rice fields. The year prior to his initial nest discovery in 1906, Wayne wrote to Brewster (1905), “I have at last found a breeding ground of Bachman’s Warbler. The locality ... is very swampy and was originally a rice field, but is now covered with a dense forest of deciduous trees with innumerable patches of low bushes and blackberry brambles.” Of his discoveries in 1906 Wayne ( 1907 ) wrote, “. . . I made a special effort to find the nest and eggs of this rare warbler, and knowing that the birds which I had seen and did not molest in 1905 would return to the same swamp to breed the following spring, I determined to devote my entire time with the hope of finding a nest.” Thus, Wayne’s first 6 nests and several later ones were probably in or on the edge of former rice fields that were undergoing natural reforestation. His statement of primeval swamp most likely referred to the tough working conditions and not to the serai stage of the forest. Although a thinned or sparse overstory is impLed, it is not clear what range in tree densities the warbler used. L nderstory conditions. — Several species of understory plants were found repeatedly supporting or concealing nests (Table 1). All identified species, with possible exception of supplejack {Berchcmia scandens) and American Hooper and Hamel • BACHMAN’S WARBLER HABITAT 377 Table 1 Frequency of Understory Plants in Descriptions of Nest Sites of Bachman’s Warbler Species Wayne Widmann Embody Holt Stevenson Total Cane 14 1 1 16 Palmetto (Sabal minor) ^ 8 8 Gallberry (Ilex coriacea)^ 8 8 Unidentified shrub 6 6 Unidentified vine 3 3 Greenbrier (Smilax spp.) 7 7 Blackberry 3 2 1 1 1 8 Grape (Vitis spp.) 2 1 3 American holly 1 1 Supplejack 1 1 Fetterbush (Lyonia lucida) 1 1 Number of nests 27 2 1 1 1 32 ^ May include S. palmetto. - May include I. glabra. holly (Ilex opaca) , form dense thickets when the overstory is open. As discussed in the previous section, the overstory appeared to be open, and a relatively dense understory was implied. Additional evidence of the density of the understory was given by Wayne who used the terms “dense swamp,” “dense thicket” and the like 11 times and mentioned that 4 nests were in a “tangle” of cane and vines. Embody (1907) also referred to a tangle of cane. Holt (1920) found a nest beside a path that had been cut through a black- berry patch. Thus, the idea of a dense understory is reinforced. The mean height of 29 nests was about .6 m and the range .3 to 1.2 m suggests that the understory was dense from near ground level to 1 m or higher. Dense understories are ephemeral, perhaps explaining the disappearance of the bird from known nesting sites in the time when it was still relatively abundant. To us, the range in density of the understory that the warbler used for nesting is one of the largest gaps in knowledge of its habitat. Another poorly defined factor is the size of the thicket chosen for placement of the nest. Widmann (1897) spoke of a 0.8 ha patch of blackberry but area is difficult to estimate without practice. We have found experienced observers tend to overestimate distance and area in stands with a dense understory. Therefore, we feel Widmann could have overestimated the size of the opening in which he found the nest. The thicket in which Stevenson (1938) found a nest was less than .02 ha. Other records do not shed light on this question. 378 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 SUMMARY Between 1897 and 1919, 40 nests of Bachman’s Warbler were found in the Southern Coastal Plain. Thirty-five were in headwater swamps and adjacent wet flats in South Carolina and 5 were in bottomlands along rivers in Missouri, Kentucky and Alabama. From descriptions of nesting habitat at 32 of the sites, it appears the bird used the portions of the bottomlands and headwater swamps that were inundated for relatively short periods compared to the lowest and wettest areas. The sweet bay — swamp tupelo — red maple association of the headwater swamps, the sweetgum — willow oak association of the wet flats and the bottomland hardwoods of the first bottoms and terraces form the complex of plant communities most probably used as nesting habitat. The overstory of areas chosen for nesting appeared to have been subjected to disturbance, either natural or man caused, that stimulated development of a relatively dense understory of shrubs, palmetto, and cane. Many of the nests found in South Carolina could have been in or on the edge of abandoned rice fields that were undergoing secondary' succession. ACKNOWLEDGMENTS We are grateful to A. E. Sanders and the Charleston Museum for providing us access to A. T. Wayne’s field notes. E. B. Chamberlain, S. A. Gauthreaux, 0. G. Langdon, B. Meanley, M. Monahan, and H. M. Stevenson reviewed the manuscript. D. A. James also made helpful suggestions. Parts of the work have been financed by grants to Hamel from the Forest Service, National Audubon Society, Charleston Natural History Society and Clemson University. We gratefully acknowledge the George M. Sutton Color Plate Fund for making possible the publication of Sidney Gauthreaux’s painting of Bachman’s Warbler which accompanies this paper. LITERATURE CITED Amkrican Ornithologists’ Union. 1975. Report of the Committee on Conservation. Auk 92 (4, Suppl.) :1B-16B. Arnow, I. F. 1908. Bachman’s Warbler in Camden Co. and breeding in Chatham Co., Georgia. Auk 25:479. Barnes, I. R. 1954. A new look at Bachman’s Warbler. Atl. Nat. 9:18-30. Brewster, W. 1905. Notes on the breeding of Bachman’s Warbler, Helminthophila bachmanii (Aud.) near Charleston, South Carolina, with a description of the first plumage of the species. Auk 22:392-394. Burleigh, T. D. 1958. Georgia birds. Univ. Okla. Press, Norman. Chamberlain, E. B. 1958. Bachman’s Warbler in South Carolina. Chat 22:73-74, 77. Dawn, W. 1958. An anomalous Bachman’s Warbler. Atl. Nat. 13:229-232. Embody, G. C. 1907. Bachman’s Warbler breeding in Logan County, Kentucky. Auk 24:41-42. Hamel, P. B., R. G. Hooper and L. M. Wright. 1976. Where is the Reverend Bachman’s Warbler? S. C. Wildl. 23:9-13. Holt, E. G. 1920. Bachman’s Warbler breeding in Alabama. Auk 37:103-104. Hughes, R. H. 1957. Response of cane to burning in the North Carolina Coastal Plain. N. C. Agr. Exp. Stn. Bull. 402. Imhof, T. a. 1973. Spring migration: Central Southern Region. Am. Birds 27:782- 785. Hooper and Hamel • BACHMAN’S WARBLER HABITAT 379 Putnam, J. A., G. A. Furnival and J. S. McKnight. 1960. Management and inventory of southern hardwoods. USDA For. Serv. Agric. Handh. 181. Stevenson, H. M. 1938. Bachman’s Warbler in Alabama. Wilson Bull. 50:36-41. . 1972. The recent history of Bachman’s Warbler. Wilson Bull. 84:344—347. Stubbs, J. 1966. Hardwood type-site relationships on the Coastal Plain. In Proc. Symposium on hardwoods of the Piedmont and Coastal Plain. Ga. For. Res. Counc., Macon. Wayne, A. T. 1901. Bachman’s Warbler (Helminthophila bachmanii) rediscovered near Charleston, South Carolina. Auk 18:274-275. . 1907. The nest and eggs of Bachman’s Warbler, Helminthophila bachmanii (Aud.), taken near Charleston, South Carolina. Auk 24:43-48. . 1912. Bachman’s Warbler in Camden Co. and breeding in Chatham Co., Georgia. Auk 29:105. WiDMANN, 0. 1897. The summer home of Bachman’s Warbler no longer unknown. Auk 14:305-310. . 1899. Bachman’s Warbler in summer. Osprey 3:13. USDA, FOREST SERVICE, SOUTHEASTERN FOREST EXPERIMENT STATION, CLEMSON, SC 29631; DEPT. OF ZOOLOGY, CLEMSON UNIV., CLEMSON, SC 29631. AC- CEPTED 22 NOV. 1976. MORPHOLOGICAL VARIATION IN NORTH AMERICAN PINE GROSBEAKS Curtis S. Adkisson The Pine Grosbeak (Pinicola enucleator) is a resident of the taiga of North America from Newfoundland to the Bering Sea, and southward in the western mountains to California, Arizona, and New Mexico. Within this range it shows geographic variation in size (Jenks 1938). The breeding range of one population in California and another in the Queen Charlotte Islands are isolated from the other populations. The limits of the forms described from the taiga are only approximately known. In this paper I present an analysis of morphological geographic variation in this species. In another paper I will describe the geographic variation in calls and songs, and attempt to use data from morphological and vocal varia- tion to draw some conclusions about the patterns of evolution in North American Pine Grosbeaks. METHODS I analyzed morphological variation using measurements of 9 characters from study skins: wing length (chord) ; upper mandible length (distal end of nostril to tip) ; lower mandible length (1, middle of ramus fork to tip; 2, exposed proximo-lateral notch at the corner of mouth to tip) ; lower mandible width at widest point; bill depth; tarsus length (outer side of proximal joint to the base of the toes) ; and tail length (base of the central pair of rectrices to tip of the longest rectrix). Weight was used in the analysis when available from specimen labels. In addition to these measurements, age, sex, plumage color, and comments of the collector were recorded. Only breeding-season adults were used in the analysis. Those taken before 1 May and after 15 August were excluded on the grounds that they might be migrants. Data were collected from 487 specimens from throughout the species’ range in North America and analyzed with analysis of variance and regression programs written by the Statistical Research Laboratory of the University of Michigan. Regression analysis and scatter plots were used to test for clinal variation. In addition, product-moment correlation coefficients were computed to test for possible concordance in all characters. RESULTS Data are presented for 9 morphological characters of birds within the following regions: the taiga from Newfoundland to western Alaska; coastal Alaska; the Queen Charlotte Islands; California; and the Rocky Mountains to northern British Columbia (Fig. 1, Table 2). I subdivided the range of the species to correspond to previously described subspecies’ ranges. Each of these regions appeared to contain separate, isolated populations, judging by 380 Adkisson • VARIATION IN PINE GROSBEAKS 38] Fig. 1. Map of the breeding range of North American Pine Grosbeaks. Dots indicate localities where birds used in this study were collected. Wing lengths are included for several isolated localities and for certain 2° blocks (arrows point to locality in center of the block). The wing length average for Newfoundland includes 5 neighboring 2° blocks. Subspecific names and ranges (delineated by lines) are from the A.O.U. Check- List (1957). the literature (Cowan 1939, Rand 1946, A.O.U. 1957) and by the geographic origin of available specimens. In addition I divided the species’ range into 2-degree blocks, and computed character means for each block containing 5 or more birds, making it possible to look for trends in variation over large areas. This technique proved of only limited use, since few of the blocks contained enough birds. Wing length means for some blocks are included in Fig. 1. Analysis of variance for the above regions shows significant geographic variation for all 9 characters (all p < .0001). Linear product-moment corre- lation coefficients of the means of each population indicate that variations in wing and tail lengths are positively correlated, as are the bill measurements. Table 1 gives a correlation matrix for the large Newfoundland sample. Weight data from most areas are unavailable. Only the Rocky Mountain sample has 382 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 Correlation Matrix Table 1 FOR 8 Size Characters in Pine Grosbeaks From Newfoundland Tail Tarsus Upper Mand. Lower Mand. 1 Lower Mand. 2 Mand. Width Bill Depth WING .787*** .111 .149 .220* .205* .233* .176 (100)1 (101) (98) (99) (99) (99) (55) TAIL .004 .098 .138 .119 .217* .149 (100) (97) (98) (98) (98) (54) TARSUS .104 .031 .105 .215* .138 (98) (99) (99) (99) (55) UPPER .572*** .522*** .314** .159 MAND. (97) (97 ) (96) (53) LOWER .697*** .403*** .047 MAND.(l) (97) (97) (53) LOWER .268** .221 MAND. (2) (97) (54) MAND. .328** WIDTH (55) *, **, *** SiKnificance levels for correlation coefficients: .05, .01, .001, respectively. ' Sample sizes in parentheses. a sufficient number of weights for correlation analysis, and in this population weight and wing length are positively correlated ir = .341, n = 31, p < .058). Pine Grosbeaks vary clinally in the Rocky Mountains and in the taiga. Nevertheless it is useful for comparisons to provide sample statistics, for subregions of the taiga as well as for the other regions, for each of the 4 least correlated size characters used in this study (Table 2j. From these data the following generalizations can be made. The longest bills are found in coastal Alaska, the shortest in the Alaskan taiga. The narrowest bills are found in the California population followed by those of the eastern taiga and the Queen Charlottes. Bill depth and width are greatest in the taiga west of Hudson Bay and least in California and the Queen Charlottes. Wing length is greatest in the western taiga and the Rocky Mountains, and least in the Queen Charlottes and eastern Maritime provinces. The tarsus is longest in coastal Alaska birds, closely followed by that of some birds from the taiga and Rocky Mountains, and is the shortest in birds from the Queen Char- lottes. The populations of the Maritime provinces, California, Queen Char- lottes, and coastal Alaska are all very distinct. Birds from the coast of Alaska in turn can be distinguished from those of the interior region using all size characters except bill width and depth. Adkisson • VARIATION IN PINE GROSBEAKS 383 Table 2 Pine Grosbeak Size Data for 8 North American Regions Wing Length ( ad. males ) ( mm ) Tarsus length (mm) N X S.D. N X S.D. Maritime Provinces 70 111.68 2.298 134 22.30 0.600 Labrador Peninsula 31 115.16 3.072 73 22.51 0.780 Taiga, Ont. to NWT 17 119.57 3.305 40 22.50 0.630 Alaskan Taiga 16 118.66 2.673 24 22.37 0.712 Coastal Alaska 6 115.23 2.331 25 22.54 0.803 Rocky Mountains 56 118.00 3.395 124 22.44 0.802 Queen Char. Islands 7 107.53 3.333 23 21.39 0.706 California 15 114.19 1.354 30 22.08 0.821 Upper Mand. Length Lower Mand. Width N X S.D. N X S.D. Maritime Provinces 130 11.37 0.484 132 9.35 0.340 Labrador Peninsula 73 11.40 0..505 70 9.78 0.340 Taiga, Ont. to NWT 40 11.69 0.647 37 10.04 0.369 Alaskan Taiga 25 10.96 0.622 25 10.19 0.318 Coastal Alaska 25 12.08 0.544 25 10.14 0.353 Rocky Mountains 124 11.83 0.633 122 9.66 0.307 Queen Char. Islands 22 11.36 0.402 22 9.50 0.286 California 29 11.45 0.418 28 8.70 0.352 The most interesting patterns of variation occur in specimens from the taiga from Newfoundland to western Alaska, and the northern Rocky Moun- tains. Birds taken from the Maritime provinces and the Gaspe Peninsula of Quebec form a homogeneous population for each of the size characters. Comparison of 2° block means from this area shows no differences. Wing length of birds from this area is among the smallest for the species. However, on the Labrador Peninsula small birds indistinguishable from those of Newfoundland, and much larger birds, have been collected. The largest and smallest adult males differ in wing length by 15%. Most of the birds taken on the Labrador Peninsula north of about 54°N are inseparable from those of the taiga population west of Hudson Bay, while most of those taken from near the St. Lawrence River to about 52 °N are inseparable from Maritime provinces birds. Regressions on latitude for wing and tail lengths in adult males are significant ip < .001). Figure 2 reveals no evidence of discontin- uities in the dine of wing length in the Labrador Peninsula sample. There is no well-defined pattern of variation in other size characters, nor is there any east-west variation in the birds of the Maritimes, Labrador, and Quebec. 3&4 THE WIL-^ON BLLLETIX * J ol. 89, \o. 3, September 1977 Y = 3^.BB -0-577 x Fig. 2. Plot of wing length on latitude in Quebec and Labrador. | ^ est of Hudion Bay. the Pine Grosbeaks are among the largest in this ' species. The region from Moosonee. Ontario, to Great Slave Lake is poorly collected, but a trend toward larger body size appears to begin in northwestern ■ Ontario 199 Total No. nests 15 21 23 22 15 14 9 5 8 4 136 No. eggs 47 61 72 69 45 44 28 17 28 14 425 No. hatched 39 46 53 55 36 26 19 12 17 8 311 Hatched/egg 0.83 0.75 0.74 0.80 0.80 0.59 0.68 0.71 0.61 0.57 0.73 No. fledged 24 28 28 34 25 17 14 9 9 4 192 Fledged/hatched 0.62 0.61 0.53 0.62 0.69 0.65 0.74 0.75 0.53 0.50 0.62 No. successful nests 10 11 13 15 12 8 6 3 5 2 85 % suceessful 0.67 0.52 0.56 0.68 0.80 0.57 0.67 0.60 0.62 0.50 0.62 provides the least dense vegetation cover, nests were constructed further from the surface of the plant and closer to the ground than were nests in the other 2 shrubs. I compared the relative levels of breeding success among nests in these 3 shrub species. There were no significant differences in clutch size, brood size, nor the number of young fledged per nest. Notwithstanding the rather distinctive nest site characteristics of the 3 shrub species used as nest substrate, there was no evidence that breeding success was related directly to any of these differences. Nests were categorized according to the magnitude of each measured nest site characteristic. Ratios of the number of young hatched to the number of eggs laid (H/E) and the number of young fledged to the number of young hatched (F/H) within each category were used as an index to breeding success. In Table 2 these indices along with the constituent data are presented relative to the height of the nest above the ground. There was a general decrease in the H/E ratio with increasing nest height. The F/H ratio increased steadily from the lower nest heights to the 160-179 cm nest height category after which breeding success decreased. A linear regression of the H/E ratio to nest height was significant (Fig. la); however, a simple relationship was not apparent for F/H ratio versus nest height (Fig. le). The H/E ratio and vegetation height (Table 3) were negatively correlated indicating a general decline in hatching success with increasing vegetation height (Fig. lb). There was, however, no obvious relationship between the F/H ratio and vegetation height (Fig. If). Direct relationships between percent nest cover and nest density and the H/E and F/H ratios were not evident (Fig. Ic, d, g, h). Caccamise • RED-WINGED BLACKBIRD NEST SUCCESS 399 ^cm) [cm) (cm) (nests/, gio m') Fig. 1. Relationships between hatching and fledging success and nest site char- acteristics. Mortality and nest site characteristics. — The 3 greatest sources of mortality to eggs and nestlings during this study were predation, abandonment, and death-in-nest (Caccamise 1976). Predation includes losses of both eggs and nestlings. Losses ascribed to abandonment include nest desertion and also apparent abandonment resulting from death of the adult. Losses from death- in-nest represent primarily nestling starvation, although a small percentage of losses in this category likely result from sources such as overcrowding Breeding Success Relative Table 3 : TO Vegetation Height at THE Nest Vegetation height (cm) <140 140- 149 150- 159 160- 169 170- 179 180- 189 190- 199 - 200- 219 >219 Total No. nests 7 11 20 19 27 14 9 12 17 136 No. eggs 22 34 57 60 90 37 28 40 57 425 No. hatched 21 24 40 45 83 29 17 24 28 311 Hatched/egg 0.95 0.71 0.70 0.75 0.92 0.78 0.61 0.60 0.49 0.73 No. fledged Fledged/ 14 11 29 27 47 17 14 17 16 192 hatched No. successful 0.67 0.46 0.72 0.60 0.57 0.59 0.82 0.71 0.57 0.62 nests 5 5 14 10 19 9 6 8 9 85 % successful 0.71 0.45 0.70 0.53 0.70 0.64 0.67 0.67 0.53 0.62 400 THE WILSON BULLETIN • Vol 89, No. 3, September 1977 Table 4 Comparisons of Mean Nest Site Characteristics Among Nests Incurring Mortality (M) and Nests Not Incurring Mortality (S) From the 3 Greatest Soluces of Nesting Mortality Sources of Mortality' Abandonment Death-in-nest Predation Nest Site Characteristics M(31)i S(135) M(48) S(64) M(39) S(97) Nest height (cm) 120 N.S.-’ 127 128 N.S. 123 136 N.S. 126 Vegetation height ( cm ) 174 N.S. 176 174 N.S. 176 185 * 173 Vegetation cover (cm) 53 N.S. 50 48 N.S. 50 50 N.S. 50 % Cover 72.7 N.S. 72.7 70.0 N.S. 70.5 75.5 N.S. 72.5 Distance to nearest nest (m) 15.3 N.S. 17.8 12.4 19.7 21.7 N.S. 16.4 Nest density C nests/ 18 10 my) 4.8 N.S. 4.1 5.0 * 3.6 4.0 N.S. 4.2 ^ Sample size, - Level of significance attained 0.05 level. using Student’s f-test; N.S. not significant; * significant at the and/or nest eviction. Since death-in-nest refers only to nestlings, those nests which did not successfully hatch young were not included in these comparisons. For each nest site characteristic I compared nests incurring losses with those nests not incurring losses from each source of mortality (Table 4). Analyses indicated that nests incurring losses attributed to the death-in-nest category were characterized by a significantly smaller mean distance to the nearest nest and a significantly greater nest density. Also nests with losses to predators were in shrubs significantly taller than nests without such losses. None of the other comparisons were significant. DISCUSSION For the Red-winged Blackbird there have been many attempts to relate nest site characteristics to various measures of breeding success (Meanley and W ebb 1963, Goddard and Board 1967, Holcomb and Twiest 1968, Robertson 1972, Holm 1973). In reanalyzing the nest height data from several of these studies and others, Francis (1971) found that some of the differences in breeding success, as related to nest height, were in fact not significant. He further suggested that the greater nesting success reported by Holcomb and Twiest (1968) in the higher nests, although significant, mav have been related to differences in nest substrate rather than nest height. In mv study, nest site characteristics differed markedly for nests placed in the 3 plant species used as nest substrate. Additionally, as there were Caccamise • RED-WINGED BLACKBIRD NEST SUCCESS 401 differences in the occurrence of the 3 shrub species according to the marsh elevation, the nests in these shrubs also reflected the differences in habitat distribution among the shrubs. However there were no detectable differences in breeding success among nests located in the 3 different plant species. Therefore it would seem that the differences in nest site characteristics specifically related to the differences in growth form among these plant species did not directly affect breeding success. It is not clear why hatching success (H/E) decreased with increasing nest height and with increasing shrub height. Since nests incurring losses to predators were located in shrubs significantly taller than nests not receiving such losses (Table 4l, it would appear that the nests in tall shrubs were more susceptible to losses to predators. hile predation was not the only factor contributing to the differential hatching success, differences in preda- tion pressure probably were important. The reasons why relationships similar to those for hatching success did not apply between fledgling success (F/H) and nest height or vegetation height are obscure. However there is a major difference between eggs and nestlings in that eggs exhibit no behavior while nestlings do. Thus nestling behavior could affect predation as well as other sources of mortality. Both measures of nest density indicated that nests incurring losses in the death-in-nest category were in areas of significantly greater nest density than nests not incurring such losses. In the current study the manifestation of nestling starvation was similar to descriptions of Red-wing nestling starvation appearing in the literature (Robertson 1973). Often starvation was noted initially when a nestling appeared somewhat smaller than its siblings, show- ing an increasing size disparitv over a period of several days and eventually either disappearing or being found dead in the bottom of the nest. At my study site territories of 1 to several females were maintained along the road borders in the dense stands of shrubs. Most foraging was done off the territories often along the numerous creeks and potholes common in the grass dominated areas of the marsh. It seems unlikely that the greater indices of nest density for nests incurring mortality from nestling starvation (Table 4) were related to overexploitation of food resources in high density areas. This is because the distance between areas of high nest density and low nest density was generally small compared to what might be considered the potential range in which a female could effectively forage. Orians (1961) pointed out that the amount of food obtained by a female on the territory is inversely related to the size of the territory; however, he also suggests that “it is doubtful whether food per se is the proximate factor by which territory size is regulated.” Since nest dispersion at my study site was a linear array of nesting clumps along the road margin, each female had access to a large 402 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 area of undefended marsh in which to forage. Considering the very large size of the foraging area relative to the number of Red-winged Blackbirds in this colony, it seems unlikely that food abundance would be significantly reduced only in the foraging areas used by females from areas of high nest density. Alternatively, the one factor that always increases directly with nest density is the potential for intraspecific interactions in the vicinity of the nest. In areas of high nest density there would be considerable opportunity for such interactions while in areas of low nest density there would be very little opportunity. Female Red-winged Blackbirds, when in polygamous associa- tions with a single male, will defend territories within the male’s territory ( Nero 1956; Nero and Emlen 1951). Thereby they exert an active role in determining the number of females able to breed in a specific territory ( Orians 1961, Holm 1973). Robertson’s results (1973), indicating similar levels of starvation between low density upland areas and high density marsh areas, could be interpreted as mitigating against the possible role of aggression in nestling starvation. However, his results are based on averages over separate colonies or groups of colonies. Because these broad comparisons were designed to contrast habitat differences they ignore variations between individual nests. Such individual nest site characterizations are the basis for my study. Since aggressive interactions between females seem to fill an important role in the mating system of the Red-winged Blackbird, excessive levels of aggression in areas of high nest density could impair the female’s ability to nourish her young. Whether this would be sufficient to increase nestling starvation is open to conjuncture. However, the contribution of intraspecific aggression in determining breeding success certainly merits further study. SUMMARY Breeding success was assessed and nest site characteristics were measured in a colony of Red-winged Blackbirds nesting on a tidal salt marsh. Breeding success was not directly related to the plant species used as nest substrate, vegetation height over the nest, or density of cover over the nest. However, negative correlations were found between hatching success and both nest height and height of the shrub used as nest substrate. Nests incurring losses from nestling starvation were characterized as being in areas of significantly greater nest density than nests not incurring such losses. Also significant differences were evident in vegetation height between nests incurring predation and nests not incurring predation. ACKNOWLEDGMENTS For their aid in the field, I would like to thank Peter J. Alexandro and Charles Wagg. This is a paper of the Journal Series, New Jersey Agricultural Experiment Station, Cook College, Rutgers-The State University of New Jersey. Caccimise • RED-WL\GED BLACKBIRD NEST SUCCESS 403 LITERATURE CITED Caccamise. D. F. 1976. Nesting mortality in the Red-winged Blackbird. Auk 93: 517-534. Francis, W. J. 1971. An evaluation of reported reproductive success in Red-winged Blackbirds. Wilson Bull. 83:178-185. CoDDARD, S. V. AND V. V. BoARD. 1967. Reproductive success of Red-winged Blackbirds in North Central Oklahoma. Wilson Bull. 79:283-289. Holcomb, L. C. and C. Twiest. 1968. Ecological factors affecting nest building in Red-winged Blackbirds. Bird-Banding. 39:14-32. Holm, C. H. 1973. Breeding sex ratios, territoriality and reproductive success in the Red-winged Blackbird (Agelaius phoeniceus) . Ecology 54:356-365. Meanley, B. and J. S. Webb. 1%3. Nesting ecology- and reporductive rate of the Red- winged Blackbird in tidal marshes of the upper Chesapeake Bay region. Chesapeake Sci. 4:90-100. Nero, R. W. 1956. A behavior study of the Red-winged Blackbird. 2. Territoriality. Wilson Bull. 68:5-37. AND J. T. Emlen, Jr. 1951. An experimental study of territorial behavior in breeding Red-winged Blackbirds. Condor 53:105-106. Orians, C. H. 1961. The ecology of blackbird (Agelaius) social systems. Ecol. Monogr. 31:285-312. AND M. F. Willson. 1964. Interspecific territories of birds. Ecology 45:736— 745. Robertson, R. J. 1972. Optimal niche space of the Red-winged Blackbird (Agelaius phoeniceus) . 1. Nesting success in marsh and upland habitat. Can. J. Zool. 50:247- 263. . 1973. Optimal niche space of the Red-winged Blackbird. 3. Growth rate and food of nestlings in marsh and upland habitat. Wilson Bull. 85:209-222. DEPT. OF ENTOMOLOGY AND ECONOMIC ZOOLOGY, RUTGERS THE STATE UNIV. OF NEW JERSEY, NEW BRUNSWICK, 08903. ACCEPTED 1 JUNE 1976. FORAGING BEHAVIOR OF THE EASTERN BLUEBIRD Benedict C. Pinkowski Avian foraging behavior is known to vary intraspecifically with habitat (Root 1967), weather iLunk 1962:15), season (Ligon 1973), prey avail- ability (Morton 1967), and from one population to another (Ligon 1968). Few studies, however, have examined the variety of factors influencing predatory behavior of a single species. I found the Eastern Bluebird (Sialia sialis) a good subject for such an investigation because this species forages in relatively open areas and is conspicuous from a distance. Moreover, bluebirds employ a variety of foraging tactics (Bent 1949:247) but typically use a lookout perch to locate prey on the ground ( Preston and McCormick 1948, Krieg 1971 ) ; several parameters of the perch-feeding technique ( perch height, predator-to-prey distance) can easily be quantified. In this paper I describe predatory behavior in the Eastern Bluebird and examine the effects of several environmental variables on bluebird foraging. METHODS Study area. — Observations were made in Macomb Co., southeastern Michigan (42°48'N, 82°59'W) during 1972 and 1973. The bluebird nesting period extended from late March to early August (Pinkowski 1975a) and most observations were made during the nesting season. Fifty nest boxes were available in the study area and bluebirds nested in these as well as in natural cavities (Pinkowski 1976a). In Michigan most bluebirds migrate south in winter; spring migrants first appear in early March with a peak arrival period occurring between 20 March and 20 April. Six pairs of bluebirds were randomly observed at all times of the day, under all types of weather conditions, and in various stages of the nesting cycle. The bluebirds foraged in old fields (Fig. 1) characterized by hawkweed {Hieracium sp.), vetch {Vida sp.), sheep sorrel {Rumex acetosella) , goats-beard {Tragopogon major), cinquefoil (Poten- tilla sp.), daisy fleahane iErigeron philadelphicus) , oxeye daisy {Chrysanthemum leucanthemum) , and various grasses. This is a low-growdng, perennial sere that appears late in field succession in Michigan (Beckwith 1954). Common foraging perches were tree limbs and branches (especially if dead or defoliated, with oaks, Quercus sp., commonly employed), fence posts, boulders, and coarse weed stalks such as mullein {Verbascum sp. ) and evening-primrose (Oenothera sp.j. Foraging bluebirds were watched from a distance without disturbance and the presence of an observer did not alter their behavior in any way (cf. Krieg 1971:5). Measurements. — Several parameters were measured on a foraging sequence, including the type of foraging tactic employed. A short “drop” to the ground (the “flydown” described for 5. sialis by Goldman 1975) was most common. Measurements made on feeding drops w'ere: perch height (vertical distance of the bird above the ground), drop base (ground level distance from a point immediately below the take-off perch to the landing location), hypotenuse of the resulting triangle (predator-to-prey distance), and 404 Pinkowski • FORAGING OF EASTERN BLUEBIRDS 405 Fig. 1. View showing the old field flora with scattered trees and shrubs in the south- eastern Michigan study area. distance between consecutive feeding perches (measured for birds moving along fence- rows having predetermined distances between fence posts) . Not all perches resulted in the bird locating prey and not all drops resulted in prey capture. A perch was con- sidered successful if a drop was made from it (prey was sighted but not necessarily captured) and a drop was considered successful if food was obtained. The latter was often impossible to determine with certainty, especially if small prey were pursued. Each observation period lasted 1-2 h. Temperature was recorded in the field at the beginning and end of each observation period and the average value was assumed for all observations made during the period. Percentage of sunshine was obtained for each observation period by noting the proportion of time that shadows were cast. Wind speed could not be measured by instrument because the birds often foraged in valleys or behind wind breaks where wind speed was quite different than elsewhere. I esti- mated wind speed at a foraging site according to the effect of wind on feeding perches and foraging bluebirds. Light winds were those not causing noticeable movement of perches (tree limbs and weed stalks) and approximated actual speeds up to 5 km/h. Moderate winds (5 to 20-30 km/h) caused perch movement but did not interfere with foraging. Strong winds (over 20-30 km/h) caused at least some perches to move rapidly and be unacceptable as lookout posts. Statistical procedures. — Percentages were examined for significant differences by a 406 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 t-test for the equality of percentages (Sokal and Rohlf 1969:607). I follow Verbeek (1975) in defining feeding tactic diversity (ETD) by the formula ETD =z -2i pOnpi, where ps is the proportion of feeding involving the i^'* feeding tactic. Unless otherwise stated, Chi-square tests on contingency tables employ Yates correction for continuity with d.f. = 1. RESULTS AND DISCUSSION Description of foraging tactics. — Dropping is the principal feeding mode of the bluebird. The ground is searched from a conspicuous perch and after locating prey, the bird sallies onto the ground and seizes its prey with the bill. Rarely is more than one food item obtained on a single drop to the ground. The food may be swallowed on the ground or brought to a perch for preparation and ingestion, but it is never held with the feet during capture and preparation. Only 21.7% of 189 small (< 1 cm long) food items were taken to a perch before ingestion compared to 88.5% of 43 large (> 2 cm long ) food items. These percentages differ significantly (t = 10.3, P < 0.001). Hunting bluebirds normally search the ground while perched upright. During inclement weather and more intensive feeding the head is lowered and the tail elevated. If low perches are not available bluebirds may perch horizontally part way up tree trunks or weed stalks to view the ground. When close to the ground bluebirds often turn the head and use monocular vision to search the ground. Binocular vision is frequently employed at relatively great heights. By changing perches when no food source is found, the bird is able to encounter a large number of possible foraging situations. Fly catching involves capturing aerial insects by short flights into the air from a perch (usually the “new perch-short flight” pattern; Leek 1971), by more extended flights (“new perch-long flight”), or by seizing aerial prey without taking flight. I found that more than one item may be obtained per flight, and several aerial insects were fed to nestlings after a single flight, hut Marshall (1957) and Krieg (1971) reported that only one item was captured per flight. Flycatching may temporarily become the only foraging lactic, as on summer evenings when aerial insects are highly visible in the long-angled sunlight (Morton 1967), after a rain, or at other times when certain prev species (e.g., swarming carpenter ants, Carnponotus sp.) are abundant. Gleaning occurs when the bird lands on and removes prey from the foliage and branches of trees or shrubs, or the main trunks of trees. Verbeek (1975), working with tyrannid flycatchers, defined gleaning as “capture of an insect sitting on any kind of substrate”; here, “gleaning” excludes prey capture on the ground. In early summer bluebirds glean small caterpillars (e.g., geometrids and pierids ) from the leaves of trees. Many hymenopterans. Pinkowski • FORAGING OF EASTERN BLUEBIRDS 407 clipterans, coleopterans, and plecopterans (see Pinkowski 1976b ) are obtained from tree trunks by gleaning. Flight- gleaning is a modification of the dropping tactic and has been described for kingbirds (Tyrannus sp.) by Smith (1966:219). The l)ird descends toward the ground after locating prey, but remains in flight while plucking prey from vegetation. It may flutter briefly while inspecting the prey, but it never does so before locating an item; this sets flight-gleaning apart from hovering, a search method observed in Mountain Bluebirds (S. currucoides) by Griddle (1927), Power (1966), and Pinkowski (1975b) but not observed in Eastern Bluebirds during this study. Flight-gleaning is employed in areas of tall weeds and therefore becomes more common as the season advances and vegetation height increases. Hopping is not a common feeding mode. Except for flycatching (new perch-long flight), it is the only foraging tactic wherein the prey is not located from a conspicuous perch. When feeding by this method a bluebird moves along the ground and feeds upon prey that is encountered after it lands on the ground. Hopping is limited to roadways, recently plowed farm- lands, lawns, burnt areas, and other disturbed habitats that have few perches and sparse ground cover. Of 2638 foraging sequences observed during March through June, 78.8% were accomplished by the dropping mode, a slightly lower percentage than that ( 87.4% ) observed by Goldman ( 1975 ) for bluebirds feeding on lawns in Ohio. Flycatching and gleaning were more common foraging tactics (10.7% and 6.8%, respectively) than hopping (2.6%) and drop-gleaning (1.1%). Feeding on fruit. — Beal (1915) found that up to 57.6% of the diet of S. sialis may consist of fruit during winter. I noted that bluebirds rely heavily on fruit sources in late summer and immediately after their arrival in early spring. Staghorn sumac {Rhus typhina) and multiflora rose {Rosa multi- flora) are the common fruits eaten in spring. Honeysuckle { Lonicera sp. ) , cherry {Prunus sp. ), and mulberry (Morus sp. ) are eaten in summer. Three distinct methods of obtaining fruit are employed; a bluebird may ( T) hover in the air while ingesting berries (analogous to and employing the same motor patterns as flight-gleaning) ; (2) perch on a limb and pluck berries from an adjacent limb (similar to gleaning) ; or (3) perch on a fruit head (e.g., staghorn sumac) and pluck fruit from directly beneath its feet (not unlike securing animal prey on the ground after a drop). Feeding tactics and season. — Although the relative frequencies of feeding tactics used by Eastern Bluebirds vary during the nesting period, dropping is the principal tactic employed in all seasons (Table 1). Frequency of the 408 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 Table 1 Seasonal Variation in Feeding Tactics of Eastern Bluebirds in Southeastern Michigan, 1972-1973“ Percentage Occurrence March April May June (N=584) (N=595) (N=770) (N=689) Dropping 99.5 86.6 88.6 43.8 Flycatching 0.4 9.4*’ 5.6 26.3 Gleaning O.I 0.4 1.2 24.2 Hopping 0.0 3.7 3.9 2.4 Drop-gleaning 0.0 0.0 0.8 3.4 “ Based on 12 birds. Most records ( 35 of 56) obtained during one observation. dropping tactic decreases in summer as vegetation height increases. A sparse ground cover is required for effective feeding by bluebirds using the drop- ping mode. After the breeding period bluebirds regularly hunted on mowed lawns around residences adjacent to the study area and evidently preferred such places to undisturbed areas containing tall vegetation. Feeding tactic diversity is lowest in March (FTD = 0.034), is higher in April and May (0.491 and 0.487, respectively), and increases markedly in June (1.261). The increase in diversity is the result of more aerial feeding late in the season as food resources are increasingly exploited in a third (vertical) dimension. Willson (1974) characterized the Eastern Bluebird as an insectivore that feeds by sallying in the low vegetation stratum (a member of the “insectivore, low, sally” guild; see Root 1967). In spring “insectivore, ground, ground glean” adequately describes the species, but by the end of summer much fruit is consumed and “omnivore, low, sally” is probably more accurate. Feeding modes and weather. — Feeding tactics were found to vary in frequency according to weather conditions. Of 89 feedings recorded in May and June during exceptionally cold (0-10°C), cloudy, and rainy or damp weather, 64 (71.9%) were accomplished by the dropping mode. A nearly identical percentage of dropping mode sequences was observed for the same period during warm (15-25°C), sunny, favorable weather (72.2%, N = 251). Flycatching was more common during favorable weather (14.7%) than during inclement weather ( 1.2%, t — 2.2, P < 0.05 ) whereas the reverse was true for gleaning (20.2% and 7.9% for inclement and favorable weather, respectively; t — 2.0, P < 0.05 ) . Flycatching is not a common feeding mode during excessively windy conditions, probably because aerial insects are reduced in number at these Pinkowski • FORAGING OF EASTERN BLUEBIRDS 409 Seasonal Variation IN Table 2 IN Foraging Measurements of Eastern Southeastern Michigan, 1972-1973“ Bluebirds Spring Summer Mean ± SD Mean ± SD Base (m) 3.22 ± 1.85 7.04 ± 5.71 Height (m) 2.02 ± 1.09 3.76 ± 2.43 Predator-prey Distance (m) 3.97 ± 1.88 8.26 ± 5.81 “ Values obtained by triangulations on the dropping tactic based on 12 birds with N = 100 for each period. The means of each measurement are significantly different (P <0.001, Mann- Whitney U-test). times (Freeman 1945). In May and June flycatching accounted for only 2.0% (N = 99) of all foraging sequences during strong winds, a significantly smaller percentage it = 2.2, P < 0.05) than that observed during moderate or light winds (13.6%, N = 1360). Thus seasonal changes in foraging tactics are similar to weather-mediated responses in that flycatching is employed more often during favorable (warm, sunny, and calm) weather and as the season advances. Presumably more aerial insects are available in favorable weather and later in the season. Factors influencing foraging height. — Measurements were made on 100 bluebird drops randomly observed in the early spring (15 March to 15 April) and 100 in the early summer (15 May to 15 June) to examine factors influencing foraging height and predator-to-prey distances. Each sample was evenly divided between males and females. Bluebirds forage closer to the ground and consume prey located nearer to their perches in spring than in summer (Table 2). Seasonal variation in foraging height may be attributable to smaller, fewer, or less active vernal insects that are more difficult to detect at greater heights. Also, as noted above, more aerial feeding occurs in summer and the birds may adjust their foraging heights accordingly. As a consequence of the greater foraging area that each perch affords in summer, fewer perches are required later in the season. Bluebirds inhabit more open habitats during summer than spring, perhaps because of a reduced dependence on perches as the season progresses. Pooled data for the spring and summer periods show positive regression when drop base (B) is plotted on foraging height (H) according to the relationship B = 1.17 + 1.20H (measurements in m). The slope of the regression differs significantly from 0 {F = 76.2, P<< 0.001, r^ = 0.28), indicating that bluebirds search areas more distant from the perch when foraging at greater heights. Evidently the area searched (the “perceptual 410 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 field”; Holling 1966, Salt 1967) more closely approximates a narrow annulus rather than all of the area within a circle as might be expected, and increases in length ( L ) according to the relationship L = 27tB = 7.4 + 7.5H. The significant relationship between foraging height and drop base also suggests that the search angle A, defined here as A = tan"^B/H, remains relatively constant. In spring and summer the search angle averages 58° (tan“^ 3.22/2.02; Table 2) and 62° (tan“^ 7.04/3.76), respectively. Over the normal range of foraging heights (1-10 m) the search angle varies from 67° (tan"^ 2.37/1) to 53° (tan"^ 13.17/10) and is surprisingly constant in view of the wide range of foraging heights and bases. Deviations from the mean search angle may occur because of the deviations from the upright posture normally assumed by perch-feeding bluebirds, different head posi- tions relative to the body (particularly as related to monocular or binocular viewing of the ground ) , or different perch inclinations relative to the ground. Positive correlations exist between foraging height and temperature (r = 0.42, P < 0.01 ) and between height and sunshine percentage (r = 0.17, P < 0.05). Lunk (1962:15) found that Rough-winged Swallows iStelgi- dopteryx rujicollis) feed close to the ground in cool, cloudy weather, and attributed this to prey response to these weather conditions. Increased sun- shine increases insect movement (Gangwere 1966), but also may enhance the visual ability of avian predators because of greater illumination. Low temperatures often occur on cloudy days, however, and it is difficult to separate the effects of sunshine and temperature on foraging behavior. No relationship was found when mean foraging heights were compared for the various wind speed categories (single factor ANOVA test, F = 1.1, d.f. = 2/197, P > 0.5). Males and females of the same species are known to partition the feeding niche by foraging at different heights (Jackson 1970). A /-test, however, revealed no significant differences in the foraging heights (/ = 0.8, P > 0.5) and drop bases (/ = 0.4, P > 0.5) of male and female bluebirds. Predator-to-prey distances were great for both males and females. Among males the maximum height recorded was 14.6 m, the maximum base was 28.3 m, and the maximum predator-to-prey distance was 29.0 m ( measure- ments from 2 drops). A female dropped from a height of 10.7 m onto a 48.8 m base to obtain prey 50.0 m away. The distances from which prey were sij2:hted were remarkable considering the small size of many of the insects involved. Perch use. — Early in the season prey are not always encountered when the ground is searched and uninterrupted bouts of continuous feeding are frequent. Observations during March and April 1972 indicated that “perch Pinkowski • FORAGING OF EASTERN BLUEBIRDS 411 success” ( the ratio of the number of perches from which a drop is executed to the total number of different perches used ) was significantly lower during March (76 of 140 perches successful, 54.2%) than April (502 of 702 perches successful, 71.5%; X“ — 15.3, P < 0.01 ). In only 3 of 28 observation periods was perch success lower than 50.0%; all occurred in March and the lowest figure observed was 31.9%. After early May bluebirds rarely failed to locate prey from a perch and alternated feeding with other activities except when feeding nestlings. Bluebirds foraging during March and April returned to the same perch after a drop on 83 of 568 occasions (14.6%). Alter hunting from an unsuccessful perch, bluebirds moved to a higher perch (as opposed to one distinctly lower) 50.0% of the time during March (N = 34), 68.3% during early April (N = 41), and 76.9% of the time during late April (N = 26). The trend to move to a higher perch later in spring is significant according to a test for a linear trend in proportions (Snedecor and Cochran 1967:246, z = 2.2, P < 0.05 ) and may occur because of increased availability of aerial prey that are searched for if prey is not located on the ground. In late spring, however, insects are more active and prey movement is less critical in the birds’ ability to locate prey. Also, by late April most bluebirds are nesting and may exhibit greater selectivity in prey consumed. Possibly a greater variety of insects can be searched for at greater heights. Time between drops during bouts of continuous feeding averaged 46.7 sec during March and April ( N = 61 ) , with a maximum of 186 sec. Time spent on a successful perch before a drop was made averaged 23.3 sec (N = 291) and was less during 'inclement weather conditions (x = 20.0 sec, N = 103 I than during favorable conditions (x = 25.2 sec, N = 188 ); the means differ significantly ( ^ = 2.2, P < 0.05) . Time spent on unsuccessful perches before the bird moved to a new perch averaged 27.2 sec but was significantly less during inclement weather (22.7 sec, N = 162) than during favorable weather (34.5 sec, N = 102; ^ = 4.6, F<0.01). Thus inclement weather imposes greater energy demands on the bird by necessitating more frequent movements (more perch changes and more drop attempts). Fewer insects are active during inclement conditions and reduced prey availability and activity evidently cause the birds to forage closer to the ground, thereby reducing the perceptual field because of the relatively constant search angle. A smaller area can presumably be searched more rapidly than a larger area. Distances traveled by birds moving from an unsuccessful perch to a new perch averaged 7.6 m iSD = 7.9 m, N = 132) and were significantly greater (Mann-Whitney U-test, P < 0.001) than the average distance of 4.1 m iSD = 5.5, ]\ = 211) traveled by birds foraging from a successful perch, based 412 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 on observations made during March and April. The mean distance traveled when moving from an unsuccessful perch is 18.0% greater than twice the average drop base observed in the spring period (Table 2), indicating that the birds move only slightly more than the minimum distance required to afford them a completely new perceptual field. By contrast, birds leaving successful perches move only 63.7% of twice the average drop base and thus search successive areas that overlap. The tendencies for bluebirds to move shorter distances and search successively overlapping areas after foraging from successful perches are similar to the findings of Smith and Sweatman (1974j, who noted that Great and Blue tits iParus major and P. caerulus) were more likely to return to previous capture sites when food was en- countered. No differences were found when distances traveled in leaving successful and unsuccessful perches were compared for males and females (P > 0.3 in each case ) . I found that pairs of bluebirds exhibited great differences in the relative sizes of areas used for foraging during the nestling period. Ten foraging ranges were examined in spring 1972 to determine the effect of perch abundance on the size of the foraging area. Although the size range of an entire foraging area was surprisingly large (4.5-38.9 ha), the size of the area containing perches was relatively constant (3.9-8.4 ha). The variances of the 2 sets of measurements (91.4 and 2.2, respectively) are significantly different (P = 41.5, P<0.01), suggesting that perch distribution may influence territory size. Factors limiting bluebird abundance. — Although the absence of nest cavi- ties may limit the number of Eastern Bluebirds (Pinkowski 1976a), the availability of perches may also be an important limiting factor in some ecological situations. Habitats having few or no perches are rarely used by Eastern Bluebirds; these areas elevate the energy demands imposed on foraging birds by necessitating more prolonged flights as the birds move from one foraging situation to another. Several aspects of this study suggested that the feeding requirements of bluebirds are stricter in spring than summer. The bluebirds I observed experienced little difficulty in obtaining food in summer and used a greater variety of habitats at that season. Of 39 nest boxes used by bluebirds at one lime or another, 23 were used in spring compared to 37 in summer. The difference in use frequency for the 2 seasons is significant = 12.2, P < 0.001) and appears related to the fact that foraging heights are less in spring and more perches are required at that season. In Michigan temperatures below Pinkowski • FORAGING OF EASTERN BLUEBIRDS 41.^ 5-o°C are common until late May and, when accompanied by overcast conditions, inhibit bluebirds from feeding on insects. Interestingly, most records of severe bluebird mortality in “winter” (Musselman 1941, Kenaga 1958) actually refer to extensive mortality in early spring (late February to early April) . Optimum conditions for bluebirds occur in areas containing an abundance of dead trees and limbs that are used as nest cavities and as foraging perches. Poor soil and a sparse ground cover help create ideal feeding conditions. SUMMARY Eastern Bluebird foraging behavior was studied in southeastern Michigan during 1972 and 1973. Bluebirds seize most prey after a short flight (“drop”) to the ground from a conspicuous perch. Other foraging tactics that may he used are flycatching, gleaning, flight-gleaning, and hopping. Frequencies of various feeding modes depend on season and weather, although dropping comprised 78.8% of the foraging sequences observed under all conditions. The base of a feeding drop increases with foraging height, suggesting a relatively constant search angle. Prey is usually located from a perch before it is pursued and habitats having a short, sparse ground cover are preferred by feeding bluebirds. Foraging height is greater in summer and during favorable, warm weather than in spring or cold, inclement weather. Bluebirds travel shorter distances to new perches if prey is sighted from the previous perch than if prey is not sighted. Males and females exhibit no differences in temporal and spatial use of perches. Perch abundance, however, influences the size of the area required by adults feeding nestlings and may be a factor limiting the distribution of bluebirds, especially in spring. LITERATURE CITED Beal, F. E. L. 1915. Food habits of the thrushes of the United States. USDA Biol. Surv. Bull. 280. Beckwith, S. L. 1954. Ecological succession on abandoned farm lands and its re- lationship to wildlife management. Ecol. Monogr. 24:349-376. Bent, A. C. 1949. Life histories of North American thrushes, kinglets, and their allies. U.S. Natl Mus. Bull. 196. Griddle, N. 1927. Habits of the Mountain Bluebird in Manitoba. Can. Field-Nat. 41:40-44. Freeman, J. A. 1945. Study in the distribution of insects by aerial currents. J. Anim. Ecol. 14:128-154. Gangwere, S. K. 1966. The behavior of Atlanticus testaceus (Orthoptera: Tetti- goniidae) on the E. S. George Reserve, Michigan. Mich. Entomol. 1:95-100. Goldman, P. 1975. Hunting behavior of Eastern Bluebirds. Auk 92:798-801. Holling, C. S. 1966. The functional responses of invertebrate predators to prey density. Mem. Entomol. Soc. Can. No. 48:1-86. Jackson, J. A. 1970. A quantitative study of the foraging ecology of Downy Wood- peckers. Ecology 51:318-323. Kenaga, E. E. 1958. Michigan bird survey, spring 1958. Jack-Pine Warbler 36:208-213. 414 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 Krieg, D. C. 1971. The behavioral patterns of the Eastern Bluebird (Sialia sialis) . New York State Mus. and Sci. Service Bull, 415. Leck, C. F, 1971. Some spatial and temporal dimensions of Kingbird foraging-flights. Wilson Bull. 83:310-311. Ligon, J. D. 1968. Sexual differences in foraging behavior in two species of Dendrocopos woodpeckers. Auk 85:203-215. . 1973. Foraging behavior of the White-headed Woodpecker in Idaho, Auk 90:862-869. Lunk, W. a. 1962. The Rough-winged Swallow Stelgidopteryx ruficollis (Vieillott : a study based on its breeding biology in Michigan. Publ. Nuttall Ornithol. Club, No. 4. Marshall, J. T., Jr. 1957. Birds of pine-oak woodland in southern Arizona and adjacent Mexico, Pacific Coast Avifauna 32:1-25. Morton, M. L. 1967. Diurnal feeding patterns in White-crowned Sparrows, Zonotrichia leucophrys gambelii. Condor 69:491-512. Musselman, T. E, 1941. Bluebird mortality in 1940. Auk 58:409-410. PiNKOw^SKi, B. C. 1975a. Growth and development of Eastern Bluebirds. Bird-Banding 46:273:-289. . 1975b. Behaviour and breeding of the Mountain Bluebird in captivity. Avic. Mag. 81:14-22. . 1976a. Use of tree cavities by nesting Eastern Bluebirds. J, WJldl. Manage. 40:556-563. . 1976b. Cedar Waxwings and Eastern Bluebirds feeding on winter stoneflies. Wilson Bull. 88:508-509. Power, H. W., III. 1966. Biology of the Mountain Bluebird in Montana. Condor 68:351-371. Preston, F. W. and J. M, McCormick. 1948. The eyesight of the bluebird. Wilson Bull. 60:120-121. Root, R. B, 1967. The niche exploitation pattern of the Blue-gray Gnatcatcher. Ecol. Monogr. 37:317-350, Salt, G. W. 1967, Predation in an experimental protozoan population (IT oodruffia- Paramecium) . Ecol. Monogr. 37:113-144. Smith, J, N. M. and H. P. A. Sweatman. 1974. Food-searching behavior of titmice in patchy environments. Ecology 55:1216-1232. Smith, W, J. 1966. Communication and relationships in the genus Tyrannus. Publ. Nuttall Ornithol. Club, No. 6. Snedecor, G. W. and W. G. Cochran. 19)7. Statistical methods. Iowa State Univ. Press, Ames. SoKAL, R. R. AND F. J. Roiilf. 1969. Biometry. W. H. Freeman Co., San Francisco. Verbeek, N. a. M. 1975. Comparative feeding behavior of three coexisting tyrannid flycatchers. Wilson Bull. 87:231-240. Willson, M, J. 1974. Avian community organization and habitat structure. Ecology 55:1017-1029. 15738 MILLAR, FRASER, All 48026. ACCEPTED 1 JULY 1976. COMPARATIVE FEEDING BEHAVIOR OF IMMATURE AND ADULT HERRING GULLS Nicolaas a. M. Vekbeek Many birds do not breed until they are 2 or more years old. Such delayed breeding is generally found among large, long-lived, non-passerine species. Lack ( 1954) suggested that delayed breeding has probably evolved in species in which reproduction at an earlier age would not be likely to succeed or might be harmful to the parents. Ashmole (1963) and Amadon (1964) suggested that young birds might be unable to catch food as efficiently as adults do. Recently, a number of field studies have shown that the ability to obtain food improves with age (see Buckley and Buckley 1974 for refer- ences). Given the inefficiency, it might be expected that the young have to compensate somehow for their lack of success, e.g. by spending more time in feeding. The object of this study was first to establish feeding efficiency in young and adult Herring Gulls (Larus argentatus)^ which show delayed breeding, and then to see how the young make up for any inefficiencies. The study was made on Walney Island, Cumbria, England, in 1973 and 1974. Herring Gulls and Lesser Black-backed Gulls (L. fuscus) breed on the southern end of the island. The colony is very dependent on nearby food sources such as the large intertidal areas of Morecambe Bay and a garbage dump on the island. METHODS At low water during spring tides extensive mussel (Mytilus edulis) beds are exposed in Morecambe Bay. Associated with the mussels is the common starfish (Asterias rubens) . Large numbers of Herring Gulls of all age classes fed upon these starfishes. Because of the distance between my observation post and tbe nearest beds (ca. 400 m) I could only distinguish (I used a 20-45 X zoom scope) between adults and immature birds (all those showing brown feathers). The immatures included birds at least 1 year old and older. The exact number of birds under observation is unknown as birds arrived and left continuously. At any one time approximately 50 birds were present, mainly adults. The gulls fed on starfishes (not mussels) by plunge diving for them, sometimes disappearing completely under water. Both immatures and adults fed in the same water apparently under similar conditions and in similar places. When I saw a gull diving I recorded whether it was successful (emerging with a starfish) or unsuccess- ful, and what the subsequent fate of the starfish was. Because the period when starfishes were available was short (about 2 hours) and infrequent (only during spring tides), I recorded as many dives as possible (different birds) rather than watching individuals over a period of time. Under these circumstances I could not obtain infor- mation on diving rates. Neither was it feasible to obtain meaningful information on the size of the starfishes taken. The data were collected during 25 hours between 3 and 21 April 1973. 415 416 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 On 5 and 6 April 1974, Hans Kruuk and I watched Herring and Lesser Black-backed gulls feeding at low tide on the crab, Hyas araneus. Gulls that managed to get a crab were often pursued by others in aerial chases. On both days we counted all the gulls feeding in a small bay, and the proportion of the various age classes represented. For each chase we recorded the number, the species, and the age classes participating as pursuers as well as the age class and the species of the pursued bird. In addition we noted whether the pursued bird lost its prey, and if possible, which pursuer obtained it. We distinguished between first-year-birds (brown young of the previous year), subadults (those showing a mixture of first-year and adult coloration), and adults. Subadult Lesser Black-backed Gulls were not seen in the general area of the colony until 19 April. We therefore assumed that all first-year-birds and subadults in our sample were Herring Gulls. In the analysis, chases in which one or more Lesser Black-backed Gulls pursued another of their species were not included. I did include those Lesser Black-backed Gulls that were chased by Herring Gulls and those that chased jointly with Herring Gulls. On 3 occasions at low water during spring tides an assistant and I counted all the gulls feeding in Morecambe Bay. The counts were made when the birds returned to the colony. We sat on the shore between the colony and the mussel beds and each of us counted the gulls that flew through his half of the sky. After 15 min I recorded our scores and the counting began anew. This was repeated until almost all gulls had returned to the colony. RESULTS Adult Herring Gulls were more successful in catching a starfish on the first dive (18 [64%J out of 28 dives) than immatures (3 [16%] out of 19 dives). These results are significantly different (x“ = 8.16, d.f. 1, P < 0.01). This has been shown too for the Brown Pelican ( Pelicanus occidentalis) ( Orians 1969) and the Sandwich Tern [Sterna sandvicensis) (Dunn 1972), but not for the adult and juvenile Royal Tern (S. maxima) (Buckley and Buckley 1974). Following a successful dive many starfishes were dropped in being brought up from the bottom, or in flight during transport from the water to the land. The gulls often did not attempt to retrieve these starfishes, perhaps because they fell in deep water. Many others were stolen when a diving gull surfaced or while it was being pursued in flight. In cases where they were not pursued in flight, adult Herring Gulls (N = 28) dropped 3 out of 28 starfishes, while immatures (N = 6) dropped 3 out of 6. These results are not significantly different (P = 0.053, Fisher exact probability test). Immature Royal Terns drop significantly more fish than adults do (Buckley and Buckley 1974), and Dunn (1972) suggests the same for Sandwich Terns. On the mussel beds a gull could lose its starfish when supplanted, when occupied defending a starfish by long-calling (Tinbergen 1959), or when a third bird took it while the owner was busy chasing another gull. The pro- portion of starfishes eaten versus those not eaten (Table 1) by adult and immature birds is not significantly different (x" — 1-65, d.f. 1,P > 0.05). I Verbeek • HERRING GULL FEEDING BEHAVIOR 417 Differences between Table 1 Adult and Immature Herring Gulls Starfish once it is Caught by Diving in the Fate of a Fate of starfish Adult Herring Gull Immature Herring Gull Eaten 28 (36%) 1 (10%) Dropped 26 (33%) 4 (40%) Stolen 23 (30%) 4 (40%) Abandoned 1 ( 1%) 1 (10%) think that this is only because my sample for the immatures is small (Table 1 ) . I gathered data only on those birds that were actually seen to dive. Many cases where birds were seen to be feeding on the beds without my knowing how they had obtained their starfish in the first place, went unrecorded. I gained the impression from these additional observations that young birds were more prone to have their starfish stolen than adults. Many gulls feed in Morecambe Bay at low water during spring tides. For instance, on 20 April we counted 20,473 gulls returning to the colony. In Fig. lA this count is plotted in relation to low tide. Similar counts were made on 3 May (19,142 gulls) and 4 July (20,304 gulls). Considering the short time that the mussel beds are exposed it is doubtful that any of these gulls had enough time to make more than 1 trip. Each of the 3 curves in Fig. lA has a major and a minor peak. The medians of any 2 curves in Fig. lA are not significantly different. During preliminary counts prior to 20 April I had noticed that adult Lesser Black-backed Gulls and immatures tended to return to the colony later in relation to low tide than did adult Herring Gulls. To analyze this I counted the proportion of immatures in 500 gulls of both species that flew past me. Once 500 gulls had been counted I repeated the procedure until almost all gulls had returned to the colony. For each 500 gulls I recorded the period over which the count was obtained and the midpoint of the period was entered in Fig. IB. The same procedure was used to obtain a curve (not shown) for the proportion of adult Lesser Black-backed Gulls among all gulls returning to the colony. I used this information to analyze the species and age composition of the returning birds. For instance, the curve of 20 April (Fig. lA) is redrawn in Fig. 1C. Clearly, the minor peaks (Fig. lA) referred to earlier are the result of the later return of Lesser Black-backed Gulls, and immatures of both species. In an aerial chase a bird carrying a crab could be pursued by 1 to 9 other gulls. Although relatively few first-year-birds were present, they participated 418 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 Fig. 1. Number of gulls returning from the mussel beds to the colony plotted in relation to low tide (1). (A) Each point on the graph is placed in the middle of a 15-min period and indicates the number of all gulls returning during that time span. (B) The number of immature Herring and Lesser Black-backed gulls per 500 gulls of all ages counted for each point on the graph. (C) Adult Herring Gulls (HG), adult Lesser Black-backed Gulls (LBG), and immature gulls ( IG ) of both species. See text for further explanation. in many more chases per bird than adult Herring Gulls (Table 2). Of the 82 recorded chases, 67 involved 1 or more of the 39 immatures present. Of the 111 adult Herring Gulls 1 or more participated in only 24 chases. Com- paring the number of successful pursuers ( i.e. those birds that managed to ol)tain the jirey from the bird they were pursuing) with the number of birds involved as pursuers (Table 2), all 3 age classes of Herring Gulls scored equally well ( x“ = 0.982, d.f. 2, P > 0.05 ) . Per bird, adult Herring Gulls were pursued significantly less than immature birds (x‘ = 25.31, d.f. 1, F < 0.001). All 3 age classes of the Herring Gull were ecjually successful in keeping their catch Avhile being pursued (x^ == 0.39, d.f. 1, P> 0.05). DISCUSSION Herring Gulls feed on a variety of foods in diverse places, including the intertidal, harbors, fields, and garbage dumps. The kinds of food obtained there require different types and degrees of skill. Starfishes are an important source of food as judged by the many birds that catch them and by the VeTbeek • HERRING GULL FEEDING BEHAVIOR 119 Table 2 Aerial Chases in a Known Number of Herring Gulls and the Frequency with Which each Age Class IS Pursued AND Takes Part as Pursuer Age class No. of birds present Pursuing birds Pursued birds No. of times pursued No. of times pursued/ birds present Successful pursued bird* No. of times involved as pursuer No. of pursuits/ birds present No. of successful pursuers- First year 10 15 1.5 11 31 3.1 5 Subadult 29 21 .7 14 70 2.4 8 Adult 111 19 .2 10 30 .3 6 ^ Those birds that managed to keep their prey. - Those birds that managed to obtain the food from the bird they were pursuing. large amounts of calcarious remains found in the colony (Shaffer 1971, pers. obs. j . Young birds should thus learn to catch them and this study shows that they do. To make up for their lack of success, several alternatives are possible. For instance, the young may try harder by spending more time diving, by diving more frequently, by being more persistent in diving repeatedly for the same starfish until successful, or by feeding on other types of food as well (i.e. by being generalists). These important aspects could not be studied under the circumstances. One other way of making up for their shortfall in obtaining food is to resort to stealing, either on the ground or in the air. In this study the immatures participated in a greater number of pursuits than one would have expected from the number of immatures present (Table 2). Adult Herring Gulls do not chase others as much as do immatures (Table 2). Apparently as the young mature, chasing becomes less important as a way of obtaining food. There may be several reasons for this. Adults may be less successful as chasers than younger birds, but this was not the case. Secondly, adults may give up sooner than immatures and very short chases involving adults may thus go unrecorded. Some evidence from aerial chases over the garbage dump in Oxford, England, in winter shows that young birds were as tenacious as adults. Apparently, the propensity of immature Herring Gulls to steal is real (Drury and Smith 1968). Moyle (1966) reports similar observations for immatures of Larus glaucescens. Because of their general inefficiency, it seems reasonable to assume that the immatures require more time than the adults to satisfy their food requirements. This is indirectly supported by the fact that most of the immatures returned to the colony later than most of the adult Herring Gulls, the bulk of which returned 45 min earlier than the immatures ( Fig. 1C ) . 420 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 The data suggest that at any given age a Herring Gull employs mostly those feeding skills that produce the best results. As the birds mature they learn new skills, improve on them, and discard or de-emphasize others. For an immature Herring Gull stealing on the ground and in aerial chases is apparently important. However, the young spend some time learning the feeding skills used by adults. Older birds do not chase as much as younger ones and I assume that for them other forms of feeding are more efficient in terms of time and energy expenditure. SUMMARY Immature Herring Gulls are less efficient in capturing starfishes by diving and possibly also in transporting them in flight than are adults. In contrast to the adults, the immatures steal much of their food from other gulls on the ground and in aerial pursuits. As the birds mature, stealing gives way to independent methods of finding food. ACKNOWLEDGMENTS It is a pleasure to thank Professor N. Tinbergen and Dr. H. Kruuk for helpful discussions and companionship in the field. Linda Verbeek and Robert Fergusson helped me count gulls. I thank Holker Estates Ltd. and the Lake District and Lancashire Naturalists’ Trusts for permission to work in the Walney gullery. Mr. Walter Shepherd, Warden of the South Walney Nature Reserve, was helpful in many ways. The work was in part supported by a grant from the Natural Environmental Research Council to Professor N. Tinbergen. LITERATURE CITED Amadon, D. 1964. The evolution of low reproductive rates in birds. Evolution 18:105-110. Asiimole, N. P. 1963. The regulation of numbers of tropical oceanic birds. Ibis 103:458-473. Buckley, F. G. and P. A. Buckley. 1974. Comparative feeding ecology of wintering adult and juvenile Royal Terns. Ecology 55:1053-1063. Dkury, W. H., Jr. and W. J. Smith. 1968. Defense of feeding areas by adult Herring Gulls and intrusion by young. Evolution 22:193-201. Dunn, E. K. 1972. Effect of age on the fishing ability of Sandwich Terns Sterna sandvicensis. Ibis 17:325-335. Lack, D. 1954. The natural regulation of animal numbers. Oxford Univ. Press, London. Moyle, P. 1956. Feeding behavior of the Glaucous-winged Gull on an Alaskan salmon stream. Wilson Bull. 78:175-190. Orians, G. H. 1969. Age and hunting success in the Brown Pelican {Pelicanus occidentalis) . Anim. Behav. 17:316-319. Shaffer. L. C. 1971. Specializations in the feeding behaviour of gulls and other birds. D. Phil. Thesis, Oxford Univ., Oxford, England. Verbeek • HERRING GULL FEEDING BEHAVIOR 421 Tinbergen, N. 1959. Comparative studies of the behaviour of gulls (Laridae) : a progress report. Behaviour 15:1-70. DEPT. OF ZOOLOGY, ANIMAL BEHAVIOUR RESEARCH GROUP, SOUTH PARKS ROAD, OXFORD, 0X1 3pS, ENGLAND (PRESENT ADDRESS: DEPT. OF BIOLOGICAL SCIENCES, SIMON FRASER UNIV., BURNABY, B.C. CANADA v5a 1s6) . AC- CEPTED 20 SEPT. 1976. REQUEST FOR ASSISTANCE WANTED: Data on the Seasonal Distribution of North American Gulls. — We are developing a procedure whereby the U.S. Air Force can predict the potential seasonal hazard to aircraft represented by gulls in parts of North America. This knowledge will be used to schedule missions around high risk areas thereby reducing the likelihood of bird-aircraft collisions. Supplemental data on local gull populations are needed from all parts of the continent. The assistance of field workers is solicited to aid us in this task. For each observation, please provide the following information: list of species present, approximate number of each species, precise locality description, dates observed, any information about causes for concentrations (e.g. sanitary landfill operation), and any details about the frequency of such concentrations in the respective areas. Information is sought from inland as well as coastal localities. Please submit reports of your gull observations to Dr. William E. Southern, Department of Biological Sciences, Northern Illinois University, DeKalb, IL 60115. Data will be gathered for a 2-year period beginning 1 September 1977. COMPARATIVE MORTALITY OF BIRDS AT TELEVISION TOV'ERS IN CENTRAL ILLINOIS James W. Seets and H. David Bohlen There have been a number of studies of mortality of migrating birds at television towers in the L .S.. including some in Illinois t Brewer and Ellis 1958. Cochran and Graber 1958. Parmalee and Parmalee 1959. Parmalee and Thompson 1963. Graber 1968 1 . Most of these considered mortality at a single tower. Kills of birds at television towers offer one means of learning the timing and geographic patterns of migration and the physiological and population traits of the migrants. The great potential of tower-kill data to provide information on migration has not been fully realized, however, because the coverage of towers has been too limited. In Illinois in 1973 there were 33 TV transmitting towers and 29 cable TV towers 152.4 m or higher. Fourteen of these were located more or less in an east-west line across central Illinois. From August to December 1972. we attempted to check 7 of the large ( 182.9 m or more in height) television towers in central Illinois (Fig. 1 1 for bird kills. We had 2 primary goals: 1 1 1 to acquire research specimens for the Illinois Natural History Survey and the Illinois State Museum, and (2 I to acquire comparative data on migration patterns across the state. METHODS The towers were checked on all mornings that followed nights with reduced visibility from fog or other precipitation, or with low cloud cover, or both. Seets and his associ- ates at the Natural Histor>* Survey checked towers from Macon County (Argenta tower) eastward, and Bohlen and his associates checked the Springfield and Bluffs towers. All intact specimens were collected, weighed, and frozen to be processed later. Crippled birds were counted but not collected. Visible evidence, such as a few feathers or other remnants of carcasses, suggested that several birds in each kill had been eaten by predators. These remnants were also counted but not collected. Great Horned Owls iBubo virginianusK soldier beetles < Cantharidae • . and sexton beetles appeared to be the principal predators and scavengers involved. Ants were seen on many of the dead birds. An effort was made to determine the precise timing of each kill, using data from weather stations at Springfield Capitol Airport. Chanute Air Force Base (Fig. 1), and from the U.S. Department of Commerce Daily eather Maps Weekly Series. The dates of kills referred to in this paper (.Table 1) are the dates the birds were collected — the mornings after the kills. The 2 towers at Springfield are only 3.2 km apart and we have considered them as one location. The central Illinois towers are in generally flat terrain. Vegetation sur- rounding the towers was either closely mowed grass or standing soybeans and corn, making 422 Seets and Bohlen • BIRD MORTALITY AT ILLINOIS TOWERS 42B B-BLUFFSy ^84.0 M 2-SPRINGFIELD, CAPITOL AIRPORT Spi-springfield, M A-ARGENTA, 324.0 M M-MONTICELLO, 319.1 M S-SEYMOUR. 299.0 M F-FITHIAN, 407.8 M 1-CHANUTE AIR FORCE BASE. RANTOUL G-GIBSON CITY. 184.4 ■ -WEATHER STATIONS •-TV TOWERS 16 0 32 64 km Fig. 1. Location of television towers checked for bird kills. it difficult to obtain exactly comparable collections from the different towers. However, we believe that collections were at least 70% complete for each tower, based on checks made by several observers both on the mornings after the kills and on subsequent mornings. All of the towers studied were similarly constructed, being triangular in cross section with at least 6 sets of cable guys at each corner. The towers ranged in height from 184.4 m to 484.0 m IFig. 1). RESULTS AND DISCUSSION On 13 dates between 2 September and 12 November 1972, 5465 birds were collected at the 7 television towers in central Illinois. Most of the birds (93.4%) were killed on 4 nights: 1-2 September, 26-27 September, 28-29 September, and 30-31 October; more than half (59.8%) were killed on the night of 26-27 September. Species Recorded in the 4 Largest Kills at Central Illinois Television Towers in the Fall of 1972i 424 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 Winter Wren {Troglodytes troglodytes) Long-liilled Marsh Wren {Telrnatodytes palustris) Sliort-hilled Marsh Wren i Cistothorus platensis) (iray Catbird {Dumetella carolinensis) Brown Thrasher (Toxostoma rufum) Species Wood Thrush [Catharus mustelina) Seets and Bohlen • BIRD MORTALITY AT ILLINOIS TOWERS 425 s •S s£ 03 a (M CO tiH - uo CSI SPI r— ( (N (M CO e- r-H 27 CSI LO - 44 m i-H - - - (N LO so - - CO (N CO CO rH o\ LO e- CSI CO pH LO a 1 — 1 LO I-H r-H (73 o (M lO 03 CM CO (N CO o o CSI O CO SO CSI < I— t CO CO CO OS 00 O 25 CO (0\ CO CO CSI 98 CO CO CO CO CO o CO CO rH o CO pH CSJ CSI (73 CO rf LO OS 00 r- Tt (X) l—i SO CO CSJ CSJ Ph CSJ e- CQ o CSI o pH SO Csl HH Os c^ CO CO O ,-H lO CSJ pH O a CO LO LO r’ c/3 OS CSI r-^ o pH (M CSI CSI (73 CO CO CSI CSJ I-H a (M csi CO LO ^ -S ^ 5? 03 03 CO 03 0< t— I 0> tJ3 3 'TJ O. 03 3 C 'S ^ 2 o S c 03 03 O > o ^ 03 O 'bi) _2 •S > ^ B 5 o o -5 3 03 OS _ S i S I I 03 3 3 ^3 3 •S $ "o u o .S 03 v -3 I-. •-i ^ 03 ^ O ^ 03 ^ ^ o; 2 > :s > .2 ^ 2, ^ ^ ^ :S 1 Qi Oh m 3^ -3 03 3 03 2 -r 03 o ^ _C 'B "O 2 'o Oh O C o o 03 03 bC C 03 H O Z Yellow Warbler (Dendroica petechia) Magnolia Warbler (Dendroica magnolia) Cape May Warbler (Dendroica tigrina) Table 1 — Continued 426 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 1 ^ (N CO On ON Tf CO ON t’ CO CO o CM LO CM ffl t" LO o 26 O - CS CO CO LO CM CO CO CL, C/3 LO CM I-O ON CO CM CO o •cf CO CM CM CO 'll CO CO LO CM CO 1— 1 LC LO CO CON 00 rH CM CO CO CM o CM NO CM o CO O rf CM NO CO o CO LO CM o c/3 CM CO CO CO O CO LO CM CO Uh CO CM On •o O CO CM CO CO CM o CM CM CON a CM LO 1-0 CM CO t'- CM LO ■rjN CM CO '”' -f lO CM CO CM CM LO c/3 CM 1^ lO CM CO < -- ON CM LO CM -- C/3 - - CM CM ON CO CM c: ^ c 3 5 ^ ^ o 1^ . . ~ ~ i- ^ 3 o ~Z - 3 5 ^ ^ S ^ •= oj p -9 — i I i 1 1 1 1 i jiSO<£z-/:c Table 1 — Continued Seels and Bohlen ■ I51R1) MOKTALITY AT ILLINOIS TOWERS 42 I I I Ti z. yT s c: I S! I bC c c 3 3 I .1 5. o S S 0 i: a c/:! "g 1 3 3 I o c s I CA ¥ -r ■jj c "g s I ^ b X I :i O 3. SI £ I I ^ • l“- II p vJ a- d II S-g ^-‘' «•= X > l< 3. a. 33 3 ^ a ji i 3 3: 3. r 3 1; -3 _3 3 bL ^ i ■^Vi III alt 111 III 3 •-' •Su a - 3-> i 428 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 A COMPARISON OF THE AS COMPARED TO KILLS AT YEARS 1958, 1962, AND 1972^ Table 2 RATIOS OF BIRDS KILLED AT TOWERS IN WESTERN ILLINOIS TOWERS IN EASTERN ILLINOIS DURING SEPTEMBER FOR THE Species 1958 1962 1972 16-17 Sept. West= 15-17 Sept. East^ 24-25 Sept. West^ 24-25 Sept. East^ 2-27 Sept. West 2-27 Sept. East Swainson’s Thrush 1.0 3.1 1.0 4.5 1.3 1.0 Gray-cheeked Thrush 1.0 1.3 1.0 5.2 1.0 1.6 Veery 1.3 1.0 3.8 1.0 Red-eyed Vireo 1.1 1.0 2.8 1.0 4.3 1.0 Tennessee Warbler 3.8 1.0 2.1 1.0 1.8 1.0 Magnolia Warbler 2.8 1.0 1.0 1.4 1.0 5.2 Bay-breasted Warbler 4.3 1.0 1.9 1.0 1.0 3.7 Northern Waterthrush 5.3 2.8 1.0 3.4 1.0 Common Yellowthroat 2.8 2.8 1.0 2.0 1.0 Bobolink 1.0 6.0 1.1 1.0 3.7 1.0 Chestnut-sided Warbler 11.9 1.0 2.1 1.0 1.0 1.9 Ovenbird 3.9 1.0 1.7 1.0 1.0 5.7 American Redstart 3.7 1.0 1.2 1.0 2.0 Total Birds 827 147 213 296 1,454 3,144 ^ In comparing the ratios, data were adjusted to correct differences in total birds killed between eastern and western towers during the given dates for each year. - Data for 1958 West are from Parmalee and Parmalee, 1959. ^ Data for 1958 East, 1962 East, West are from Graber, 1968. Inspection of our data shows no consistent relationship between tower height, terrain, or tower location and number of birds killed for the kills we studied. Kills were neither consistently high nor low at any particular tower. We believe that the number of birds killed at a given tower on a given night is related primarily to local weather conditions and to the number of birds flying. Kills in relation to weather factors. — The kills occurred following the passage of cold fronts, usually when conditions of low overcast (550 m or less) and reduced visibility (< 8 km) prevailed; however, 4 of the kills occurred when the lowest overcast was 1220 m to 1830 m and the visibility was 11.3 km or more. On the 4 nights when 93% of the birds were killed, ceilings were 550 m or less. All kills occurred within 32 h (usually within 6 h) after the passage of cold fronts, when the winds were from the north. We do not know exactly when during the night the kills occurred. It may be important to know the precise timing of the kills when comparing bird losses at the different towers (Table 2), because the time factor may have a bearing on the species killed. Species recorded in the small kills at Central Illinois television towers in the fall of 1972^ ~~~~~~ SPP SPI S F SPI SPI SPI SPI SPI A SPI 4 14 14 14 21 28 10 12 18 28-29 10-12 Species Sept. Sept. Sept. Sept. Sept. Sept. Oct. Oct. Oct. Oct. Nov. Seets and Bohlen • BIRD MORTALITY AT ILLINOIS TOWERS 429 (M (M i-H CO CO — I cOCSIrH <-H CM CM I— I I— I 03 ”3 ;s ’5i) 03 o o o 3 CJ sm ■S & c ■' cn ^ O 3 03 o m X 43 - bX) C c ‘o ^ "O c/i U cfi 3 3 Is H ^ J ^ -3 b ^ O ^ .S -n 3 !-h "O OJ 03 O > O 43 -3 O S s .2 ^ ^ ^ -3 V 1:3 T3 > 43 3 . Cd 3 ^ 22 3 ■i 1 1 1 1 ^ „ bC > (3 « g 33 C 33 -5 o C bC 43 03 3 Tahle 3 — Continued 430 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 4J cq LC eg (M I— I r-l rH r-l i-H ^ eg r-i "-i eg r-H ^ „ CC T?- r-. (M T3 T = ~ is — :3 u C 2 _ .2 :r 5 £ I 2 ^ cj t . ^ I > ^ ^ i ^ E ■E £ ‘- E £ i c Z a; w S£ _ 2 i -i c St t •E i = in z: — -e .'. St = - r 3 ~ T I - * c - I I X 3 §■> “• £ § I ^ I I 3 2. ^ 2 XI *r z. ^ z.J^ _ i; X ^ 5 4 X ^ r c •- C u — _ icU*ntificalion of localitios, soe Fiy. Seets and Bohlen • niKI) MORTALITY AT ILLINOIS TOWERS 431 In analyzing the relationship of the kills to weather, we have emphasized the data for the large kills in order to have sufficient numbers for comparison. Comparative data for the large kills are presented in Table 1 and the small kills are summarized in Table 3. Comparison of kills. — Graber’s (1968) radar transect data on tbe migra- tion in central Illinois in September show that the number of nocturnal migrants is fairly uniform at different locations across the state. Thus, if weather conditions were the same at all towers, we would expect the number of birds killed at each tower to be similar, but, in fact, there are great differences in the numbers of birds killed at different towers on the same night (Table 1). On 27 September, for example, kills ranged from 107 birds to 992 at different towers of comparable height. There were very different numbers of birds killed at towers as close (19.3 km) as Seymour and Monti- cello — 127 versus 992. Such differences cannot be explained without more detailed weather records than are presently available. Although the data from radar transects for central Illinois indicate a fairly uniform distribution of total night migrants across the state in September, this does not necessarily indicate a uniform statewide distribution of each species. Because large television towers are well distributed across central Illinois, the kills at those towers provide a means of comparing the species composition of the flights of migrants in the eastern and western segments of the state. In making such a comparison, we find that some species appear relatively more numerous in the kills on the western side of the state while other species are more prevalent on the eastern side. Chi-square analysis of the data for species involving 20 or more individuals indicated that the species composition did, in fact, differ significantly between the eastern and the western towers (x“ = 736; P < 0.001). The analysis seems to indicate a difference in the relative numbers of birds of individual night migrating species between the east and west sides of Illinois. In a few cases, species were not present in kills on one side of the state but were represented on the other side. These instances included the Common Flicker (6), Great Crested Flycatcher (2), Alder Flycatcher (6), and Yellow Warbler (17), all found at western towers but not at eastern towers. Only the Cape May Warbler (8) and Grasshopper Sparrow (T) were present at eastern towers but not at western towers on nights when kills occurred on both sides of the state. Other species were present in kills on both sides of the state but differed significantly in numbers from one side to the other (Table 2). It is worth- while to compare our 1972 data on the species killed at western versus eastern towers with Graber’s ( 1968) and Parmalee’s data for 1958 for the 432 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 same region (Table 2). In several species (for example, see thrushes) the pattern was consistent in all years. Rare and/or infrequent species. — A few species deserve special comment, either because of their rarity, or because of their infrequent occurrence in tower kills. Yellow Rail: Single Yellow Rails were found among tower kills at Spring- field on 27 and 29 September and at Monticello on 27 September, marking the first time that the Yellow Rail was found among tower kills on the eastern side of the state. Red-bellied Woodpecker: The kill of night migrants on 27 September 1972 at the television tower at Bluffs included a Red-bellied Woodpecker, a supposedly non-migratory species. Black-throated Blue Warbler: On 27 September 1972, 17 Black-throated Blue Warblers were killed on the eastern side of the state and 3 were killed on the western side. Three were killed on the western side on 29 September (2) and 10 October (1). These kills were high compared with the number of Black-throated Blue Warblers (1 or 2 a season) seen in the field in Illinois. Henslow’s Sparrow: One was found among tower kills on the eastern side of the state for the first time on 28-29 October 1972, at Argenta. Sharp-tailed Sparrow: The Sharp-tailed Sparrow was found for the first time in kills at eastern towers on 27 September at Fithian (1) and Monticello (1). Field Sparrow: On 31 October 1972 a Field Sparrow was found for the first time among eastern tower kills at Fithian (1), Monticello (5), and Argenta ( 1 ) . Acknowledgments Appreciation is extended to the following members of the Illinois Natural History Survey: Dr. R. R. Graher for his many helpful suggestions and encouragement, Dr. Glen C. Sanderson and Helen C. Schultz for editorial help, and W. L. Anderson for his critical review of the manuscript. We would also like to express our thanks to Dr. William R. Edwards for his help in the statistical analysis of our data. Literature Cited Brewer, R. and J. A. Ellis. 1958. An analysis of migrating birds killed at a television tower in east-central Illinois, September 1955-May 1957. Auk 75:400-414. Cochran, W. W. and R. R. Grader. 1958. Attraction of nocturnal migrants by lights on a television tower. Wilson Bull. 70:378-380. Grader, R. R. 1968. Nocturnal migration in Illinois — different points of view. Wilson Bull. 80:36-71. Seets and Bohlen • BIRD MORTALITY AT ILLINOIS TOWERS 433 Parmalee, P. W. and B. G. Parmalee. 1959. Mortality of birds at a television tower in central Illinois. And. Bull. 111:1-4. Parmalee, P. W. and M. D. Thompson. 1963. A second kill of birds at a television tower in central Illinois. Aud. Bull. 128:13-15. ILLINOIS NATURAL HISTORY SURVEY, URBANA 61801 (JWS), AND ILLINOIS STATE MUSEUM, SPRINGFIELD 62706 (hDB). ACCEPTED 1 JUNE 1976. EFFECT OF FLOCK SIZE ON FORAGING ACTIVITY IN WINTERING SANDERLINGS James Silliman, G. Scott Mills, and Stephen Alden Birds in flocks may increase the proportion of time spent feeding and thus food intake hy dividing the time spent watching for predators among flock memhers. This advantage of foraging in flocks has been supported by mathe- matical analysis (Pulliam 1973) and by experimental work with aviary birds (Powell 1974). Field work with Wood Pigeons [Columba palumbus) (Mur- ton et al. 1971) showed that single birds had lower feeding rates and spent more time looking around than flock birds, but Murton (1971) interpreted this as indicating that single birds were seeking to join flocks for reasons unrelated to predator protection. Page and Vdiitacre (1975) found that predation is substantial on wintering shorebirds and that they are less susceptible to predation when in flocks, but no evidence exists that flocking shorebirds increase their foraging activity. To test this possibility, we exam- ined foraging activity in relation to flock size in wintering Sanderlings {Calid- ris alba). Because Barash ( 1974) found that chickadees in flocks have fewer aggressive encounters than single birds, we looked for similar behavior among Sanderlings. METHODS Data were collected hy 9 investigators from 26 to 28 November 1974 at Punta Santa Rosa 37 km northwest of Kino. Sonora, Mexico. The beaches on the south side of the point are sandy while those on the north, at the mouth of a channel, are composed of algae-covered rocks approximately 4-6 cm in diameter. Sanderlings were the most com- mon shorebirds present. Investigators dispersed along the beaches in 4 groups. Each group consisted of 1 re- corder and 1 or 2 observers; group memhers and roles were changed frequently. Each group was ecjuipped with a spotting scope, stopwatch, mechanical counter, and binoculars. The first bird observed in a flock was selected hy counting hack from the lead bird using random numbers less than 10. Successive birds were picked hy counting hack a random number from the position of the last bird observed. When the count exceeded the number of remaining birds, counting continued with the lead bird. For purposes of analysis we decided to take approximately ecjual numbers of observations of birds in 3 classes on rocky and sandy beach: singles, flocks of 2-10 shorebirds, and flocks of greater than 10. Shorebirds other than Sanderlings were included in the total flock size, since all birds con- tribute to the possible reduction of predator alert time. \^'e recorded the total seconds out of 1 min that a Sanderling appeared to he foraging • hereafter called foraging time). The observer timed foraging activity with a stopw'atch while the recorder monitored 1 min intervals. e also recorded the number of feeding movements in 1 min (hereafter called foraging rate) using a mechanical counter. To test for possible differences in foraging method due to substrate type or flock size, we 434 Sillirnan et al. • SANDERLING FLOCK SIZE AND FORAGING 435 Table 1 Foraging Times* of Sanderlings in Different Flock Sizes on Different Habitats Flock size Habitat Singles 2-10 >10 N X s- N X s- N X s- Sandy- 33 53.8 79.0 54 49.3 191.0 67 53.3 151.0 Rocky 34 49.9 67.6 45 51.1 157.3 55 53.2 90.7 Combined 67 51.8 75.8 99 50.1 174.7 122 53.3 122.9 * Total seconds in 1 min spent foraging. classified foraging movements as probes if the bill penetrated the surface, or pecks if it did not. We estimated the number of movements on the few occasions that they were too rapid to be counted directly. Thirty-two of the total 475 min of foraging data in- clude birds observed between 30 and 45 sec whose rates were prorated to 1 min. No sleep- ing birds or birds observed for less than 30 sec were included in the data analysis. Ag- gressive interactions were recorded only for those birds selected for foraging observations. Due to large variances, we transformed the data as the square root of (X + -5) to normalize them for statistical tests. Statistical analysis was done with the aid of the University of Arizona computer services using the SPSS statistics programs. RESULTS Sanderling flocks on sandy beaches tended to be small and move rapidly whereas those on rocky beaches were slower and sometimes large enough to include both sleeping and foraging birds. We collected no quantitative data on the relative frequency of flock sizes, hut flocks of 2 to 10 birds seemed most common. Single birds were fairly common but usually did not remain single for long before being joined by others. There was no appreciable change in mean time spent foraging due to par- Foraging Rates* of Sanderlings in Table 2 Different Flock Sizes on Different Habitats H bitat Flock size Singles 2-10 >10 X X s- X X s- X X s- Sandy 28 26.9 401.4 40 51.0 1708.3 31 37.8 881.8 Rocky- 22 43.7 627.8 41 57.0 889.9 25 61.9 1832.4 Combined 50 34.3 561.1 81 54.1 1287.0 56 48.6 1426.4 * Number of movements per minute. 436 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 Polynomial Analysis of Variance- Table 3 —Foraging Rate with Flock Size on Sandy Beach Source Degrees of freedom M. S. Between 2 25.54* Linear Term 1 10.33 Quadratic Term 1 42.05* Within 96 5.32 * F probability < .05 ticipation in flocks by Sanderlings (Table 1). Analysis of variance showed no significant relationship between flock size and foraging time in either habi- tat (sandy P = .22, rocky P = .54) or in both combined (P = .22). However, the foraging rate of Sanderlings in flocks (X = 51.8) was con- siderably higher than that of single birds (X = 34.3), and this difference was significant by a 1-tailed t-test (t = 3.15, P = .001). The increase in foraging rate was not associated with a change in foraging method. Of a sample of 504 movements by single birds, 92% were probes, while 90% of 7091 movements by birds in flocks were also probes. Mean foraging rate tended to increase as flock size increased on rocky beach (Table 2), but this trend was not significant by analysis of variance (P = .48). There was a marked decline of foraging rate in flocks greater than 10 on sandy beach, shown to be significant by polynomial analysis of variance (Table 3). Mean foraging rates of birds on rocky beach were higher in all cases than on sandy beach (Table 2). This difference was significant by a 2-tailed t-test for single birds (t = 2.93, P = .005) and very nearly so for flocks (t = 1.97, P = .051). Ninety % of foraging movements were probes on both rocky and sandy beach (N = 2111 and 5484 respectively). We found a positive correlation between aggressions per bird-minute and flock size (r = .212, significance of r = .0001). Increased aggressive in- Table 4 Aggressions OF Sanderlings in Different Flock Sizes on Different Habitats Flock size Habitat Singles 2-10 >10 Sandy .076* (53) .122 (90) .376 (96) Rocky .019 (53) .092 (76) .100 (80) * Aggressions per bird observed per minute. Number of bird-minutes in each category in paren- theses. Silliman et al. • SANDERLING FLOCK SIZE AND FORAGING 437 teractions in larger flocks were due to increased aggressions per bird (Table 4) and increased numbers of birds participating. Aggressions were more frequent in all size classes on sandy beach. DISCUSSION The increased foraging rate of Sanderling in flocks is equivalent to in- creased food intake if the proportion of successful feeding movements remains relatively constant, as Goss-Custard (1970a) found for Redshank [Tringa totanus ) . We attribute the lack of a corresponding increase in foraging time to our inability to measure the brief pauses between foraging movements. The increased foraging rate of Sanderlings in flocks could be attributed to causes other than less time spent looking for predators. Krebs (1974) sug- gested that herons in flocks fed at a faster rate than solitary individuals be- cause flocks form at patches of abundant food. However, Sanderling flocks and single birds foraged in the same areas and flocks moved cohesively along the beach. Murton (1971j and Krebs (1974) have argued that single birds spend less time foraging because they are searching for flocks to join. This seems unlikely in the case of Sanderlings since flocks were seldom far from foraging single birds. Finally, Sanderlings in this study did not change their foraging method when in flocks to achieve the increase. We conclude that increased foraging rate may be related to less time spent searching for preda- tors between feeding movements. Our data indicate that Sanderlings do not join flocks to reduce aggressive encounters, as Barash (1974) found for chickadees. Recher and Recher (1969) found that Semipalmated Sandpipers {Calidris pusilla) likewise in- crease aggressive encounters in flocks. The decreased foraging rate of birds in large flocks on sandy beach did not occur on rocky beach and may have been caused by limited food on sandy beach. Sanderlings on sandy beach had lower feeding rates, more aggres- sive encounters, and higher flock speed than those on rocky beach, suggesting that food was less abundant on sandy beach. Higher aggression among shore- birds has been associated with lower food availability (Recher and Recher 1969) , as has higher flock speed of woodland passerines (Morse 1970 ) . Large flocks on habitats with limited resources may deplete locally available prey thereby reducing the average feeding rate (Goss-Custard 1970b). Protection from predators does not diminish as flock size increases, but competition for food where resources are limited may determine an optimum flock size. ACKNOWLEDGMENTS Planning and fieldwork for this study were done with the cooperation of Ruby Allen, Steve Hilty, Terry Johnson, Ingrid Porton, Harriet Smith and Bonnie Swarhrick. Tom 438 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 Starmer provided guidance with statistical procedures. We thank Stephen Russell. H. Ronald Pulliam. DaHd G. Ainley and Da\id P. Barash for critically reading an earlier draft of this manuscript. LITER-\TURE CITED Barash. D. P. 1974. An advantage of winter flocking in the Black-capped Chickadee, Parus atricapillus. Ecology 55:674—676. Goss-Custard. J. D. 1970a. The responses of Redshank iTringa totanus (L.)) to spatial variations in the density of their prey. J. Anim. Ecol. 39:91-113. . 1970b. Feeding dispersion in some overwintering wading birds. In Social behavior in birds and mammals ( Crcok. J. H. ed. i . Academic Press, London. Krebs. J. R. 1974. Colonial nesting and social feeding as strategies for exploiting food resources in the Great Blue Heron iArdea herodias) Behaviour 50:99-134. Morse. D. H. 1970. Ecological aspects of some mixed species foraging flocks of birds. Ecol. Monogr. 40:119-168. Merton. R. K. 1971. hy do some species of birds feed in flocks? Ibis 113:534—536. , A. J. Isaacson, and N. J. Westwood. 1971. The significance of gregarious feeding behaviour and adrenal stress in a population of \^'ood Pigeons Columba palumbus. J. Zook. Lond. 165:53-84. Page. G. and D. F. \^Hitacre. 1975. Raptor predation on ^rintering shorebirds. Con- dor 77 : 73-83. Powell. G. V. N. 1974. Experimental analysis of the social value of flocking by Star- lings iSturnus vulgaris) in relation to predation and foraging. Anim. Behav. 22: 501-505. Pl'lliam. H. R. 1973. On the advantages of flocking. J. Theor. Biol. 38:419—422. Recher. H. R. and j. a. Recher. 1959. Some aspects of the ecology of migrant shore- birds. H. Aggression. Wilson Bull. 81:140-154. DEPT. OF ECOLOGY AND EVOLUTIONARY BIOLOGY. UNIV. OF ARIZONA, TUCSON 85721. ACCEPTED 26 MAR. 1976. ACTIVITY PATTERNS OF FEMALE RUFFED GROUSE DURING THE BREEDING SEASON Stephen J. Maxson Ruffed Grouse [Bonasa umhellus) are difficult to observe for extended periods in the wild. Consequently, despite the considerable research atten- tion this bird has received ( e.g. Bump et al. 1947, Gullion and Marshall 1968, and others ) , few precise data are available concerning its activity patterns. Recent radiotelemetry studies in Minnesota ( Archibald 1973 and several studies cited therein ) have increased our knowledge of this aspect of the Ruffed Grouse’s behavior but activity patterns of female Ruffed Grouse during the breeding season remain poorly documented. This paper reports a radio- telemetry study of female Ruffed Grouse activity patterns from pre-incubation through post-incubation periods at the University of Minnesota’s Cedar Creek Natural History Area located about 45 km north of Minneapolis, Minnesota. METHODS Field observations were made from 1 April-30 June 1971 and 1 April-7 July 1972. The Ruffed Grouse population was at a peak level during this investigation. Spring counts of drumming males totaled 30 and 28 on the square mile study area in 1971 and 1972 respectively. Female Ruffed Grouse were captured hy baited lily-pad traps (Gullion 1965), nest traps (Weller 1957), dip-netting on the nest ( Robel et al. 1970), and hy nightlighting ( Huempfner et al. 1975). Hens captured on their nests were handled in the field to mini- mize the time they were kept off the eggs. All others were placed in burlap hags and transported to the Cedar Creek laboratory where they were weighed, sexed, aged, leg- banded, and equipped with a 24-26 g transmitter similar in harness design to one de- scribed by Brander (1968). Expected transmitter life ranged from 50-75 days but usually birds were recaptured and fitted with new transmitters before this time. All birds were released at the point of capture. Radio-marked grouse were monitored with an automatic radio-tracking system ( Coch- ran et al. 1955). Two towers 0.8 km apart support directional receiving antennas con- tinually rotating at 1% rpm. During each antenna revolution, radio signals emanating from the transmitter equipped grouse were received by the antennas. Signals were relayed to a centrally located laboratory and, after amplification and modification, were recorded on 16 mm film as degree bearings for each tower. These bearings were used to determine the location of the bird by triangulation. With the use of a microfilm reader, an activity designation (active, inactive, or unknown) was determined at 15 min intervals for each bird. Marshall and Kupa (1963) determined that the pitch of the radio signal changed as a grouse moved about. This change in signal pitch associated with activity is reflected as irregularities in signal peaks on the film record ( Sargeant et al. 1965). During each 15 min period (except during incubation) a grouse was considered active if 4 or more signals exhibited these irregularities, or if a change in bearing of 1 or more degrees 439 440 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 Table 1 Percentage of Time Female Grouse Were Active Bird No. Age Pre-incubation Incubation With Post-inc Brood ubation Without : Brood No. Days % Time Active No. Days '% Time Active No. ' Days % Time Active No. Days '% Time Active 1657 J 28 44.5 27 4.5 20 52.7 1690 J 36 51.2 25 4.8 — — — — 1691 A 35 47.3 26 3.5 26 51.5 — — 1695 J 36 48.7 25 3.9 9 58.4 — — 1698 J — — — — — — 29 67.5 1699 J — — — — 8 56.9 — — 2200 A — — — — 30 58.5 — — 2201 J — — — — — — 29 60.5 2202 — — — — — — — 18 51.0 2210 J 45 47.9 10 4.9 — — — — 2219 J — — 21 4.1 — — — — 2235 A — — 25 3.1 — — — — 2238 J 46 37.3 26 4.4 — — — — 2239 J 43 55.6 26 5.8 30 54.4 — — 2241 J 37 41.7 17 4.1 — — 43 59.0 2246 A 36 48.4 26 3.4 18 47.1 — — 2248 J — — 25 4.7 — — — — Mean 38 46.9 23 4.3 20 54.2 30 59.5 occurred for either tower. During the incubation period activity changes were de- termined to the nearest minute. Data for each l)ird were divided into pre-incuhation, incubation, and post-incubation periods and were analyzed in 2 ways using the University of Minnesota’s Control Data 6600 computer system. First, the sampled activity for each day was plotted giving every 15 min interval an activity symbol ^active, inactive, or unknown). This illustrated day- to-day periods of activity. Second, the percentage of time a bird was active during each 15 min interval over a given time period (pre-incubation, incubation, or post- incuhation) was plotted, giving a composite 24-h day comprised of data for all days during a time period. This illustrated activity trends throughout the specified period. All times given are C.S.T. RESULTS Activity data were obtained from 17 female Ruffed Grouse during the study. Table 1 summarizes the percentage of time the birds were determined to be active during the pre-incubation, incubation, and post-incubation periods. Percent activity tended to be greater during the post-incubation period than the pre-incubation period, perhaps due to increased daylength. Hens with Maxson • GROUSE ACTIVITY 441 NO 2241 NO. 2246 TIME Fig. 1. Temporal distribution of percentage of 15-min intervals active during the pre-incubation period for Hens 2241 and 2246. broods tended to have a lower percent activity than those without broods. During incubation, activity was greatly reduced and limited to a few short feed- ing periods each day. No relationship between age (juvenile 10-12 months old, adult 22 months or older) and activity levels was evident except during incubation when the 3 adult hens monitored exhibited the 3 lowest activity levels. There was no apparent relationship between color phase of the birds and activity levels. 442 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 NO. 2241 , SUNRISE SUNSET^ ^ — I — I — I — I — S — I — I — I — I — I — I — I — I I I I I I I I I I r 00 0600 1200 1800 2400 TIME NO 2246 00 0600 1200 1800 2400 TIME Fk;. 2. Daily periods of activity, visits to the nest site, and probable egg laying periods for Hens 2241 and 2246 during the pre-incubation period ( =: inactive, • • • = active, = ben at or near nest site, * = nest visits during which egg laying probably occurred, blank = no data). Pre-incubation. — The temporal distribution of percentage of 15 min in- tervals active during the pre-incubation period was determined for 9 hens I Maxson 1974). Fig. 1 gives 2 examples of these data. Activity seldom oc- curred at night. Peaks of activity were closely associated with sunrise and sunset. The evening peak was greater than the morning peak for all hens. Maxson • GROUSE ACTIVITY 443 Activity levels during the day fluctuated and were variable among birds. Daytime activity seldom fell below the 50% level and sometimes exceeded the dawn-dusk peaks for short periods. Fig. 2 illustrates daily activity for 2 hens. Daytime activity, for all hens combined, began prior to sunrise on 302 of 306 (98.7%) grouse-days and ceased after sunset on 296 of 310 (95.5%) grouse-days. Activity usually be- gan 30-60 min before sunrise and in most cases ceased 15-45 min after sun- set. The exact timing of activity onset and cessation varied slightly from day to day and was likely influenced by weather changes. Egg laying. — By coordinating activity and location data for 8 hens from which nearly continuous telemetry records were obtained, it was possible to determine when they had visited their nest sites prior to the beginning of in- cubation. I assumed that eggs were laid during some of these visits, although I could not determine exactly when egg laying occurred during a nest visit. Fig. 2 indicates all occasions whan Hens 2241 and 2246 were known to be at or near their nest sites as well as probable egg-laying visits. Both hens had 13-egg clutches. Visits to the nest followed a definite pattern. Hens seldom visited the nest site prior to the onset of egg laying, suggesting that nest construction was not very time-consuming. Once egg laying began, hens were rarely near the nest site except during presumed laying visits. During laying visits they normally remained inactive on the nest from 1 to several hours. As the clutch approached completion, hens tended to remain on the nest for longer periods. Egg-laying visits occurred 1-5 days in succession at intervals ranging from 25-30 h. Eggs were laid slightly later during each day of a laying sequence. When the next egg of a sequence appeared to be due sometime after the end of evening activity, the egg was not laid until the following morning thus be- ginning another sequence with the usual 25-30 h interval between eggs. The number of days in a laying sequence varied both among birds and for the same individual as well. As examples, the laying sequences for several hens were the following: Hen 2210, 1-2-2-3-4; Hen 2239, 2-3-3-S; Hen 2241, 3-2-4-4; Hen 2246, 1-3-5-4 (numbers indicate the number of consecutive days during which an egg was laid while hyphens indicate the skipping of a day between eggs). Similar overall patterns of egg laying were observed for all hens monitored. Since renesting by Ruffed Grouse in the wild has been proven on only 1 occasion (Barrett 1970:79-81), evidence is presented here that Hen 1695, which had only 8 eggs and did not begin incubating until 20 May (several days later than the other hens ) , successfully renested. Eig. 3 illustrates the activity of this hen during the pre-incubation period. Although some gaps occur in these data, the probable laying times of 7 eggs were determined. 444 THE WILSON BULLETIN • Vo/. 89, No. 3, September 1977 NO. 1695 Fig. 3. Daily periods of activity, visits to the nest site, probable egg laying periods, and probable visits to a previous nest site for Hen 1695 during the pre-incubation period ( = inactive, • • • = active, r= hen at or near nest site, * := nest visits during which egg laying probably occurred, :=: hen at or near probable first nest site, blank = no data) . The remaining egg may have been laid during the data gap on the morning of 17 May. The first egg of the clutch was apparently laid on 9 May. Prior to this date a pattern of activity similar to that occurring during egg laying was evident. Inactive periods possibly associated with egg laying are indi- cated in the figure. The telemetry record indicates that these were all at a single location approximately 160 m from the nest. This suggests that an earlier nest had been established on 25 April ( about the same time other hens were beginning to lay) and that up to 10 eggs had been laid. On 8 May, prior to the onset of incubation, they were likely destroyed by a predator. The sec- ond nest was apparently established the following day on 9 May. Examination of the nest visit pattern suggests that the original clutch size would have been 14 ( assuming that the first egg was laid on 25 April and the 14th was laid on 13 May). No nest visits occurred on 14 and 15 May. This was the only hen which skipped more than 1 day between laying se- (luences. Probably this time lag was required lor formation of additional eggs. Rematiiig may have been necessary as well. Incubation. — Activity patterns during the incubation period were de- termined for 12 hens. Field observations and the telemetry record indicated that hens normally left the nest only to feed. The number of nest absences per day varied from 1-5 but was most often 2 or 3 (Table 2). While no hen Maxson • GROUSE ACTIVITY 445 Table 2 Number of Days Hens Had from 1 TO 5 Nest Absences^ Total No. Nest Absences Per Day No. 1 2 3 4 5 1657 5 11 — — 1690 — 4 8 7 — 1691 1 6 9 4 — 1695 1 11 5 2 2 2210 — 6 1 — — 2219 — 1 6 — — 2235 — 11 2 — — 2238 — 2 8 2 — 2239 1 11 4 2 — 2241 — 5 7 — — 2246 — 11 7 — 2248 — — 7 6 2 TOTAL 3 73 75 23 4 1 Includes only days on which nearly constant telemetry records were obtained. had the same number of absences per day throughout the incubation period, some (e.g. Hen 2235) were fairly consistent in making 2 feeding trips per day whereas others (e.g. Hen 2238) usually made 3 trips. The length of 590 nest absences was determined to the nearest minute and the total minutes off the nest per day was calculated for 177 grouse-days. Table 3 summarizes these data and indicates intra- and inter-bird variability. Eight birds had mean absence lengths of 18-24 min. Hen 2239 had 1 unusually long absence of 197 min. This was more than twice the length of the longest absence recorded for any of the other birds. Possibly this hen was disturbed by a predator while feeding and failed to return to the nest in the usual amount of time. Excluding the 197 min absence, Hen 2239 had absences ranging from 14-65 min (mean 37 min ) . On days hens were absent only twice, the last absence tended to be longer than the first (44 of 73 grouse-days (60%)). In contrast, of 102 days when hens were absent more than twice, the last absence of the day was the longest on only 42 (41%) occasions. The total number of minutes off the nest varied from day to day for indi- vidual hens. Overall, juvenile hens spent more time off the nest than adults (juvenile mean 66 min, adult mean 46 min) suggesting that adults are more efficient incubators. Among birds, there was no consistent relationship be- tween mean absence length and mean total minutes off the nest per day. Also, hens did not consistently increase or decrease nest attentiveness as incuba- tion progressed except during the last day or 2 when eggs were in the process 446 THE WILSOxN BULLETIN • Vol. 89, No. 3, September 1977 Table 3 Summary of Nest Absences During the Incubation Period Bird No. No. of Recorded Absences^ Absence Length (min) Total Time Off Nest Per Day (min)- Mean Range Mean Range 1657 49 28 17- 77 70 47- 93 1690 65 19 8- 56 57 40- 77 1691 59 15 7- 32 41 18- 58 1695 64 21 6- 42 58 41- 96 2210 19 34 20- 82 65 46- 81 2219 32 24 11- 37 62 46- 76 2235 38 24 12- 59 46 34- 76 2238 50 22 13- 41 66 45- 93 2239 47 40 14-197 90 28-242 2241 37 23 5- 32 60 43- 81 2246 57 21 8- 37 52 33- 77 2248 73 19 7- 33 70 53-100 1 Includes only absences where departure and return were determined to the nearest minute. 2 Includes only those days on which nearly constant telemetry data were obtained. of hatching. At that time the birds seemed restless and were frequently active at the nest site for short periods. Daily activity of 3 hens is illustrated in Fig. 4. All hens demonstrated a shift in the temporal distribution of the first activity period of the day as in- cubation progressed. Activity usually began prior to sunrise during the early stages of incubation hut later the start of activity was delayed as much as 4^5 h after sunrise. The beginning of this shift in activity ranged from 20-30 May in 1971 and from 16-25 May in 1972. Synchrony among hens was evident in 1972 when 5 of 7 began this shift during 22-25 May. The shift began during different stages of the incubation period (from the 4th to 19th day ) for different hens. No relationship was evident between the onset of the activity shift and any trend in average hourly temperature, wind velocity and direction, or amount and time of occurrence of precipitation. A possible relationship between plant phenology and the activity shift was noted in 1972 when phenologic events were studied in detail. During the week of 18-25 May the leaves of most trees and shrubs reached full size. The rapid leafing out resulted in a “closing in” of the forest canopy. During this period the ferns which made up much of the understory grew rapidly and nearly attained full size (approx. 1ml. The net effect was a marked reduc- tion in the amount of light reaching the forest floor. Synchronization of circadian activity rhythms has been shown to be strongly affected by the day- Maxson • GROUSE ACTIVITY 447 NO. 1695 MAY 20 25 - 30- JUNE 5 10 “T 1 1 1 r— I 1 1 — r 00 0600 T 1 1 r 1200 TIME T — I — I — 1800 ■T — r 2400 NO. 2238 MAY 20 - JUNE 25- 30- 5- 10 - SUNRISE SUNSET I I — I — I — r*— 1 — I — I — I — I — I — I — I — I — I — I — I — I — I — I — 1 — I — I — I — f 00 0600 1200 1800 2400 TIME NO. 2246 TIME Fig. 4. Daily periods of activity during the incubation period for Hens 1695, 2238, and 2245 ( = inactive, • • • = active, = known disturbances by humans, blank = no data) . 448 THE WILSON BULLETIN • Vol 89, No. 3, September 1977 NO. 2239 NO. 2241 Fig. 5. Temporal distribution of percentage of 15-min intervals active during the post- incubation period for Hen 2239 (with brood) and Hen 2241 (no brood). night cycle of illumination (Aschoff 1966). If the onset of morning activity during incubation is associated with light intensity at the nest, activity would likely start increasingly later as the tree canopy leafed over and herbaceous vegetation grew up around the nest. Maxson • GROUSE ACTIVITY 449 On several occasions precipitation noticeably affected normal activity pat- terns. On 19 May 1971 a snow-rain storm from about 04:15-10:30 left a temporary accumulation of snow on the ground. Of 3 hens incubating at that time 2 did not leave the nest to feed until 14:00-14:15. The third re- mained on the nest until 17:30. Rain occurred during most of the morning hours on 29 May 1972. Several hens including 2238 and 2246 (Fig. 4) de- layed or omitted the normal morning feeding period. On another occasion, at the onset of a hard rain shower, several hens which had been feeding re- turned to their nests almost immediately. This behavior is adaptive since egg temperatures are maintained during periods when rapid chilling would prob- ably occur if the hen was absent. Post-incubation. — Temporal distribution of percentage of 15 min intervals active during the post-incubation period was determined for 7 hens with broods and 4 broodless hens which had lost clutches to predators (Maxson 1974). As examples of these data Fig. 5 illustrates the activity patterns of Hen 2239 (with brood) and Hen 2241 (no brood). With but a single excep- tion (Hen 2202), once activity peaked in the morning the birds maintained a high percentage of activity throughout most of the daylight hours until ac- tivity ceased in the evening. These daytime levels of activity were often higher than those observed during the pre-incubation period. Evening activity peaks tended to be greater than early morning peaks, but the pattern was not so con- sistent as during pre-incubation. A difference in timing of the dawn-dusk peaks between hens with broods and those without broods was readily apparent. Brood hens did not attain the morning activity peak until 1-2% h after sunrise. The evening peak oc- curred 15 min-2 h prior to sunset. Broodless hens reached morning activity peaks 15-30 min after sunrise and evening peaks from 15 min before to 15 min after sunset. Daily activity during the post-incubation period is illustrated for Hens 2239 and 2241 in Fig. 6. Hen 2239 began activity 2-3 h after sunrise during the first few days following hatching of the chicks. Thereafter, the onset of ac- tivity became progressively earlier until it approached sunrise on the 13th day. An almost identical pattern was exhibited by all brood hens. For the post-incubation period as a whole, brood hens began activity after sunrise on 101 of 114 (88.6%) grouse-days. Activity ceased prior to sunset on 104 of 115 (90.4%) grouse-days. In contrast, broodless hens initiated daytime activity prior to sunrise on 74 of 105 (70.5%) grouse-days and ceased activity after sunset on 55 of 83 (66.3%) grouse-days. No doubt the delaying of morning activity onset and early evening activity cessation by brood hens is related to brooding of the chicks during cooler 450 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 NO. 2239 TIME I4g. 6. Daily periods of activity during the post-incubation period for Hen 2239 (with 1)100(1 1 and Hen 2241 (no brood) ( = inactive, • • • = active, blank =: no data). periods of the day. Brooding at these times was most pronounced during the first few days when chicks were least able to maintain body temperatures. Young chicks moving about during early morning hours would be exposed to wetting and rapid chilling due to heavy dew normally present. As the chicks grew older and more independent of the brooding hen, activity patterns of these hens gradually began to resemble those of broodless hens. Maxson • (i ROUSE ACTIVITY 451 DISCUSSION A major concern of telemetry studies is that the transmitter may cause be- havioral changes in the study animals. Since many telemetry studies are con- ducted on animals difficult to observe in the wild, effects of the transmitter on behavior are difficult to determine. Boag (1972) stated that activity levels as well as food intake by captive female Red Grouse ( Lagopus 1. scoticus ) were lower among radio-marked birds than controls, especially during the first week after transmitter attachment. At Cedar Creek properly fitted transmitter harnesses had no noticeable effects on behavior of Ruffed Grouse. The har- nesses fitted so well that within a short time after release only the whip an- tenna on the bird’s back was visible. During field observations it was not pos- sible to distinguish marked from unmarked birds unless the whip antenna could be seen. Observations of captive grouse at Cedar Creek by Huempfner and Maxson ( unpubl. data ) also failed to reveal behavioral differences be- tween marked and unmarked birds. Data concerning Ruffed Grouse activity are available from several telemetry studies. Huempfner ( unpubl. data ) found that mid-day activity during the winter months (especially when birds were able to snow-burrow roost) was substantially lower than mid-day activity levels recorded during the present study. Brander (1965) reported that radio-marked Ruffed Grouse, in the absence of sufficient snow for burrowing, roosted at sunset and left the roost about 30 min before sunrise during March and April. His birds were inactive from mid-morning until mid-afternoon especially during March. The inactive periods probably reflect the need to conserve energy during the winter months. Schladweiler ( 1965 ) stated that Ruffed Grouse generally began activity 30-60 min before sunrise and ceased activity 30-60 min after sunset during the breeding season. Once activity began his birds tended to remain active most of the day without any prolonged inactive periods. Weather disturbed them little except when chicks were small. Probably a transition from pro- longed mid-day inactive periods (Brander 1965, Huempfner unpubl. data) to few inactive periods during daylight hours ( Schladweiler 1965, the present study) occurs during spring as temperatures increase and snow cover dis- appears. The pattern of activity found in the present study ( Figs. 1 and 5 ) is similar to the 2-peak activity pattern reported for several species of day-active birds by Aschoff (1966). The pattern of dawn peaks usually being lower than the dusk peaks was also evident in the data of other Gedar Creek researchers, e.g. Archibald (1973) for male grouse during spring and Pierson and Tester (unpubl. data) for several grouse of both sexes during November. This may prove to be the normal pattern throughout the year in this species. These 452 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 morning and evening activity peaks are probably associated with feeding periods. Archibald (1973) observed male grouse feeding in trembling as- pen {Populus tremuloides) clones during these times. On several occasions in the present study, I observed female grouse feeding in male aspen clones during the morning and evening activity peaks. Egg laying. — Egg laying patterns similar to those of the present study (Figs. 2 and 3) have been found in domestic chickens {Callus gallus) (War- ren and Scott 1935, Scott and Warren, 1936), and pheasants {Phasianus colchicus) (Labisky and Jackson 1966). Warren and Scott (1935) stated that differences in time intervals between eggs in a sequence were probably due to differences in the time an egg spends in the uterus. Scott and War- ren (1936) determined that skipping of one or more days between laying sequences was caused by a delay in ovulation of the first egg of a sequence rather than by overnight retention of a fully developed egg in the uterus. Probably the Ruffed Grouse follows a similar pattern of delaying ovulation between laying sequences. The tendency for hens to remain inactive on the nest during laying visits for longer periods as the clutch neared completion was also noted in Spruce Grouse (Canachites canadensis franklinii) (McCourt et al. 1973). Incubation. — Bump et al. (1947:288-289), Kupa (1966), and Schlad- weiler ( 1968) reported that incubating Ruffed Grouse usually left the nest for short periods, only a few times per day, as I found in the present study (Table 2, Fig. 4) . Skutch (1962) stated that, with the exception of certain nidifugous species, most birds which incubate alone cover their eggs 60-80% of the daytime. Tetraonids of at least several species exhibit nest attentiveness greater than 80%. In the present study, hens averaged 95.7% of the incubation period on the eggs. Kupa ( 1966) stated that Ruffed Grouse hens spent an average of 23 h and 12 min (96.7%) on the nests each day. Lennerstedt (1966) re- ported that a Gapercaillie (Tetrao urogallus) hen, during 12 days of its incu- bation period, was off the nest only 4.9% of the time. McCourt et al. (1973) stated that 2 incubating Spruce Grouse spent 93% of daylight hours on the nest. The shift in timing of the first activity period of the day exhibited by all hens in the present study (Fig. 4) has not been noted by other researchers. Lennerstedt ( 1966) reported that activity periods of an incubating Caper- caillie were fairly evenly distributed throughout the day except from 18:00- 23:00 when no absences occurred. Although his study was conducted under conditions of continuous daylight the data indicate that onset of activity periods between 03:00-06:00 gradually shifted from slightly after 03:00 to Maxson • GROUSE ACTIVITY 453 about 05:20 over the period of study. This shift is similar to that observed during my study but more data are needed to determine if the pattern is con- sistent among birds. SUMMARY Seventeen female Ruffed Grouse were equipped with radio transmitters and monitored with an automatic radio tracking system. Activity data were divided into pre-incubation, incubation, and post-incubation periods for each hen. During the pre-incubation period peaks of activity were closely associated with sun- rise and sunset. The evening peak was greater than the morning peak for all hens. Ac- tivity normally began 30-60 min prior to sunrise and ceased 15-45 min after sunset. Day- time activity seldom fell below the 50% level. Egg laying patterns were determined for 8 hens. Hens seldom visited the nest site prior to the onset of egg laying and, once laying began, were rarely near the nest except during presumed laying visits. During laying visits hens typically remained inactive on the nest from 1 to several hours. Laying visits occurred 1-5 days in succession at 25-30 h inter- vals. When the next egg of a sequence appeared to be due sometime after the end of evening activity the egg was not laid until the following morning. Evidence is presented that one hen successfully renested. During incubation hens most often left the nest 2 or 3 times per day to feed. Most birds averaged 18-24 min per absence and 57-70 min off the nest per day. All hens ex- hibited a change in timing of the first activity period of the day as incubation progressed. This activity change may be related to plant phenology. During post-incubation, once activity began in the morning, birds usually maintained a high percentage of activity throughout most of the daylight hours until activity ceased in the evening. Evening activity peaks tended to be higher than morning peaks. Brood hens did not attain the morning peak until 1-2% h after sunrise. The evening peak occurred 15 min-2 h prior to sunset. In contrast, broodless hens reached morning peaks 15-30 min after sunrise and evening peaks from 15 min before to 15 min after sunset. This difference between hens with and without broods is related to brooding of the chicks during the cooler portions of the day by the brood hens. ACKNOWLEDGMENTS I wish to express sincere appreciation to David F. Parmelee, John R. Tester, and Wil- liam H. Marshall for their invaluable advice and assistance throughout the course of this study. For the design, construction, and maintenance of the telemetry equipment I am indebted to Ralph Schuster, Richard Reichle, and Valerian Kuechle. I am grateful to Richard Huempfner, Gary Erickson, and D. Andrew Saunders for assistance with the field work. The cooperation of the many personnel of the Cedar Creek Natural History Area is gratefully acknowledged. This investigation was supported by the U.S. Atomic Energy Commission (COO-1332-108) . I thank Lewis W. Oring and George-Ann Maxson for critically reviewing the manuscript. LITERATURE CITED Archibald, H. L. 1973. Spring drumming activity and space use of Ruffed Grouse. Ph.D. thesis, Univ. of Minnesota, Minneapolis. 454 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 Aschoff, J. 1956. Circadian activity pattern with two peaks. Ecology 47:657-662. Barrett, R. W. 1970. Behavior of Ruffed Grouse during the breeding and early brood periods. Ph.D. thesis, Univ. of Minnesota, St. Paul. Boag, D. a. 1972. Effects of radio packages on behavior of captive Red Grouse. J. Wildl. Manage. 36:511-518. Brander, R. B. 1955. Factors affecting dispersion of Ruffed Grouse during late win- ter and spring on the Cloquet Forest Research Center, Minnesota. Ph.D. thesis, Univ. of Minnesota, St. Paul. . 1958. A radio-package harness for game birds. J. Wildl. Manage. 32:630-632. Bump, G., R. W. Darrow, F. C. Edminster, and W. F. Crissey. 1947. The Ruffed Grouse: life history, propagation, management. New York State Conserv. Dept., Al- bany. Cochran, W. W., D. W. Warner, J. R. Tester, and V. B. Kuechle. 1965. Automatic radio-tracking system for monitoring animal movements. BioScience 15:98-100. Gullion, G. W. 1965. Improvements in methods for trapping and marking Ruffed Grouse. J. Wildl. Manage. 29:109-116. AND W. H. Marshall. 1968. Survival of Ruffed Grouse in a boreal forest. Living Bird 7:117-167. Huempfner, R. a., S. j. Maxson, G. J. Erickson, and R. J. Schuster. 1975. Recap- turing radio-tagged Ruffed Grouse by nightlighting and snow-burrow^ netting. J. Wildl. Manage. 39:821-823. Kupa, j. j. 1956. Ecological studies of the female Ruffed Grouse iBonasa umbellus L.) at the Cloquet Forest Research Center, Minn. Ph.D. thesis, Univ. of Minnesota, St. Paul. Labisky, R. F. and G. L. Jackson. 1966. Characteristics of egg laying and eggs of yearling Pheasants. Wilson Bull. 78:379-399. Lennerstedt, I. 1966. Egg temperature and incubation rbythm of a Capercaillie ( Telrao urogo/lus L.) in Swedish Lapland. Oikos 17:169-174. Marshall, W. H. and J. J. Kupa. 1963. Development of radio-telemetry techniques for Ruffed Grouse studies. Trans. N. Am. Wildl. Nat. Resour. Conf. 28:443-456. Maxson, S j. 1974. Activity, home range, and habitat usage of female Ruffed Grouse during the egg-laying, incubation, and early brood periods as determined by radio- telemetry. M.S. thesis, Univ. of Minnesota, Minneapolis. McCourt, K. H., D. a. Boag, and 1). M. Keppie. 1973. Female Spruce Grouse ac- tivities during laying and incubation. Auk 90:619-623. Robel, R. j., j. N. Briggs, J. J. Cebula, N. J. Silvey, C. E. Viers, and P. G. Watt. 1970. Greater Prairie Chicken ranges, movements, and habitat usage in Kansas. J. Wildl. Manage. 34:286-305. Sargeant, a. B., j. E. Forbes, and I). W. Warner. 1965. Accuracy of data obtained through the Cedar Creek automatic radio-tracking system. Minnesota Mus. Nat. Hist. Tech. Rept. No. 10. ScHLADWEiLER, P. 1965. Movements and activities of Ruffed Grouse iBonasa umbel- lus L. ) during the summer period. M.S. thesis, Univ. of Minnesota, St. Paul. . 1968. Feeding behavior of incubating Ruffed Grouse females. J. Wildl. Manage. 32:426-428. Scott, H. M. and D. C. Warren. 1936. Influence of ovulation rate on the tendency of the fowl to produce eggs in clutches. Poult. Sci. 15:381-389. Skutch, a. F. 1962. The constancy of incubation. Wilson Bull. 74:115-152. Maxson • GROUSE ACTIVITY 455 Warren, D. C. and H. M. Scott. 1935. The time factor in egg production. Poult. Sci. 14:195-208. Weller, M. W. 1957. An automatic nest-trap for waterfowl. J. Wildl. Manage. 21: 456-457. DEPT. OF ECOLOGY AND BEHAVIORAL BIOLOGY, UNIV. OF MINNESOTA, ST. PAUL, MN. 55101 (present address: dept, of biology, UNIV. OF NORTH DA- KOTA, GRAND FORKS, N.D. 58202). ACCEPTED: 7 APRIL 1976. WEIGHTS AND FAT CONDITION OF SOME MIGRANT WARBLERS IN JAMAICA A. W. Diamond, P. Lack, and R. W. Smith In the Old World, deposition of fat before migrating has been described in both summer and winter quarters of many Palaearctic bird species ( refer- ences in Pearson 1971), but in the New World there has been only one study outside continental North America (Rogers and Odum 1966). During a study of the annual cycles of forest birds in Jamaica ( Diamond 1974), we caught over 400 parulid warblers of 19 species (see also Diamond and Smith 1973, Lack and Lack 1973 j. This paper describes the variation in weight and visible fat condition of 302 of the 7 most commonly caught species; mean weights of all the migrant warblers caught are given in an Appendix. War- blers were caught between October 1970 and May 1971 by A. W. D. and P. L., in August and September 1971 by A. W. D. and from September 1971 to April 1972 by R. W. S. and S. Gowen. Nomenclature follows Bond (1971) for birds, and Adams (1972) for plants. TRAPPING SITES AND METHODS The 4 main trapping sites (Fig. la) were as follows: Port Henderson Hill. — Altitude 155 m. Low xeric scrub (“dry limestone scrub forest” of Asprey and Robbins 1953), much disturbed by cutting for charcoal, rarely exceeding 3 m high, and dominated by red birch (“Gumbo Limbo”) {Bursera simaruba) and the tall cactus Stenocereus hystrix; Mona Woods. — Altitude 185 m. A small patch of secondary riverine forest, with a canopy 15 to 18 m high and a dense undergrowth of shrubs and creepers. Most birds were caught beside a stream leading out of the Mona Reservoir; Irish Town. — Altitude 770 m. A small garden on the crest of a ridge, wdth thick scrub on the slopes and secondary forest, mostly of native trees, in the valley on one side; Green Hills. — Altitude 1080 m. The garden of the Institute of Jamaica Field Station, surrounded by montane forest, on the northern (windward) side of the western end of the Blue Mountain range. All the birds were caught in mist nets, mostly between dawn and noon but some, espe- cially at Irish Town, in the evening. All birds were weighed to the nearest 0.25 g on a “Pesola” spring balance with a range of 0-50 g. Subcutaneous fat was estimated as follows ( Diamond 1974) : each of 4 areas of the body (furculum, axilla, abdomen, and rump) was scored independently, on a scale from 0 (no fat) to 3 (fat mounded), giving a summed possible range of scores of 0-12. Fat score and weight were correlated ( Spearman’s rho r= 0.3, p < .01), but the relationship is weak enough that weight and fat score did not always vary in parallel. Insufficient data were obtained for a detailed analysis of diurnal changes in weight or fat in any one species. However when all 7 species were treated together, weights taken before 10:00 were significantly less (by about 5%) than those taken after 10:00 456 Diamond et al. • WARBLER WEIGHT AND FAT 457 Fig. 1. (a) Sketch map of Jamaica. Inset: location of netting sites described in text. (b)-(d) Changes in weight and visible subcutaneous fat through the winter for (b) Black-and-white Warbler, (c) Worm-eating Warbler, and (d) Black-throated Blue Warbler. Individual weights are shown as solid circles; a solid line connects monthly means. Open circles show monthly median fat scores. Weight and fat are corrected for diurnal variation (see text). (p < .001, 2-tailed t-test). Fat scores before 10:00 also averaged 60% lower than those after 10:00 (p < .001, 2-tailed median test). Unless noted otherwise, all weights and fat scores in this paper are “morning” ones, those after 10:00 having been corrected by subtracting 5% (weight) or 60% (fat score). Weight loss of birds kept overnight was greater than 5%, averaging 8.9% (12 birds) ; 5% represents the average weight difference over a time period varying from 0 to 12 h, and so is less than the weight lost overnight, i.e. over 12 h. In 4 species it has been possible to compare the weights from Jamaica with weights from a North American breeding area, the Powdermill banding station in Pennsylvania (referred to below simply as “Pennsylvania”) ; Powdermill weights from June and July only have been used, since birds caught then are likely to be local breeders rather than migrant birds. The other 3 species were not caught regularly in the summer at Powder- mill so for comparison we have used weights given by Baldwin and Kendeigh (1938) and Wetherbee (1934) although these, like our Jamaican weights, may include some migrant birds. SPECIES ACCOUNTS Black-and-white Warbler {Mniotilta varia) . — These were common through- out the island, and were seen from early September through May. This was 458 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 one of 2 species caught regularly at all 4 trapping sites. Fat score declined gradually through the winter (Fig. lb); weight remained steady, but weights of retrapped birds (Table 1 ) appeared to decrease slightly through the winter. The few birds caught in April were slightly heavier than those trapped in March, and the single bird caught in May had the maximum possible fat score and was about 25% heavier than any bird caught during the winter. Winter weights were significantly lower than the mean weight (10.9 g) of 15 birds caught in Pennsylvania in June and July ( 2-tailed t-test, p < .001), and several February weights were below the lowest weight (8.5 g) recorded by Drury and Keith (1962) in migrant birds. Worm-eating Warbler { Helmitheros vermivorus) . — Widespread from late September to April. This rather skulking species fed mainly in undergrowth and was caught at all the trapping sites except Port Henderson. The few birds caught in September were lean; fat scores rose thereafter, except in December, but weights remained steady through the winter until March, when both weight and fat score rose sharply (Fig. Ic). One individual retrapped in late March was 21% heavier than in early February. Black-throated Blue Warbler { Dendroica caerulescens) . — This species ar- rived later than most others, none being caught before October; it was caught in all the trapping sites except Port Henderson. Weight and fat declined slightly from October through February, rising thereafter to a maximum in May (Fig. Id). The weights of the 2 birds retrapped in May were 46% and 62% higher than their respective winter weights (Table 1). Several birds caught in October and November were considerably heavier than most caught during the winter. Two birds weighed in February were close to the fat-free weight of 7.6 g given by Connell et al. (1960). Prairie Warbler (Dendroica discolor). — Found mainly in the lowlands. Prairie Warblers were caught between late August and April, most commonly at Port Henderson hut also at Mona Woods and Irish Town. Weight and fat were high in autumn and spring, low from October through March (Fig. 2 a). The average weight of 3 immatures in summer, 7.2 g (Wetherbee 1934), is higher than most winter weights. Ovenbird (Seiurus aurocapillus) . — This and the Black-and-white Warbler were the only species caught regularly at all trapping sites. Most birds were lean in September and October ( Fig. 2b), and weights increased from Sep- tember through November, then remained steady until March. The few indi- viduals caught in April were the fattest and heaviest Ovenbirds caught during the study, and one bird caught twice in April increased in weight by 41% in 20 days. Most winter weights were below the mean weight of 19.7 g given by Wetherbee (1934) and Baldwin and Kendeigh (1938) for 15 birds be- tween May and August, but were not significantly different from those of 16 Diamond et a/. • WARBLER WEIGHT AND FAT 459 Table 1 Weight Changes in Retrapped Individual Warblers Black-and-white Warbler A* 18 Oct— 9.75** F 17 Nov— 9.5 K 30 Nov— 9.5 8 Jan— 9.0 16 Dec— 9.0 21 Feb— 9.25 12 Jan— 9.0 25 Mar— 9.25 B 31 Oct— 9.25 8 Mar— 9.0 1 Nov— 8.5 L 24 Jan— 9.75 6 Nov— 9.25 G 20 Nov— 10.25 30 Jan— 9.75 11 Dec— 9.5 6 Feb— 9.5 C 31 Oct— 9.0 4 Dec— 9.25 H 1 Dec— 9.25 M 20 Feb— 10.0 14 Feb— 9.25 16 Dec — 8.5 25 Mar— 8.75 8 Feb— 8.25 D 3 Nov— 10.0 E 18 Oct— 9.5 29 Mar— 10.25 I 16 Dec — 9.0 28 Nov— 9.0 15 Feb— 8.75 E 6 Nov— 10.5 0 7 Nov— 10.25 31 Jan— 10.25 J 1 Feb— 9.0 4 Dec— 9.5 9 Feb— 8.5 Worm-eating Warbler A 1 Nov— 12.5 23 Jan— 12.5 B 14 Nov— 12.0 5 Mar— 12.0 C 18 Oct— 13.25 16 Nov— 13.75 D 14 Nov— 11.5 29 Dec— 12.5 24 Mar— 12.0 E 6 Feb— 12.5 25 Mar— 15.25 F 17 Oct— 12.75 27 Nov— 12.5 Black-throated Blue Warbler A 31 Oct— 9.75 D 14 Feb— 8.25 H 3 Nov— 8.25 24 Jan— 9.0 20 Feb— 8.75 16 Nov— 9.25 30 Jan — 9.0 24 Nov— 8.75 14 Feb— 9.25 E 24 Nov— 8.75 8 Jan— 9.75 5 May— 12.75 B 24 Nov— 8.5 I 17 Dec— 8.75 7 Feb— 7.5 F 13 Mar— 10.0 25 Apr— 10.75 23 Dec— 9.0 C 7 Dec— 9.5 J 24 Dec— 7.0 6 Feb— 9.25 G 24 Oct— 11.0 30 Apr— 10.0 5 May— 11.5 460 THE WILSON BULLETIN • Vol 89, No. 3, September 1977 Table 1 (Continued) Ovenbird A 31 Oct— 16.5 c 12 Jan— 19.5 F 29 Oct— 20.0 30 Nov— 17.5 29 Dec — 17.0 15 Feb— 19.5 12 Nov— 20.25 20 Jan— 17.0 D 24 Mar— 19.5 G 11 Dec— 19.0 31 Jan— 15.25 31 Mar— 20.25 23 Dec— 19.5 B 23 Oct— 19.0 E 28 Nov— 18.25 H 23 Jan— 19.0 8 Apr— 17.5 12 Dec— 18.5 30 Jan— 19.25 28 Apr— 24.5 I 24 Nov— 19.25 24 Dec— 20.25 Common Yellowthroat A 23 Dec— 9.5 D 19 Nov— 9.0 G 2 Oct— 11.25 11 Jan— 9.75 9 Feb— 9.25 16 Feb— 9.25 11 Dec— 10.25 B 1 Feb— 10.25 H 11 Dec— 10.5 9 Feb— 9.75 E 29 Dec— 9.0 30 Jan— 9.0 28 Apr— 10.0 C 16 Feb— 10.5 I 11 Dec— 10.5 24 Mar— 9.75 F 8 Nov— 9.75 25 Apr— 11.5 8 Apr — 8.75 Prairie Warbler A 17 Nov— 6.5 B 17 Nov — 6.25 12 Jan— 6.5 15 Feb — 6.5 8 Mar— 6.5 American Redstart A 1 Nov — 7.0 B 14 Feb— 7.0 29 Dec— 6.5 24 Jan— 6.5 5 Mar— 7.0 * Dates and weights for each lettered group indicate successive captures of one individual. ** Weight in grams. birds caught in Pennsylvania in June and July and were well above the mean fat-free weight of 16.0 g given by Rogers and Odum (1966). One bird, retrapped 5 times at Irish Town, dropped in weight suddenly in January and was not caught again (Table 1) ; its last recorded weight was about 20% below the average winter weight. Diamond et al. • WARBLER WEIGHT AND FAT 461 Fig. 2. As Fig. 1 (b)-(d). (a) Prairie Warbler, (b) Ovenbird. (c) Common Yel- lowthroat. (d) American Redstart. Common Yellowthroat (Geothlypis trichas) . — Present from September through April, Yellowthroats favored dense undergrowth and long grass; most were caught in Mona Woods, where both these habitats abound, but a few were trapped at Irish Town and Green Hills. Weight and fat were as high in autumn as in spring (Fig. 2 c). Most winter weights were below the 10.4 g mean of 24 summer weights given by Wetherbee (1934) and Baldwin and Kendeigh (1938), but did not differ significantly from those of 60 birds weighed in Pennsylvania in June and July. American Redstart (Setophaga ruticilla). — We caught this species between late August and mid-May (although according to Bond (1971) it is found throughout the year in the Greater Antilles) in all habitats but most com- monly at Port Henderson. Autumn weights were very variable, and many were higher than the few spring weights obtained (Fig. 2d). Both fat score and weight declined from September through November; fat apparently de- clined again from January through March, while weight remained constant, but this apparent difference may be due to the small number of birds scored for fat during this period. Most winter weights were within the range of fat- free weights (6.6 to 7.1 g) given by Rogers and Odum (1964), and most 462 THE WILSON BULLETIN • Vol. 89, No. 3, September 1977 were below the lowest weight (7.1 g) recorded by Drury and Keith (1962) in migrant birds; the mean weight of birds caught in Jamaica was signifi- cantly lower than the mean (8.6 g) of 25 birds caught in June and July in Pennsylvania (2-tailed t-test, p < .001). DISCUSSION All the species described here increased in both weight and fat score prior to the spring migration. As has been found in North American migrants leaving Central America (Rogers and Odum 1966) and in Palaearctic mi- grants leaving Africa (Pearson 1971), very few fat individuals were caught in spring, and populations disappeared very soon after the first fat birds were recorded; this could be due either to birds laying down fat very quickly, or to their departing while still lean, or to a change in feeding be- havior which makes the birds more difficult to catch. That some Palaearctic migrants in Africa do lay down fat very quickly is well known, but Pearson (1971) thought that some leave their wintering areas, particularly those well to the south of the Sahara, in a relatively lean condition. The same may be true in Jamaica, since Cuba would make a convenient stepping-stone for birds making for Florida, a further 300 km to the north, but we caught too few birds in spring to be able to decide among these possibilities. Rogers and Odum (1966) found that many migrants arriving in Central America in autumn were extremely lean and may have begun to use non-fat tissue as fuel. Some species were very lean on arrival in Jamaica { Helmitheros vermivorus^ Seiurus aurocapillus ) , but most were quite heavy on arrival and some { Dendroica discolor^ Geothlypis trichas, Setophaga ruticilla) were as fat then as in the spring. Pearson (1971) interpreted the presence of heavy, fat birds in autumn as evidence of migration through the area. He pointed out that this could be confirmed only by retrapping birds which arrive with very little fat, put on weight quickly and then leave, but only very intensive trapping could reveal this pattern. It is also possible that winter residents may lose weight after arrival in autumn, having put on more fat than was used to reach the winter quarters. This seems to happen more often in spring than in autumn, at least in the New World (Rogers and Odum 1966) ; among Palaearctic species there is less evidence, but Reed and Sedge warblers { Acrocephalus scirpaceus and A. schoenobaenus ) frequently lose weight after arriving in their summer quarters in Britain (G. Hirons, pers. comm.). Pre- sumably it is advantageous for a migrant to put on as much fat as possible before a long flight, even if some of it is not needed if flying conditions prove favorable. Most birds which winter in the Antilles arrive there from the north (Drury Diamond et al. • WARHLEK WEIGHT AND FAT 463 and Keith 1962). Some of these species winter in northern South America as well as the West Indies, but others, such as Prairie and Black-throated Blue warblers, winter almost exclusively on Caribbean islands. Most Prairie and Black-throated Blue warblers which arrive in Jamaica are probably winter residents. It is possible, however, that some Prairie Warblers pass through Jamaica on their way to winter quarters further east in the Caribbean, since Jamaica lies only a little to the south of the direct line between the western end of the Prairie Warbler’s breeding range and islands in the eastern Carib- bean. American Redstarts, Yellowthroats and Black-and-white Warblers all winter in northern South America as well as in the Antilles, and the heavy individuals of these species which were caught in autumn may have included some migrating birds. Few heavy Worm-eating Warblers were caught in autumn; this species does not winter in South America (Bond 1971) so most of the birds caught in Jamaica are probably winter residents, though a few may be on migration to Central America. The Ovenbird was the only species in which there was evidence that any birds arrived in a depleted condition; no particularly heavy birds were caught in autumn, although the species does winter in northern South America and some migration through Jamaica might have been expected. A notable feature in most species was the low level of fat carried through the winter; median fat scores for the months November through February were between I and 4. Leek (1972) suggested that migrants in Panama were under greater feeding pressure than residents, particularly in bad weather, and if this were also true in Jamaica, it might explain the low fat levels car- ried by migrants during the winter. However, in Jamaica at least, there is little if any competition between migrants and residents (Lack and Lack 1973), and most resident species are fatter during the winter than during the breeding season (Diamond 1974). Residents do not show any spring fatten- ing comparable to that of migrants. Most begin to breed at the same time as the migrants lay down fat, presumably in response to the increase in insect numbers which begins in February and March. In addition to the small fat deposits carried by most species in winter, 2 species, the Black-and-white Warbler and American Redstart, were also lighter than birds weighed in summer on their breeding grounds. Comparisons be- tween Jamaican and North American weights must be treated with caution, since the origin of the Jamaican wintering birds is not known. It is worth pointing out that warblers may well be leaner and lighter in winter than in summer; a similar suggestion was made by Moreau (1944) in comparing the weights of Palaearctic migrants in Europe with those in Africa. In both cases the comparison is difficult because there are so few published weights of breeding birds. 464 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 SUMMARY Weights and fat scores are described for 7 species of migrant parulids mist-netted in Jamaica in 1970-72. Weights were variable in autumn in most species, some of which may have been on migration; most autumn arrivals were probably winter residents. Fat levels and weights were low during the winter and 2 species (Black-and-white Warbler and American Red- start) were significantly lighter than birds caught in June and July in Pennsylvania. Mean weights of all species increased in spring, though few birds were trapped then and it is possible that some individuals left Jamaica with low fat levels. ACKNOWLEDGMENTS A. W. Diamond was supported by a grant to the late Dr. D. Lack from the Natural Environment Research Council, which is gratefully acknowledged. Dr. Lack initiated and encouraged this work and he, Drs. C. and A. Kepler, and Dr. M. P. Harris have improved the manuscript by their comments. We thank the Institute of Jamaica for permission to use their Field Station at Green Hills, and the Zoology Department of the University of the West Indies, Mona, for their hospitality. Mrs. E. Diamond helped with most of the field work. Dr. Mary H. Clench very kindly made available the weights of warblers caught at the Powdermill Nature Reserve banding station of Carnegie Museum of Natu- ral History, Pittsburgh, Pa. and improved the manuscript by her pertinent comments. LITERATURE CITED Adams, C. S. 1972. Flowering plants of Jamaica. Univ. of the West Indies, Mona, Jamaica. Asprey, G. F. and R. G. Robbins. 1953. The vegetation of Jamaica. Ecol. Monogr. 23:359-412. Baldwin, S. P. and S. C. Kendeigii. 1938. Variations in the weights of birds. Auk 55: 416-467. Bond, J. 1971. Birds of the West Indies. Collins, London. Connell, C. E., E. P. Odum, and H. Kale. 1960. Fat-free weights of birds. Auk 77: 1-9. Diamond, A. W. 1974. Annual cycles in Jamaican forest birds. J. Zool., Proc. Zool. Soc. Lond. 173:277-301. AND R. W. Smith. 1973. Returns and survival of banded warblers wintering in Jamaica. Bird-Banding 44:221-224. Drury, W. H. and J. A. Keith. 1962. Radar studies of songbird migration in coastal New England. Ibis 104:449-489. Lack, D. and P. Lack. 1973. Wintering warblers in Jamaica. Living Bird 11:129-153. Leck, C. F. 1972. The impact of some North American migrants at fruiting trees in Panama. Auk 89:842-850. Moreau, R. E. 1944. Some weights of African and of wintering Palaearctic birds. Ibis 86:16-29. Pearson, D. J. 1971. Weights of some Palaearctic migrants in southern Uganda. Ibis 113:173-184. Rogers, I). T. and E. P. Odum. 1964. Effect of age, sex, and level of fat deposition on major body components in some wood warblers. Auk 81:505-513. Diamond et al. • WARBLER WEIGHT AND FAT 465 AND . 1966, A study of autumn postmigrant weights and vernal fattening of North American migrants in the tropics. Wilson Bull. 78:415-433. Wetherbee, K. B. 1934. Some measurements and weights of live birds. Bird-banding 5:55-63. A. W. D. AND P. L.: EDWARD GREY INSTITUTE OF FIELD ORNITHOLOGY, SOUTH PARKS ROAD, OXFORD 0X1 3pS, ENGLAND. R. W. S. : IVY LODGE, THE CRICKET GREEN, STOKENCHURCH, BUCKS., ENGLAND. (PRESENT ADDRESS A.W.D.: ZOOLOGY DEPT., UNIV. OF NAIROBI, P.O. 30197, NAIROBI, KENYA.) ACCEPTED 20 JAN. 1976. 466 THE WILSON BULLETIN • VoL 89, No. 3, September 1977 Mean Weights (g) Appendix OF Migrant Parulid Warblers IN Jamaica n X S2 max Black-and-white Warbler $ 14 9.25 0.2 10.0 Mniotilta varia 9 30 9.5 0.2 13.0 Swainson’s Warbler 18 15.0 1.1 17.5 Limnothlypis swainsonii Worm-eating Warbler 37 13.0 0.8 15.25 Helmitheros vermivora Tennessee Warbler 3 8.5 0.1 - Vermivora peregrina Northern Parula $ 7 7.0 0.7 - Parula americana 9 10 7.0 0.3 - Magnolia Warbler 6 7.75 0.9 - Dendroica magnolia Cape May Warbler 13 9.25 0.3 10.0 Dendroica tigrina Black-throated Blue Warbler 6 7 9.0 0.2 - Dendroica caerulescens 9 19 8.75 0.4 12.75 Black-throated Green 9 7.5 0.4 - Warbler Dendroica virens Blackburnian Warbler 1 8.75 - - Dendroica fusca Yellow-throated Warbler 2 8.5 - - Dendroica dorninica Pine Warbler 1 10.5 - - Dendroica pinus Blackpoll Warbler 1 13.5 - - Dendroica striata Prairie Warbler $ 17 7.5 1.3 11.25 Dendroica discolor 9 18 6.75 0.3 8.0 Palm Warl)ler 3 9.5 0.2 - Dendroica palmariim Ovenbird 63 19.25 3.2 25.5 Seiurus aurocapillus Northern Watertbrusb 10 15.75 0.6 20.75 Seiurus noveboracensis Common Yellowthroat 6 30 9.75 0.3 11.25 Geothlypis trichas 9 18 9.25 0.5* 11.5 American Redstart $ 21 7.5 0.8 11.5 Setophaga ruticilla 9 28 7.5 1.3 10.25 * Differences between sexes significant different at .01 level, ** Differences significant at .001 level (2-tailed t-test). GENERAL NOTES Wing molt of the Kittlitz’s Murrelet. — Knowledge of the relation between timing of molt and other annual cycle events in birds is important to an understanding of breeding seasons. Although no study of the natural history of the Kittlitz’s Murrelet { Brachyramphus brevirostris) has been conducted, published field observations and information on molt obtained from museum specimens presented here permit the timing of molt in relation to breeding to be outlined. I examined 213 specimens from the collections of the Denver Natural History Museum (DNHM), Museum of Comparative Zoology at Harvard Univ. (MCZ), Univ. of British Columbia (UBC), Univ. of California Museum of Vertebrate Zoology (MVZ), Univ. of Michigan Museum of Zoology (UMMZ), the United States National Museum (USNM), the American Museum of Natural History (AMNH), California Academy of Sciences (CAS), Chicago Natural History Museum (CNHM), Carnegie Museum (CM), Peabody Museum of Natural History (PMNH), Univ. of Alaska (UA), and the Univ. of Kansas Museum of Natural History (KU). The status of each primary of the 14 specimens in molt was recorded by assigning to each growing feather a score, according to its growth stage, of 1 (empty or pin feather) 2, 3, or 4. Full-grown new feathers score 5, so that in newly molted birds, each feather is scored 5. A completed molt score is 50 since I quantified feather replacement only on the left wing (Table 1). Bent (U.S. Natl. Mus. Bull. 107, 1919) and Demente’ev and Gladkov (Birds of the Soviet Union, Vol. 2. IPST, Jerusalem, 1968) stated that adult Kittlitz’s Murrelets undergo 2 seasonal molts, a partial prealternate molt in spring and a complete prebasic molt in late fall or early winter. Kozlova (Fauna of the USSR: Birds 2:1-140, IPST, Jerusalem, 1961) noted that although the dates of the complete prebasic molt are not known, the partial prealternate molt is completed by late May. The prealternate molt occurs apparently rapidly between mid-April and mid-May. Laing (Victoria Mem. Mus. Bull. 40, 1925) took a male in basic plumage on 22 March 1924 in Chignik Bay, Alaska. A male (MCZ 250797) taken on 9 April 1916 near Bethel, Alaska, is in its basic plumage with no wing or body molt and little wear of the remiges. A female (Bailey, Colorado Mus. Nat. Hist., Pap. Ser. 8, 1948) taken on 28 April 1922 near Wales and a female (USNM 92184) collected near Cape Etolin, Alaska, on 3 April 1883 are in their basic plumages as are 2 males (KU 40294, 40295) taken on 15 April 1960 near Point Hope. Four males (AMNH 757401, CNHM 456901, MCZ 320031, MVZ 14534) and 2 females (AMNH 753398, 753399) taken in Glacier Bay on 16 May 1913 are in their alternate plumages. The simultaneous wing molt does not begin until late August (see Table 1) ; the birds are rendered flightless. A female (DNHM 19287) taken by A. M. Bailey near Barrow on 26 July 1936 is of interest. Its primaries were molted and being replaced by new ones and the bird was flightless. It had failed possibly at breeding and had initiated a rapid and early wing molt. Premature body molting due to apparent breeding failure has been reported in auklets {Aethia spp.) by Bedard (Can. J, Zool. 47:1025-1050, 1969) and in the Marbled Murrelet (B. marmoratus) by Sealy (Bird-Banding 46: 141-154, 1975). The breeding season (egg-laying to fledging of young) of the Kittlitz’s Murrelet in Alaska spans the period from early June to mid-August (Thayer, Condor 16:117-118, 1914; Bailey, Auk 44:1-23, 1927; Ford, Auk 53:214, 1936; Thompson et ah. Auk 83: 349-351, 1966; Bailey, Condor 75:457, 1973; J. Bedard, pers. comm.). 467 468 THE WILSON BULLETIN • VoL 89, No. 3 Table 1 Molt of the Left Primaries of the Kittlitz’s Murrelet“ Specimen Date, sex, locality, molt status Primary number 6 10 CNHM 159075 26 Sept. 1929, $ Barrow: score 25 2 2 3 3 3 3 2 3 2 2 CNHM 159076 17 Sept. 1941, $ Barrow: score 21 1 1 2 2 3 3 3 2 2 2 CNHM 159077 26 Sept. 1929, $ Barrow: score 29 1 2 2 3 3 4 4 4 3 3 CNHM 159078 26 Sept. 1929, ? Barrow: score 30 1 2 2 3 4 4 4 4 3 3 DNHM 19287 26 July, 1936, $ ? ? score 22 2 2 2 2 3 3 2 2 2 2 LACM 78605 fall, 1962, $ Barrow: score 26 1 1 1 2 3 3 4 4 4 3 PMNH 1460 17 Sept. 1941, $ Barrow: score 18 2 2 2 2 2 2 1 2 2 1 PMNH 9278 25 Aug. 1936, ? Barrow: score 8 1 1 1 1 1 1 1 1 0 0 PMNH 9279 25 Aug. 1936, ? Barrow: score 14 1 1 2 2 2 2 2 2 0 0 PMNH 9285 17 Sept. 1941, 9 Barrow: score 0 0 0 0 0 0 0 0 0 0 0 UMxMZ 125204 25 Aug. 1936, 9 Barrow: score 16 2 2 2 2 2 2 2 2 0 0 UMMZ 125205 25 Aug. 1936, $ Barrow: score 7 1 1 1 1 1 1 1 0 0 0 UMMZ 125206 17 Sept. 1941, $ Barrow: score 18 2 2 2 2 2 2 2 2 1 1 UMMZ 125207 17 Sept. 1941, $ Barrow: score 21 2 2 2 2 2 2 2 2 3 2 “ Explanation of symbols: 0, feather that is old; 1, empty socket or pin feather; 2, growing feather with vane up to one-third grown; 3, growing feather with vane between one-third and two-thirds grown; 4, growing feather with vane more than two-thirds grown but not full length; 5, feather full length, but still with blood in calamus. Thus, the prebasic molt does not overlap breeding; the young have fledged by the time this molt begins in late August. The separation of breeding and prebasic molt has been recorded also in the Marbled Murrelet and Ancient Murrelet {Synthliboramphus antiquiis) by Sealy (Bird-Banding 46:141-154, 1975; Condor 78:294-306, 1976). I am indebted to A. M. Bailey, N. A. Din, N, K. Johnson, R. A. Paynter, Jr., R. W. Storer, and R. L, Zusi for permitting me to examine specimens in their care. D. Amadon, M. H. Clench, D. D. Gibson, R. M. Mengel, R. T. Orr, C, G. Sibley, K. E, Stager, and M. A. Traylor kindly loaned specimens. P. S. Humphrey critically read an earlier draft of the manuscript. Travel to museums was made possible by grants from the Frank M. Chapman September 1977 • GENERAL NOTES 469 Memorial Fund of the American Museum of Natural History and the National Research Council of Canada. — Spencer G. Sealy, Dept, of Zoology, Univ. of Manitoba, Winnipeg, Canada. Accepted 9 Apr. 1976. Incidence of rnnt eggs in llie Canada Goose and Semipalinated Sandpiper.- There are few published reports of runt (dwarf) eggs in nature (Rothstein, Wilson Bull. 85:340-342, 1973) and little is known about the rate at which they occur in a given population. In 1973, while working under contract for the Canadian Wildlife Service on North Twin Island in James Bay, we examined about 950 eggs of various species. These included about 500 eggs (122 nests) of the Canada Goose {Branta canadensis) and 29 eggs (8 nests) of the Semipalmated Sandpiper {Calidris pusilla) . In one Canada Goose nest, found on 19 May, there were 3 normal eggs (x 82.9 X 56.5 mm, 148 g) and a runt (46.4 X 35.8 mm, 39 g). The runt was only 26% of normal weight and unusually spherical. After boiling it was opened and found to contain a rather fibrous yolk, 5 mm in diameter. We did not disturb the normal eggs and their number had not changed by 25 May. Another Canada Goose nest, found on 16 June, contained 2 runt eggs (61.1 X 35.2 mm, 39.8 g; 56.0 X 34.3 mm, 35.0 g), but no normal eggs. These runts had no yolks and were, in Palmer’s terminology (Handbook of North American Birds, Yale Univ. Press, New Haven, Conn., 1:13, 1962), “long elliptical.” The female goose was apparently incubating the eggs in a normal manner and, unless it was a replacement clutch, had probably been so doing for nearly the full term — as other clutches were already hatching. One Semipalmated Sandpiper’s nest contained 3 normal eggs (x 29.8 X 21.5 mm, approx, vol. 70.5 cc) and a runt egg (22.2 X 16.1 mm, approx, vol. 29.5 cc) of normal shape and color, but a volume only 42% normal. The normal eggs hatched 3 July but the fate of the runt is unknown. Based on the above figures, the rate of occurence of runt eggs is 0.6% for the Canada Goose or 0.4% if the 2 runts found in one nest are considered a single instance, 3.4% for the Semipalmated Sandpiper and 0.4% for all eggs examined by us in 1973. Unfortunately the samples are not random, because if no runts had been found there would have been no report. Museum samples are also liable to be biased upwards, because of a tendency for the unusual to be collected. If, therefore, we are to obtain reliable estimates of the rate of incidence of runt eggs in general and perhaps to make comparisons between species and populations it will be necessary for those handling large numbers of eggs to keep, at least approximate, records of the number of eggs they examine, even if no abnormality is found. Barth (Zool. Mus. Univ. Oslo, Contrib. 81, 1967) found only 1 runt among 4560 eggs (0.02%) in 4 species of gulls {Larus) and Ricklefs (Bird-Banding, 46:169) one runt in about 2000 eggs (0.05%) of the Starling [Sturnus vulgaris). We cannot recall previously finding a runt in the many eggs examined. — T. H. Manning and Brenda Carter, RR 4 Merrickville, Ontario, Canada, KOG I NO. Accepted 22 Apr. 1976. Late fledging date for Harris’ Hawk. — On 29 November 1975, as part of an Arizona Raptor Study Committee project, we banded two nestling Harris’ Hawks (Parabuteo unicinctiis) approximately 40 km north of Phoenix, Maricopa County, Ari- zona. These 2 birds subsequently fledged sometime between 2 and 4 December 1975. This is the latest recorded fledging date for the species. Previously recorded late dates are: a nest with fledged young in October and 470 THE WILSON BULLETIN • VoL 89, No. 3 November (LeSassier and Williams, Wilson Bull. 71:386-387, 1959) ; nests with young fledged in September (Pache, Wilson Bull. 86:72-74, 1974) ; and nest with recently fledged young 26 October 1975 (Mader, Auk in press). The Arizona nest was located in excellent Harris’ Hawk habitat. Additionally, 1975 was a year of high desert cottontail (SylvUagus auduboni) numbers, and the caretaker of the nearby golf course was systematically shooting these mammals and not retrieving them. Coyotes {Canis latrans) were observed carrying off the carcasses, and the Harris’ Hawks may also have been using this source of food; cottontail skulls, tails, and legs were found in the nest and around the base of the nest site. A few feathers of Gambel’s Quail {Lophortyx gambelii) , and the tail of a Harris’ antelope ground squirrel (Ammo- spermophilus harrisii harisii) were also collected from the nest. The nest was built in a Saguaro (Carnegiea gigantea) , about 7 to 8 m from the ground. Because of the large size and lack of cup in the nest, we believe it to have been rebuilt or added to several times. Klimosewski first saw it in the winter of 1974-75, and in the spring of 1975 he saw an adult female sitting on the nest; however, no young were fledged from this presumed nesting attempt. The next indication of use was on 22 November 1975, when 2 large young were seen in the nest. Two males and a female (sex determined by comparative size), all in adult plumage, were in attendance at this nest both in the spring and in November/December and would support the conclusion that this late nesting was at least the second attempt by the same group. A nest-helping system has been recorded for Harris’ Hawks by Mader (Living Bird, 14:59-85, 1975). — Eleanor L. Radke, P.O. Box 446, Cave Creek, AZ 85331 and John Klimosewski, 1810 N. 16th Ave., Phoenix, AZ 85007. Accepted 15 March 1976. The spatial distrlhiition of wintering IHaek-bellied Plovers. — The Black-bellied Plover (Pluvialis squatarola) is a common winter resident along much of the coastal United States. Individuals in foraging flocks of wintering Black-bellied Plovers are generally (juite scattered. This is in contrast to most other winter shorebirds (e.g., Sanderlings, Crocethia alba; Semipalmated Plovers, Charadrius semipalmatus; Knots, Calidris canutus; and Ruddy Turnstones, Arenaria interpres) which frequent the same beaches in fairly compact flocks. This note discusses the spatial distribution of wintering Black-bellied Plovers along tbe Gulf coast beaches of Sanibel Island, Florida. From 25 through 30 December 1975, 1 made 13 surveys of Black-bellied Plovers on Sanibel Island, each time pacing off the distance betw^een adjacent plovers. I measured 201 inter-plover distances, sampling only sections of beach bordered by vegetation. I avoided stretches of beach with many people and all areas where there were dogs, for the plovers seemed to avoid both situations. To avoid sampling regions where recent disturbance (e.g., a dog running along the beach) may have caused all the plovers to leave the area temporarily, I did not record any inter-plover distances which were greater than 270 m. The groups of inter-plover distances were homogeneous (Kruskal- Wallis test, P 7> .975), so all samples were combined. The null hypothesis that the 201 observed distances are indistinguishable from a ran- dom distribution of plovers along tbe beach was tested against the alternative hypothesis that observed distances were more evenly spaced than a random distribution of plovers wmuld produce; tbe plovers were obviously not clumped. The random distances were generated from the equation, 1 ”• 1 7V„, _ V . s /=! 5-/+ 1 September 1977 • GENERAL NOTES 471 where N,n is the proportional length of the wth segment of a line divided into S random lengths (MacArthur, Proc. Natl. Acad. Sci. U.S. 43:293-295, 1957). To account for the fact that I sampled only inter-plover distances which were less than 270 m, I used S = 239 to produce 201 random distances which were less than the proportional equivalent of 270 m, and 18 random distances which would not have been recorded because they were greater than the proportional equivalent of 270 m. {S := 238 or 240 also produces 201 ap- propriate random distances, and the ensuing statistics are similar to those presented here.) The 2 frequency distributions are presented in Table 1; the null hypothesis, that Black- bellied Provers are randomly distributed within the scattered foraging flocks, is strongly rejected (G-test, P<.001). The plovers are somewhat evenly dispersed within the foraging flocks. Observed Inter-plover Table 1 Distances Compared with Expected Random Distances Distance ( m ) Observed Expected 0-29 22 43 30-59 36 37 60-89 43 29 90-119 36 24 120-149 31 20 150-179 18 16 180-209 7 13 210-239 4 10 240-269 4 9 201 201 In an attempt to investigate how such spacing was maintained, I paid close attention to 4 sets of 2 Black-bellied Plovers which were less than 3 m apart. Additionally, on 3 occasions, I was successful in “herding” together 2 plovers which had originally been separated by more than 30 m. In 30 min of observation on each of the 7 pairs, I never observed any behavior (aggressive or otherwise) which seemed responsible for the spacing. In all cases, the plovers which were close together would simply slowly move apart. Apparently, the birds space themselves by mutual avoidance rather than by aggressive actions. The fairly large standard deviation (57.4 m) around the mean inter- plover distance (95.9 m) also suggests a low-key spacing behavior. In contrast to the subtlety of the intra-specific behaviors which produced the spacing, it was not at all uncommon to see a plover peck at and chase away other species of shorebirds which wandered by. This is in contrast to the finding of Recher and Recher (Wilson Bull. 81:140-154, 1969) that intra-specific aggression was much more common than inter- specific aggression in foraging flocks of migrant shorebirds. Goss-Custard (pp 3-35 in Social Behavior in Birds and Mammals, J. H. Crook, ed.. Academic Press, London, 1970) described 2 main types of shorebird foraging flocks: compact and widely scattered. He suggested that flocking while foraging facilitates detection of predators (e.g.. Page and Whitacre, Condor 77:73-83, 1975), and that compactness of the flock is dependent on whether or not feeding efficiency is decreased 472 THE WILSON BULLETIN • Vol. 89, No. 3 by compact flocking. The observation that Black-bellied Plovers are somewhat evenly spaced within these foraging flocks is consistent with the idea that such scattered flocking is an attempt to avoid intra-specific interference. Mitchell A. Byrd, Bruce S. Grant, Stewart A. Ware, Barbara S. Warren, and two anonymous reviewers made ver>^ helpful comments on an earlier draft of this note. H. Wade and Barbara R. Stinson provided room and board on Sanibel Island. My sincere thanks to all of the above.^ — Christopher H. Stinson, Dept, of Biology, College of William and Mary, Williamsburg, VA 23185. Accepted 13 Apr. 1976. Predation and dispersion of Herring Gull nests.— Tinbergen (1960, The Her- ring Gull’s World, Harper and Row, New York) reported that Herring Gulls (Laras argentatus) deserted most nests from which red fox (Vulpes vulpes) took eggs. The adults so affected reportedly renested at the borders of the colony, and their deserted territories were incorporated into territories of adjacent pairs. This “spreading out phenomenon,” as it was termed, was believed to function as a passive defense by dis- persing the nests making their location by predators more difficult. We noted a different response in the reactions of Herring Gulls to red fox (Vulpes fulva) predation on South Manitou Island in northern Lake Michigan (Leelanau Co., Mich.). During studies of productivity at this colony in 1974, Shugart marked and mapped the location of 51 nests in a strip transect (10 m X 215 m) encompassing about 15% of the central nesting area. Eggs in the 51 nests were marked. Hatching began 18 May and newly hatched chicks were banded within 1-2 days of their hatching date. Shugart made the following observations. In 23 of the 51 nests during the first week of hatching, 18 chicks were killed by fox, 16 other chicks disappeared and were probably taken by fox, and 9 small chicks apparently died from exposure during nightly fox visits to the colony. Evidences for the fox predation were the presence of fox tracks on the perimeter of the colony and canine tooth punctures in the chick carcasses following the nights in question. Seven unhatched eggs that remained in the predated nests were found broken and addled outside of nests several days after the chicks were killed, disappeared, or died. The latter indicated that incubation of the original remaining eggs did not continue after the nests were predated. Within 2-8 days after the death of the first Herring Gull chicks, Shugart observed that grass and twigs were being added to the predated nests or that new nests were being constructed near the original nests. Eight 134%) of the original predated nests had additional eggs laid in the same nest cup. Of the remaining pairs, 14 (61%) apparently laid in newly constructed nests 1 to 9 m (x =: 2.05 m, SD = 1.21 m) from the originally predated nests. The distance between initial nests in the sample area averaged 4.88 m fSD = 2.15 m) which is significantly more (t = 4.580, P 0.05) than eggs from peripheral nests. Protein, carl)ohydrate, and lipid weights and their energy values from both locations were equal (Table 2). These results support the finding that embryos of equivalent age from the center and periphery of the Granite Island colony (Ryder and Somppi, Wilson Bull. 89:243-252, 1977) showed no significant differences in developmental characteristics and size. It appears that the differences in hatching success in relation to nest location in our colony may not be due solely to differential quantities of proteins, carbohydrates, and lipids in the yolks. The results do not preclude the possibility that differences exist in the types and quantities of essential amino acids and/or other compounds which may be important in determining egg hatchability. Additionally, low egg success in peripheral areas may reflect lower parental attentiveness than in central regions. We thank L. Somppi, C. Ryder and T. Carroll for assistance in collecting and measur- ing eggs in the field. Financial support for this and related research on gull ecology was provided by the National Research Council of Canada and a Lakehead University President’s Research Grant. We appreciate the cooperation and interest of R. Trowbridge for allowing us to base field operations at Bonavista. — John P. Ryder, Dept, of Biology, Lakehead Univ., Thunder Bay, Ontario, P7B 5E1, Donald E. Orr and Giiomi H. Saedi, Dept, of Chemistry, Lakehead Univ., Thunder Bay, Ontario, P7B 5E1. Accepted 25 Mar. 1976. Roof-nesting by Common Terns. — During the summer of 1975 a pair of Common Terns {Sterna hirundo) nested on the flat roof of a building on Great Gull Island, New York (at the eastern end of Long Island Sound). Gill (Auk 70:89, 1953) reported Common Terns nesting on a boat on Long Island. I find no reference in the literature to Common Terns nesting on buildings. Least Terns {S. albifrons) have been reported nesting on roofs in Florida (Fisk, Am. Birds, 29:15-16, 1975). On 12 July 1975 I first noticed a Common Tern sitting on the roof of 1 of the old army buildings, now used as sleeping quarters on Great Gull Island. On 13 July I climbed onto the roof and found 2 warm eggs in a shallow depression where I had seen the adult tern sitting. A loose layer of pebbles on the flat surface of the roof covered most of the tar and roofing paper. The nest depression was shielded on 1 side by a piece of roofing paper and was partly lined with small pieces from a rotting board lying on the roof about 1 m from the nest. While I was on the roof one of the adult terns dove at me. A tern was last seen incubating on 25 July during a storm. On 26 July and on following days no birds were seen on the nest. On 18 August 1 egg was left in the nest. I opened it and found an embryo which I judged to be 11 to 12 days old using the criteria of Hays and LeCroy (Wilson Bull. 84:187-192, 1971). On Great Gull Island Common Terns often nest on the crumbling concrete of the old fort which covers most of the island (Cooper et ah, Proc. Linn. Soc. 71:108-118, 1970). Most of the concrete surfaces are effectively at ground level. At times terns have nested on concrete lookout platforms at least 2 m above the ground. This roof nest was about 4 m above the ground. The roof’s pebble surface gave the nest a substrate similar to the island’s pebble beaches. During the period when the roof-nest terns probably chose their nest site, many of the traditional nesting areas were overgrown or still being defended. A resulting shortage of nesting habitat may have caused the selection of the roof as a nest site. I do not think that the desertion of the eggs on the roof was due to any particular disadvantage in the nest site, rather, it may have been caused by factors which influenced the desertion of many nests on the night of the storm of 25-26 July. 476 THE WILSON BULLETIN • Vol. 89, No. 3 Roof-nesting, like the use of other man-made structures on Great Gull Island, demonstrates the adaptability of Common Terns in their choice of nest sites. It will be interesting to see whether the use of roofs for nesting continues and increases in future seasons. I am grateful to Helen Hays and to Kenneth C. Parkes for their comments on the manuscript. This note is contribution No. 42 from the Great Gull Island Project. — Anne E. Mac- Farlane, 325 E. 72nd St., New York 10021. Accepted 20 April 1976. Rapid chick separation in Whip-poor-wills. — This note describes a poorly known aspect of Whip-poor-will {Caprimulgus vociferus) behavior and emphasizes the possible importance of nestling behavior to survival. While hiking through second-growth deciduous forest in Jasper County, Illinois, on 5 May 1972, I flushed a female adult Whip-poor-will from 2 eggs resting in a shallow leafy depression. The nest site, “nest,” and eggs were typical of published descriptions for the species. During the next 13 days I visited the site 5 times and always found the female incubating at precisely the same location with the eggs slightly rearranged within the nest. On 22 May (4 days from the last visit) the female allowed me to approach to 1 m before flushing. As she flushed, 2 chicks simultaneously separated in opposite directions to a distance of about 15 cm from each other. Their separation occurred so rapidly and unexpectedly to me that I am uncertain whether the chicks were flipped apart l)y the female with her feet as she flushed, or whether they separated under their own power. I noted no discrete Imps. That one chick rather forcefully tumbled forward to rest, left me with the immediate impression that it had been propelled. The chicks remained perfectly motionless, and their eyes remained closed during several minutes of observation. Two days later, as the female flushed, the chicks separated about 40 cm from each otlier by a series of rapid but perceptible hops. They moved in exactly opposite direc- tions as l)efore. I was impressed again by the rapidity of their separation, by their motionlessness after a simultaneous and quick stop, and by the effectiveness of their camouflage. The cliicks’ eyes were first noted to he open on 27 May when the chicks hopped apart about 65 cm along perpendicular paths as the female flushed. On 31 May only 1 chick hopped from the nest (to about 60 cm). The second chick “froze” within the nest. On this visit I saw the male adult and droppings around the nest for the first time. The male appeared at the moment of typical distraction be- havior 1)> the female (sharp “thurp” calls; posturing with dropped wings, fanned tail and erect head; injury-feigning skirmishes through the leaves). The original nest site was abandoned on 2 June and was littered with droppings. I unexpectedly flushed the brooding male about 8 m away, but was looking in the wrong direction to observe the chicks directly as he flushed. They rested about 1 m apart and faced in opposite directions. The male exhibited distraction behavior similar to that of the female. The male was brooding at this same site on 4 June, but neither chick moved when he flushed. On 6 June the male was brooding the chicks about 15 m from the original nest site. All 3 flushed together. The chicks each flew in straight lines about 45° from one an- other to a distance of about 12 m. One chick landed in a branch 2 m up, and the other landed on the forest floor. The male immediately placed himself between me and September 1977 • GENERAL NOTES 477 the chick on the branch and exhibited distraction behavior. The female did not appear. Neither adult nor chicks could be found in the vicinity the following day. There are 4 references to possible rapid chick separation in Whip-poor-wills in the literature (Bent, U.S. Natl. Mus. Bull, 176, 1940; Fowle and Fowle, Can. Field-Nat. 68:37, 1954; Raynor, Bird-Banding 12:98-104, 1941; Tuttle, Bird-Lore 13:235-238, 1911), but the behavior is described nowhere in detail nor interpreted. The adaptive advantage of rapid chick separation is undoubtedly the increased probability that at least 1 of the chicks will survive nest disruption by a predator. I believe rapid chick separation is one more element of an anti-predator repertoire of adaptations in Whip-poor-wills which includes, in addition, cryptic coloration, brood site movement, and adult distraction behavior. — Eric L. Dyer, Station 17, Vanderbilt Univ. Hospital, Nashville, TN 37232. Accepted 30 July 1976. An intraspecific mortal attack. — On the morning of January 6, 1976, I was looking out my window as 2 female House Sparrows (Passer domesticus) dove (hurtled) into the grass nearby. One held the other by the neck and after a few seconds the struggling victim lay still. The attacking sparrow, still on top of the nearly lifeless one, began to strike hammering blows with its bill on the head of the victim. Several sparrows flew near, and all flew off leaving the motionless body on the ground. Minutes later a House Sparrow returned, jumped on the dead sparrow and again struck it on the head several times, then flew away. On 8 January I observed a similar incident involving female House Sparrows. The attacking sparrow held the neck of the struggling one, which eventually got loose. Both flew off, one pursuing the other. — ^Vera Lee Grubbs, 3816 Elmer Lane, Shreveport, LA. 71109 Accepted 1 Mar. 1976. Riifous-sidcd Towhccs mimicking Carolina Wren and Field Sparrow. — Eastern populations of the Rufous-sided Towhee iPipilo erythrophthalmus) do not exhibit any marked local dialects, and the high percentage of unique song patterns in the songs of a local population suggests that what a bird hears when it is developing its song does not play an important role in determining the song patterns developed (Borror, Condor 77:183-195, 1975). It is thus of considerable interest to encounter eastern towhees whose songs (or song parts) are excellent mimics of other species. This paper is a report on the songs of 2 towhees (of several hundred I have recorded), one using an introduction consisting of Carolina Wren song phrases, and the other singing Field Sparrow songs. Both birds were seen when recorded. Mimicry of Carolina Wren. — On 27 July 1975 I recorded a towhee near Murray, Kentucky (OSU recording No. 13679, with 67 songs), some of whose songs had an introduction consisting of (or containing) from 1 to 3 song phrases of a Carolina Wren (Thryothorus ludovicianus) . The recording contained 5 different song patterns, 4 of which are shown in Fig. 1 (A, B, E, F) ; 2 (A and E) were normal songs for this population (a 2-note introduction followed by a trill) but 2 of the other 3 had Carolina Wren phrases in the introduction (B and F in Fig. 1), and a 5th contained only 2 Carolina Wren phrases (of the type in F, without the buzzy note and final trill). Most of the songs of the B pattern were sung in alternation with songs of the A pattern, while most songs of the F pattern were sung consecutively, only occasionally alternating with 478 THE WILSON BULLETIN • Vol. 89, No. 3 songs of the E pattern. There was only 1 song of the 5th pattern, which was in a series of F pattern songs. Two Carolina Wrens were heard near where this towhee was recorded, but neither sang songs of the patterns sung by the towhee. The wren phrases in pattern B of this towhee are of a pattern that is fairly common in the wren; I have found it in the songs of 27 birds (of over 380 recorded, from 16 states) — 20 in central Ohio, 4 in southern Ohio, 2 in southwestern Kentucky (one about 50 km from the Murray towhee) , and 1 near Chincoteague, Va. The towhee songs of pattern B all contained 3 Carolina Wren phrases, uttered at the rate of 3.64 per sec; the wren songs of this pattern contained from 1 to 8 phrases (average of 345 songs, 5.10), uttered at rates of 3.23 to 4.61 (average 3.72) per sec. Sonagrams of the final phrases of 2 of these Carolina Wren songs are shown in Fig. 1 {C and D) . Carolina Wren phrases of the type in F (Fig. 1) are less common than those in B; I have found such phrases in only 9 birds — 6 in central Ohio, 1 in West Virginia, 1 in southwestern Kentucky (about 25 km from the Murray towhee), and 1 near Tallahassee, Florida. The towhee songs contained only 1 or 2 (average, 1.6) of the wren phrases, uttered at the rate of 2.27 per sec; the wren songs of this type contained from 2 to 5 phrases (average of 88 songs, 3.50), uttered at rates of 2.15 to 2.40 (average, 2.28) per 6- H - 4 1 a- a r o A U 6- lA 5- a. w 4 ■ CL t' - ' i U 2- 4' u >• 0 B 0 -I ^ 6- Z b' r Ur 0 UJ a. c 6- 5 aoo 025 050 D 075 V- % \ I t t G ' a’ ' time in seconds Fig. 1. Sonagrams of songs of a Rufous-sided Towhee (#13679, Murray, Ky., 27 July 1975), and final phrases of Carolina Wren songs. A, towhee, 13679-29; B, towhee, 13679-28; C, Carolina Wren, 12733-19, Blendon Woods, Franklin Co., Ohio, 19 March 1974; D, Carolina Wren, 12785-2, Georgesville, Ohio, 18 April 1974; E, towhee, 13679-58; F, towhee, 13679-55; G, Carolina Wren, 13651-20, Murray, Ky. (about 25 km from 13679), 8 July 1975; H, Carolina Wren, 3863-9, Blendon Woods, Franklin Co., Ohio, 18 April 1959. September 1977 • (ilENEHAL NOTES 479 6- •v “V X V V V cc Ui Q. 2“ -I 6- u u 5- 3 4- ^ 3- ? 2- > 5 6- o s- u a: 4- u. 3- 2- 6- 5- 4- 3- 2- 0.00 "WWWW V *v c V'^VUVUWWWU NVWWWWW 025 050 E Q75 100 1.25 1.50 1.75 200 2.25 TIME IN SECONDS Fig. 2. Sonagrams of Rufous-sided Towhee (A) and Field Sparrow (B-E) songs. A, towhee, 3146-10, Blendon Woods, Franklin Co., Ohio, 2 April 1958; B, Field Sparrow, 9840-4, Blendon Woods, 6 April 1969; C, Field Sparrow, 3897-1, Blendon Woods, 24 April 1959; D, Field Sparrow, 3830-1, Blendon Woods, 1 April 1959; E, Field Sparrow, 9271-1, Blacklick Woods, Franklin Co., Ohio, 29 March 1968. sec. Sonagrams of the final phrases of 2 of these Carolina Wren songs are shown in Fig. 1 (G and H). Mimicry of Field Sparrow. — On 2 April 1958 I recorded a towhee in Blendon Woods, Franklin Co., Ohio (OSU recording No. 3146, with 13 songs), singing Field Sparrow {Spizella pusilla) songs. When 1 first heard this bird I thought it was a Field Sparrow, and it was not until 1 saw the bird and watched it sing that I realized that these songs 480 THE WILSON BULLETIN • VoL 89, No. 3 were being sung by a towhee. All tbe songs in the recording were of the same type or pattern; the songs were distinctly 2-parted and the notes were down-slurred, with the 3 or 4 notes in the first part uttered at the rate of 3.54 per sec, and the 4 to 8 in the second part uttered at the rate of 12.90 per sec. Individual songs contained 8-12 notes (average, 10.7), and averaged 1.46 sec in length (Fig. 2, A). Field Sparrows are common in the area where this towhee was recorded. Field Sparrow songs are subject to considerable variation, and while one rarely finds 2 birds singing identical songs, the various song patterns may be classified in a number of major groups. Goldman (Ph.D. Thesis, Ohio State Univ., 1972), in a study of 197 Field Sparrows, mostly from central Ohio, recognized 40 major song pattern types; the songs of this towhee were of a type he found in 8 of the birds he studied. Other people might recognize a different number of major song pattern types in Field Sparrows; in my studies I have recognized 15 major types, and the songs of this towhee were of a type that I found in about 9% of the birds I studied; 4 songs of this type are shown in Fig. 2{B-E). The songs of this towhee are shorter and contain fewer notes than most Field Sparrow songs; the songs Goldman studied averaged 2.64 sec in length, with an average of 23.20 notes. The final notes in tlie towhee songs are uttered more slowly than those in most Field Sparrow songs; Goldman found the note rate in the last part of the song to range from 5.9 to 35.7 notes per sec, and averaged 16.97 per sec. Discussion. — Except for the Mockingbird (Mimus polyglottos) , mimicry of 1 species by another in wild North American passerines appears to be quite rare. It has been reported in several species, but in only a few cases has the report been supported by audio-spectrographic analyses 0.05) was indicated for either the high (2.0 m, 369 observations) or the low (1.5 m, 351 observations) perch structure (Table 1). Of 6 species for which there were sufficient data to conduct a chi- Pkrcii Hkight Sklkc.'I Table 1 ION 15 Y Birds IN AN Alfalfa Field Perch use (Nuinher of observations) 1.5 M 2.0 M Chi square value'’ Species 01)served Expected-' Observed ] Expected •'* American Kestrel ( Falco sparveri us ) 8 9.6 12 10.4 0.51 Eastern Meadowlark ( Sturnella magna ) 6 20.4 34 19.6 20.73* Red-winged Blackbird ( Agelai us phoeniceus ) 154 157.0 180 177.0 0.11 .‘Savannah Sparrow iPusserculus sundwichensis ) 132 120.5 114 125.5 2.15 Grasshopper Sparrow ( Ammodramus suvannurum ) 23 13.0 3 13.0 15.38* Vesper Sparrow ( Pooecetes gramineus ) 28 27.0 26 27.0 0.07 Preference by all sj)ecies 351 345.6 369 374.4 0.16 « Expected ratios based on the frequency of high and low perches within territories of the indi- vidual species. Chi-sqnare analysis, testing the hypothesis (a = 0.05) that there exists no preference for a specific perch height; tested with one degree of freedom. * Significant, (P<0.05). September 1977 • GENERAL NOTES 487 square test (expected ratios based on the frequency of high and low perches within the territories of the individual species), only 2 species, the Grasshopper Sparrow and the Eastern Meadowlark, showed a specific preference for 1 of the 2 sizes. The Grasshopper Sparrow, according to Smith (U.S. Natl. Mus. Bull. 237, Part 2, 1968) and Wiens (Ecology 54:877-884, 1973) will normally select the highest perch available. The Grasshopper Sparrow, in this study, used (P < 0.05) the lower perches. Furthermore, in every territory established by this species, both large and small structures were present and therefore available for use. Only the Eastern Meadowlark exemplified the generalization that higher perches will be used instead of lower ones. Of a total of 40 observations, the higher perches were selected (P<0.05) 34 times. Also using the perches but recorded 4 or fewer times were Mourning Dove (Zenaida macroura) , Short-eared Owl {Asia jlammeus) , Eastern Kingbird iTyrannus tyrannus) , Barn Swallow {Hirundo rustica) , American Robin (Turdus migratorius) , Starling {Sturnus vulgaris). Bobolink (Dolichonyx oryzivorus) , and Brown-headed Cowbird iMolothrus ater) . The general conclusion to be drawn appears to be that the above grassland-associated birds will use any elevated perch structure, at least up to 2 m tall. Some species may exhibit more specific tendencies, possibly preferring the highest available perch or preferring perches of a certain height range above the vegetation. This note represents a part of a larger study completed as partial fulfillment for the Master of Arts degree at Western Michigan University. I am indebted to Dr. Richard Brewer for his critical review of this note. Thanks goes to Mr. Ray Vliek, without whose land and cooperation this study would not have been possible. Support for this research was provided by fellowships from the Upjohn grant for graduate studies in Biology and a Western Michigan University Graduate Student Research Grant. — Keith G. Harrison, Dept, of Biology, Western Michigan Univ., Kalamazoo 49008. Present address: Michiana Area Council of Governments, County-City Bldg., South Bend, IN 46601. Accepted 15 Apr. 1976. SPECIAL REVIEW John Ostrom’s Studies on Archaeopteryx, The Origin of Birds, and the Evolution of Avian Flight Archaeopteryx lithographica is the most significant fossil species in the class Aves. Studies on Archaeopteryx have formed the basis for 3 generally accepted ideas about avian evolutionary history: (1) birds have their origins in reptiles, specifically within the pseudosuchian Thecodontia, (2) birds evolved their adaptive way of life — flight — from bipedal, arl)oreal ancestors, and (3) the origins of more modern avian taxa were post-Jurassic in time. Archaeopteryx being considered on the “main-line” of avian evolution or “close to it.” Archaeopteryx has held this central position because it is the oldest fossil with obvious avian affinites and is represented by a number of well- preserved specimens. It is not surprising therefore that in addition to considerable work on the morphology of Archaeopteryx, many workers liave attempted to reconstruct from that morphology much about the ecology and behavior of this species; and by inference these findings have been assumed to represent the avian ancestral condition. This controversial literature leads one to the important (juestion of just how far historical analysis, which is highly inferential, can depart from the available evidence and yet remain “respectable” science, or alternatively to what extent can historical narrative explanation he regarded as “good” science? Indeed, this is one of the critical questions of paleontological methodology and I will return to it later. The recent discovery of additional specimens of Archaeopteryx has created new interest in this species and in the larger problems mentioned above. Dr. John Ostrom of Yale University, a leading student of dinosaurs, was studying pterosaurs in 1970 and discovered that one specimen was actually referable to Archaeopteryx. His study of the other known specimens resulted in a series of paj)ers (1970, 1972, 1973, 1974, 1975a, 19751), 1976a) culminating in a large descriptive review (1976b). Ostrom’s work focuses on two juohlems, the origin of birds and the origin of flight, and I believe that his solutions to these i)rohlems will almost certainly change contemporary viewpoints about Archaeopteryx and early avian evolution. 1 find most of his arguments persuasive, not in the sense that they arc* necessarily true, hut that they “explain” far more than previous hypotheses. This is an important distinction, because arguments against Ostrom’s viewpoints have not focused so much on his analysis of morphology and the hypotheses (h'lived directly from it, hut rather on alternative hypotheses that appear to have little evidentiary support. Ostrom (1976b) presents a detailed com{)arative anatomical discussion, either refuting or calling into serious (piestion many previous ideas about the morphology of Archae- opteryx. For example: (1) the pubis was probably not directed sharply backward as the Berlin specimen seems to indicate; (2) the hand and forelimh skeletons are not especially hirdlike, and on the basis of comparisons with reptiles, Ostrom argues that the digits are nuiid)ers I. II, III; by implication this would he true for modern birds as well and thus contradicts the conclusions of some embryologists who have identified the digits as II, III, and IV; (3) x-ray studies indicate that there was proximal fusion of the metatarsals, and two proximal tarsal elements were co-ossified with the tibia and fibula, and at least two distal tarsals were fused to the metatarsus; (4) Ostrom believes that elements previously considered parts of the sternum are misidentified, and he concludes that an ossified sternum was not present. Ostrom has made other notable discoveries about the morphology of Archaeopteryx 488 September 1977 • SPECIAL REVIEW 489 that have significance when comparisons are made to modern birds and fossil reptiles. It is here that ornithologists owe Ostrom a particular debt, for it is doubtful whether any avian morphologist or paleontologist has as complete understanding of reptilian morphology as he does. This special knowledge has permitted him to undertake a broad comparative investigation, far surpassing efforts by previous authors. Which brings us to his two major conclusions, first about the relationships of Archaeopteryx and birds to reptilian taxa, and second, about the origin of flight. Since its discovery in 1861, nearly all authorities have recognized the avian affinities of Archaeopteryx, yet its morphology was recognized as basically reptilian. Ostrom identifies two features as clearly indicating a relationship to birds: the possession of feathers and the fusion of the clavicles into a furcula. He also mentions other “birdlike” features, but does not make a strong attempt to use them as evidence of relationship. For example, the fusion of metatarsal elements and the fusion of tarsal elements with the tibia-fibula and metatarsals are characters shared with birds, though the theropod Syntarsus exhibits similar conditions. The tibia and fibula have become slender in Archaeopteryx, and especially in later birds, and both bones are elongated relative to the femur. Furthermore, Ostrom considers the orientation of the pubis to be intermediate between theropods and birds. All these features would seem to corroborate a phyletic relationship between birds and Archaeopteryx that excludes other vertebrates (but see comments below on Syntarsus). Although not expressed entirely in Hennigian terminology, Ostrom’s analysis very much follows the principles of phylogenetic systematics. He recognizes that “proof” of homology is not possible, therefore “the only reasonable working hypothesis remaining is that . . . resemblances are homologous in the absence of contrary evidence” (1976b: 100). Contrary evidence would be, of course, support for a phylogenetic hypothesis requiring convergence in the characters under consideration. Ostrom also supports the notion that only derived character-states can be used to indicate relation- ships, shared primitive character-states lacking such information. For the most part, I believe Ostrom’s conclusions and statements about systematics are well-founded and expressed in a logically consistent framework of phylogenetic reasoning. Unfortunately, we are all captives of our past — intellectually and psychologically — and Ostrom is no exception. When examined closely some of his statements lack clear meaning, but it must be said that all previous persons writing about Archaeopteryx fell into the same trap. As an example, consider the old argument about whether Archaeopteryx is (1) an aberrant form off the “main-line” of avian evolution, (2) on the “main-line” of avian evolution, or (3) the “direct ancestor” of birds (the second and third arguments are sometimes interchangeable) . I doubt that there is a more pointless issue in the study of avian evolution than this. Ostrom ( 1975b :521) believes that considering Archaeopteryx as a “main-line” (never defined by him) transitional form is fundamental to his arguments about aymn-Archaeopteryx affinities to theropods. Nothing could be further from the truth, because the conclusion to be drawn from his studies is that birds + Archaeopteryx are more closely related to theropod dinosaurs (and perhaps only a few genera of theropods) than to other reptiles; this conclusion can be reached whether Archaeopteryx is “main-line” or not. Thus, Ostrom has not yet escaped from some unnecessary doctrine of his paleontological training: “I personally believe Archaeopteryx lies very close to bird origins and probably is directly ancestral to all later birds” ( 1975a :61). It is questionable whether “very close” has any precise semantic or biological meaning as used here or whether “probability levels” are at issue. Nowhere does Ostrom (nor do paleontologists in general) present ways in which hypotheses of 490 THE WILSON -BULLETIN • VoL 89, No. 3 ancestral-descendant relationships can be tested. One necessary condition for an ancestor is that the ancestral species (only species can be ancestors) must have all primitive character-states relative to its descendants. Indeed, Ostrom considers the greatly shortened ischium as a unique (derived) feature of Archaeopteryx. If A. lithographica were the direct ancestor of birds it would be necessary to postulate reversals in the avian lineage for each derived character-state of A. lithographica, and this is less parsimonious than assuming that these derived states were evolved after the speciation event giving rise to birds on the one hand and Archaeopteryx on the other. But to repeat, I fail to see that these types of arguments have a fundamental bearing on Ostrom’s major conclusions. Following the discovery of Archaeopteryx in 1861 many morphologists and paleontolo- gists wrote about the similarities of Archaeopteryx to different reptilian taxa. Many of these workers identified various dinosaur groups as the possible ancestors of Archaeop- teryx and birds, and this view gained some acceptance. Later R. Broom and G. Heilman argued that dinosaurs and the Archaeopteryx-avian lineage were both derived from the same common ancestor, the pseudosuchian thecodonts. With few exceptions (particularly P. Lowe and N. Holmgren) this viewpoint of avian origins has been accepted dogma for over 50 years, and Ostrom’s historical analysis (1976b: 168-173) adds yet another example within avian systematics where a particular idea about relationships is main- tained on the basis of authority rather than documented evidence. Ostrom resurrects the theory of dinosaurian origins and convincingly demonstrates that the morphology of Archaeopteryx is extremely similar to that of theropod dinosaurs and very different from other reptilian groups, including pseudosuchians — on such comprehensive work are new dogmas born and sustained! Basically, Ostrom’s argument is that theropods and Archaeopteryx share many derived character-states within reptiles and therefore his hypothesis of common ancestry seems well supported. Ostrom appears to have done a masterful job in this analysis, although it will take a specialist familiar with reptilian anatomy to be the final arbiter. Hecht (1976:357-360) criticizes Ostrom’s list of shared derived characters because he believes they are “adaptive” or “fusion-reduction” characters and therefore of “low weight.” Hecht’s contentions are straws in the wind, for all significant taxonomic characters are “adaptive,” and derived characters are either the result of common ancestry or independent origin (convergence). The only way to distinguish between these alternatives is by reference to a particular phylogenetic hypothesis. At best, all that Hecht offers is support for Bock (1965), who at that time accepted a pseudosuchian ancestry. It is imj)ossible here to summarize all the similarities between Archaeopteryx and theropods, but they are so substantial that without feathers Archaeopteryx would have been classified as a theropod. The similarities are strongest in the morphology of the forelimb, pectoral girdle, vertebral column, and skull. Of particular interest is the possibility, not discussed by Ostrom, that Archaeoptryx may be related to only one or a few genera of theropods rather than the entire group. For example, Syntarsus and Ar- chaeoptryx appear to share some specializations absent in other theropods. This problem deserves further attention because it has obvious relevance for the analysis of avian origins and higher taxa in general. As Ostrom correctly points out, the morphology of theropods and Archaeopteryx is the key to understanding the origin of birds and their flight mechanism, and his inferences from that morphology have led him to an unconventional, and controversial, hypothesis for the origin of avian flight. The literature on the origin of flight is an amalgam of good science and speculations bordering on science fiction. The latter are sometimes passed off as scientific because it is “historical narrative explanation,” but the essential September J977 • SPECIAL REVIEW 491 problem remains how to “describe” (one somehow hesitates to call this “explanation”) the events of an admittedly interesting historical occurrence and yet not succumb to making inferences that exceed the available evidence. The main outline of the different theories on avian flight extends hack to 19th century workers. Basically there are two: flight originated from terrestrial, cursorial bipeds or from arboreal bipeds that passed through a gliding phase. The latter has been generally accepted by contemporary biologists and has had its clearest exposition by Bock (1965, 1969). Ostrom (1974, 1976a) resurrects the theory of terrestrial origin, but with a new twist. He argues as follows: (1) morphologically Archaeopteryx is a theropod, thus functional inferences should be based on that morphology with little emphasis given comparisons with modern birds; (2) Archaeopteryx was an active biped, and the hindlimh provides no indication of special adaptations for an arboreal habit; (3) the forelimh of Archaeopteryx, like that of closely related theropods, was a grasping appendage with strong powers of adduction and was adapted for predation and not climbing; (4) feathers evolved along with high metabolic activity and as a thermoregulatory control mechanism; (5) contour feathers of the forelimh were modified to aid in capturing prey and only later evolved flight functions. Although many workers will he skeptical at first exposure to Ostrom’s ideas — after all, the wings look like those of modern birds — once one confronts the totality of the evidence, his ideas become more and more acceptable compared to the alternatives that have been suggested. Ostrom is willing to construct hypotheses about the origin of avian flight only on the evidence presented by Archaeopteryx. Previous hypotheses — particularly the arboreal theory of flight — have been biased by expectations that Archaeopteryx should function as a bird. But Archaeopteryx was a theropod dinosaur with feathers. Personal prejudicies against the use of wings as capture devices should not he based on one’s experiences with living birds — where the wing is clearly flight adapted — but on what might be expected of a feathered theropod. If non-feathered theropods were using forelimbs for predation, then might not Archaeopteryx have done likewise? In one of his most interesting papers, Ostrom (1976a) suggests that in Archaeopteryx the pectoralis minor (supracoracoideus) was a depressor of the arm, not an elevator as in birds, that the trunk skeleton was flexible, that the pectoral girdle was not rigidly fixed, that the sternum was probably cartilaginous, and that the forelimh skeleton ex- hibited no specializations usually attributed to avian flight. Archaeopteryx apparently could not elevate the humerus above shoulder level nor could the hand be folded back against the forearm. On the other hand, Ostrom claims that forelimh functions in- cluded rapid extension of the manus, powerful anteroventral flexion of' the forearm toward the midline, and the capacity for extreme hyperextension of the wrist — all adaptations expected in a predator. It seems to me that Archaeopteryx cannot be used to support an arboreal origin of avian flight. Archaeopteryx does not appear to have arboreal adaptations, and one wonders whether a species previously adapted for cursorial locomotion could move into the trees without such adaptations. Moreover, Archaeopteryx does not appear to possess a morphology indicating flight or even parachuting-gliding ability. If workers insist on building hypotheses based only on what we presently know, then we may be compelled to accept a terrestrial origin of avian flight. To invoke additional, unknown proto-avian stages is tantamount to the erection of ad hoc hypotheses. Nowhere does Ostrom deny the possibility of an arboreal origin (nor would I), but such a conclusion must await further discoveries. Surely there must have been a radiation of feathered coelurosaurs, and perhaps some of these were arboreal — but Archaeopteryx was not. 492 THE WILSON BULLETIN • VoL 89, No. 3 Historical narrative explanation typically does not involve direct deduction of historical events from natural laws, hence some philosophers of science claim that historical narration is not explanation, but merely description. Be that as it may, systematic hypotheses can be evaluated on the basis of how well they account for the available data and how consistent they are with known properties of organisms (phys- iology, genetics, etc.). In the case of the origin of avian flight. Archaeopteryx is about the only real evidence we have; to his credit, Ostrom is unwilling to extend him- self much beyond that evidence. I am not trying to create a bandwagon over Ostrom’s papers, but they are exciting. Some of his findings may eventually be refuted, but there is no doubt that much of his meticulous work will last and that our ideas on avian evolution will be significantly influenced by his results. Ornithologists owe this non-ornithologist a great deal for this contribution. — Joel Cracraft. LITERATURE CITED Bock, W. J. 1965. The role of adaptive mechanisms in the evolution of higher levels of organization. Syst. Zool. 14:272-287. . 1969. The origin and radiation of birds. Ann. New York Acad. Sci. 167:147- 155. Hecht, M. K. 1976. Phylogenetic inference and methodology as applied to the verte- brate record. Evol. Biol. 9:335-363. Ostrom, J. H. 1970. Archaeopteryx: Notice of a “new” specimen. Science 170:537-538. — . 1972. Description of the Archaeopteryx specimen in the Teyler Museum, Haarlem. Koninkl. Nederl. Akad. van Wettenschappen, ser. B, 75:289-305. . 1973. The ancestry of birds. Nature 242:5393. . 1974. Archaeopteryx and the origin of flight. Quart. Rev. Biol. 49:27-47. . 1975a. The origin of birds. Ann. Rev. Earth Planet. Sci. 3:55-77. . 1975b. On the origin of Archaeopteryx and the ancestry of birds. Proc. Centre Nat. Rech. Sci. Collog. Int. 218 Problemes Actuels de Paleontologie- Evolution des Vertebres, pp. 519-532. . 1976a. Some hypothetical anatomical stages in the evolution of avian flight. Smithson. Contrib. Paleobiol. 27:1-21. . 1976b. Archaeopteryx and the origin of birds. Biol. J. Linn. Soc. 8:91-182. ORNITHOLOGICAL LITERATURE Proceedings of the 16th International Ornithological Congress, Edited by H. J. Frith and J. H. Calaby. Australian Academy of Science, Canberra, A.C.T., Australia, 1976: xvii 765 pp., many black-and-white photographs, drawings, and charts. $50.00 Aust. — The 16th I.O.C. was held in Canberra, Australia, from 12-17 August 1974. The scientific program included 11 symposia with 61 papers, and an additional 131 papers in general sessions. Abstracts of the latter were published in The Emu, vol. 74, Supplement, 1975. The present volume contains only the papers from the symposia, and thus is a very incomplete representation of the meeting. In general it appears that the symposia papers are more reviews than research reports, while the general session papers are mainly of the latter type. There are some exceptions, how- ever, Although the symposia cover a variety of topics, there is a strong emphasis on ecological and biogeographical subjects, especially in relation to the southern continents. Most of the papers present well prepared integrative summaries of the then current status of their subjects, but several differ little from other writings by their authors. A number of papers review the biology of Australasian birds, which must have been particularly interesting to non-Australians attending the congress. The volume opens with several I.O.C. committee reports, and the presidential address by Jean Dorst on “Historical factors influencing the richness and diversity of the South American avifauna.” Symposium No. 1 is titled Origins of Australasian Avifauna, and includes papers on paleogeography and paleoclimatology (J. Cracraft) ; the fossil record (P. V. Rich) ; protein evidence (C. G. Sibley) ; adaptive radiation in Meliphagidae (A. Keast) ; and the origins of Australian waterfowl as evidenced by their reproductive photoresponses (J. Kear & R. K. Murton). The symposium is unified by its considera- tion of the roles of immigration and isolation in the development of the Australasian avifauna. Symposium No. 2, Biology of Southern Hemisphere Species, includes a general dis- cussion by A. Keast; a study of the natural history of the emu (S.J.J.F. Davies) ; and a consideration of Australasia and the origin of the Phalacrocoracidae (G. F. van Tets). Papers on Australasian specialities include a comparative field study of Scrub-birds and Lyrebirds (G. T. Smith) ; evolution of the bowerbirds and birds of paradise (R. Schodde) ; osteology of Grallinidae, Cracticidae, and Artamidae (A. McEvey) ; and a review of the New Zealand Wattlebirds (G. R. Williams). A general conclusion from this symposium is that less is known about the biology and relationships of Australasian birds than about those of most other regions. Symposium No. 3 concerns The Value of Various Taxonomic Characters in Avian Classification. W. J. Bock gives a general review of recent advances in avian sys- tematics. Other papers include analyses of various kinds of taxonomic characters: ex- ternal morphology (L. L. Short) ; fossil birds (P. Ballman) ; the digestive system (V. Ziswiler) ; behavior (K. Immelman — abstract only) ; oology (W. Meise) ; and bio- acoustic characters (V. Ilyichev) . I found this symposium disappointing. In his intro- duction E. Mayr identifies its main concerns as (1) the relationships of aberrent or isolated taxa, and (2) the mutual relationships of major avian groups. Few authors addressed these problems directly. It is remarkable that there is almost no discussion of the various systematic philosophies currently vying for predominance. Perhaps this is because the symposium was defined in terms of characters rather than the methods of their analysis. Symposium No. 4, Breeding Birds in Southern Continents, includes studies on en- 493 494 THE WILSON BULLETIN • Vol 89, No. 3 dogenous controls of reproductive rhythms (E. Gwinner & V. Dorka) ; breeding seasons of Australian waterfowl (L. W. Braithwaite) ; onset of breeding in African hornbills (A. C. Kemp); breeding of African birds in non-arid habitats (G. L. MacLean), and environmental control of breeding and movements in Australian birds (H. A. Nix). Symposium No. 5 concerns Biology of Crowned Sparrows {Zonotrichia) in Two Hemispheres. Papers by B. B. deWolfe, J. R. King, M. L. Morton, F. Nottebohm, A. H. Meier, and D. S. Earner discuss the various environmental and internal factors con- trolling reproductive and related behaviors in several species. Symposium No. 6 is titled Structure of Feathers. It includes papers on gross feather structure ( E. Rutschke) ; structural adaptations of feathers (P. Stettenheim) ; structural colors (J. Dyck) ; taxonomic and evolutionary aspects of feather proteins lA. H. Brush) ; keratin synthesis ( D, J. Kemp et ah); and the molecular structure of feather keratins ( R. B. D. Fraser and T. P. Macrae). The symposium demonstrates that feathers,’ the most distinctive characteristics of birds, are the subject of active research at levels ranging from gross anatomy to biochemistry, and leads to the expectation of important advances in the near future. Symposium No. 7, Physiological and Behavioural Adaptations to Arid Zones includes a general review by W\ R. Dawson, and papers on the birds of Africa and South America (G. L. MacLean) ; Australia tS. J. J. F. Davies) ; Australian ducks (L. W. Braithwaite) ; and Sandgrouse (G. L. MacLean). Symposium No. 8 considers the Systematics of Australian Passerine Birds. It includes studies of Australian and Pacific Island w^arblers (A. Keast) ; some monotypic genera of Australian oscines ( R. Schodde & J. L. McKean); the Quail-thrushes (J. Ford); and protein studies of various forms ( C. G. Sibley). It is clear that new approaches to systematic analysis promise to clarify the relationships of many enigmatic Australian passerines. Symposium No. 9 examines the Evolution of Island Land Birds. A. Keast discusses general principles in relation to the specific case of isolated forest outliers in southern Australia, and F. Salomonson gives a theoretical analysis of the main problems con- cerning avian evolution on islands. Other papers include studies of population variation on islands ( P. R. Grant); land-hridge islands (J. M. Diamond); the species-area relation within archipeligos (T. W. Schoener) ; and plant succession and avifaunal structure (C. Ferry et ah). Symposium No. 10 deals with Co-operative Breeding in Birds. I am told by one who was there that many considered this to be the most exciting topic at the meeting. This is not so apparent from the papers, which deal mainly with surveys of co-operative breeding in various regions, including Australia (I. Rowley) ; Africa (L. G. Grimes) ; America )G. E. Woolfenden) ; and Eurasia (A. Zahavi). The symposium is important in bringing together in one place an enormous amount of data on the occurrence of co-operative breeding, but now a general synthesis is needed. Symposium No. 11 concerns Seabirds: Distribution, Speciation and Ecological Diversi- fication at Sea. Papers cover birds of the tropical “middle seas” (K. H. Voous) ; the North Atlantic Ocean ( W. R. P. Bourne) ; South America and the North West Atlantic ( R. G. B. Brown) ; the Australian sector of the Southern Ocean (G. W. Johnstone and K. R. Terry) ; and a general review by M. D. F. Udvardy. Altogether this is a valuable collection of papers on a diverse group of topics, although as noted above it is only part of the proceedings of the 16th 1.0. C., rather than the com- plete collection of papers that the title suggests. It should be a useful addition to any ornithological library. — Robert J. Raikow. September 1977 • ORNITHOLOGICAL LITERATURE 495 Bird Life. By Ian Rowley. The Australian Naturalist Library, Taplinger, New York, 1975. 284 pp., 28 plates of color and black and white photographs, 34 line drawings, maps and tables. $14.95. — This excellent book with a rather unfortunate title is part of a new series on Australian natural history, modeled after the Collins New Naturalist Library. If Rowley’s contribution is typical of this new series, Australians and Australia buffs have a great deal to look forward to. Most introductory ornithology books are written by Americans or Europeans for North- ern Hemisphere readers. Rowley has now supplied Australians with an antipodean equivalent: an introduction to the ecology and behavior of birds, using examples from the Australasian fauna. It is also a highly interesting summary of recent field research on Australian birds, drawing from material that hitherto has been hidden in technical reports or journal articles not available to most readers — especially those in the Northern Hemisphere. The first 5 chapters comprise a general introduction: how birds have evolved and adapted to Australia’s unique environmental conditions. This is followed by 16 chap- ters of research-oriented accounts of single species or groups of related species that illustrate the introductory statements. Ten of these chapters deal with resident or sedentary birds such as the Superb Blue Wren {Maliirus cyaneus), Australian Raven iCorvus coronoides), Australian Magpies {Gymnorhina spp.). Kookaburra iDacelo gigas) , Miners (Manorina spp.), and Mallee Fowl {Leipoa ocellata) . The last six discuss migrant species (using an overly strict definition of the term) such as Tas- manian Mutton-birds (Puffinus tenuirostris) and nomadic species such as Brolgas {Grus rubicundus) and White-tailed Black Cockatoos {Calyptorhynchus baudini) . Rowley begins most accounts by stating who conducted the research, where and how it was done, and whether or not it is still in progress. He then summarizes the results in an interesting manner, successfully avoiding both the too technical and the too ele- mentary. Much of the excitement of this book lies in reading about active research; one cannot help wondering whether some of the questions raised have been answered since the book was completed. For those who wish to read further about a given species, the accounts are well referenced. The book ends with a chapter on economic ornithology and conservation problems, an appendix on methods of study (especially banding), another on books, journals, and societies, eight pages of references, and a good ten-page index. As an American interested in Australian birds, but with only a few weeks of first- hand experience with the fauna, I found Rowley’s book fascinating. He has done a particularly good job in describing the varied social systems — breeding regimens and movements — showing how they are adapted to the Australian climatic and ecological conditions that are so different from those in the Northern Hemisphere. Also of interest is the comparatively small number of migratory species in the Australian avi- fauna (approximately 8%) vs. residents (66%) and nomads (26%). The few trans- equatorial migrants are mostly waders and seabirds (no passerines). Other species migrate north-south within the Australo-Papuan region or between Tasmania and main- land Australia. A few may also be altitudinal migrants, but this is still poorly known. An American reader will be struck by the very different components of the Australian avifauna: the large number of nectivorous birds (lorikeets, sunbirds, silvereyes, and some 69 species of honeyeaters) , and the many nomadic species. IMost nomads breed seasonally, but may change the breeding locale from year to year. A few seem to be able to breed whenever conditions are favorable. Nomadic species are not just arid land birds affected by erratic rainfall in the interior of the continent. They are also forest 496 THE WILSON BULLETIN • VoL 89, No. 3 and farmland species that follow a changing food supply. These latter include flower feeders (such as lorikeets and honeyeaters) , insectivores (currawongs) , fruit eaters (dicaeids), seed eaters (some cockatoos and grass finches), and scavengers (some corvids) . It is unfortunate that most ornithological research in Australia has had to be con- cerned only with economically important species: game birds, agricultural pests, etc. The CSIRO-conducted research has been of generally high quality but its limited scope has left serious gaps in the general knowledge of the avifauna. Fortunately private individuals and groups, particularly banders, are becoming increasingly active, so the serious research effort is becoming better balanced. In a few instances Rowley’s discussions have been outdated in information, or he has limited himself too much on a topic. His treatment of navigational theories is slightly behind the times, and in the section on zoogeography he begins with the Pleistocene, omitting mention of the important recent studies of plate tectonics and their effect on the origins of the Australian fauna. The book also should have been edited more tightly for grammatical errors: on p. 163 the word “data” is used both as a singular and plural in the same paragraph; on p. 108 one finds “orientate”; farther and further apparently are considered interchangeable; and in places the usage of commas, colons, and semicolons is quixotic. The only serious fault of the book lies in the poor placement of the plates in relation to the text. For example, the plate showing nestling growth rates of Australian Ravens and 2 photographs of White-winged Choughs is inserted in the middle of the Kooka- burra chapter, 24 pages after the raven account and 6 pages after the end of the choughs. The raven text is “illustrated” instead by a wren plate that belongs in the previous chapter. As the plates are not evenly spaced through the book, they could easily have been better coordinated with the text. It is also annoying to have the plates referred to only by number rather than by page, especially when they are so badly scattered. These minor objections aside, considering his stated aims, Rowley has done an ex- cellent job. This hook will he enjoyed by anyone even slightly interested in the Australian avifauna or in the intricate adaptations of birds to their environment. The title is sadly undescriptive of the contents, but perhaps through listings such as “Bird Life [Australian]” that I saw recently in a dealer’s catalog, this book will find its in- tended, and deserving, readership outside Australia. — Mary H. Clench. Avian Physiology, 3rd Edition. Edited by Paul B. Sturkie. Springer-Verlag, New York, Heidelberg, Berlin, 1976: xiii -f 400 pp., 77 tables, 102 figures. $23.80. — The prospect of having a current treatment of avian physiology in one volume is an attractive one, for the primarily physiological chapters of “Avian Biology” (D. S. Earner and J. R. King, editors; Academic Press) are distributed over 4 volumes and a 3 year interval (1972-75). Sturkie’s book is thus timely and worth considering as to its adequacy as a reference work for ornithologists requiring physiological information in their studies. Authoritativeness has been assured in this volume through assemblage of a group of authors who have contributed chapters on their respective fields of inter- est: Nervous System (T. B. Bolton) ; Sense Organs (M. R. Kare and J. G. Rogers, Jr.) ; Blood: Physical Characteristics, Formed Elements, Hemoglobin, and Coagulation ( D. Sturkie with P. Griminger) ; Heart and Circulation: Anatomy, Hemodynamics, Blood Pressure, Blood Flow, and Body Fluids (P. D. Sturkie) ; Heart: Contraction, Conduction, Electrocardiography ) P. D. Sturkie); Respiration (M. R. Fedde) ; Regula- September 1977 • ORNITHOLOGICAL LITERATURE 497 tion of Body Temperature (G. C. Whitlow) ; Energy Metabolism (G. C. Whitlow) ; Alimentary Canal: Anatomy, Prehension, Deglutition, Feeding, Drinking, Passage of Ingesta, and Motility (P. D. Sturkie) ; Secretion of Gastric and Pancreatic Juice, pH of Tract, Digestion in Alimentary Canal, Liver and Bile, and Absorption (P. D. Sturkie) ; Carbohydrate Metabolism (R. L. Hazelwood) ; Protein Metabolism (P. Griminger) ; Lipid Metabolism ( P. Griminger) ; Kidneys, Extrarenal Salt Excretion, and Urine ( P. D. Sturkie); Hypophysis (P, D. Sturkie); Reproduction in the Female and Egg Formation (P. D. Sturkie with W. J. Mueller) ; Reproduction in the Male, Fertilization, and early Embr>mnic Development (P. D. Sturkie with H, Opel) ; Thyroids (R. K. Ringer) ; Parathyroids, Ultimobranchial Bodies, and the Pineal (R. K. Ringer and D. C. Meyer); Adrenals (R. K. Ringer); Pancreas (R. L. Hazelwood). The chapters are well supplied with tables and figures that effectively supplement the text. Each of the chapters concludes with a list of the references cited in it, which should facilitate access to the original literature. The coverage of this literature extends mainly through 1974; only a handful of references later than this are included. The various chapters are all quite informative. I found those on respiration; regula- tion of body temperature; carbohydrate, protein, and lipid metabolism; excretion; reproduction in the female; and various aspects of endocrinology of particular interest. The involvement of so many authors leads to only a few inconsistencies. For example, one derives quite different impressions of the state of knowledge concerning the avian pineal from the statements by P. D. Sturkie ( p. 287 ) and by R. K. Ringer and D. C. Meyer (pp. 365-368). My principal dissatisfaction with the third edition of “Avian Physiology” concerns the insufficient attention it devotes to the physiology of wild birds. The bulk of the available information on avian physiology does pertain to the domestic fowl, but a substantial body of literature exists on wild birds, which could permit broadly com- parative treatments of a number of topics, something done very well in the volumes of “Avian Biology.” Adequate use of this literature has been made in only a few of the chapters, notably those dealing with respiration, regulation of body temperature, energy metabolism, and excretion, which have a fairly strong comparative orientation. Further use of such an orientation would have led to improved coverage of a variety of processes important in the lives of birds, e.g., molt, migration (including its navi- gational aspects), winter fattening, reproductive timing, circadian periodicities, respira- tory gas exchange and moisture loss in eggs. What comparative coverage does exist in many of the chapters is complicated by use of imprecise common species names (e.g., “sparrow”) and/or omission of scientific names. “Avian Physiology” will be a useful reference and text for individuals concerned with domesticated species. It also will be helpful to ornithologists seeking information on topics for which the domestic fowl is an adequate model species. However, the utility to this latter audience would have been increased by more extensive treatment of wild birds. — William R. Dawson. The Birds of the Malay Peninsula, Vol. 5. By Lord Medway and David R. Wells, illus. by H. Gronvold. H. F. & G. Witherby Ltd., London, 1976: 448 pp., 25 color plates. £25. — Since the authors of this volume were my colleagues and since I had a deep personal interest from the inception of their work, I take pleasure in reviewing this important addition to the recorded knowledge of the birds of the Malay Peninsula. The careful understatements in the preface only hint at the struggle and pathos in- 498 THE WILSON BULLETIN • Vol. 89, \o. 3 volved in the development and completion of the 5 volume work first conceived by H. C. Robinson, and later continued by F. N. Chasen, and in an unpublished form, by E. Banks. All of the first 3 authors died before the set could be completed, and its publication extended over a period of 5 decades. The first 4 volumes were published in 1927, 1928, 1938, and 1939. ^ hen, in 1964, the plates for the final volume were discovered in the British Museum, David Wells told me that he and Lord Medway were contemplating completing the series both with enthusiasm and trepidation. They wished to duplicate the format of the previous 4 volumes, but realized that the costs would be much above the original. I do not know what the original volumes sold for, but they had become collector’s items by 1958 when I bought my copies. There was no handbook or field guide for Malayan birds at that time, but Smythies “Birds of Borneo” encompassed most of the species and was very useful. The recent “A field guide to the birds of S. E. Asia” by King and Dickinson helps to fill the field guide needs for that area. It is necessary for the reader to have at hand the previous 4 volumes before approach- ing the 5th critically. The authors have done a commendable job of recreating the format. Any criticism leveled at type form, color plates, book size, etc. must be tempered by knowledge of the objectives involved. Use of the volume and its many local names is made easier by having a gazatteer of localities in the introductory section. Following the introduction are 3 chapters by David Wells (Resident Birds), Lord Medway (Migratory Birds), and Ian C. Nisbit (The Eastern Palearctic Migration System in Operation) based upon their research and observations. In these they summarize much of what has been learned of Malayan birds since the 1938 volume was published. These introductory chapters take up the first 77 pages of the book. The remainder, over 325 pages, is devoted to a discussion of the 576 species recorded from Malaya. In this the taxonomy is brought up to date, progression of molt is indicated where known, migration information such as arrival and departure dates are given, and in- formation on such topics as nesting, number of eggs, and a description of the voice is provided. Detailed biological descriptions are not given except where such data were inadequately reported in previous volumes. By this method a great mass of information is added to that already provided in the first 4 volumes, and knowledge is brought up to 1973 when the manuscript was closed for editing and publication. The authors are to be highly commended for this authoritative and carefully prepared volume of great historical interest as well as value to the ornithology of Southeast Asia. H. Elliott McClure. Falcons Return. By John Kaufmann and Heinz Meng. Wm. Morrow and Co., New York, 1975: 128 pp., 106 black and white photos. S5.95. — Subtitled “Restoring an endangered species,” this book is an account of the Peregrine Falcon and details its life history, its widespread destruction due to DDT poisoning, and efforts to restore the species by introducing captive-bred Peregrines back into the wild. The first section is a description of the birds’ external anatomy, nest sites, courtship, flight, and other aspects of Peregrine biology. The text is illustrated with black and white photographs ranging in quality from fair to excellent. Those of the Holt’s Ledge eyrie are particularly good and show all aspects of the breeding biology of Peregrines. This section records the demise of the birds and concludes with a nontechnical discussion of pesticide poisoning and food chains. September 1977 • ORNITHOLOGICAL LITERATURE 499 Falconry is the subject of tlie second section. The history of the sport is described, and many falconry terms are defined. Capture, training, and flying of eyasses (young Peregrines removed from the nest) are described, and lure flying and luinting are ex- amined. Although some conservationists may cringe at the mention of falconry, the authors contend that the only chance of bringing the Peregrine hack depends upon the skillful use of the techniques of falconry. They suggest that hacking, an old method of raising eyasses so that they remain basically wild, hut learn to fly and hunt on their own, is a possible means of replenishing the natural population. The third section is devoted to the efforts of Heinz Meng to breed and raise Peregrines in captivity. The account of Meng’s devoted and skillful work as the chicks’ father is unnecessarily detailed and frequently melodramatic. Efforts by workers at Cornell University’s Laboratory of Ornithology are also mentioned. Their capability of raising more than 200 Peregrines a year and placement of captive-hred chicks in eyries where parent birds fail to breed due to eggshell thinning, offer some glimmer of hope. However, the continued use of DDT and other pesticides in poorer countries of Central and South America where many birds winter continues to threaten the existence of Peregrines and other raptors. The hook concludes with a brief bibliography. Although this hook is for general readers, particularly those with an interest in wildlife, it will also interest some ornithologists. It is a simply, but well-written and informative account of the Peregrine Falcon. — David R. Maurer. Geographic and Climatic Relationships of Avifaunas with Special Reference to Comparative Distribution in the Neotropics. By Paul Slud. Smithsonian Contribu- tions to Zoology, No. 122, 1976; iii + 149 pp. — This massive study presents an extremely detailed analysis of bird distribution patterns throughout the world, although emphasis is placed on birds of the Neotropics. In addition, the bulk of the world’s islands for which species lists exist are also examined. Dr. Slud has divided his book (for it is sufficiently lengthy to be so labeled) into four major sections: the Passerine-Nonpas- serine Relationship; the Suboscine-Oscine Relationship; the Passerine-Nonpasserine Suboscine-Oscine Relationship; and Requirements for Further Research. A too-detailed discussion as to why water birds should not he included in the analyses which will follow, and further interesting discussion on the separation of migrants and native land birds, lead into the first major subdivision detailing various properties of the world’s Passerine-Nonpasserine relationships. Regarding migrants, the author’s belief that migrants “complement the residents” (p. 9) rather than “compete with them” (p. 9) is not really borne out by Fig. 4, p. 10. Instead, migrants seem to have a difficult time existing in the complex and apparently highly competitive tropical rainforest as if there were no empty niche space to go around, or as if the highly specialized and competitive residents keep temporary migrants from coexisting in the community. Slud himself notes (p. 9) that, “Both in Africa and South America it is the richest biotope, the equatorial rain belt, that acts as a barrier which many migrants do not cross, very few enter, and the remaining ones skirt or pass over in order to winter in the southern third of the continent.” Admittedly there is much rain forest north of Colombia (where the migrant percentage diverges from the native fauna, see Fig. 4, p. 9), but the data can realistically be interpreted both ways, as arguing for and against competitive interactions. In fact, the inverse correlation of percent migrants with total avifauna (r = 0.99, p. 8) also supports a competition hypothesis. The author points out the similarity of the Passerine-Nonpasserine ratio in habitats 500 THE WILSON BULLETIN • VoL 89, No. 3 throughout the world and notes that it is generally 2:1, with the ratio being higher in temperate areas and lower in tropical regions. The ratio then logically (and nicely) rises with altitude in the tropics. Figure 6 clearly illustrates the positive relationship of Passerine-Nonpasserine ratios with what is apparently overall physical environmental complexity, with the highest U.S. values being concentrated in the Southwest and adjoining Mexico. Slud discusses a “peninsular effect” of lowered ratios and specifically mentions Florida (0.89 P-N ratio) and the Yucatan Peninsula (1.03). He ignores Baja California (a physiographically complex peninsula) which has the highest ratio on the North American land mass! Perhaps one of the most interesting sections is that dealing with birds on islands. “Insular avifaunas tend to correlate in size with the area of the island, but only in a general way” (p. 21). “Insular biotas, however, conform to no universal standard and their compositions are each the unique result of interplay among many factors that are differentially peculiar to islands: this makes islands synecologically nonintercomparable.” ( p. 21). With these statements a reader is led to think that perhaps the author has been isolated from the open literature for ten years and that the theories of island biogeogra- phy have slipped by him. But no, Slud has kept up with the literature. He is merely launching a low key attack on a number of studies that have dealt with avian insular biology, in particular, early work by Diamond, Terborgh and Faaberg, and MacArthur and Wilson. Slud is not speaking theoretically, but presents data on turnover rates, the effect of distance and colonization rates on islands. In particular, he reanalyzes earlier reported results with apparently more complete data and arrives at different conclusions. Slud notes, for example, that there is no evidence yet for believing that small far islands and large near islands have similar extinction rates of bird species. He also points out errors in previously published calculations of colonization rates and notes the difficulty of defining the species pool of potential colonizers in biogeographic studies. The author very likely places too much emphasis on anomalous little Cocos Island in the tropical eastern Pacific which has undergone no species turnover in about 70 years (no extinctions and no successful colonizations), and has an endemicity value of possibly 50%. One does not read this well-reasoned section and feel that the basic island biogeographic theory is shaken, for it is built on evidence from numerous and varied fields of study. One does, however, see a hard-working field biologist who gathers data carefully for years before publishing a monograph, challenging members of the “MacArthurian school” to bring a more solid data base into theoretical constructs. The “quick and dirty” techniques of idea biology have made large and exciting contributions to modern ecological theory, but such papers are open to attack by the very limited nature of their data. Perhaps fewer useless theories (which often tie up reasearchers who are com- mitted to supporting or refuting such will-o-the-wisp hypotheses) would clutter the litera- ture if more detailed empirical results were obtained before publishing. However, the spark of imagination is necessary if predictive syntheses are to be made and if one is going to be able to see the forest, rather than the individual trees, the rule, rather than the exception. Trends in the Suboscine-Oscine ratio are discovered (it declines from about 60% in South America to about 20% in Mexico: it is lower in the dry tropics than in the humid tropics). The comparative Passerine-Nonpasserine :Suboscine-Oscine ratio is prob- ably more useful. In particular, such a ratio can indicate whether or not a site is in the humid tropics (a low ratio), and allows comparisons of values obtained from high- lands or higher latitudes (a higher ratio). As Slud notes, the ratio is most useful in site-oriented situations. The usefulness of ratios in surveys where time is limited (and September 1977 • ORNITHOLOGICAL LITERATURE 501 some useful results can still be obtained) is pointed out in Figs. 32 and 33. The overall tone of the “Further Research” section seems to be artificially tacked onto the bulk of the publication, although the section is in itself informative. The data base for this study is presented in detailed tables at the end of the publication. Basically, this is an enjoyable, if ponderous, work. There is much food for thought. The style is possibly too reportorial and I wish that the author had gone on to pursue other ramifications of the data, particularly from the evolutionary standpoint. One recalls (again) the Mac Arthur technique. Here are the data, now we need more ideas. — Michael A. Mares. The Biogeochemistry of Blue, Snow, and Ross’ Geese. By Harold C. Hanson and Robert L. Jones. Southern Illinois University Press, 1976: xviii + 281 pp., 266 figures, 45 tables. $15.00. — The purpose of this study is to show how elemental analyses of goose feathers can be used to determine the local origin of the birds. The book is divided into 8 chapters which cover a discussion of geographical origins of wild geese, sampling, analytical procedures, data analyses, soil and plant relationships, geology, soils, and feather mineral patterns. Data are also presented on the differences in feather mineral patterns and the origins of migrant and wintering geese. An adequate review of the previous literature on the chemical composition of feathers is also presented. Finally, the authors are able to discuss the biogeochemistry of feathers and wild geese and their chemical and mineral environment. Detailed data for 12 elements in feathers (calcium, magnesium, sodium, potassium, phosphorus, iron, zinc, manganese, copper, boron, silicon, aluminum) are presented and discussed, while a less detailed discourse on sulfur is presented in an appendix. The data were obtained by analyzing the vane portions of the pimary feathers, since these are more highly mineralized than the shaft. Optical emission spectroscopy was employed on ashed and subsequently liquefied samples using reference plant samples as standards. Although it is not possible in a short review to mention all the important observations and conclusions obtained from this vast and important study, a few comments might serve to provide some concept of the variety of information that has been gleaned from this project. Sodium is the dominant ion affecting the levels of absorption and excretion of calcium, magnesium, and potassium. Canada Geese, for example, have enough calcium and magnesium in their environment to exceed the excretory losses stimulated by their high sodium intake. The calcium content of feathers was the most important aid in dis- tinguishing geese from the various colonies of Blue and Lesser Snow Geese. Coastal geese can be distinguished from inland populations of western Canada Geese on the basis of the high concentration of phosphorus in their primary feathers. A favored food of Canada Geese is one species of Equisetum, a zinc accumulator, which may account for the high zinc content of Canada Geese from the Belcher Islands. There is at least one species of Equisetum which is reputed to be a gold accumulator which Hanson and Jones might consider employing as a tracer as they continue to expand their study further. In such an extensive study interelemental relationships arise which are quite important. Aluminum and silicon are closely related as is potassium with both these elements, reflecting their involvement with micaceous clays. Iron and manganese coherence are reminiscent of their association in soils as well as ore bodies. In addition, close 502 THE \^TLSON BULLETIN • Vol. 89, \o. 3 significant relationships Avere observed between iron, silicon, and aluminum as well as between zinc and copper, reflecting the metalliferous areas over which birds have passed. It is clear from this study that the chemical examination of feathers provides a sensitive record of environmental and metabolic relationships of elements. The concentration of these elements in feathers indicates the breeding grounds of geese of known and unknown origins. This book is well worth reading, heavily illustrated, and nicely printed. It presents much valuable data of interest to avian biologists as well as general biogeochemists. — Ursula M. Cowgill. Birds and Their ays. By Alexander Dawes Du Bois with Charlotte A. Du Bois. T.S. Denison & Co., Minneapolis, Minnesota, 1976. 184 pp., 81 black and white photos. S8.95. — Intended as a companion to the author’s earlier work. Glimpses of Bird Life, this hook consists of two parts. Part I is an anecdotal and haphazard account of various species encountered during the author’s lifelong pursuit of birds (one early account is on a horse drawn mail stage). Birds were observed in many parts of North America, although most observations are from around the author’s Minnesota home or near Cornell Lniversity. Topics such as nesting, care of young, feeding, song, and sociality are discussed. Part II is a more detailed description of the lives of a dozen favorite species. ith the exception of a blurred kingfisher on p. 16, a spotted photo on p. 127, and a repetitive series of woodpecker holes, the photographs are of good quality and add to the text. These pictures, taken from a blind by the author, are illustrative of his devotion to birds and photography. A major criticism is with the constant anthropomorphic interpretation of birds’ actions: a bluebird "had sung to instill courage in his two fearful young, to give them promise of safety, and hope of food as a reward.” A Red-headed Woodpecker showed so much grief at his wounded mate, “that I shot him also, out of compassion.” Some of the wording is awkward: “All these hazards birds have no way of coping with,” while other material is stilted: a bathing tanager is likened to “a flame trying to extinguish itself.” This book will have a limited appeal since it is neither visually impressive enough for the coffee table trade nor rigorously scientific. — David R. Maurer. The Bluebird: How You Can Help Its Fight For Survival. By Lawrence Zeleny. An Audubon Naturalist Library- Book, Indiana Lniversity Press, Bloomington, 1976: 170 pp., 7 color plates, 33 text figs., 3 tables. S7.95. — In this excellent book for conser\a- tionists and naturalists, Lawrence Zeleny gives a succinct and informative account of bluebird life history- encompassing the breeding ranges of the three species, food habits, courtship behavior, nesting and care of the young, and migration. He makes the reader acutely aware of the problem of declining bluebird populations due to the effects of man’s destruction of their natural habitat, and competition from Starlings and House Sparrows. Other contributing factors are a decline in bluebird winter food supplies, adverse weather conditions, and the indirect effects of insecticides. He also provides a listing of plants (by botanical and common name) that yield supplies of winter berries enjoyed by bluebirds. After defining the scope of the problems, Zeleny presents solutions by which man can assist the bluebird in maintaining and increasing its present population size. He gives September 1977 • ORNITHOLOGICAL LITERATURE 503 detailed directions and illustrations for the construction and placement of suitable roosting and nesting boxes. Based on a typical bluebird nesting timetable, the author explains how to monitor the nesting sites, and how to protect the bluebirds from predatory mammals, snakes, birds, and other cavity-nesting competitors. He also discusses various insects that are troublesome for bluebirds during the nesting period, and offers a table of insecticides that can be employed in the destruction of these pests, but that are unlikely to be noxious to the birds. He also gives the history, objective, construction, and management of bluebird trails that have been successful in helping the bluebirds to survive. Throughout the book Zeleny ascribes the human attributes of love, happiness, sorrow, and altruism to the bluebird. The dispassionate scientist may criticize this anthro- pomorphic connotation, but after reading the heartwarming and delightful accounts of the author’s handraising a brood of orphaned bluebirds to adulthood and seeing them and their offspring return to nature, one can only admire and think of the true beauty of God’s creatures. This is a truly enjoyable book that I recommend most enthusiastically. — Edward V. Swierczewski. ORNITHOLOGICAL NEWS AARON M. BAGG STUDENT MEMBERSHIP AWARDS— 1977 Student Membership Awards in the Wilson Ornithological Society have been made available through funds generously donated in the memory of the late Aaron M. Bagg, former president of the Society. The Student Membership Committee has designated the award recipients for 1977 as follows: Theresa M. Allen, University of Texas at Arlington; Jonathan L. Atwood, California State University at Long Beach; Albert Aulette, Michigan State University; Bruce M. Beehler, Princeton University; Keith L. Bildstein, Ohio State University; Erik J. Bitterbaum, University of Florida; Roderick N. Brown, McGill University; Kelly B. Bryon, Sam Houston State University, Texas; William D. Clark, University of Illinois; Leon J. Folse, Jr., Texas A&M University; Eric D. Forsman, Oregon State University; Kimball L. Garrett, University of California at Los Angeles; Ralph J. Gutierrez, University of California at Berkeley; Geoffrey G. Hogan, Brock University, Ontario; Ronald L. Kalinoski, Syracuse University; Walter D. Koenig, University of California at Berkeley; Scott M. Lanyon, State University of New York at Geneseo; Howard Levenson, Humbolt State University, California; Douglas W. McWhirter, Michigan State University; Michael C. Moore, Indiana University; Gerald R. Meyers, Kent State University; Barry R. Noon, State University of New York at Albany; Douglas O. Norman, State University of New York at Stony Brook; Gary L. Nuechterlein, University of Minnesota; Richard T. Reynolds, Oregon State University; Douglas G. Richards, University of North Carolina; Kim M. Riddell, University of Florida; Wanda K. Rola-Pleszcynska, University of Toronto; Mark R. Ryan, Iowa State University; Josef K. Schmutz, University of Alberta; Bonita M. Smith, Miami University, Ohio; Donald L. Sparling, University of North Dakota; Gail E. Spealer, University of Florida; Michael N. Weinstein, California Polytechnic State University — Douglas James, Chairman, Student Membership Committee. 504 THE WILSON BULLETIN • VoL 89, No. 3 EBBA Research Gram The Eastern Bird-Banding Association is offering a S250 Memorial Grant in aid of research. The applicant must be an undergraduate or graduate student and must be using bird banding as part of his or her research. Applications must be completed before 1 March 1978. For further information ^vTite to the Chairman; Dr. Bertram G. Murray, Jr., c/o Biology-Livingston, Kilmer Campus, Rutgers University, New Brunswick, N.J. 08903. Aaron M. Bagg Student Membership Awards Student Membership Awards in The Wilson Ornithological Society are available because of funds generously donated in the memory of Aaron M. Bagg, a former president of the Society. Application forms for the awards to be granted in 1978 may be obtained from James R. Karr, Dept, of Ecology, Ethology, and Evolution, Vivarium Building, Univ. of Illinois, Champaign 61820. The deadline for applying is 1 December 1977. An Aaron M. Bagg Student Membership Award provides a 1-year gratis new membership in The Wilson Ornithological Society for selected exceptional students in the field of orni- thology.— James R. Karr, Chairman, Student Membership Committee. Erratum. — On p. 187, paragraph 5. line 1 of the March issue (Vol. 89, No. 1), the word “distant" should be “distinct." September 1977 • THE PRESIDENT'S PAGE 505 THE PRESIDENT’S PAGE The Wilson Ornithological Society comprises a balanced blend of members from the ranks of both amateurs and professionals united by a common interest in avian biology. It is advantageous to preserve this balance. But, it is particularly important too for the Society to develop programs that will attract student members because these constitute the potential ornithologists of the future. A few years ago the Society adopted the present policy permitting student members, upon graduating, to apply toward a regular life mem- bership the total dues paid when they were students. Maybe this has not been publicized enough, or perhaps the financial status of a new graduate prevents the grasping of opportunities no matter how attractive. Whatever the reason, not many students have accepted this life membership option and its financial benefits. More recently, the Aaron M. Bagg Student Membership Awards have provided pre-paid first-year mem- berships to especially talented students, and this program has been very successful. Also, the Alexander Wdlson Prize has recognized the best student paper presented at annual meetings. Finally, research support to graduate students has been allocated annually in the form of Louis Agassiz Fuertes Grants. But more needs to be accomplished and I will regularly communicate to the member- ship concerning these actions, and also concerning other new executive initiatives of general society interest, or on other matters, particularly those requiring views expressed from the membership. This time I want to stress the matter of assistance to student research through the Fuertes Grants. In recent years the Fuertes Fund Committees have become more and more vocal concerning the difficulty of the task in selecting 1 or 2 recipients from among the increasing number of excellent applications received. There simply have not been enough funds available to award all those who deserve the aid and recognition. Of course, it is unrealistic to expect that there will ever be enough money available to reward every deserving person, but a higher proportion should receive recog- nition than now is the case. There are some who have argued that indeed the recognition is really the only im- portance of the grants because the amounts awarded are not presently enough to be essential to graduate research. True enough, it is an honor to receive these grants and the recipients can be proud to include mention of them in their resumes the rest of their careers. On the other hand, I know of many cases in which an amount of a few hundred dollars, such as a Fuertes Grant, has been extremely important to the successful com- pletion of both master’s and doctoral research programs. With the continuing decline in federal and state funds to support such research, through default this activity becomes increasingly thrust upon other resources such as the Fuertes Fund. Thus, I recommend that new funds be sought to increase the number of Fuertes Awards given annually. I recognize that the primary responsibility of an organization such as the Wilson Ornithological Society is to maintain the viability of a respected research journal, and that almost all operating funds are allocated for that purpose. If the number of Fuertes Grants are to be increased this means finding ways to independently supple- ment that fund. Therefore, I have appointed a special committee to study this matter and make recommendations to the Executive Council at the West Virginia meeting next May. This committee is chaired by Dr. C. J. Ralph. A second charge to the committee concerns revising the application and selection process. I invite members to send their opinions on either matter to Dr. Ralph addressed to the Institute of Pacific Islands Forestry, 1151 Punchbowl Street, Honolulu. Hawaii 96813. — Douglas James. PROCEEDINGS OF THE FIFTY-EIGHTH ANNUAL MEETING { George A. Hall, Acting Secretary At the invitation of the Department of Zoology and the Department of ildlife and Fisheries of Mississippi State University, the Oktibbeha Audubon Society, the Mississippi Ornithological Society, and the Noxubee National Wildlife Refuge, the Fifty-eighth Annual Meeting of the \^’ilson Ornithological Society was held on the campus of Mississippi State University near Starkville, Mississippi, from Thursday 19 May to Sunday 22 May 1977. The Executive Council met on Thursday evening in Harned Hall. Business sessions, and scientific papers' sessions were held in McCool Hall on Friday and Saturday. On Thursday evening members were entertained at a reception at the Lakeside Country Club. A session of 3 motion pictures < Atchafalaya, At the Crossroads, Life in a \^'eaverbird Colony) was held on Friday evening. The Annual Banquet was a buffet dinner held on Saturday evening at the Mississippi State Union. After a few announcements including the announcement of the award winners the group moved to McCool Hall for the showing of a most impressive movie on the Harpy Eagle presented by Neil Rettig, Wolfgang Salb, and Alan Degan. On Friday and Saturday mornings, field trips were held to the Noxubee National Wildlife Refuge, where the feature was a nesting Red-cockaded Woodpecker who per- formed beautifully for all. and to the Cliftonville Heronry, a colony of Little Blue Herons and Cattle Egrets. On Sunday a somewhat longer trip was taken to Noxubee Refuge, and many of the members participated in a canoe trip on the Tombigbee River. A display of ornithological art was on exhibition at the University library and in the MSU Union. Trips were arranged to a local clock factory and to some antebellum homes. A star gazing session was also held on Thursday and Friday nights. FIRST BUSINESS MEETING The session was called to order at 9:15, Friday 20 May by Local Chairman J. A. Jackson. He introduced Dr. J. C. McKee, Vice-President for Research and Graduate Study at Mississippi State University, who made a short speech of welcome. President Andrew J. Berger then responded for the \^'ilson Society and called to order the first business meeting. The minutes of the 1976 meeting at Ithaca, New York were approved. Acting Secretary G. A. Hall summarized the actions taken by the Executive Council on Thursday evening: 1. The Council heard reports from the officers and committee chairmen, copies of which follow. 2. The Council unanimously re-elected Dr. Jerome A. Jackson as editor of the ilson Bulletin. 3. The 1978 meeting will be held at Jackson’s Mill, West Virginia, on 4-7 May 1978. No definite commitments for subsequent meetings have been made but tentatively the 1979 meeting will be in Duluth. Minnesota, the 1980 meeting will be a joint meeting with the Cooper Society, and the 1981 meeting may be in Charleston, South Carolina. 506 September 1977 • ANNUAL REPORT 507 4. The Council is exploring the proposals made by the Council of the A.O.U. to produce a combined membership list for the 3 ornithological societies. Treasurer Ernest E. Hoover summarized the report of the Treasurer. The full report follows. THE WILSON ORNITHOLOGICAL SOCIETY REPORT OF THE TREASURER Year Ending December 31, 1976 GENERAL EUNDS Balance as of last report December 31, 1975 RECEIPTS Membership Dues Active for 1976 Active for 1977 Total Active Sustaining for 1976 Sustaining for 1977 Total Sustaining Subscriptions to The Wilson Bulletin For 1976 For 1977 Total Subscriptions Advance Renewals — Sales of Back Issues of The Wilson Bulletin Interest and Dividends on Savings & Investments Income from General Endowment Fund Income from G. M. Sutton Colorplate Fund .... Interest on Endowment Savings Account Interest on Regular Savings Account Total Interest and Dividends Royalties from Microfilming Back Issues of The Wilson Bulletin Contributions from Authors and Others Transfer from Regular Savings Account $21,088.95 $ 4,203.00 11,640.00 $15,843.00 135.00 570.00 705.00 1,963.50 6,580.00 8,543.50 419.50 1,140.00 4,758.71 1,761.14 423.97 368.85 7,312.67 185.24 1,878.00 7,000.00 Total Receipts $43,026.91 DISBVRSEMENTS The Wilson Bulletin (Printing & Engraving) 27,578.23 The Wilson Bulletin (Mailing & Maintenance) 2,351.08 Colorplate Processing Expense — 2,201.80 Editor’s Expense 1,703.26 Secretary’s Expense 86.09 Treasurer’s Expense 1,636.38 Committee Expense 302.32 Annual Meeting Expense 300.67 508 THE WILSON BULLETIN • VoL 89, No. 3 International Council for Bird Preservation 30.00 Miscellaneous Expense 2.00 Total Disbursements $36,191.83 Excess of Receipts Over Disbursements $ 6,835.08 GENERAL CASH FUND Checking Account $18,978.39 Savings Account $ 2,464.86 Balance in Old Kent Bank and Trust Co., Grand Rapids, Michigan, December 31, 1976 $21,443.25 JossELYN Van Tyne Memorial Library Fund Balance as of Last Report December 31, 1975 $ 804.27 RECEIPTS Sale of Duplicates and Gifts 1,041.70 DISBURSEMENTS Purchase of Books 1,437.76 Balance in Old Kent Bank and Trust Co., Grand Rapids, Michigan, December 31, 1976 408.21 Louis Agassiz Fuertes Research Fund Margaret Morse Nice Fund Edwards and W.O.S. Paper Funds Balance as of Last Report December 31, 1975 — - $ 313.00 RECEIPTS Contributions 2,634.00 DISBURSEMENTS Grant-In Aid To Charles R. Brown $100.00 To Richard 0. Bierragard 200.00 To Susan Hannon 200.00 To Stephen T. Emlen 200.00 To Arthur J. Wiseman 100.00 To Stephen Borecky - 100.00 Total $ 900.00 Balance in Old Kent Bank and Trust Co., Grand Rapids, Michigan, December 31, 1976 $ 2,047.00 Aaron Moore Bagg Student Membership Award Fund Balance as of Last Report December 31, 1975 $ 400.00 RECEIPTS Contributions $ 200.00 September 1977 • ANNUAL REPORT 509 DISBURSEMENTS Student Membership Grants $ 176.00 Balance in Old Kent Bank and Trust Co., Grand Rapids, Michigan, December 31, 1976 $ 424.00 Endowment Funds General Endowment Fund Balance in Endowment Savings Account, Old Kent Bank and Trust Co., Grand Rapids, Michigan as of Last Report, December 31, 1975 $ 6,990.00 RECEIPTS Life Membership Payments $ 2,550.00 Balance in Endowment Savings Account, Old Kent Bank and Trust Co., Grand Rapids, Michigan, December 31, 1976 $ 9,540.00 Investments Held as of December 31, 1976 United States Government Bonds $ 5,393.75 International Bank Bonds 9,810.00 Canadian Provincial Bonds 4,100.00 Corporate Bonds 19,975.00 Convertible Corporate Bonds 2,625.00 Convertible Preferred Stocks 14.245.00 Common Stocks 53,733.50 Uninvested Principal 1,885.70 Total Investments Total General Endowment Fund December 31, 1976 George Miksch Sutton Colorplate Fund Investments Held as of December 31, 1976 International Bank Bonds $ 1,090.00 Canadian Provincial Bonds 4,587.50 Corporate Bonds 10,000.00 Common Stocks 9,050.00 Total Investments $ 24,727.50 Total Combined Wilson Ornithological Society Endowment Funds December 31, 1976 $146,035.45 Ernest E. Hoover, Treasurer Chairman C. E. Braun summarized the report of the Conservation Committee. The full report will appear later in The Wilson Bulletin. Editor J. A. Jackson gave his report. $111,767.95 $121,307.95 REPORT OF THE EDITOR 1976 From 1 January through 31 December 1976, I received 183 new manuscripts; this figure does not include book reviews and news items. I received 78 new manuscripts during 1977 through 17 May. Rejection rate for manuscripts received during 1975 and 510 THE ILSO\ BULLETIN • Vol. 89, No. 3 processed to some definite outcome was 34.9%, Eor manuscripts received thus far in 1977, the rate has been 37.5%. Turn-around time for both notes and papers is presently about 15 months from date of acceptance to date of publication and is improving. Processing time ( time from receipt to acceptance of papers published in volume 88 — in- cluding time needed for revision by the author) averaged 87.8 days for notes and 147.4 days for major papers. Time from receipt to rejection of manuscripts averaged 47.6 days. Through the generosity of George M. Sutton, it was possible to include a color plate in each issue of volume 88, Volume 88 included 721 pages and was the largest volume published to date; this increased size was made possible by generous contributions from a number of Society members. Editing of volume 88 was facilitated by the assistance of nearly 200 referees, by the hard work of my editorial assistants, and by the patience, hard work, and expertise of my secretary, Lyda Eubanks. I gratefully acknowledge the efforts of all of these individuals. Jerome A. Jackson, Editor The composition of the following committees had been previously announced: Nominat- ing— 0. S. Pettingill, Chairman, H. L. Batts, P. S. Street; Resolutions — R. D. Burns, Chair- man, S. A. Gauthreaux. S. F. Spofford; Auditing — J. F. Ponshair, G. M. Wickstrom, President Berger appointed R. C. Banks, J. A. Mosher, R. W. Schreiber, and R. C. Whit- more as a committee to judge student papers for the Alexander Wilson Prize. A list of new members was posted for inspection by the membership. REPORT OF THE MEMBERSHIP COMMITTEE — 1976-77 A major task for the Membership Office this past year was the preparation of the new membership list. Our experience with the 1974 list had led us to believe that we had a system that could generate the product with a minimum of time and effort. Such was not the case! Despite frecjuent correspondence with the Treasurer and diligent efforts to keep up-to-date files, we found many errors — and our membership found more following publication. Although we continue to believe that the membership list is a proper function of this office rather than of the Treasurer, it is clear that a better form of filing must he devised. Since the last annual meeting, we have added 240 new members. Against these we count losses of 11 through death and 209 by suspension, leaving a net gain of 20. The number of suspensions is unusually large and includes professional and foreign members of long standing. We hope their loss will prove temporary. Fifty-one members used the space provided on the dues notice to suggest one or more possible new members. We do not have complete figures, hut at least 17 of these sug- gestions led to successful recruitments. As a side interest this year, we kept track of membership migration. Our membership proved to he highly mobile. Two-hundred-fifty-five of them moved; 19 moved twice. Because each address change involves a certain amount of paper work and expense, we would like to encourage those members who know that they may move about frequently but have a reasonably stable base of operations, e.g. an academic department, to use that locus as their address for the Society. Abbot S. Gaunt, Chairman September 1977 • ANNUAL REPORT 511 REPORT OF THE STUDENT MEMBERSHIP COMMITTEE — 1976 Letters requesting membership nominations of college students interested in ornithology and inviting application from exceptional students to be considered for Aaron M. Bagg Student Membership Awards were sent to all members affiliated with educational insti- tutions in the Wilson Ornithological Society and American Ornithologists’ Union (duplications eliminated). This totaled approximately 850 letters mailed. The Cooper Ornithological Society was not included because the new membership list was issued after the mailing deadline. This activity resulted in 147 student nominations including 61 nominated for membership awards of which 34 were selected as award recipients. The award recipients will be announced in The Wilson Bulletin. Also, letters requesting nominations of promising pre-college students were mailed to the natural history and conservation organizations and institutions in 9 northeastern states. This resulted in 36 pre-college nominations. In keeping with the practice of the past, this procedure will progress from region to region on a yearly basis. In total, combining college and pre-college students, the committee processed 183 student nominations. Evaluation of the number of these that actually joined was not pursued. Douglas James, Chairman REPORT OF THE CONSERVATION COMMITTEE 1976 Activities in 1976-77 centered around review of management practices on National Wildlife Refuges in the United States and preparation of a report dealing with the problems involved. The committee’s report on refuge management practices will be pub- lished in a future issue of The Wilson Bulletin. Considerable demand for copies of previous Conservation Committee reports on eagles. Sandhill Cranes and sagebrush avifauna continued in 1976-77 with over 50 requests be- ing processed. The Conservation Committee contributed to the Environmental Impact Statement analyzing livestock grazing in Wyoming, provided reference sources and sug- gestions concerning effects on birds of spraying with insecticides for control of spruce budworm in New Brunswick, and contributed to a request for data on the status of Sandhill Cranes in British Columbia. By far, most correspondence received in 1976 77 concerned refuge management practices. Clait E. Braun, Chairman Keith W. Harmon Jerome A. Jackson Carroll D. Littlefield REPORT OF THE LIBRARY COMMITTEE 1976 During calendar 1976, the Josselyn Van Tyne Memorial Library continued to grow and prosper. Janet Hinshaw (very capably relieved for a time by Elizabeth Strauch) continued to make good progress with the ongoing problems of sorting, filing, shelving, and the correction of remaining discrepancies in lists and records, along with the day-to- day affairs of mailing, receiving, and correspondence, and handling of back issues of The Wilson Bulletin. The New Book Fund was put to good use in some 44 judicious purchases of books and journals, while being considerably bolstered by the sale of duplicates. Sixty-eight 512 THE ILSON BULLETIN • Vol 89, No. 3 loans were made to 50 members: in all, 204 books, journals, reprints, translations, and photocopies. Through 110 exchanges for the Bulletin, we received 135 journals, news letters, and reprints. Complimentary subscriptions and gifts raised the total of periodicals received to about 160. Donations are once more most gratefully acknowledged: 1068 items in all, from 32 members and organizations. These included 682 reprints, 268 journals, 84 books, 1 thesis, 5 translations, 18 reports, abstracts, and pamphlets, from the following: W. H. Behle, A. J. Berger (over half the items, most being reprints), G. R. Brody estate (50 items), California Dept. Fish and Game (by R. M. Jurek), C. T. Collins, J. Cooper, Delaware Museum of Natural History, P. B. Hamel, F. Haverschmidt, E. Hoover, D. W. Johnston, L. Kelson, C. Kendeigh, LGL, Ltd. (by W. J. Richardson), Linnaean Society of N.Y. (by R. W. Dickerman), H. G. Lumsden, H. F. Mayfield, M. E. Morse estate (42 items), R. B. Payne, A. R. Phillips, R. A. Romanes, W. Southern, P. Stettenheim, Mrs. W. C. Stone, J. G. Strauch, Tamarack Press, Welder Wildlife Foundation, L. Wolf, Col. L. R. Wolfe, M. Wood, Yale University Press, R. L. Zusi. Your support is most gratifying. May it continue. Contribute as you are able; purchase duplicate items offered for sale; most of all, make use of our fine Library collections for your own profit and pleasure. William A. Lunk, Chairman SECOND BUSINESS MEETING President A. J. Berger called the Second Business Session to order at 15:00 on 21 May 1977. The following report of the Auditing Committee was read and accepted. auditor’s report We have examined the treasurer’s records, bank statements, cancelled checks, account books, and other financial records of the Society covering transactions occurring during the past fiscal year. The financial status of the Society is substantially as set forth in the Treasurer’s Report dated 31 December 1976. Our examination ascertained that all income has been applied to the proper funds and no expenditures have l)een made except as authorized. James E. Ponshair, Member George M. Wickstrom, Member The list of new members previously posted was voted on and the persons listed were formally accepted as members of the Society. Chairman O. S. Pettingill presented the following slate of officers as proposed by the Nominating Committee; President, Douglas A. James; First Vice-President, George A. Hall; Second Vice-President, Abbot S. Gaunt; Secretary, James Tate, Jr.; Treasurer, Ernest E. Hoover; Elective Member of the Council, term to expire in 1978, Sidney A. Gauthreaux. Jr.; Elective Member of the Council, term to expire in 1980, Clait E. Braun. This report was accepted, and there being no other nominations from the floor, the Secretary was instructed to cast a unanimous ballot for the slate. Chairman E. D. Burns and member S. A. Gauthreaux of the Resolutions Committee presented the following resolutions which were adopted by the membership. September 1977 • ANNUAL REPORT 513 WHEREAS, the United States Government is currently funding a number of water projects of doubtful economic value, of questionable safety, and destructive to natural environments, and, WHEREAS, many of these water projects currently under construction would elimi- nate or severely damage much essential habitat for certain low density or localized bird populations, particularly in the Southwest, and, WHEREAS, the economic and environmental values of these projects are currently being carefully, reexamined by the President of the United States, THEREFORE, BE IT RESOLVED that the Wilson Ornithological Society opposes the construction of all water projects that are detrimental to the environment and of dubious value to human welfare, and urges that full and careful consideration be given to environmental concerns before approval of future water projects, and BE IT FURTHER RESOLVED that the Society commends President Carter for pro- posing the review and possible cancellation of unsound water projects. WHEREAS, certain agencies within the Federal Government have become increas- ingly interested in research on and management of non-game species of wildlife, par- ticularly birds, and WHEREAS, there is pending legislation that wmuld increase funds for work on non-game wildlife, by Federal and State Agencies, THEREFORE, BE IT RESOLVED that the Wilson Ornithological Society commends the Fish and Wildlife Service, the Forest Service, and the Bureau of Land Management for their increased interest in non-game species of wildlife and, BE IT FURTHER RESOLVED that the Society urges passage of the pending legis- lation that would further increase support for work on non-game species of wildlife by Federal and State Agencies. WHEREAS, the recent general interest and concern for the environment has frequently resulted in a number of published accounts of the locations of breeding and roosting areas of rare and endangered species in journals, newspapers, environmental impact statements, and other places, and WHEREAS, this has too often resulted in increasing visitations by humans to certain locations and, thereby, further threatens the future survival of certain rare and en- dangered species, THEREFORE BE IT RESOLVED that the Wilson Ornithological Society strongly dis- courages publication of detailed descriptions and or photographs of breeding and roosting locations of rare and endangered species, and that authors, editors and compilers of such information make every attempt to protect these areas from undue human interference, AND BE IT FURTHER RESOLVED that each and everyone of us be reminded that we are stewards of our natural world, and whether it be visiting, studying, photographing, or providing directions to the locations of rare and endangered species, that we be ever mindful of the consequences of our actions. WHEREAS, the Fifty-eighth Annual Meeting of the Wilson Ornithological Society convened on the campus of Mississippi State University has been a smashing success, and WHEREAS, the scientific program has been of high quality, the field trips were well organized and thoughtfully conducted, and in addition, the culinary talents of the Local Committee on Arrangements plus their gracious southern hospitality was out-performed only by the cooperativeness of the local Red-cockaded Woodpeckers, BE IT THEREFORE RESOLVED that the Wilson Ornithological Society gratefully acknowledges and thanks the Local Committee on Arrangements and its Chairman, Jerry 514 THE WILSON BULLETIN • VoL 89, No. 3 Jackson, for their many efforts and long hours of toil which have made this meeting a memorable success. The meeting adjourned at 15:25. At the annual banquet the following awards and prizes were announced: Louis Agassiz Fuertes Awards Ernest E. Stevens, “The significance of the geographic variation in the rictal flange color of the parasitic Brown-headed Cowbird iMoIothrus ater)’^ Dale Lewis, “Environmental influence on the population structure and social behavior of Plocepasser mahali” Margaret Morse Nice Award David P. Hendricks, “Breeding range and distribution of the Brown-capped Rosy Finch” Edwards Prize Glen A. Fox, “Eggshell quality: its ecological and physiological significance in a DDE-contaminated Common Tern Population” Edwards Prize, Second Place Douglas W. Mock, “Pair-formation displays of the Great Blue Heron” Alexander W ilsoii Prize David R. Maurer, “The appendicular myology and phylogenetic affinities of the alcediniform Coraciiformes, the trogons and sub-oscines” At this time it is appropriate to list the winners of the Alexander Wilson Prizes for 1975 and 1976 which have not been previously published in the Proceedings. 1975 Douglas W. Mock, “Vocabulary shifts during pair-formation in Great Blue Herons” 1976 Stephen R. Borecky, “The appendicular myology and phylogenetic affinities of tlie birds of paradise ( Paradisaeidae) and the howerhirds (Ptilonorhynchidae) ” PAPERS SESSION Travis McDaniel, Noxubee National Wildlife Refuge, Mississippi, An introduction to Noxubee National Wildlife Refuge. Ricliard Bradley, Florida State Museum, Gainesville, Florida, Hybridization in Calypte hummingbirds. Richard C. Banks, U.S. Fish and Wildlife Service, Washington, D.C., Nomenclature of the Black-bellied Whistling (Tree) Duck. j. W. Hardy, University of Florida, Reproductive behavioral ecology of the Southern San Bias Jay Cyanocorax s. sanhlasiana. Glen E. Woolfenden, University of South Florida, Tampa, Growth and survival of young Florida Scrub Jays. G. Thomas Bancroft, University of South Florida, Tampa, Molt and breeding in Florida Scrub Jays. D. Bruce Barbour. University of South Florida, Tampa, Territorial vocalizations of the Florida Scrub Jay. Lester L. Short, American Museum of Natural History, New^ York, Introductory Remarks. Walter J. Bock. Columbia University, New^ York, Morphology of the feeding apparatus of woodpeckers. H, Winkler, Austrian Academy of Sciences. Vienna, Vocal and other behavior in the Strickland's W oodpecker. September 1977 • ANNUAL REPORT 515 Ernest E. Hoover, 1044 Wel)ster St. NW, Grand Rapids, Michigan, Iris color changes in the Hairy Woodpecker. Lawrence Kilham, Dartmouth Medical School, Llanover, New Hampshire, Nesting behavior of Yellow-bellied Sapsucker. Luis F. Baptista, Occidental College, Los Angeles, California, Revision of the Mexican Piculus complex. Jerome A, Jackson, Mississippi State University, Mississippi State, Home range, inter- specific competition, and management for the Red-cockaded Woodpecker. Lester L. Short, American Museum of Natural History, New York, Burdens of the picid hole-nesting habit. Alexander Cruz, University of Colorado, Boulder, Colorado, Ecology of the Jamaican Woodpecker ( Melanerpes radiolatus). Randall Breitwisch, University of Miami, Coral Gables, Florida, The ecology and behavior of the Red-bellied W oodpecker, Melanerpes carolinus, in South Florida. Joseph B. Williams, Pepperdine University, Malibu, California, Competition among bark- foraging birds in central Illinois: Experimental evidence. W, Wilson Baker, Tall Timbers Research Station, Tallahassee, Florida, Roosting behavior of Red-cockaded Woodpeckers in north Florida. Robert G. Hooper, Michael R. Lennartz, and Richard F. Harlow, Southeastern Forest Experiment Station, Clemson, South Carolina, Territory and home range sizes of the Red-cockaded Woodpecker. Virginia Kirby, University of Arizona, Tucson, Adaptive modifications in the ribs of woodpeckers ( Picidae ) . Richard N. Conner, Virginia Polytechnic Institute and State University, Blacksburg, Bill and body size differences in woodpeckers: An alternate view. Curtis S. Adkisson, Virginia Polytechnic Institute and State University, Blacksburg, A comparison of the vocal behavior of Red and White-winged crossbills. Irvine D. Prather, Virginia Polytechnic Institute and State University, Blacksburg, Behavioral relationships between Black Vultures and Turkey Vultures. John N. Mugaas, Southwestern at Memphis, Memphis, Tennessee, Microclimatic analysis of Black-billed Magpie habitat using equivalent blackbody temperatures: Implications for thermal budgeting and behavioral thermoregulation. Larry J. Miller, Louisiana State University, Baton Rouge, The effects of altered photo- period upon the migratory orientation in the White-throated Sparrow, Zonotrichia alhicollis. Bette J. Schardien, Mississippi State University, Mississippi State, A comparative study of nestling development in Mockingbirds and Brown Thrashers. David R. Maurer, University of Pittsburgh, Pittsburgh, Pennsylvania, The appendicular myology and phylogenetic affinities of the alcediniform Coraciiformes, the trogons and the sub-oscines. Ronald D. Drohney, University of Missouri-Columbia, Puxico, Missouri, Feeding ecology of Wood Ducks in Missouri. Alexander Cruz, University of Colorado, Boulder, Colorado, Adaptive evolution in the Jamaican Blackbird, Nesopsar nigerrimus. Elliot J. Tramer and Thomas R. Kemp, University of Toledo, Toledo, Ohio, Foraging ecology of migrant warblers and vireos in the highlands of Costa Rica. James R. Karr, University of Illinois, Champaign, Intercontinental variation in the evolution of tropical rainforest avifaunas. Noel 0. Warner, Florida State University and Tall Timbers Research Station, Tallahassee, 516 THE WILSON BULLETIN • VoL 89, No. 3 Avian diversity and habitat in Florida: An analysis of a peninsular diversity gradient. Robert C. Whitmore and E. James Harner, West Virginia University, Morgantown, Anal- ysis of multivariately determined community matrices using cluster analysis and multi- dimensional scaling. Chandler S. Robbins and Danny Bystrak, U.S. Fish and Wildlife Service, Laurel, Mary- land, Recent changes in bird populations revealed by breeding bird survey. Joseph M. Meyers, University of Georgia, Athens, Effect of selected transmissionline rights-of-way treatments on forest bird communities. T. Scott Taylor, University of Missouri-Columbia, Puxico, Missouri, Avian use of moist soil impoundments. James G. Dickson and Charles A. Segelquist, Southern Forest Experiment Station, Nacogdoches, Texas, Breeding bird populations in pine and pine-hardwood forest stands in east Texas. ATTENDANCE ALABAMA: Birmingham, Fred Carney, Walter F, Coxe, Joseph A. Imhof, Thomas A. Imhof, Elberta G. Reid, Robert R, Reid, Jr.; Jacksonville, Bill Summerour; Tuscaloosa, Richard K. Crawford; University, David T. Rogers, Jr. ARIZONA: Tucson, Virginia Kirby. ARKANSAS: Fayetteville, Douglas James. CALIFORNIA: Los Angeles, Luis F. Baptista, Ralph W. Schreiber; Malibu, Joseph B. Williams. COLORADO: Boulder, Alexander Cruz; Fort Collins, Clait E. Braun. FLORIDA: Coral Gables, Randall Breitwisch; Gainesville, Erik J. Bitterbaum, Richard Bradley, John W. Hardy, Barbara Kimball, Sarah Sloane, Thomas A. Webber; Lake Placid, Fred E. Lohrer; Orlando, Richard V. Demmer; Tallahassee, Wilson W. Baker, Noel E. Warner; Tampa, G. Thomas Bancroft, Douglas B. Barbour, Jim Rodgers, Glen B. Woolfenden; Winter Haven, Peggy MacQueen. GEORGIA: Athens, Joseph M. Meyers, Atlanta, Franklin McCamey; Augusta, Emil K. Urban. HAWAII: Honolulu, Andrew J. Berger. ILLINOIS: Champaign, James R. Karr; Momence, Mr. and Mrs. William T. Lory; Rock- ford, John T. Bergstrom. INDIANA: Richmond, Timothy Brush, Alan Simon. KANSAS: Hays, Charles A. Ely, Pat Lattas, Ren Lohoefener. KENTUCKY: Ethel Woolfenden, Lester B. Woolfenden. LOUISIANA: Baton Rouge, J. W. Eley, Dick Ferrell, Robert B. Hamilton, Robert S. Kennedy, Dwight J. LeBlanc, Dr. and Mrs. George H. Lowery, Jr., Christina Lusk, Larry Miller, Mr. & Mrs. Bob Newman, Robert E. Noble, John P. O’Neill, John S. Sylvest, Dan Tallman, Erika Tallman; Eunice, Harland D. Guillory; Franklin, Jack H. Deshotels; Monroe, David Kee; Shreveport, Horace H. Jeter. MAINE: Wayne, Olin S. Pettingill, Jr. MARYLAND: Boyds, Mary Schaefer; Cumberland, Bruce Lawson; Frostburg, James Mosher, Susan Mosher; Laurel, Chandler S. Robbins, Eleanor C. Robbins; Lonaconing, William J. Devlin. MASSACHUSETTS: Manomet, Kathleen S. Anderson. MICHIGAN: Ann Arbor, Janet Hinshaw, Stephen Hinshaw; Grand Rapids, Ernest Hoover; Jackson, Robert A. Whiting; Pleasant Lake, Hubert P. Zernickow, Norene E. Zernickow. September 1977 • ANNUAL REPORT 517 MINNESOTA: Duluth, P. B. Hofslund. MISSISSIPPI: Cleveland, Mr. and Mrs. J. S. White; Clinton, DeAnne Smith, Marita Smith; Columbus, Thelma Barnes, William S. Parker; Greenville, Ed Alexander, Ginger Alexander; Gulfport, Jay Toups, Judith Toups; Jackson, Stephen W. Peterson, Annie C. Turcotte, W. H. Turcotte; Kosciusko, Rebecca W. Davis, Walter V. Davis, Ray E. Weeks; Mississippi State, Martha B. Hays, Jerome A. Jackson, Nancy Jackson, Wilma Mitchell, Patricia Ramey, Bette Schardien, Wayne Weber, Wendy Weber, Pat Shindala, Glenn Liming, Mike Christensen, Ardis Christensen, David Werschkul, Sue Werschkul, Oskar Zernickow, Keith Parsons, Joe Ferguson, Frances Fortenberry, Martha Ward, Tom Morrow, Glenn Clemmer, Sherry Clemmer, Ann McWhorter, Bob Esher, Kathy Esher, Courtney Hackney, Lois Kilgore, Lawrence Croft, Elsie Croft, Frances Windham, Arlie Wilson, Ethel Wilson; Monticello, Carl Bauer, Florence Bauer; Shaw, Nona Herbert; Starkville, E. W. Permenter, Bonnie Turner, Bill Hughes, Nellie Hughes, Florence Dunn, Burton S. Webster, Marjorie Webster, Caroline Bennett, Mrs. Gifford Bull, Julia Broyles, Travis McDaniel, Willena Ratliff; Vicksburg, Louis P. Cashman, Jr. MISSOURI: Cape Girardeau, William R. Eddleman, Marie Heye, Paul L. Heye; Columbia, Jeffrey Brawn, Larry Vangilder; Puxico, Ronald Drobney, Judy Sherpelz, T. Scott Taylor. NEW HAMPSHIRE: Lyme, Jane Kilham, Lawrence Kilham. NEW JERSEY: East Millstone, John Jubon, Mary Jubon; Mt. Holly, Katherine Price. NEW YORK: Mamaronick, Robert Arbib; New York, Susan R. Drennan, Lester L. Short. NORTH CAROLINA: Chapel Hill, Helmut C. Mueller. OHIO: Columbus, Abbot S. Gaunt, Sandra Gaunt; Gambler, Robert D. Burns; Lake- wood, Nancy R. Klamm, William A. Klamm; Waterville, Harold Mayfield, Virginia Mayfield. PENNSYLVANIA: Pittsburgh, David R. Maurer; W ashington, Judi Ickes, Roy Ickes. SOUTH CAROLINA: Clemson, Sidney A. Gauthreaux, Jr., Michael R. Lennartz; Seneca, Robert G. Hooper. TENNESSEE: Martin, David Pitts, Marion Pitts; Maryville, Ralph J. Zaenglein; Memphis, Diane Bean, Ben B. Coffey, Jr., Lula C. Coffey, Helen Dinkelspiel, Henry Dinkelspiel, John Mugaas, Lynn Mugaas, Martha Waldron; Nashville, Morris D. Williams. TEXAS: Baytown, Mary K. Jones; Nacogdoches, James G. Dickson, Debbie A. Ellis, Charles D. Fisher, James Kroll; Sherman, Charles R. Brown, Sam D. Wolfe, HI. VIRGINIA: Blacksburg, Curtis S. Adkisson, Richard N. Conner, Irvine D. Prather. WASHINGTON, D.C. : Richard Banks, Ralph M. Browning. WEST VIRGINIA: Charleston, Anne Shreve, Harvey Shreve, Jr.; Morgantown, George A. Hall, Robert C. Whitmore. WISCONSIN: Milwaukee, Charles M. Weise. FOREIGN COUNTRIES: Austria, Hans WYnkler; Kenya, Jenny Horne. September 1977 ■ ANNUAL REPORT 519 « cd DS 3 o 0) ^ ^ O ^ fO rj , i-H csi ^ 0) (-J (H r/^ LO ^ o ■' CN sJ' •- _a . [jh - 'S . . c ^ tJ- CM ^ 'S C" . M S S 2 g ^ ^ & s“« y . ^- s 5 j -; c * fe ..• cfi CM . E C/) E JS s “ '2 ^ . ’^ E ^ ^ CQ 55 (1^ a; . S I • I-H i _C Q On • ^ ^ "o lO 2 j § S “ I J s ^ ° cd i "S ^ ° ^ ^ « OJ o c/T bn 1> bC . o P^ ^ ‘^^CC " a a 2 < t c\ c. s o 2 |> c" 3 tp- C CO C^ ^ ^ ^ ~ O C ^•' C = . 1 ^ ^ ^ i ^ C "2 E — J ^ ^ • 2 E ;r; ^ 00 a; ^ M pQ U ^ cd 5 2 , ^ *2; K^- r- < . ^_; S ^ . • 3 1^ . LO " >R2 2 LO . ^ o ^ E , CE ;;; -E CO jy CJ^ E.ES . '— I aj . !U « I. B. Barbour, 113. G. 520 THE WILSON BULLETIN • Vol. 89, No. 3 SUGGESTIONS TO AUTHORS Manuscripts. — Manuscripts intended for publication in The Wilson Bulletin should be neatly typewritten, double-spaced i especially tables and ‘literature cited”), with at least 3 cm margins all around, and on one side of good quality paper. Do not use eraseable typing paper. All pages should be numbered. Two copies should be submitted. Xero- graphic copies are acceptable if they are clearly readable and on good quality paper. Copies on heav>*, slick paper, as used in some copy machines, are not acceptable. Tables. — Tables are expensive to print and authors should consider carefully whether or not a table is really necessary or adds to the paper. Tables should be designed so as to be narrow and deep rather than wide and shallow. Double space all entries in tables, including titles. Do not use vertical rules. Tables should be typed on separate sheets and placed at the end of the MS. Figures. — All illustrations should be prepared (particularly insofar as the lettering goes) so as to be readable when reduced in size. The final size will usually be 11.4 cm wide. Illustrations larger than 22 X 28 cm will not be accepted, and these should be reduced photographically before submitting. Legends for all figures should be typed on a separate sheet. Photographs should be clear, of good contrast, and on glossy paper. Draw- ings should be in India ink on good drawing board, drafting paper, or blue-lined graph paper. All lettering on drawings should be done with a lettering instrument or the equiva- lent. Designate the top of each illustration and identify (on the back in soft pencil) with author’s name, and figure number. Submit 2 duplicates or readable xerographic copies of each figure so that originals don’t have to be sent to the reviewers. Style and Format. — For general matters of style in preparing a scientific article, authors should consult the “CBE Style Manual,” 3rd ed.. Am. Inst. Biol. Sci., W'ashington, D.C. 1972. All MSS should be submitted in the general format used in recent issues of the Bulletin. Avoid footnotes, and avoid more than 2 levels of subject subheadings. Except in rare circumstances lead papers should be followed by a summary, not to exceed 10% of the length of the paper. Summaries should be informative when standing by themselves. Most units should be given in the metric system, and compound units should be given in one-line form ( i.e. cm-sec'") . The continental system of dating ( 21 March 1972) and the 24 hour clock ( 09:00 and 22:00) should be used. References. — In long MSS, if more than 5 papers are cited, these should be included in a terminal “Literature Cited” section. Include only references actually cited, and include only material available in the open literature (“In-house” technical reports and the like should not be cited). The style of citation can be obtained from recent issues of the Bulletin. For abbreviations of periodical names use the list given in the most recent issue of “BIOSIS,” Bioscience Information Service, Philadelphia, Pa. If in doubt, do not abbre- viate serial names. All references in “General Notes” and in long papers containing fewer than 5 references should be cited internally e.g. (James, Wilson Bull, 83:215-236, 1971) or James (Wilson Bull. 83:215-236, 1971). Nomenclature. — Common names and technical names of birds should be those given in the 1957 A.O.U. Check-list (and such supplements as may appear) unless justification is given for departing from this list. For bird species in Middle and South America the Bulletin uses the common names appearing in Eisenmann, “Species of Middle American Birds,” 1955 and Meyer de Schauensee “The Species of Birds of South America,” 1966. Common names of birds should be capitalized. This issue of The Wilson Bulletin was published on 30 September 1977. The Wilson Bulletin Editor Jerome A. Jackson Department of Zoology P.O. Drawer Z Mississippi State University Mississippi State, MS 39762 Editorial Assistants Bette J. Schardien Patricia Ramey Bonnie J. Turner Martha Hays Wayne C. Weber Review Editor Robert Raikow Color Plate Editor William A. Lunk Department of Life Sciences 865 North Wagner Road University of Pittsburgh Ann Arbor, MI 48103 Pittsburgh, PA 15213 Suggestions to Authors See Wilson Bulletin, 87:144, 1975 for more detailed “Suggestions to Authors.” Manuscripts intended for publication in The Wilson Bulletin should be submitted in dupli- cate, neatly typewritten, double-spaced, with at least 3 cm margins, and on one side only of good quality white paper. Do not submit xerographic copies that are made on slick, heavy paper. Tables should be typed on separate sheets, and should be narrow and deep rather than wide and shallow. Follow the AOU Check-list (Fifth Edition, 1957) and the 32nd Supplement (Auk, 90:411-419, 1973), insofar as scientific names of U.S. and Canadian birds are concerned. Summaries of major papers should be brief but quotable. Where fewer than 5 papers are cited, the citations may be included in the text. All citations in “General Notes” should be included in the text. Follow carefully the style used in this issue in listing the literature cited ; otherwise, follow the “CBE Style Manual” (1972, AIBS). Photographs for illustrations should have good contrast and be on gloss paper. Submit prints unmounted and attach to each a brief but adequate legend. Do not write heavily on the backs of photographs. Diagrams and line drawings should be in black ink and their lettering large enough to permit reduction. Original figures or photographs submitted must be smaller than 22 X 28 cm. Alterations in copy after the type has been set must be charged to the author. Notice of Change of Address If your address changes, notify the Society immediately. Send your complete new address to the Treasurer, Ernest E. Hoover, 1044 Webster St,, N.W., Grand Rapids, Michigan 49504. He will notify the printer. The permanent mailing address of the Wilson Ornithological Society is: c/o The Museum of Zoology, The University of Michigan, Ann Arbor, Michigan 48104. Persons having business with any of the officers may address them at their various addresses given on the back of the front cover, and all matters pertaining to the Bulletin should be sent directly to the Editor. CONTENTS W' NESTING HABITAT OF BACHMAN’s WARBLER — A REVIEW Robert G. Hooper and Paul B. Hamel 373 , MORPHOLOGICAL VARIATION IN NORTH AMERICAN PINE GROSBEAKS Curtis S. AdkisSOn 380 | BREEDING SUCCESS AND NEST SITE CHARACTERISTICS OF THE RED-WINGED BLACKBIRD j Donald F. Caccamise 3% J FORAGING BEHAVIOR OF THE EASTETN BLUEBIRD Benedict C. Pinkowski 404 COMPARATIVE FEEDING BEHAVIOR OF IMMATURE AND ADULT HERRING GULLS ^ Nicolaas A. M. Verbeek 415 ' j COMPARATIVE MORTALITY OF BIRDS AT TELEVISION TOWERS IN CENTRAL ILLINOIS I James W. Seets and H. David Bohlen 422 I EFFECT OF FLOCK SIZE ON FORAGING ACTIVITY IN WINTERING SANDERLINGS * James Silliman, G. Scott Mills, and Stephen Alden 434 ! ACTIVITY PATTERNS OF FEMALE RUFFED GROUSE DURING THE BREEDING SEASON . Stephen J. Maxson 439* 0 WEIGHTS AND FAT CONDITION OF SOME MIGRANT WARBLERS IN JAMAICA A. W. Diamond, P. Lack, and R. W. Smith GENERAL NOTES WING MOLT OF THE KiTTLiTz’s MURRELET Spencer G. Sealj INCIDENCE OF RUNT EGGS IN THE CANADA GOOSE AND SEMIPALMATED SANDPIPER T. H. Manning and Brenda Carter LATE FLEDGING DATE FOR HARRIS’ HAWK Eleanor L. Radke and John Klimosewski THE SPATIAL DISTRIBUTION OF WINTERING BLACK-BELLIED PLOVERS Christopher H. Stinson PREDATION AND DISPERSION OF HERRING GULL NESTS Gary W. Shugart and William C. Scharf EGG QUALITY IN RELATION TO NEST LOCATION IN RING-BILLED GULLS John P. Ryder, Donald E. Orr, and Ghomi H. Saedi ROOF-NESTING BY COMMON TERNS Anne E. MacFarlane RAPID CHICK SEPARATION IN WHIP-POOR-WILLS Erie L. Dyer AN INTRASPECIFIC MORTAL ATTACK Vera Lee Grubbs RUFOUS-SIDED TOWHEES MIMICKING CAROLINA WREN AND FIELD SPARROW Donald J. Borror HEAT LOSS FROM THE NEST OF THE HAWAIIAN HONEYCREEPER, “AMAKIHI” G. C. Whittow and A. J. Berger SPREAD OF THE GREAT-TAILED GRACKLES IN SOUTHWESTERN LOUISIANA H. Douglas Pratt, Brent Ortego, and Harland D. Guillory POPLAR LEAF-STEM GALL INSECTS AS FOOD FOR WARBLERS AND WOODPECKERS Sally Hoyt Spofford PERCH HEIGHT PREFERENCES OF GRASSLAND BIRDS Keith G. Hanison '■“I 467 1 '*1 469 I 469 ' 470 ,1 472 473 I 475 .j 476 (j 477 477 I 430 I 483 I 485 ; SPECIAL REVIEW — JOHN OSTROM’S STUDIES ON ARCHAEOPTERYX, THE ORIGIN OF BIRDS, AND THE EVOLUTION OF AVIAN FLIGHT Joel Cracraft 488 ORNITHOLOGICAL LITERATURE 492 ORNITHOLOGICAL NEWS 503 THE president’s PAGE 505 PROCEEDINGS OF THE FIFTY-EIGHTH ANNUAL MEETING 506 520 SUGGESTIONS TO AUTHORS u The WIson Bulletin PUBLISHED BY THE WILSON ORNITHOLOGICAL SOCIETY VOL. 89, NO. 4 DECEMBER 1977 PAGES 521-678 MUS. COMP. ZOOU. LIBRARY The Wilson Ornithological Society Founded December 3, 1888 Named after ALEXANDER WILSON, the first American Ornithologist. President — Douglas A. James, Department of Zoology, University of Arkansas, Fayetteville, Arkansas 72703. First Vice-President — George A. Hall, Department of Chemistry, West Virginia Univer- sity, Morgantown, W. Va. 26506. Second Vice-President — Abbot S. Gaunt, Department of Zoology, Ohio State University, Columbus, Ohio 43210. Editor — Jerome A. Jackson, Department of Zoology, P.O. Drawer Z, Mississippi State Uni- versity, Mississippi State, Mississippi 39762. Secretary — ^James Tate, Jr., P.O. Box 2043, Denver, Colorado 80201. Treasurer — Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. Elected Council Members — Sidney A. Gauthreaux, Jr. (term expires 1978) ; James R. Karr (term expires 1979) ; Clait E. Braun (term expires 1980). Membership dues per calendar year are: Active, $10.00; Sustaining, $15.00; Life memberships, $200 (payable in four installments). The Wilson Bulletin is sent to all members not in arrears for dues. The Josselyn Van Tyne Memorial Library The Josselyn Van Tyne Memorial Library of the Wilson Ornithological Society, housed in the University of Michigan Museum of Zoology, was established in concurrence with the University of Michigan in 1930. Until 1947 the Library was maintained entirely by gifts and bequests of books, reprints, and ornithological magazines from members and friends of the Society. Now two members have generously established a fund for the purchase of new books; members and friends are invited to maintain the fund by regular contribution, thus making available to all Society members the more important new books on ornithology and related subjects. The fund will be administered by the Library Committee, which will be happy to receive suggestions on the choice of new books to be added to the Library. William A. Lunk, University Museums, University of Michi- gan, is Chairman of the Committee. The Library currently receives 104 periodicals as gifts and in exchange for The Wilson Bulletin. With the usual exception of rare books, any item in the Library may be borrowed by members of the Society and will be sent prepaid (by the University of Michigan) to any address in the United States, its possessions, or Canada. Return postage is paid by the borrower. Inquiries and requests by borrowers, as well as gifts of books, pamphlets, reprints, and magazines, should be addressed to “The Josselyn Van Tyne Memorial Library, University of Michigan Museum of Zoology, .Ann Arbor, Michigan.” Contributions to the New Book Fund should be sent to the Treasurer (small sums in stamps are acceptable). A complete index of the Library’s holdings was printed in the September 1952 issue of The Wilson Bulletin and newly acquired books are listed periodically. The Wilson Bulletin The official organ of the Wilson Ornithological Society, published quarterly, in March, June, September, and December. The subscription price, both in the United States and elsewhere, is $15.00 per year. Single copies, $4.00. Subscriptions, changes of address and claims for undelivered copies should be sent to the Treasurer. Most back issues of the Bulletin are available and may be ordered from the Treasurer. Special prices will be quoted for quantity orders. All articles and communications for publications, books and publications for reviews should be addressed to the Editor. Exchanges should be addressed to The Josselyn Van Tyne Memorial Library, Museum of Zoology, Ann Arbor, Michigan. Known office of publication : Department of Zoology, Mississippi State University, Mississippi State, Mississippi 37962. Second class postage paid at Mississippi State, Mississippi and at additional mailing office. Allen Press, Inc., Lawrence, Kansas 66044 Lesser Prairie Chicken (Tympanuchus pallidicinctus), photographed 26 Aprii 1975 in Roosevelt County, New Mexico by Keith Giezentanner. THE WILSON BULLETIN A QUARTERLY MAGAZINE OF ORNITHOLOGY Published by the Wilson Ornithological Society VoL. 89. No. 4 December 1977 Pages 521-678 THE LESSER PRAIRIE CHICKEN’S INELATABLE NECK SACS George Miksch Sutton In the fall of 1932, in the vicinity of Arnett, Ellis County, western Okla- homa, I first saw and handled the Lesser Prairie Chicken {Tympanuchus pallidicinctus) . The species was common at that time in the “shinnery oak” country thereabouts, especially on the Davison Ranch a few kilometers south- east of the city (Sutton, Ann. Carnegie Mus., 24:11-12, 1934). A detailed watercolor sketch that I made of the head and foot of an adult male bird shot that year on 6 October shows the neck sac to be light cinnamon-buff, a color that contrasted sharply with the bright, slightly ochraceous orange- yellow of the comb above the eye. Noting that Florence M. Bailey (Birds of New Mexico, New Mexico Dept, of Fish and Game, Santa Fe, 1928:207) described the neck sacs as “y^How in the breeding season,” I fell to believing that the sacs brightened to yellow in spring and summer and reverted to cinnamon-buff (or some such com- paratively dull shade) in fall and winter. When, in 1936, I spent about 6 weeks at Arnett (early May to mid-June), I observed Lesser Prairie Chickens almost daily, for in certain parts of the “shinnery country” thereabouts they were common. Repeatedly I made a point of driving to one or more “gobbling grounds,” where I observed the performing males at remarkably close range from my car. Convinced that the neck sacs were not at all yellow, I made a point of collecting 2 males early in the day on 25 May and drawing them in detail before the colors of the fleshy parts had had a chance to fade. In each of these drawings the neck sac is light, somewhat reddish brown, again in sharp contrast to the bright orange-yellow of the comb above the eye. I made a third detailed sketch of an adult male bird later that summer but did not record the date. To my surprise I found that authors continued to describe the color of the neck sacs inadequately. Friedmann {in Ridgway, U.S. Natl. Mus. Bull. 50, Pt. IX, 1941:220), who may have been following Bailey (supra), stated 521 522 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 that the “gular sacs” were “yellow in the breeding season.” Ligon, in “New Mexico Birds and Where to Find Them” (Univ. New Mexico Press, Albu- querque, 1961:89), called the sacs “orange-colored.” Peterson, in his “Field Guide to the Birds of Texas” (Houghton Mifflin Co., Boston, 1960:74), came much closer to accuracy when he described the sacs as “dull red rather than orange,” though I continue to feel that red is the wrong word. Robbins, Bruun, and Zim, in their “A Guide to Field Identification Birds of North America” (Golden Press, N.Y., 1966:86), described the sacs as “reddish,” but the illustration on the opposite page is hopelessly misleading since the color shown there approaches pale lilac or violet. I suspect that Arthur Singer, whose excellent drawings illustrate this work, was advised to make clear that the color of the neck sac was very different from that in the Greater Prairie Chicken i T. cupido) , and he may have heard or assumed that the proper color was close to that of the neck sac of the Sharp-tailed Grouse (Pedioecetes phasianellus) . In any event, now that I have been observing the Lesser Prairie Chicken off and on for 45 years, I am convinced that its neck sacs are never yellow or purple at any season; nor are they orange, the color-word that best describes the sacs of the Greater Prairie Chicken. I would call them tan, were not that word so widely used commercially for a variety of shades. The accompanying colorplate, which is based on photographs taken by my friend and former student, Keith Giezentanner, now Development Supervisor for the New Mexico Department of Game and Fish at Santa Fe, shows the color of the inflated air sacs admirably. The photographs were taken on the morning of 26 April 1975 at a well established booming ground 13 km east of Milnesand, Roose- velt County, southeastern New Mexico. STOVALL MUSEUM OF SCIENCE AND HISTORY, UNIV. OF OKLAHOMA, NORMAN 73019. ACCEPTED 28 JUNE 1977. NESTING HABITAT OF CANADA GEESE IN SOUTHEASTERN MICHIGAN Richard M. Kaminski and Harold H. Prince Habitat selection by birds is guided by instinctive and learned responses to stimuli from the physical environment, conspecifics, and other species within the environment (Hilden 1965). Whitmore (1975) reviewed studies that described species preferences and differences in habitat use based on certain features of the landscape and vegetation ; however, most earlier studies were largely qualitative and failed to reveal which parameters were most important among several that affect habitat selection. Recent studies of passerines, employing multivariate analyses (Anderson and Shugart 1974, Cody 1968, James 1971, Sturman 1968, Whitmore 1975), have revealed dif- ferences between species-specific habitat types within particular communities. Crawford and Bolen (1976) used multiple regression analysis to correlate factors of vegetation and land-use with spring and fall population levels of Lesser Prairie Chickens {Tympanuchus pallidicinctus) . Little attempt has been made to quantitatively show differences within species between used and unused portions of the available habitat. Although Klebenow (1969) attempted unsuccessfully to differentiate (using discrim- inant function analysis) between habitat that was used and not used by Sage Grouse [Centrocercus urophasianus) for nesting and brood rearing, similar studies with other bird species, including waterfowl, are unavailable. This study investigates factors which separated used from unused nesting wetlands and nesting sites of Canada Geese {Branta canadensis) in southeastern Mich- igan in order to better understand nesting habitat selection by this species. STUDY AREA, METHODS, AND ANALYSIS The study area (9065 km^) lies within the Huron River Valley of southeastern Lower Michigan which Hanson (1965) includes as part of the breeding range of giant Canada Geese {B. c. maxima). A morainic topography, resulting from the Wisconsin glacier, contains numerous kettle hole lakes and marshes. Kaminski (1975) presented a more detailed description of the study area’s wetlands and vegetation. Morphological measurements (culmen length and width, tarsus length, middle-toe length, and body weight) of molting geese (1 year and older) were made to determine subspecies identity of the Huron River Valley flock. Mean values for these measurements were similar to those documented by Hanson (1965) for giant Canada Geese (Kaminski 1975). Between 15 April and 25 April 1974 for 8 days (08:00-16:00), we conducted a helicopter survey of the study area in order to estimate numbers of nesting Canada Geese. Quarter sections (65 ha each) were chosen as the sampling unit; the boundaries of which were easily identified from the air. Topographic maps of the study area were used to enumerate all quarter sections containing any wetland (pond, lake, river, marsh, and waste treatment 523 524 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 lagoon j that could potentially provide nesting habitat for Canada Geese. A total of 6275 quarter sections contained at least 1 of these wetlands. A 5% sample (n = 310) was randomly selected (using a table of random numbers) and positioned on county maps by their appropriate legal description and then systematically searched for nesting geese. Ground searches of other wetlands revealed additional nests for study. Wetlands that contained nesting geese were characterized by a shoreline development index (Reid 1961) which is based on shoreline configuration (a value of 1 denotes a perfectly round shoreline), percent residential and/or recreational shoreline occupancy, area of permanent open water, and area of emergent vegetation within the nesting quarter section. These data were obtained from aerial photos and from an inventory of Michigan’s lakes prepared by Humphrys and Green (1962). Williams and Nelson (1943), Miller and Collins (1953), and others suggested that Canada Goose nesting sites should be elevated to provide good visibility, afford protection, be near water, and provide a firm foundation. On the basis of these criteria, appropriate parameters were measured to evaluate the magnitude of difference between muskrat ( Ondatra zibethica) lodges and islands selected as nesting sites, and similar unused sites. Nest site type dictated the parameters that were measured. Parameters measured on and around muskrat lodge nest sites were: (1) width of lodge top, (2) percent occurrence of cover, (3) lodge height above standing water, (4) distance from the lodge to open water, (5) average height of emergent aquatic vegetation, and (6) distance from the lodge to the nearest shoreline. The same measurements were recorded for the nearest muskrat lodge devoid of nesting Canada Geese. We assumed that the geese had a choice between the sites independent of social interactions between nonspecific pairs. This assumption did not appear to be violated because of the low average density (0.08/65 ha) of nests in 1974. Data for percent occurrence of cover and height of vegetation were collected at 0.1 m intervals along transect lines (0.05 m X 10 m) extending from the base of each lodge in the 4 cardinal directions. Only vegetation (dead annuals plus live and dead perennials) that was presumed to be available to Canada Geese selecting nest sites and that intersected and/or overshadowed the transect line was counted. Parameters measured on islands used by nesting geese and islands not used were: (1) % slope at the highest point on the island, (2) density of vegetation, (3) distance from the island to the nearest shoreline, (4) island length, and (5) average height (up to 3 m) of all understory vegetation. Percent slope was measured with a Haga altimeter. Distance measurements were made with a range finder. A density board, described by DeVos and Mosby (1969), was used to estimate the density of vegetation. Four readings, corresponding to the cardinal directions, were taken within 3 m of the shore-water interface on all islands plus at the nest site on islands used by nesting Canada Geese. Replicated measurements (taken within 1 circular plot (0.03 ha) circumscribing the nest and within 1 randomly placed plot positioned adjacent to the shore on islands not used by nesting Canada Geese) were used to estimate vegetation height. Data from nest sites were analyzed using a multivariate discriminant function analysis modified from Cooley and Lohnes (1971). The goal of discriminant function analysis is to maximize among-group variation thereby assigning individuals to a group on the basis of data peculiar to the group ( Lachenbruch 1975). Green (1971) presented an excellent discussion on the statistical theory- and ecological application of discriminant function analysis. In our analysis, one discriminant function was calculated because g-1 ig ~ number of groups contrasted) was less than p. the number of elements of the vector variable (Cooley and Lohnes 1971) and it accounted for 100% of the among-group vari- ance. Variation about reported mean values is denoted by 95% confidence limits. All Kaminski and Prince • CANADA GOOSE NESTING HABITAT 525 percent data were transformed using arcsine values (Sokal and Rohlf 1969) prior to analysis. RESULTS AND DISCUSSION Twenty-six active nests were located during the survey of quarter sections. We estimated there were 526 ±231 active nests on the study area at the time of the survey. The design of the aerial survey did not exclude any wetland size class; hence quarter sections containing wetlands were surveyed in rela- tion to their abundance. As a result, the survey concentrated on searching small wetlands (Fig. 1). Wetlands with nesting Canada Geese had shoreline development values averaging 1.4 ± 0.2 (n = 30). This type of shoreline configuration (nearly circular) is common to most wetlands in southeastern Michigan. Shoreline development values for nesting wetlands differed sig- nificantly (P < 0.01) when stratified by nest site type (muskrat lodge, island, or floating mat of vegetation) suggesting that the presence of suitable nest sites was more important to Canada Geese selecting nesting wetlands than was the shape of the shoreline. The area of emergent aquatic vegetation (predominately Typha latifolia and Scirpus spp.) within nesting quarter sections ranged from 0 to 40 ha and did not appear to directly influence habitat choice by nesting geese. Nesting wetlands having little or no emergent vegetation contained one or more islands which were virtually inaccessible to mammalian predators, alleviating the necessity for nest concealment by emergent cover. Nesting wetlands covered by more emergent vegetation usually contained muskrat lodges which were the most frequently used nest site type in the study area (Kaminski 1975). Cooper (1973) stressed the important commensal relationship between muskrats and the use of emergent cover by nesting Canada Geese at Marshy Point, Manitoba. Twelve (40%) nesting wetlands had 10% or more of their shorelines occupied residentially and/ or recreationally, suggesting that Canada Geese will tolerate some human habitation when selecting nesting wetlands in southeastern Michigan. The most important factor affecting use of wetlands by nesting geese appeared to be the area of permanent open water. Ninety-two percent of all nests located during the aerial survey were situated on wetlands having 2 or more hectares of open water (Fig. 1). The greatest proportion (42%) of nesting pairs used wetlands for nesting that contained more than 25 ha of open water. This is similar to Hanson’s (1965) observations that although Canada Geese demon- strate a wide adaptability for various nesting habitats, these must be available in large blocks and contain bodies of water of moderate to large size. The difference between selected nest sites and ones not used by Canada Geese was evaluated using a discriminant function analysis. A multivariate analysis of variance yielded a highly significant (P < 0.001) discrimination 526 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 AREA OF OPEN WATER (HA) Fig. L Percentages of Canada Goose nests in relation to the area of open water asso- ciated with nesting wetlands in 1974. I)etween both categories of muskrat lodge and island sites. Width of musk- rat lodge top had the highest scaled eigenvector coefficient (Table 1) indicat- ing it was most influential in separating lodges used by nesting geese com- pared to ones not used. All lodges used by nesting Canada Geese exceeded 1 m in top width while only 2 met this criterion in the unused category. Rienecker (1971) observed that Canada Geese more readily accepted artificial nesting structures having large (0.9 m-1.2 m) platforms. Although percent occurrence of cover, surrounding muskrat lodges, was not significantly differ- ent (P > 0.05) among lodges with goose nests, percent occurrence of cover was significantly different (P < 0.05) among lodges not used by nesting geese. This suggests that Canada Geese selected muskrat lodges for nest sites that were surrounded by a similar amount of cover. Although percent occurrence of cover ranked second in discriminatory ability (Table 1), it contributed similarly to the discriminant function along with lodge height above standing water and distance from the lodge to open water. These 4 parameters are probably important cues used by Canada Geese in selecting muskrat lodges as nest sites and should he measured when field evaluating lodges as potential nest sites for this species. Discriminant scores for all muskrat lodges were computed using a grand mean of 50 (S.D. = 10). Histograms of these scores depict the relative difference between muskrat lodges used and not used by Kaminski and Prince • CANADA GOOSE NESTING HABITAT 527 Table 1 Mean (95% C.I.) and Scaled Eigenvector Coefficients for Parameters Measured on AND Around Muskrat Lodges Used and Not Used by Canada Geese as Nesting Sites IN 1974 Parameter Used lodges (n = 23) Unused (n = 23) Scaled eigenvector coefficient^ Width of lodge top (m) 1.6 (1.4- 1.8) 0.88 (0.80-0.96) -2.535 % occurrence of cover 35.1 (32.4^39.4) 30.2 (25.0-35.5) -0.786 Lodge height above water (m) 0.34 (0.30-0.38) 0.27 (0.21-0.33) -0.718 Distance from lodge to open water (m) 17.5 (9.6-25.4) 25.7 (5.7-45.7) 4-0.556 Average height of vegetation (m) 0.82 (0.75-0.89) 0.80 (0.70-0.90) -0.294 Distance from lodge to nearest shoreline (m) 58.7 (39.3-78.1) 58.9 (39.0-78.8) -0.289 Root of W-^A = 1.046 Wilk’s lambda - 0.489; df = 6,39; F = 6.79; (P < 0.001) 1 The largest absolute value is most important. nesting Canada Geese (Fig. 2). Although lodges within the 46-55 range could not he clearly assigned to 1 of the 2 groups with much confidence, each distribution is comparatively distinct with used lodges occupying the lower ranges of discriminant scores. The minimal overlap between the distributions suggests that those lodges selected by nesting geese were superior nesting sites. Five parameters were measured on islands used and not used by nesting Fig. 2. Histograms of discriminant scores from parameters measured on and around muskrat lodges used and not used by Canada Geese as nesting sites in 1974. 528 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 Table 2 Mean (95% C.I.) and Scaled Eigenvector Coefficients for Parameters Measured on Islands Used and Not Used by Canada Geese as Nesting Sites in 1974 Parameter Used islands (n = 37) Unused (n = 37) Scaled eigenvector coefficient^ Island relief (% slope) 15.7 (13.4^18.1) 8.4 (6.2-10.9) +24.352 Island vegetation density (%) 45.7 (36.1-55.3) 62.9 (53.3-72.0) -16.607 Vegetation density at nest site (%)^ 17.1 (11.8-23.1) — Distance from island to nearest shoreline (m) 73.2 (61.1-85.3) 61.4 (45.1-77.7) + 11.788 Island length (m) 65.9 (39.0-92.8) 85.1 (50.8-119.4) -11.501 Average height of vegetation (m) 1.7 (1.5- 1.9) 2.0 (1.6- 2.4) -10.836 Root of W-^A = 0.439 Wilk’s lambda = 0.695; df = 5,68; F = 5.96; (P < 0.001) 1 The largest absolute value is most important. 2 Not included in discriminant function analysis. Canada Geese (Table 2) . Percent slope of island relief had the highest relative power for discrimination being 7% greater on the average for islands used by nesting geese compared to unused islands. Hanson and Eberhardt (1971) observed that Canada Geese did not use islands that had low profiles for nesting in the Columbia River of Washington. Islands having more relief not only facilitate nest vigilance but render nests less vulnerable to fluctuating water levels. The density of vegetation was significantly lower (P < 0.01) on all islands used by nesting geese compared to islands not used. Further- more, the density of vegetation at the immediate nest site was significantly lower (P < 0.01) than the average vegetation density on the remaining area of the nesting island. Sherwood (1968) reported that most Canada Geese nesting at the Seney National Wildlife Refuge in northern Michigan selected islands that were free of dense, high brush which enhanced visibility and accumulated less snow. Barry (1962), Cooper (1973), and Ryder (1967) observed that snow cover on the breeding grounds delayed nest initiation in Atlantic Brant {Brant a bernicla), Canada Geese, and Ross’ Geese {Chen rossii) respectively; because suitable nest sites were not available. Although all variables contributed cumulatively to the discriminant function, distance from the island to the nearest shoreline, island length, and the height of vegetation differed slightly in their order of magnitude (Table 2) suggest- ing a reduced contribution to the separation. Percent slope of island relief and the density of island vegetation were the most important parameters, among those measured, affecting island use by nesting Canada Geese. Increas- ing island relief and thinning dense stands of vegetation should improve the Kaminski and Prince • CANADA GOOSE NESTING HABITAT 529 Fig. 3. Histograms of discriminant scores from parameters measured on islands used and not used by Canada Geese as nesting sites in 1974. suitability of islands for nesting in southeastern Michigan. Discriminant scores, forming frequency distributions (Fig. 3), from both island groups show the greatest overlap in the 46-65 range making it difficult to accurately predict if an island having a score within this range will be used by nesting Canada Geese in southeastern Michigan. The less distinct separation between these frequency distributions may reflect the preference that Canada Geese show for insular nest sites throughout their breeding range. Although an absolute separation was not obtained in either case, the analyses show that certain physiognomic characteristics delineated selected nest sites from sites not chosen. Those parameters, most significant in the discrimina- tion, were probably important proximate cues (Hilden 1965) affecting site selection by Canada Geese. Klopfer and Hailman (1965) stated that if a bird species recognizes and distinguishes between suitable and unsuitable habitats, its reproductive efficiency could be enhanced. This should theo- retically contribute to the fitness of reproducing individuals. Information obtained in this study is valuable for predicting potential nest site availability, for providing guidelines in the manipulation of habitat, and for the effective construction and positioning of artificial nesting structures for Canada Geese. Experimental manipulation of nest site quality along with the density of breeding pairs as they affect site selection would be a logical 530 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 advancement of this study. Similar research with other species whose nest sites lend themselves to discriminant function analysis would augment our understanding of factors affecting species-specific nest site selection and provide an opportunity to evaluate their strategies of habitat selection. SUMMARY Nesting habitat of Canada Geese in southeastern Michigan is described. Most nesting pairs (92%) preferred wetlands that contained 2 or more hectares of open water. Data were collected from both muskrat lodges and islands used and not used by Canada Geese as nesting sites. These data were analyzed using a discriminant function analysis to determine which factors best separated used from unused nesting sites. Top width of muskrat lodges and percent slope of island relief along with the density of island vegetation were most important in the discrimination. This approach provides a quantitative technique for evaluating the potential availability of nesting habitat along with revealing species- specific nest site preferences. ACKNOWLEDGMENTS W'e wish to thank G. Martz for assistance with design and implementation of the study and F. VanSimaeys for piloting the helicopter. We are indebted to D. L. Beaver, L. W. Gysel, B. D. J. Batt, and A. Afton for reviewing the manuscript and to W. Conley for assistance with the discriminant function analysis. This study was financed by the Michigan Agricultural Experiment Station and the Wildlife Division of the Michigan Department of Natural Resources. It represents part of a M.S. thesis submitted to Michigan State University by the senior author. This is Michigan Agricultural Experi- ment Station Journal Article Number 7669. LITERATURE CITED Anderson, S. H. and H. H. Shugart, Jr. 1974. Habitat selection of breeding birds in an east Tennessee deciduous forest. Ecology 55:828-337. Barry, T. W'. 1962. Effect of late seasons on Atlantic Brant reproduction. J. Wildl. Manage. 26:19-26. Cody, M. L. 1968. On the methods of resource division in grassland bird communities. Am. Nat. 102:107-147. Cooley, W. W. and P. R. Lohnes. 1971. Multivariate data analysis. Wiley, New York. Cooper. J. A. 1973. The history and nesting biology of the Canada Geese of Marshy Point. Manitoba. Ph.D. thesis, Univ. of Massachusetts, Amherst. Crawford. J. A. and E. G. Bolen. 1976. Effects of land use on Lesser Prairie Chickens in Texas. J. Wildl. Manage. 40:96-104. DeVos, a. and H. S. Mosby. 1969. Habitat analysis and evaluation. Pp. 136-172 in Wildlife Management Techniques < R. H. Giles, ed.). The Wildl. Soc., Washington, D.C. Green, R. H. 1971. A multivariate statistical approach to the Hutchinsonian niche: bivalve molluscs of central Canada. Ecology 52:543-556. Hanson, H. C. 1965. The giant Canada Goose. Southern Illinois Univ. Press, Carbon- dale. Kaminski and Prince * CANADA GOOSE NESTING HABITAT 531 Hanson, W. C. and L. L. Eberhardt. 1971. A Columbia River Canada Goose popu- lation, 1950-1970. Wildl. Monogr. 28. Hidden, 0. 1965. Habitat selection in birds. Ann. Zool. Fenn. 2:53-75. Humphrys, C. R. and R. F. Green. 1962. Michigan lake inventory. Michigan State Univ., East Lansing. Dept, of Resour. Dev. Bull. Nos. 1-83. James, F. C. 1971. Ordinations of habitat relationships among breeding birds. Wilson Bull. 83:215-236. Kaminski, R. M. 1975. Nesting giant Canada Geese of southeastern Lower Michigan. M.S. thesis, Michigan State Univ., East Lansing. Klebenow, D. a. 1969. Sage Grouse nesting and brood habitat in Idaho. J. Wildl. Manage. 33:649-661. Klopfer, P. H. and J. P. Hailman. 1965. Habitat selection in birds. Pp. 279-303 in Advances in the Study of Behavior, Vol. 1. (D. S. Lehrman, R. A. Hinde, and E. Shaw, eds.). Academic Press, New York. Lachenbruch, P. a. 1975. Discriminant analysis. Hafner Press, New York. Miller, A. W. and B. D. Collins. 1953. A nesting study of Canada Geese on Tule Lake and Lower Klamath National Wildlife Refuges, Siskiyou County, California. Calif. Fish Game 40:5-16. Reid, G. K. 1961. Ecology of inland waters and estuaries. Van Nostrand Reinhold Co., New York. Rienecker, W. C. 1971. Canada Goose nest platforms. Calif. Fish Game 57:113-123. Ryder, J. P. 1967. The breeding biology of Ross’ Geese in the Perry River region. North- west Territories. Can. Wildl. Serv. Rep. Ser. 3. Sherwood, G. A. 1968. Factors limiting production and expansion of local populations of Canada Geese. Pp. 73-85 in Canada Goose Management (R. L. Hine and C. Shoen- feld, eds.), Dembar Educ. Res. Serv., Madison, Wise. SoKAL, R. R. AND F. J. Rohlf. 1969. Biometry. W. H. Freeman Co., San Francisco. Sturman, W. a. 1968. Description and analysis of breeding habitats of the chickadees, Pams atricapillus and P. rufescens. Ecology 49:418-431. Whitmore, R. C. 1975. Habitat ordination of passerine birds of the Virgin River Valley, southeastern Utah. Wilson Bull. 87:65-74. Williams, C. S. and M. C. Nelson. 1943. Canada Goose nests and eggs. Auk 60: 341-345. DELTA WATERFOWL RESEARCH STATION, RURAL RT. 1, PORTAGE LA PRAIRIE, MANI- TOBA, rIn 3a1, and dept, of fisheries and wildlife, MICHIGAN STATE UNIV., EAST LANSING 48824. ACCEPTED 11 NOV. 1976. RESIDUES OF ENVIRONMENTAL POLLUTANTS AND SHELL THINNING IN MERGANSER EGGS Donald H. White and Eugene Cromartie There is little information regarding the types and concentrations of environmental pollutants in mergansers. However, the reports that are avail- able indicate that residues of certain toxic chemicals are high in eggs and tissues of Common (Mergus merganser) , Red-breasted {Mergus serrator), and Hooded mergansers [Lophodytes cucullatus) from southern Canada, Michigan, and Wisconsin (Fimreite et al. 1971, Vermeer and Armstrong 1972, Faber and Hickey 1973, Vermeer et al. 1973, Fimreite 1974). Also, significant eggshell thinning has been detected in Common and Red-breasted mergansers from Wisconsin and Michigan (Faber and Hickey 1973). Mergansers feed mostly on fishes and invertebrates (Munro and Clemens 1939, Timken and Anderson 1969, Bellrose 1976) and are more susceptible to chemical contamination than species feeding at lower trophic levels. This study was conducted (1) to determine the levels of environmental pollutants in merganser eggs, mainly those of Hooded Mergansers, as factors contributing to possible population declines and (2) to compare eggshell thickness with eggs of earlier collections. Hooded Merganser eggs were more readily avail- able than those of Common or Red-breasted mergansers since Hooded Mer- gansers commonly use nest boxes on many federal and state refuges through- out their breeding range. METHODS AND MATERIALS Federal and state biologists assisted us in collecting clutches of merganser eggs in 1973 and 1975 (see Table 1 for collecting sites). Cooperators were sent insulated con- tainers for shipment of eggs with instructions to collect only fresh whole clutches. So that hens would have time to renest, eggs were collected early in the nesting season and kept refrigerated until shipment to the Patuxent Wildlife Research Center, Laurel, Mary- land. Most of the clutches were complete and only a few contained addled eggs or eggs with developing embryos. Each egg was opened at its equator after determining its weight, length, breadth, and volume (by water displacement). One egg from each clutch was randomly selected for chemical analysis since eggs within a particular clutch usually contain similar residue levels (Klaas and Swineford 1976). The egg contents were stored frozen in chemically cleaned jars until analysis. Eggshells were dried at room temperature for at least 30 days, then weighed and measured with the shell membranes left intact. Three thickness measurements were taken randomly around the equator using a Starrett Model 1010 M micrometer and a mean shell thickness was calculated for each egg. Similar procedures were used in measuring museum egg collections except that measurements were taken at the blow-hole of each 532 White and Cromartie • POLLUTANTS IN MERGANSER EGGS 533 egg. Mean clutch thickness values were calculated by averaging clutch means from each locality within a collecting region and not by averaging individual egg measurements (Klaas et al. 1974) ; this method gives an indication of average shell thickness by popu- lation as opposed to individual hens. In comparing clutch means, all shell measurements used were from fresh or early incubated eggs, therefore stage of incubation did not significantly bias the data. Historical collections of merganser eggs were measured at the American Museum of Natural History, Museum of Vertebrate Zoology at Berkeley, Philadelphia Academy of Sciences, and the Western Foundation of Vertebrate Zoology. Contents of each egg were homogenized with a Virtis homogenizer. A 10 g aliquot was mixed with anhydrous sodium sulfate in a blender and extracted for 7 h with hexane in a Soxhlet apparatus. An aliquot of the extract was cleaned up by gel permeation chromatography or on a florisil column. Pesticides and polychlorinated biphenyls (PCB’s) were separated into 3 fractions on a Silicar column and analyzed by gas chromatography. The limit of quantification was 0.1 ppm for pesticides and 0.5 ppm for PCB’s on a wet- weight basis. The analytical procedures have been described in detail by Cromartie et al. (1975). Residues in 10% of the samples were confirmed with a gas chromatograph/mass spectrometer. All residues were corrected for moisture loss as suggested by Stickel et al. ( 1973) . Lipid content in eggs of Hooded, Red-breasted, and Common mergansers averaged 16%, 16%, and 14%, respectively. Residue arithmetic means and geometric means were very similar, therefore only arithmetic means were reported. Egg contents were analyzed for total mercury at the Environmental Trace Substances Research Center, Columbia, Missouri. An aliquot of the homogenized sample was digested under reflux conditions with concentrated nitric acid. Stannous chloride was added to reduce the ionic mercury to elemental mercury which was measured photometrically on an atomic absorption spectrophotometer. The limit of quantification was 0.02 ppm on a wet-weight basis. RESULTS Hooded Merganser eggs. — Residues of DDE, DDT, DDD, dieldrin, PCB’s, and mercury in Hooded Merganser eggs are presented in Table 1. Residues varied greatly within and among localities. Of 96 eggs, DDE was found in 92, dieldrin in 22, and PCB’s in 82. Eggs collected in 1973 from Necedah National Wildlife Refuge (NWR), Wisconsin had the highest mean of DDE; the highest mean of DDE for 1975 occurred in eggs from Iroquois NWR, N.Y. Sample sizes from these localities were small (2) however, and may not reflect overall contamination levels in the breeding populations. All collections from the Northeast (Maine, New Hampshire, New York, Vermont) had mean levels of DDE greater than 0.1 ppm. The highest mean level of dieldrin was in eggs from the Upper Mississippi NWR, Iowa. Usually dieldrin was detected in only 1 or 2 eggs from a locality, therefore mean levels are not indicative of all the breeding birds. The highest mean PCB level was found in eggs from New Hampshire; all eggs from this locality contained PCB’s as did eggs from the other localities in the Northeast. Of the 90 Hooded Merganser eggs analyzed for mercury, 89 contained detectable resi- dues. The highest mean level of mercury occurred in eggs from Big Lake NWR, Arkansas. Table 1 Residues and Shell Thickness of Hooded Merganser Eggs^ 534 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 Zi c (73 d +1 S' +1 o +1 g On ^ O o d +1 s o +1 g o -S +1 ^ ^ I CSI o +1 5 csi d o _ +1 e CM +1 1 CM o d lO On QJ I— ( O O d II g CM d II g LO ^ LO d CM d II s CO o o d ^ +1 2 o^ CO LO d 2 bC cd .t: CD ^ J ? I I CO d II g LO CM d II g LO ^ o o d ^ 4-1 S Tl CM O lO d LO 2 5 il O o II O ON d II ^ CS Oi b£ C •Z tu 32 -I S 2 r-H OJ 1 2 hJ o 13 « .H o II O o C\ d II CM II g O - O) -I CO ^ Residues, ppm wet weight White and Cromartie • POLLUTANTS IN MERGANSER EGGS 535 14; S o u P3 < o +1 II o +1 5 LO o +1 § ft « £ 2 C3 K £ o fO o +1 CS CM CO o +1 g CO CM O CM ^ O + 1 o o o o +1 s On ^ CO uO 0 o^ o o +1 s CO ^ o CM C2 + 1 § O o o +1 s Ci2 CN Z a ^ I ^ o o + 1 0 CO ^ LO o o + 1 0 CO ^ On d> O Q 2 o ^ +1 0 CO o CM o o + 1 CM CO o II O ^ 'Zl Cv o c^ o d +1 CO O d +1 0 CO o +1 «2 CO UO d o Tt d +1 0 CO CM + 1 S CM d +1 0 0 ^ LO d d ^ ON ^ + 1 ij (n CJ (T) o> ^ c X c w C " . sc • ''UH <1> 3 S'S'Sl M ^ O 303 ir'c (u ^3 ft ^Jft^ « c « *''0 u « S S 3 ii I I 3303 03 . tl-SS'Sg-3 i'li'il p 3 3 " 3 S 3 = II ■^"SldO 536 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 Table 2 Residue Comparisons of Pooled Samples of Hooded Merganser Eggs FROM 3 Regions, 1975 Residues, ppm wet weight^ Regional pool N2 DDE PCB’s Mercury Northeast ( Maine, New Hampshire, New York, Vermont) 25 1.77 ± 0.29" (24)" 3.84 ± 0.08" (25) 1.01 ± 0.12" (24) Midwest (Iowa, Michigan, North Dakota) 20 0.79 ± 0.16” (18) 1.97 ± 0.52" (20) 0.64 ± 0.07” (20) South-central (Arkansas, Missouri, Tennessee) 28 0.78 ± 0.10” (28) 0.86 ± 0.04” (21) 0.62 ± 0.12” (28) 1 Values are means ± standard errors. 2 N = total no. of samples in the pool. ^ No. of samples in the pool having detectable residues. Significant differences exist between means having different superscripts (P < 0.01, f-test). In addition to the chemicals listed in Table 1, certain other toxicants were detected in some Hooded Merganser eggs at much lower levels. Heptachlor epoxide was found in 6 eggs, 1 each from Iowa and New Hampshire and 2 each from Maine and Michigan, but residues were low, ranging from 0.14 to 0.48 ppm. One egg each from Vermont and Michigan contained mirex at 0.08 and 0.18 ppm, and 4 of 10 eggs from Maine contained mirex, ranging from 0.15 to 0.66 ppm. Chlordane isomers were detected in 6 eggs, 2 from New Hampshire and 4 from Maine, ranging from 0.09 to 1.8 ppm. Toxaphene occurred in only 2 eggs, at 0.17 ppm in an egg from Seney NWR, Michigan and 0.10 ppm in an egg from Big Lake NWR, Arkansas. Hexachlorobenzene (HCB) was detected at 0.19 ppm in an egg from Seney NWR, Michigan. We pooled residue data from localities wdthin major regions if they were not statistically different (P > 0.05) from one another in order to compare DDE, PCB’s, and mercury in Hooded Merganser eggs on a regional basis. Residues in eggs from Oregon and Idaho were not pooled because of very small sample sizes and mean differences. Means of pooled residues are shown in Table 2. DDE residues were significantly higher (P < 0.01, Utest ) in eggs from the Northeast than in those from the Midwest or South-central states. However, these are relative comparisons based on specific localities and are not representative of whole areas. PCB’s were significantly higher (P < 0.01) in the Northeast and Midwest samples than in those from the South-central region, probably resulting from heavy industrial use of PCB’s in those regions. Mercury was significantly higher (P < 0.01) in eggs from the Northeast White and Cromartie • POLLUTANTS IN MERGANSER EGGS 537 Table 3 Residues and Shell Thickness of Red-breasted and Common Merganser Eggs, 1975^ Red-breasted Merganser (Door County, Wisconsin ) Common Merganser ( Door County, Wisconsin) Common Merganser ( Seney NWR, Michigan ) N" 18 2 1 Mean shell thickness (mm) 0.302 ± 0.004 0.314 ± 0.006 0.346 (178)" (16) (6) DDE 15.73 ± 1.39 24.44 ± 4.72 9.85 (18)^ (2) DDT 0.62 ± 0.10 0.70 ND (18) (1) DDD 0.40 ± 0.04 0.34 ± 0.08 ND (17) (2) Dieldrin 1.00 ± 0.11 0.64 ± 0.10 1.39 (18) (2) Heptachlor epoxide 0.31 ± 0.04 0.27 ± 0.09 0.17 (18) (2) Mirex 0.42 ± 0.15 ND" ND (8) Chlordane isomers 0.57 ± 0.06 0.82 ± 0.32 0.34 (18) (2) HCB 0.11 ± 0.01 ND ND (16) Toxaphene 0.26 ± 0.04 ND ND (3) PCB’s 44.67 ± 6.06 79.43 ± 9.02 24.19 (18) (2) Endrin 0.33 ND ND (1) Mercury 0.56 ± 0.06 0.56 ± 0.26 0.52 (18) (2) 1 Values are means ± standard errors. 2 N = total no. of clutches collected; one egg per clutch was analyzed. 3 Total no. of eggs in clutches that were measured. * No. of eggs having detectable residues. ^ ND = not detected. than in those from the Midwest or South-central regions, although eggs from a locality in Arkansas had the highest overall mean of mercury (Table 1). Red-breasted and Common merganser eggs. — In general, residues of DDE, dieldrin, and PCB’s were higher in Red-breasted and Common merganser eggs, and other chemicals occurred more frequently, than in eggs of Hooded Mergansers (Table 3). A high of 29 ppm DDE was detected in a Common 538 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 E E uf CO UJ Z u z h- z u »- 3 u Z < .65 - .60 .55 - V J I 2 3 PCB'S,ppm Fig. L Relationship of PCB’s in Hooded Merganser eggs and mean clutch thickness on a population basis. (Spearman’s rank correlation, r = -0.63, P < 0.01) Merganser egg from Door County, Wisconsin; a Red-breasted Merganser egg from the same locality had a high of 113 ppm PCB’s. Residue levels in eggs of the 2 species were similar except that mirex, HCB, and endrin were not detected in Common Merganser eggs. Mercury residues were similar to those found in Black-crowned Night Heron { Nycticorax nycticorax) eggs from Lake St. Clair ( Stendell et al. 1976). Eggshell measurements and shell thickness changes. — Clutch thickness means by locality are shown in Table 1 for Hooded Mergansers, and in Table 3 for Red-breasted and Common mergansers. There was wide geo- graphic variation in mean thicknesses among Hooded Merganser clutches (Table 1) ; clutch means ranged from a low of 0.546 mm in New Hampshire to a high of 0.630 mm in Arkansas (Minnesota mean excluded because of only 2 eggs in the sample). According to Klaas et al. (1974), variation in eggshell thickness among clutches l at least in some species) depends on many factors, including: differences related to clutch size and stage of incubation, genetic and physiological differences among females, differences in diet of the various females, and differences in environmental conditions between years. There was a negative correlation (r = -0.28, Spearman’s rank correlation, White and Cromartie • POLLUTANTS IN MERGANSER EGGS 539 Table 4 Comparisons of Eggshell Measurements from Early Museum Collections and 1973, 1975 Collections Mean clutch thickness ± standard error, mm Species Collection region 1880-1927 1973-197.5 Percent change Hooded Merganser Iowa, Michigan, Minnesota, North Dakota, Wisconsin 0.628 ± 0.025 (6/55)" 0.576 it 0.C05 (28/174) -8.3 Red-breasted Merganser Michigan, Wisconsin, south Manitoba 0.367 ± 0.001 (8/105) 0.302 ± 0.004 (18/178) -17.7“ Common Merganser Michigan, Wisconsin, south Manitoba 0.426 ± 0.011 (3/33) 0.326 ± 0.015 (3/22) -23.5“ 1 Total number of clutches/total number of eggs within clutches. Percent change highly significant (P < 0.001, analysis of variance). Snedecor and Cochran 1967) between DDE residues and mean clutch thick- ness for each locality, but the relationship was not significant (P > 0.05). However, PCB’s and mean clutch thickness also were negatively correlated (r = -0.63) on a locality basis and the relationship was highly significant (P < 0.01) ( Fig. 1). Regression analysis showed that a significant relationship (r = 0.61, P < 0.05) existed between DDE and PCB’s, but when DDE and PCB’s were combined and compared with mean clutch thickness by population, there was no significant (P > 0.05) relationship. It appears that PCB’s were contribut- ing more to the negative relationship between residues and mean clutch thickness than DDE. The results must be viewed with caution however, because of the wide variety of factors that may contribute to geographic variation in shell thickness of Hooded Merganser eggs. Controlled experi- mental studies are needed to clarify these findings. In order to test for possible eggshell thinning, we compared eggshell measurements from our collections with those of early museum collections (Table 4). Data from each major region were combined if the clutch means did not differ significantly ( P > 0.05) from one another. For Hooded Mergansers, we were able to obtain comparable data only from certain midwestern states, therefore comparisons with collections from the Northeast and South-central regions could not be made. Clutches of Hooded Merganser eggs were 8.3% thinner than earlier collections from the same general area ( Table 4) however, the difference was not significant ( P > 0.05). Highly significant shell thinning (P < 0.001) was detected in the Red- breasted Merganser eggs. The clutch means were 17.7% thinner than those 540 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 from early museum collections from the same general area (Table 4). Faber and Hickey (1973) found that Red-breasted Merganser eggs from Wisconsin in 1969 were 17.0% thinner than pre-1947 collections from the same area. Although their comparisons were made on an individual egg basis rather than on clutch means, the results were strikingly similar to ours. Highly significant shell thinning (P < 0.001) also was detected in Common Merganser eggs. The 1975 collections were 23.5% thinner than early museum collections from the same region (Table 4) ; only 3 clutches were available from each time period for comparisons, but the results proved significant. Faber and Hickey (1973) also reported shell thinning in Common Merganser eggs from Michigan and Wisconsin; collections made in 1970 were 15.8% thinner than pre-1947 collections. DISCUSSION In general, residues of organochlorines in Hooded Merganser eggs were considered to be low. Geographical differences were detected however; residues of DDE, PCB’s, and mercury were higher in the Northeast than in the other regions. It is doubtful that these relatively low levels of organo- chlorines could be responsible for overall population declines of Hooded Mergansers, but some hens with moderate to high levels of toxicants could experience poor reproductive success. Also, mercury averaged about 1 ppm in eggs from the Northeast and about 0.63 ppm at other localities; these levels may be sufficient to cause aberrant behavior in ducklings (Heinz 1975) . Most of the Red-breasted and Common merganser eggs contained poten- tially dangerous levels of DDE and possibly PCB’s. In addition, mercury and a wide array of other toxic organochlorines were present in the eggs. Both species exhibited significant eggshell thinning. Mergansers are top predators, and consequently are highly subjected to contamination from their food sources. This is especially true for Red-breasted and Common mer- gansers, which feed almost exclusively on fishes of various types. Hooded Mergansers take fewer fish and more invertebrates than the other species; this may explain the lower residues in their eggs. The effects cumulative concentrations of toxic chemicals in eggs might have on reproduction and survival of wild birds are unknown. A toxicant load of the magnitude found in Red-breasted and Common mergansers might cause reproductive failure or otherwise threaten the survival of merganser populations. Experimental studies with DDE have shown that reproductive impairment may be induced in Mallards {Anas platyrhynchos) and Black Ducks {A. rubripes) at residues similar to those found in some of the Red- breasted and Common mergansers ( Longcore et al. 1971, Haegele and Hudson 1974). Also, in experimental studies with methylmercury, Heinz (1975) White and Cromartie • POLLUTANTS IN MERGANSER EGGS 541 found that aberrant behavior of Mallard ducklings resulted when eggs ac- cumulated 1 ppm or less of mercury. In our study, residues of mercury in Red-breasted and Common merganser eggs averaged 0.56 ppm and ranged up to 1.2 ppm. Field studies are needed to document reproductive success of mergansers in the Door County, Wisconsin area. Further, these data demonstrate the persistence and prevalence of DDT, DDE, and dieldrin in the environment, even though the use of technical DDT was suspended in December 1972 and the use of dieldrin was suspended in October 1974. SUMMARY Clutches of merganser eggs were collected in 1973 and 1975 to determine whether levels of organochlorines and mercury^ might be responsible for possible population declines and to compare eggshell measurements with those of early museum collections. One egg per clutch was selected randomly for chemical analysis. Overall, residues of DDE, PCB’s, and mercury were low in Hooded Merganser eggs; locality means for DDE ranged from 0.07 to 13.2 ppm, PCB means ranged from 0.44 to 4.91 ppm, and mercury means ranged from 0.16 to 1.49 ppm on a wet-weight basis. Residues of DDE and PCB’s appeared to be high in Red-breasted and Common merganser eggs. DDE averaged 15.7 ppm in Red-breasted Merganser eggs and PCB’s averaged 44.6 ppm; Common Merganser eggs contained an overall mean of 19.5 ppm DDE and 61.0 ppm PCB’s. Hooded Merganser eggs from the Midwest had thinned 8.3%, but the change was not significant. Highly significant shell thinning was detected in Red-breasted and Common merganser eggs; Red-breasted Merganser eggs were 17.7% thinner and those of Common Mergansers were 23.5% thinner than museum collections. ACKNOWLEDGMENTS We are grateful to numerous biologists and refuge managers of the U.S. Fish and Wildlife Service and state agencies of Maine, New Hampshire, Vermont, and Missouri for their help in collecting samples. Special thanks are extended to James Elder, Charles Kjos, Tom Erdman, Leigh Fredrickson, Howard Spencer, James Dorso, Harold Nevers, and Tom Myers for their collecting efforts. We appreciate the cooperation of personnel at the Western Foundation of Vertebrate Zoology, Museum of Vertebrate Zoology at Berkeley, Philadelphia Academy of Sciences, and the American Museum of Natural History for allowing the measurements of museum egg collections. Russel Dyrland provided technical assistance. G. Hensler, H. Ohlendorf, S. Wiemeyer, M. Haegele, and E. Klaas provided critical reviews of the manuscript. Clementine Glenn compiled the tables and typed the manuscript. LITERATURE CITED Bellrose, F. C. 1976. Ducks, geese, and swans of North America. Stackpole Books, Harrisburg, Pa. Cromartie, E., W. L. Reichel, L. N. Locke, A. A. Belisle, T. E. Kaiser, T. G. Lamont, B. M. Mulhern, R. M. Prouty, and D. M. Swineford. 1975. Residues of organo- chlorine pesticides and polychlorinated biphenyls and autopsy data for Bald Eagles, 1971-72. Pestic. Monit. J. 9:11-14. 542 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Faber, R. A., and J. J. Hickey. 1973. Eggshell thinning, chlorinated hydrocarbons, and mercury in inland aquatic bird eggs, 1969 and 1970. Pestic. Monit. J. 7:27-36. Fimreite, N. 1974. Mercury- contamination of aquatic birds in northwestern Ontario. J. Wildl. Manage. 38:120-131. , W. N. Holworth, j. a. Keith, P. A. Pearce, and I. M. Crunchy. 1971. Mer- cury in fish and fish-eating birds near sites of industrial contamination in Canada. Can. Field-Nat. 85:211-220. Haegele, M. a., and R. H. Hudson. 1974. Eggshell thinning and residues in Mallards one year after DDE exposure. Arch. Environ. Contam. Toxicol. 2:356-363. Heinz, C. 1975. Effects of methylmercury on approach and avoidance behavior of Mallard ducklings. Bull. Environ. Contam. Toxicol. 13:554-564. Klaas, E. E., H. M. Ohlendorf, and R. C. Heath. 1974. Avian eggshell thickness: variability and sampling. Wilson Bull. 86:156-164. , AND D. M. SwiNEFORD. 1976. Chemical residue content and hatchability of Screech Owl eggs. Wilson Bull. 88:421-426. Longcore, j. R., F. B. Samson, and T. W. Whittendale. 1971. DDE thins eggshells and lowers reproductive success of captive Black Ducks. Bull. Environ. Contam. Toxicol. 6:485-490. Munro, j. a., and W. a. Clemens. 1939. The food and feeding habits of the Red- breasted Merganser in British Columbia. J. Wildl. Manage. 3:47-53. Snedecor, D. W., and W. C. Cochran. 1967. Statistical methods. 6th edition. Iowa State Univ. Press, Ames, Iowa. Stendell, R. C., H. M. Ohlendorf, E. E. Klaas, and J. B. Elder. 1976. Mercury in eggs of aquatic birds. Lake St. Clair — 1973. Pestic. Monit. J. 10:7-9. Stickel, L. F., S. N. Wiemeyer, and L. J. Blus. 1973. Pesticide residues in eggs of wild birds: adjustment for loss of moisture and lipid. Bull. Environ. Contam. Toxicol. 9:193-196. Timken, R. L., and B. W. Anderson. 1969. Food habits of Common Mergansers in the northcentral United States. J. Wildl. Manage. 33:87-91. Vermeer, K., and F. A. J. Armstrong. 1972. Mercury in Canadian prairie ducks. J. Wildl. Manage. 36:179-182. , F. A. J. Armstrong, and D. R. M. Hatch. 1973. Mercury in aquatic birds at (day Lake, Western Ontario. J. Wildl. Manage. 37:58-61. U.S. FISH AND WILDLIFE SERVICE, PATUXENT WILDLIFE RESEARCH CENTER, GULF COAST FIELD STATION, P.O. BOX 2506, VICTORIA, TX 77901, -AND U.S. FISH AND WILDLIFE SERVICE, PATUXENT WILDLIFE RESEARCH CENTER, LAUREL, MD 20811. ACCEPTED 8 AUG. 1977. BREEDING BIRD SURVEY COUNTS AS RELATED TO HABITAT AND DATE Wayne C. Weber and John B. Theberge The Breeding Bird Survey (hereafter BBS) is a standardized technique designed to measure year-to-year changes in numbers of breeding birds (Robbins and Van Velzen 1967, 1969; Van Velzen and Robbins 1971). It has been carried out over much of North America each year since 1966. The factors causing variability in BBS counts, such as time of day, weather, and time of year, have been briefly discussed by Robbins and Van Velzen (1967). However, no detailed analysis has yet been made of the effects of these factors, nor of the relationship between BBS counts and habitat. In this paper, we describe some of these relationships for an area of southern Ontario, Canada. We used the BBS to study breeding bird populations during 1971 in Waterloo County, Ontario (now the Regional Municipality of Waterloo). Our chief aim in conducting the study was to obtain an index to bird popu- lations against which future changes could be measured and compared with changes in land use or other factors. Our purposes in this paper are: (1) to show that the BBS method, when considered together with land use data, is useful in relating bird populations to habitat; (2) to describe some of the bird-habitat relationships evident in our study area; and (3) to outline some of the problems in using the BBS as a technique for estimating bird populations, and particularly to evaluate the effect of time of year on numbers of birds recorded. The field work for the study was done by Weber, but both of us participated in its planning and in the analysis of results. STUDY AREA AND METHODS Study area. — Waterloo County, located in southern Ontario about 100 km west-southwest of Toronto (Fig. 1), has an area of 1336 km^ and a human population of 254,037 (1971 Canadian census). There are 2 large metropolitan areas in the county — Kitchener- Water- loo, with a population of 151,000, and Galt-Preston-Hespeler (recently amalgamated under the name of Cambridge) , with a population of 62,000 — as well as several smaller towns and villages. Urban growth in the county is extremely rapid (an increase of 46.4% from 1961 to 1971). The area consists of old glacial outwash plains and rolling moraines, with an elevation of about 240 to 430 m; it lies entirely within the drainage basin of the Grand River. Soils range from coarse glacial sands and gravels to fine alluvial deposits along the rivers. The area is in a transition zone between the Great Lakes-St. Lawrence and Deciduous Forest Regions (Rowe 1972) ; the climax forests were dominated by sugar maple (Acer saccharum) and American beech (Fagus grandijolia) , with some conifers such as eastern 543 544 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 Fig. 1. Waterloo County, Ontario, showing locations of Breeding Bird Survey routes. Inset map shows general location of county in southern Ontario. hemlock (Tsuga canadensis) and eastern white pine {Pinas strobus) . Only about 10% of the land is now forested (less than in most surrounding areas), and most of this consists of small second-growth woodlots, often in pockets of swampy or poorly-drained soil unsuit- able for farming. The county supports a fairly intensive agriculture dominated by dairying and the raising of crops such as corn, oats, barley, wheat, and hay. We divided habitats in the county into 4 major categories: fields, forest, urban habitats, and “miscellaneous” habitats (including wetlands and gravel pits). These were subdivided into 20 habitat types. This classification was intended not to correspond with plant Weber and Theberge • BREEDING BIRD SURVEYS 545 communities, but to reflect major physical and vegetational features of the habitat which are probably important to birds. Our habitat types were: Fields. — These included: (1) pasture, hay fields, and alfalfa fields; (2) brushy pasture (not grazed or mowed for several years, usually with numerous shrubs or small trees) ; (3) cornfields; (4) other grains — mainly oats and barley (often mixed), some wheat; (5) other crops, mainly potatoes; and (6) bare earth. Forest. — These included: (1) upland deciduous forest — mainly sugar maple-American beech forest, in various successional stages, but mostly young; (2) upland coniferous forest — plantations of red pine {Pinus resinosa) and eastern white pine; (3) upland mixed forest — like upland deciduous, but with eastern hemlock or eastern white pine also present (deciduous trees always dominant) ; (4) riparian deciduous forest — mainly willows {Salix spp.), also balsam poplar (Populus balsamijera) , American elm {Ulmus americana) , etc., along streams; (5) swamp coniferous forest — mainly northern white- cedar {Thuja occidentalis) and/or tamarack (Larix laricina) ; (6) swamp mixed forest — red maple {Acer rubrum) , American elm, black ash (Fraxinus nigra), tamarack, north- ern white-cedar, eastern hemlock, etc.; (7) orchards (included under forests for lack of a better alternative) . Urban. — These included: (1) commercial — business districts, i.e., stores and offices; (2) industrial — factories, warehouses, railway yards, etc. (newer areas often interspersed with fields) ; (3) residential — both “estate” areas with widely-spaced houses and many trees, and more typical areas with more houses and fewer trees; (4) cemeteries and parks — usually with many trees. Miscellaneous. — These included: (1) lakes and ponds; (2) marshes — both cattail {Typha lati folia) and shrub- willow {Salix spp.) marshes; (3) gravel pits. Methods. — The BBS technique was developed by Chandler S. Robbins of the U.S. Fish and Wildlife Service from similar methods used for many years by wildlife biologists in surveys for American Woodcock, Ruffed Grouse, and other gamebirds. A survey route consists of 50 stops spaced at 0.8 km {~V2 mile) intervals; thus each route is 39.4 km (24.5 miles) long. The survey is begun h before local sunrise. The observer spends 3 min at each stop and records all birds heard at any distance, and all seen within 0.4 km (% mile). In the continent-wide BBS, supervised by the U.S. Fish and Wildlife Service and Canadian Wildlife Service, each route is covered only once a year; in southern Canada, this may be done between 1 June and 7 July. For more details, see Robbins and Van Velzen (1967). We set up 4 BBS routes in Waterloo County, spanning the county from east to west at intervals of about 12 km (Fig. 1). Each route was surveyed 8 times between 18 May and 16 July 1971. Direction of coverage was reversed in alternate weeks. Although surveys were continued for 8 weeks, only 5 weeks’ results (28 May to 4 July) were used in the analysis (see Discussion for reasons). In conjunction with the bird surveys, we estimated the area covered by each of the 20 habitat types along the survey routes. At each stop, the percentage covered by each type within a 0.4 km radius was estimated in the field to the nearest 10%. These data were then summed to give totals for each route. We also noted the presence and importance at each stop of hedgerows (rows of trees or shrubs) , scattered trees, farm buildings, and streams. Based on the habitat composition and the importance of hedgerows and scattered trees at each stop, we assigned it an “edge rating,” as a rough index to the amount of forest-field edge present. These ratings ranged from 0, for little or no edge, to 2, for much edge. For example, a stop where forest and fields each covered 30% or more of the area was assigned a 2, whether or not hedgerows and scattered trees were present. If a stop was 100% fields but deciduous 546 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Table 1 Habitat Composition Along Breeding Bird Survey Routes Habitat Survey route^ 1 2 3 4 Overall FIELDS 85.2% 70.8% 70.0% 67.8% 73.5% Pasture 30.0 26.4 29.2 22.4 27.0 Brushy pasture - 0.2 1.8 5.2 1.8 Corn 22.6 22.6 24.6 26.8 24.2 Other grains 28.4 17.4 13.4 13.0 18.1 Other crops - 4.0 0.2 - 1.1 Bare earth 4.2 0.2 0.8 0.4 1.4 FOREST 10.4 21.2 8.2 16.8 14.2 Upland deciduous 4.8 12.0 4.4 9.0 7.6 Upland coniferous 0.2 - 0.2 0.4 0.2 Upland mixed - 3.4 0.4 0.4 1.1 Riparian deciduous 2.0 4.4 0.8 0.8 2.0 Swamp coniferous 1.4 1.0 0.4 0.6 0.9 Swamp mixed 2.0 0.2 1.2 5.6 2.3 Orchard - 0.2 0.8 - 0.3 URBAN 3.4 6.6 20.4 14.6 11.3 Commercial - - 2.2 2.0 1.1 Industrial 0.2 0.8 5.8 1.6 2.1 Residential 3.2 5.8 12.2 10.2 7.9 Cemeteries and parks - - 0.2 0.8 0.3 MISCELLANEOUS I.O 1.4 1.4 0.8 1.1 Lakes and ponds - 0.4 - - 0.1 Marsh 0.4 1.0 - 0.8 0.6 Gravel pits 0.6 - 1.4 - 0.5 ^ For brief descriptions of habitat tyyes, see “Study Area.” 2 Route 1 — Linwood to North Woolwich (see Fig. 1); Route 2 — Ariss to New Prussia; Route 3 — New Hamburg to Breslau; Route 4 — Galt to Haysville. hedgerows were important, the edge rating was 2; if both hedgerows and scattered trees were present but unimportant (covering a small area, or far from the observation point), the edge rating was I ; and if both were absent, the rating was 0, Edge ratings for individual stops were then summed to give a total for each route, ranging from a minimum possible 0 to a maximum possible 100. RESULTS Habitat composition along survey routes. — The percentage of area occupied by each habitat on the 4 survey routes is shown in Table 1. Fields occupied an average of 74% of the area, ranging from 68% on Route 4 to 85% on Route 1. Forest occupied only 14% of the area overall, but was more impor- Weber and Theberge • BREEDING BIRD SURVEYS 547 Total Numbers of Birds Table 2 (Selected Species) Recorded Along Breeding Bird Survey Routes, 28 May to 4 July 1971 Species Survey route 12 3 4 FIELD SPECIES Killdeer 102 85 74 88 Horned Lark 124 65 49 34 Bobolink 123 106 53 77 Savannah Sparrow 470 346 336 275 URBAN SPECIES Chimney Swift 28 41 55 60 Purple Martin - 3 20 - FOREST SPECIES Black-capped Chickadee 2 20 4 13 Veery 1 12 2 6 Red-eyed Vireo 25 35 23 30 FOREST-EDGE SPECIES Gray Catbird 5 18 2 34 Brown Thrasher 3 14 6 13 Yellow Warbler 17 49 9 34 Song Sparrow 163 213 138 190 tant on Routes 2 and 4 (21% and 17%, respectively) than on Routes 1 and 3 (10% and 8%). Urban habitats took up 20% on Route 3 and 15% on Route 4, and averaged 11% over the 4 routes. Comparison of bird numbers among routes. — Weber recorded 101 bird species (not including 6 migrants and non-breeders) on the 4 routes between 28 May and 4 July. For many species, differences in total numbers among routes showed a close relationship with habitat composition; some of these species are included in Table 2. (See appendix for scientific names of all birds mentioned in this paper.) Several birds characteristic of fields (Kill- deer, Horned Lark, Bobolink, Savannah Sparrow) were most abundant on Route 1, which was 85% fields; another species, the Upland Sandpiper, occurred only on Route 1. Two highly urban species, the Purple Martin and Chimney Swift, reached peak numbers respectively on Routes 3 and 4, the routes with most urban habitat. Numbers of Field Sparrows paralleled the extent of brushy pasture on the survey routes. Many forest and forest-edge species were numerous on Routes 2 and 4, which had many wooded areas, but scarcer on Routes 1 and 3. Fig. 2 shows 548 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 CARDINAL • GREAT CRESTED FLYCATCHER A EASTERN WOOD PEWEE O ROSE - BREASTED GROSBEAK A Fig. 2. Comparison of numbers of birds recorded, 28 May to 4 July, with percent forest cover on survey routes. graphs for 4 forest species — Great Crested Flycatcher, Eastern Wood Pewee, Cardinal, and Rose-breasted Grosbeak — whose numbers showed particularly close relationships with percent forest cover. Fig. 3 does the same for 4 forest-edge species — Mourning Dove, Common Flicker, House Wren, and Indigo Bunting — using the “edge rating” for each route instead of percent forest cover. Edge ratings were 50, 67, 54, and 66 for Routes 1, 2, 3, and 4 respectively. These ratings refer only to forest-field edge; other types of edge (urban-field, urban-forest) were far less extensive. As the amount of edge on each route was roughly proportional to the amount of forest, bird species whose numbers closely reflected edge ratings would also closely reflect percent forest cover. To determine whether a bird was best considered a forest or forest-edge species, we relied both on pub- lished information and on our own observations during the study. Of our “forest” birds, the Great Crested Elycatcher and Cardinal also occur to some extent in non-forested habitats; but Dow (1970) in Ontario and Emlen (1972) in Texas found that Cardinal densities increased with vegetation density. Hespenheide (1971) considered the Eastern Wood Pewee a forest- edge species, but in comparison with, for instance, the Eastern Kingbird, a more typical edge species, we would still consider the Wood Pewee a forest bird. Bird species display a complete spectrum from those preferring dense forest to those inhabiting treeless fields, and the distinction between “forest” and “forest-edge” species must sometimes be arbitrary. The lines in Figs. 2 and 3 were fitted by eye. Those for forest birds (Fig. 2) were drawn through the origin, on the assumption that numbers Weber and Theberge • BREEDING BIRD SURVEYS 549 MOURNING DOVE • HOUSE WREN A INDIGO BUNTING A Fig. 3. Comparison of numbers of birds recorded, 28 May to 4 July, with edge ratings on survey routes. of forest birds should decline to 0 only when forest cover approaches 0. However, edge rating is a much less precise measure than percent forest cover, and is relative rather than absolute. Notice that the lines for Common Flicker and Indigo Bunting (Fig. 3) decline to 0 birds with an edge rating of about 40. Comparison of bird numbers in different habitats. — The data in Table 3, comparing the abundance of the commoner bird species in each major habitat category, were obtained by a stop-by-stop tabulation of the number of birds at selected stops over the 28 May to 4 July period. A total of 47 stops in fields, 11 in forest, and 19 in urban areas were used; thus only 77 of the total of 200 stops were included in this analysis. For fields, we included only those stops which were 100% fields; but for forest and urban habitats, because of their small extent, we included all stops which were 60% or more forest or urban, respectively. The 40 species recorded in largest numbers accounted for 97.1% of the total birds recorded at the selected stops. Of these 40, 6 species, comprising 43.0% of total individuals, were considered characteristic of urban areas and farm buildings; 18 (30.4% of individuals) were forest-edge species; 7 (21.8% of individuals) were field species; 7 (3.2% of individuals) were forest species; and 2 (1.6% of individuals) were water-associated species. Because of the nature of the habitats, the majority of stops selected for analysis contained some “edge.” The field stops, although none included any forest or urban habitat, nearly all contained some hedgerows and scattered trees which attracted numerous “edge” birds. Both the forest and urban 550 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Table 3 Abundance of Birds in Different Habitats (Birds per IOO Stops) Habitat Species Overall Fields Forest Urban I. Starling (U)^ 416.3 514.9 67.3 384.2 2. House Sparrow (U) 371.6 528.9 101.8 412.6 3. Red-winged Blackbird (W, FI) 160.1 131.1 78.2 22.1 4. Common Crackle (E) 151.3 137.0 103.6 178.9 5. Savannah Sparrow (FI) 142.7 211.9 29.1 38.9 6. Rock Dove (U) 95.3 173.6 25.5 85.2 7. American Robin (E) 88.8 66.8 89.1 117.9 8. Common Crow (E) 77.9 74.9 85.5 42.1 9. Song Sparrow (E) 70.4 52.3 89.1 18.9 10. Brown-headed Cowbird (E) 55.7 49.8 41.8 38.9 II. American Goldfinch (E) 41.8 36.2 36.4 28.4 12. Eastern Meadowlark (FI) 36.0 27.2 12.7 22.1 13. Bobolink (FI) 35.9 46.8 5.4 - 14. Vesper Sparrow (FI) 35.2 46.4 32.7 1.1 15. Killdeer (FI) 34.9 48.1 9.1 10.5 16. Chipping Sparrow (E) 32.4 26.0 9.1 58.9 17. Bank Swallow (W) 29.4 17.0 14.5 6.3 18. Horned Lark (FI) 27.2 48.9 1.8 5.3 19. Mourning Dove (E) 26.7 15.3 36.4 24.2 20. Barn Swallow (U) 21.4 29.8 5.4 6.3 21. Cedar Waxwing (E) 19.6 8.9 41.8 37.9 22. Cardinal (FO) 18.4 7.7 41.8 16.8 23. Chimney Swift (U) 18.4 5.1 16.4 116.8 24. Northern Oriole (E) 18.0 11.1 27.3 11.6 25. House Wren (E) 17.1 2.6 29.1 10.5 26. Eastern Kingbird (E) 13.9 13.2 9.1 4.2 27. Great Crested Flycatcher (FO) 12.1 3.0 43.6 4.2 28. Red-eyed Vireo (FO) 11.3 4.2 50.9 6.3 29. Yellow Warbler (E) 10.9 - 7.3 8.4 30. Cliff Swallow (U) 9.7 2.6 - 6.3 31. Eastern Wood Pewee (FO) 9.4 0.4 54.5 5.3 32. Blue Jay (FO) 9.1 0.9 25.5 2.1 33. Warbling Vireo (E) 8.8 1.3 7.3 6.3 34. Common Flieker (E) 8.8 3.0 9.1 9.5 35. Indigo Bunting (E) 6.5 1.3 32.7 11.6 36. Rose-breasted Grosbeak (FO) 6.3 0.9 38.2 1.1 37. Gray Catbird (E) 5.9 - 14.5 3.2 38. Spotted Sandpiper (W) 5.1 7.2 7.3 - 39. Willow Alder flycatcher (E)" 4.8 0.4 7.3 1.1 40. Black-capped Chickadee (FO) 3.9 0.4 32.7 1.1 1 Letters in parentheses after species name designate major habitat type considered “typical” for species (i.e., where it reaches highest densities). E = forest-edge; FI = fields; FO = forest; U = urban areas and farm buildings; \V = water (lakes, streams, and their edges). - Both Willow and Alder flycatchers were present along the routes, in about equal numbers, but were not always recorded separately. Weber and Theherge • BREEDING BIRD SURVEYS 551 Table 4 Composition of Breeding Avifauna IN Different Habitats Group of birds Fields Habitat Forest Urban FIELD BIRDS No. of species^ 7 3 4 Individuals per 100 stops^ 560.4 140.0 93.6 Individuals as % of total 24.0% 11.1% 5.5% FOREST-EDGE BIRDS No. of species 11 11 15 Individuals per 100 stops 491.5 612.8 604.0 Individuals as % of total 21.1% 48.3% 35.1% FOREST BIRDS No. of species 1 8 1 Individuals per 100 stops 7.7 319.9 16.8 Individuals as % of total 0.3% 25.2% 1.0% URBAN AND FARM-BUILDING BIRDS No. of species 4 3 5 Individuals per 100 stops 1247.2 194.6 1005.1 Individuals as % of total 53.5% 15.3% 58.4% WATER-ASSOCIATED BIRDS No. of species 2 - - Individuals per 100 stops 24.2 - - Individuals as % of total 1.0% - - TOTAL INDIV. (25 commonest spp.) 2331.0 1267.3 1719.5 TOTAL INDIV. (all spp.) 2375.3 1600.0 1790.5 1 Out of 25 commonest species in each habitat. stops included other habitats, mostly fields; thus “edge” was also present there: Only 72.7% of the area at “forest” stops was actually forested, and only 78.9% of the area at “urban” stops was actually urban. Only one of the 200 stops was 100% forest. At the 47 stops in fields (Table 4), only 7 of the 25 commonest species (24.0% of individuals) were true “field” birds, nesting on the ground and carrying out all other activities in fields. Three of the 4 commonest species — House Sparrow, Starling, and Rock Dove — were associated with, and nested in, farm buildings. These 3 are often considered urban birds (Weber 1972), but in Waterloo County, their total numbers in rural areas almost 552 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Comparison of Numbers Table 5 OF Edge Birds in Fields With and Without Deciduous Hedgerows Species Number of birds per 100 stopsi Fields Fields with Fields with- overall hedgerows out hedgerows Mourning Dove^ 15.3 14.3 12.7 Eastern Kingbird^ 13.2 18.6 14.5 Common Crow^ 74.9 74.3 61.8 American Robin^ 66.8 65.7 50.9 Cedar Waxwing^ 8.9 5.7 16.4 Red-winged Blackbird^ 131.1 155.7 112.7 Northern Oriole^ 11.1 15.7 5.5 Common Crackle® 137.0 115.7 125.5 Brown-headed Cowbird® 49.8 64.3 50.9 American Goldfinch® 36.2 45.7 29.1 Chipping Sparrow® 26.0 18.6 21.8 Song Sparrow® 52.3 62.9 40.0 TOTALS — 12 edge species 622.6 657.2 541.8 TOTALS — all species 2375.3 2332.4 2314.1 1 Data based on 47 stops for fields overall; 14 stops for fields with hedgerows; and 11 stops for fields without hedgerows. 2 Species characteristic of deciduous hedgerows. 3 Species characteristic of coniferous hedgerows. certainly exceeded those in cities, even if their densities were lower. Forest- edge birds were also important in fields (11 out of 25 species, 21.1% of individuals) . In forest, only 8 of the 25 commonest species, and 25.2% of individuals, were true forest birds; forest-edge birds (11 species) accounted for 48.3%. This is a result of the unavoidable inclusion of some fields in the forest stops analyzed, plus the edge created by the road rights-of-way. Even farm-building birds (3 species, 15.3% of individuals) and field birds (3 species, 11.1% of individuals) crept into the top 25 forest species. In urban habitats, only 5 of the top 25 species were typical urban birds, but they made up 58.4% of individuals. Forest-edge birds accounted for 15 species, though only 35.1% of individuals; their importance is not surprising, as many urban areas (at least residential areas) consist, in effect, of almost continuous “edge.” Five species of field and forest birds also entered the urban list, but were relatively unimportant. Effect of deciduous hedgerows on bird numbers in fields. — Forest-edge birds are numerous in fields, as we have noted. However, the Ontario Weber and Theberge • BREEDING BIRD SURVEYS 553 Comparison of Numbers Table 6 OF Farm-building Birds in Buildings Fields With and Without Farm Number of birds per 100 stops^ Fields Fields with Fields without Species overall farm buildings farm buildings Rock Dove 173.6 234.3 92.0 Barn Swallow 29.8 30.0 28.0 Starling 514.9 512.9 560.0 House Sparrow 528.9 695.7 300.0 TOTALS — 4 farm-building species 1247.2 1472.9 980.0 TOTALS — all species 2375.3 2573.0 2300.0 1 Data based on 47 stops for fields overall; 14 stops for fields with farm buildings; and 5 stops for fields without farm buildings. Department of Agriculture has advocated more intensive use of farmland, including removal of hedgerows. To evaluate the significance of hedgerows to birds, we compared numbers of birds at 14 stops in fields where deciduous hedgerows were important with those at 11 stops in fields where they were lacking (Table 5). All stops containing coniferous hedgerows were excluded from this analysis. Twelve species of “edge” birds totalled 541.8 individuals per 100 stops without hedgerows, and 657.2 (21.1% higher) with hedgerows; 9 of the 12 were commoner with hedgerows. Nevertheless, even where hedgerows were absent, many “edge” birds were supported by scattered trees or by forest- field edge beyond the 0.4 km radius (from which birds were counted if heard) . Of the 3 edge species not positively associated with deciduous hedgerows, 2 (Common Crackle and Chipping Sparrow) preferred coniferous hedgerows, which were excluded from this analysis. The third species, the Cedar Wax- wing, was commoner without hedgerows for reasons unknown — perhaps merely the small sample size. Effect of farm buildings on bird numbers in fields. — Like hedgerows, farm buildings have a great effect on numbers of birds recorded in fields. Table 6 compares bird numbers at 5 stops where no farm buildings were present within 0.4 km with those at 14 stops where farm buildings were important (close to the observation point, or 2 or more farmsteads present within 0.4 km). Total numbers of 4 “farm-building” species were 50.3% higher with farm buildings than without them (1473 versus 980 per 100 stops) ; Rock Doves and House Sparrows were more than twice as abundant. Starlings 554 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Table 7 Diversity and Density of Birds IN Different Habitats Xo. of Total no. Species Individuals Habitat stops of species per stop per stop Fields 47 56 9.61 23.7 Forest 11 71 11.60 16.0 Urban 19 50 7.62 17.9 would undoubtedly have shown the same pattern had surveys been done earlier in the season, before wandering flocks of juveniles appeared. DISCUSSION Diversity and density of birds in different habitats. — Diversity will be discussed only in terms of numbers of species. Out of 101 species ( excluding migrants and non-breeders) recorded on the survey routes from 28 May to 4 July, we recorded 50 species at urban stops, 56 at field stops, and 71 at only 11 forest stops (Table 7). Another indication of diversity is the mean number of species per stop, which varied from 7.6 in urban habitats to 11.6 in forest. Although these figures may be inflated by the inclusion of some edge habitat in each category, forests clearly have more species than either fields or urban habitats. A similar pattern was found by Speirs et al. ( 1967, 1970, 1975) in a comprehensive census-plot study of bird populations in Ontario County, Ontario, about 130 km east-northeast of Waterloo County. They found a total of 30 species on 11 10-ha study plots in fields; 79 species on 11 forest plots; and 52 species on 10 urban plots. Their low species count in fields is explained by the fact that they largely excluded trees, shrubs, and farm buildings (Speirs and Orenstein 1967) ; for example, they recorded no Rock Doves, Bank Swallows, Common Crows, House Sparrows, or Northern Orioles in fields. As the BBS does not measure absolute density, the trends in avian density suggested by our data are misleading. From Table 7, it would appear that the highest densities (individuals per stop) are in fields. This results merely from the observer’s ability to see and hear birds at much greater distances in fields than elsewhere. In forest and urban areas, trees and buildings impede the detection of distant birds, and noise from traffic and other sources further reduces detectability in urban areas. Speirs et al. (1970) give mean total bird densities for Ontario County of 240 pairs per 100 ha in fields, 613 in forest, and 1005 in urban areas; the same trend undoubtedly holds true in Waterloo County. Even allowing for the birds added by farm buildings and W eber and Theberge • BREEDING BIRD SURVEYS 555 hedgerows (largely excluded by Speirs et al.), fields unquestionably have lower densities than any other habitat. Critique on the method. — The BBS technique is not a reliable indicator of the relative abundance of different species because of differences in con- spicuousness among species. Emlen (1971) has quantified conspicuousness as the coefficient of detectability (CD) — the proportion of individuals in an area which is ordinarily detected by an observer. Not only does CD differ greatly among species, but the CD value for each species varies with habitat. For instance, though we made no measurements, our guess is that the mean detection distance in forest is about % that in fields. As a result, differences among habitats in a species’ numbers may be over- or underestimated. One advantage of the BBS is that it inevitably samples “edge” habitats as well as “pure” habitats; in fact, it is considerably biased toward edge habitats, as roadsides usually create an edge situation. In contrast, the usual approach in census-plot studies is to include only “pure” habitats, and to deliberately avoid mixed habitats and “edge.” As an illustration of this, the Common Crow, a typical edge species which ranked 8th in abundance on our surveys, was not even listed among the commoner species in Ontario County by Speirs et al. (1970), whose plot censuses covered all the major pure habitat types. Pure habitats, unmixed with edge, do not cover any extensive areas in south- ern Ontario. Thus the BBS records a segment of the bird population hardly touched by traditional census-plot methods. The factors causing variability in BBS counts are discussed by Robbins and Van Velzen (1967:6-12). These include the observer, time of day, weather, and time of year. As all our surveys were conducted by one observer, only the other 3 factors need concern us here. Most species of birds sing less frequently as the morning progresses, al- though the rate of decrease varies with the species (Robbins and Van Velzen 1967:11). This becomes particularly noticeable when the direction of cover- age is reversed in alternate weeks, as we did. A cogent example is the number of Mourning Doves recorded on Route 1. Mourning Doves sing frequently for about an hour after sunrise, but much less frequently thereafter. Most of the forest-edge on Route 1, hence most of the Mourning Doves, were near the east end of the route. When the survey was begun at the east end, a mean of 12.5 Mourning Doves was recorded. When it was begun at the west end, only 4.3 were recorded; the birds at the east end had stopped singing by the time the observer arrived there. BBS routes are generally not surveyed during rain, steady drizzle, or fog, or when winds exceed Beaufort force 3 (19 km/h). Within these constraints, however, weather affects counts less than we had anticipated. A case in point is the survey of 1 July, which was begun under marginal weather conditions 556 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Table 8 Weekly Counts (All Survey Routes Combined) OF Common Bird Species Species Weeki CV2 1 2 3 4 5 6 7 8 Killdeer 70 63 61 74 76 75 96 58 .10 Rock Dove 102 177 147 248 237 144 244 272 .26 Mourning Dove 39 64 42 45 55 61 38 71 .18 Chimney Swift 26 39 37 31 44 33 24 14 .14 Eastern Kingbird 39 30 29 27 23 30 41 46 .11 Great Crested Flycatcher 15 29 23 30 20 19 7 7 .21 Eastern Wood Pewee 4 15 21 21 17 20 20 16 .14 Horned Lark 63 60 56 55 58 43 48 37 .12 Bank Swallow 36 56 40 68 43 87 164 192 .33 Barn Swallow 54 60 33 27 39 55 67 90 .33 Cliff Swallow 0 16 4 7 20 50 113 52 .94 Blue Jay 40 48 13 11 12 7 10 12 .92 Common Crow no 152 143 150 171 163 147 154 .07 House Wren 15 31 31 25 42 42 32 34 .22 American Robin 156 189 157 163 195 184 196 226 .09 Cedar Waxwing 8 9 60 42 36 49 47 30 .49 Starling 485 688 1000 775 955 745 1133 1611 .16 Red-eyed Vireo 6 22 17 27 21 26 12 17 .18 Yellow Warbler 19 22 22 22 23 20 12 10 .05 House Sparrow 607 657 726 726 816 791 1006 936 .08 Bobolink 90 73 65 65 79 75 57 38 .09 Eastern Meadowlark 82 64 63 81 79 73 67 53 .12 Red-winged Blackbird 335 304 304 338 344 311 271 327 .06 Northern Oriole 63 41 37 33 44 25 17 23 .21 Common Crackle 346 298 300 286 381 253 287 522 .16 Brown-beaded Cowbird 122 124 117 104 110 102 85 55 .08 Cardinal 26 37 33 40 35 39 23 29 .07 American Goldfinch 143 124 67 66 74 87 76 107 .29 Savannah Sparrow 213 230 257 276 318 346 336 310 .16 Vesper Sparrow 69 60 66 70 84 72 69 61 .13 Chipping Sparrow 53 55 63 71 74 61 58 66 .12 Song Sparrow 130 135 120 147 153 149 150 175 .10 TOTAL INDIVIDUALS (all species) 3838 4227 4344 4360 4920 4484 5206 5873 TOTAL SPECIES 92 87 82 82 79 81 82 79 1 Weeks are as follows: Week 1, 18 to 21 May; Week 2, 28 May to 1 June; Week 3, 4 to 9 June; Week 4, 10 to 18 June; Week 5, 19 to 23 June; Week 6, 28 June to 4 July; Week 7, 7 to 10 July; Week 8, 12 to 16 July. 2 CV = coefficient of variation (standard deviation divided by mean) for Weeks 2 to 6 (28 May to 4 July). Weber and Theberge • BREEDING BIRD SURVEYS 557 (low clouds, wind 16 km/hj and was halted by heavy rain after 12 stops (it was completed the next day). On these 12 stops, 234 birds of 38 species were recorded, compared with a mean of 248 birds of 36 species — almost identical — on 5 previous coverages of this section of the route. We conclude, as does Anthony J. Erskine (pers. comm.), that weather during a survey generally has little effect on counts if rain and strong winds are avoided. Time of year had a very noticeable effect on counts for most species. Table 8 shows week-by-week total counts for the 32 commonest species. A Friedman non-parametric analysis of variance (Siegel 1956:166) showed that time of year had a significant effect on counts (p < .05). Much of the variation was contributed by Week 1 (18 to 21 May) and Weeks 7 and 8 (7 to 16 July) ; for most species, counts during these 3 weeks tended to be either higher or lower than those during Weeks 2 to 6 (28 May to 4 July). For 30 of the 32 commonest species, at least 1 of the counts during Weeks 1, 7, and 8 lay outside the range of those in Weeks 2 to 6; for 15 of the 32, all 3 counts in Weeks 1, 7, and 8 lay outside this range. During Week 1 (18 to 21 May), high counts were recorded for several species (e.g. Blue Jay, Bobolink, and Northern Oriole), probably because they were still migrating in numbers. Interestingly, a sizable Blue Jay migration was noted on 18 and 19 May, the same dates when Weir (1972) reported an influx at Prince Edward Point, Ontario, about 305 km to the east. On the other hand, numbers of several insectivorous species (e.g. Great Crested Flycatcher, Eastern Wood Pewee, House Wren, Cedar Waxwing, and Red-eyed Vireo) were low, presumably because many individuals had not yet arrived from the south. During Week 2 (28 May to 1 June), Blue Jays were still migrating, and most Cedar Waxwings still had not arrived. Even during Week 3 (4 to 9 June), a few migrants were recorded. The presence of migrants in June may be unusual, however, as April and May 1971 were abnormally cold in southern Ontario, and bird migration was noticeably delayed as a result (Fairfield 1971, Goodwin 1971, Weir 1972). During Weeks 7 and 8 (7 to 16 July), a number of species (e.g. Great Crested Flycatcher, Red-eyed Vireo, Yellow Warbler, Bobolink, Cardinal) were recorded less often because they had stopped singing or sang less often. Most of these are species usually detected by ear. In fact, at least 3 species (Horned Lark, Brown Thrasher, and Northern Oriole) had noticeably de- creased their song frequency even by Week 6 (28 June to 4 July) . In contrast, a number of visually-conspicuous species (e.g. Eastern Kingbird, Bank and Barn swallows, American Robin, Starling, House Sparrow) showed peak counts in Weeks 7 and 8; this is attributable to the presence of fledged young and of noisy, highly visible family groups or flocks. Restricting our attention to Weeks 2 to 6 (28 May to 4 July), we found 558 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 that counts varied little for most species, although the coefficients of variation ranged from .05 for the Yellow Warbler to .94 for the Cliff Swallow. Two species, the Blue Jay and Cedar Waxwing, showed high coefficients (.92 and .49) only because migratory movements occurred in Weeks 2 and 3; later counts of these species were quite consistent. There was a tendency for highly-localized or colonial species (e.g. Cliff and Bank swallows. Rock Doves) to have high coefficients, although there were exceptions to this. Neverthe- less, the median coefficient of variation for the 32 species was only .135, indicating that, for most species, one count in the period 28 May to 4 July is almost as reliable as 5 counts. We conclude from these data that the period 28 May to 4 July is best for conducting Breeding Bird Surveys in southern Ontario. This is 3 or 4 days earlier than the period of 1 June to 7 July recommended by the U.S. Fish and Wildlife Service for southern Canada, but southern Ontario lies farther south than other parts of southern Canada, and undoubtedly the nesting season is correspondingly earlier. Finally, we wish to offer some suggestions concerning the continent-wide Breeding Bird Survey. We believe that the value of the Survey would be greatly enhanced by the collection of data similar to ours on land use along survey routes. Land use data could be collected either on the ground, by individual Survey cooperators, or possibly by centralized interpretation of data from high-level aerial photography. Such data need not be collected annually, but perhaps only once every 3 or 4 years. The main stated purpose of the Survey is to measure year-to-year changes in the abundance of breeding birds ( Robbins and Van Velzen 1967, Erskine 1970). We suspect that changes in land use will be the most important single factor responsible for long-term changes in bird numbers; but without infor- mation on land use along the actual survey routes, it will be difficult to determine whether changes in numbers have resulted mainly from land use changes or from other, more subtle causes like pesticides. This is especially true in areas sparsely sampled by BBS routes, such as most of the western United States, where land use along BBS routes may not reflect land use over the area as a whole. Before information on land use can be gathered, however, a classification of habitats usable throughout North America is needed. This classification must reflect important features of both natural and man-altered habitats, and must be easily comprehensible to amateur ornithologists, but its development would be well worth the effort. Even if it does not prove practicable to collect land use data on a continent- wide basis, we hope that our approach will be useful to others who wish to study changes in bird populations in a localized area such as the one we studied. Weber and Theberge • BREEDING BIRD SURVEYS 559 SUMMARY We used the Breeding Bird Survey technique to study breeding lurd populations in relation to habitat in Waterloo County, Ontario, in 1971. Four survey routes across the county were each covered 8 times between 18 May and 16 July. In conjunction with these surveys, we devised a classification of habitat types and estimated the coverage of each type at each sampling point. We compared bird numbers among survey routes, and found that numbers of several species were closely related to the extent of particular habitat types. We also compared bird numbers in 3 major habitat categories (fields, forest, and urban areas), based on results from selected sampling points. Because of the nature of the sampling and of the habitats themselves, all 3 contained a high proportion of forest-edge birds. Our data support those of others showing that forests have the most species of birds and urban areas fewest, and are consistent with a pattern of densities highest in urban areas and lowest in fields. In a critique on the method, we looked at the effects of time of day, weather, and especially time of year on bird counts. Counts in the third week of May were high for some species which were still migrating in large numbers, and low for others which were still arriving. Counts after 4 July were high for some visually-conspicuous species which congregate in family groups or flocks, and low for other species because of a decrease in song. Between 28 May and 4 July, however, counts varied little for most species. We conclude that interpretation of the significance of changes in bird numbers shown by Breeding Bird Surveys would be facilitated if complementary data on land use were gathered. We recommend the development of a classification of habitats usable through- out North America, and its application in conjunction with the continent-wide Breeding Bird Survey. ACKNOWLEDGMENTS Several people provided helpful advice and information during our work, especially Robert S. Dorney of the School of Urban and Regional Planning, University of Waterloo, as well as several Planning graduate students (particularly Derek Coleman and Ray Smith) and a number of local naturalists including Craig A. Campbell, Larry Lamb, and Willard Schaefer. Anthony J. Erskine and Chandler S. Robbins advised us on methods before the study began, and gave helpful comments on an early draft of the manuscript, as did Jerome A. Jackson and J. Murray Speirs. We wish to thank all these people, and also the National Research Council of Canada, which financed the study through a grant to Theberge. LITERATURE CITED Dow, D. D. 1970. Distribution and dispersal of the Cardinal, Richmondena cardinaUs, in relation to vegetational cover and river systems. Am. Midi. Nat. 84:198-207. Emlen, j. T. 1971. Population densities of birds derived from transect counts. Auk 88:323-342. — 1972. Size and structure of a wintering avian community in southern Texas. Ecology 53:317-329. Erskine, A. J. 1970. The Breeding Bird Survey in Canada, 1966-1969. Can. Wildl. Serv., Prog. Notes No. 15. 560 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 Fairfield, G. M. 1971. The Toronto spring warbler migration study. Ont. Field Biol. 25:34-41. Goodwin, C. E. 1971. Ontario-Western New York Region (Ontario). Am. Birds 25: 735-739. Hespenheide, H. a. 1971. Flycatcher habitat selection in the eastern deciduous forest. Auk 88:61-74. Robbins, C. S. and W. T. Van Velzen. 1967. The Breeding Bird Survey, 1966. U.S. Fish. Wildl. Serv., Bur. Sport Fish. Wildl., Spec. Sci. Rep., Wildl. No. 102. AND . 1969. The Breeding Bird Survey, 1967 and 1968. U.S. Fish. Wildl. Serv., Bur. Sport Fish. Wildl., Spec. Sci. Rep., Wildl. No. 124. Rowe, J. S. 1972. Forest regions of Canada. Can. Dep. Environ., Can. For. Serv., Publ. No. 1300. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York. Speirs, J. M., G. Markle, and R. G. Tozer. 1970. Populations of birds in urban habitats, Ontario County, 1969. Ont. Field Biol. 24:1-12. AND R. Orenstein. 1967. Bird populations in fields of Ontario County, 1965. Can. Field-Nat. 81:175-183. AND . 1975. Bird populations in forests of Ontario County, 1966-1968. Ont. Field Biol. 29:1-24. Van Velzen, W. T. and C. S. Robbins. 1971. The Breeding Bird Survey, 1969. U.S. Fish Wildl. Serv., Bur. Sport Fish. Wildl., Admin. Rep. Weber, W. C. 1972. Birds in cities: a study of populations, foraging ecology, and nest- sites of urban birds. M.Sc. thesis, Univ. British Columbia, Vancouver. Weir, R. I). 1972. Spring migration at Prince Edward Point, Ontario. Can. Field-Nat. 86:3-16. FACULTY OF ENVIRONME.NTAL STUDIES, UMV. OF WATERLOO, WATERLOO, ONTARIO n2l 3g1, CANADA ( PRESENT ADDRESS OF WCW: DEPT. OF ZOOLOGY, MISSIS- SIPPI STATE UNIV., MISSISSIPPI STATE 39762 ). ACCEPTED 15 JULY 1976. APPENDIX: SCIENTIFIC NAMES OF BIRDS MENTIONED IN TEXT AND TABLES Ruffed Crouse, Bonasa umhellus; Killdeer, Charadrius vociferus; American Woodcock, Philohela minor; Upland Sandpiper, Bartramia longicauda; Spotted Sandpiper, Actitis macularia ; Rock Dove, Columba livia; Mourning Dove. Zenaida macroura; Chimney Swift, Chaetura pelagica; Common Flicker, Colaptes auratus; Eastern Kingbird, Tyrannus tyran- nus; Great Crested Flycatcher, Myiarchus crinitus; Willow Flycatcher, Empidonax traillii; Alder Flycatcher, Empidonax alnorum; Eastern Wood Pewee, Contopus virens; Horned Lark, Eremophila alpestris; Bank Swallow, Riparia riparia; Barn Swallow, Hirundo rustica; Cliff Swallow, Petrochelidon pyrrhonota; Purple Martin, Progne subis; Blue Jay, Cyanocitta cristata; Common Crow, Corvus brachyrhynchos; Black-capped Chickadee, Parus atricapillus; House Wren, Troglodytes aedon; Gray Catbird, Dumetella carolinensis; Brown Thrasher, Toxostoma rufum; American Robin, Turdus migratorius; Veer\', Catha- rus juscescens; Cedar W'axwing, Bombycilla cedrorum; Starling, Sturnus vulgaris; Red- eyed Vireo, Vireo olivaceus; Warbling Vireo, Vireo gilvus; Yellow Warbler, Dendroica petechia; House Sparrow, Passer domesticus; Bobolink, Dolichonyx oryzivorus; Eastern Meadowlark, Sturnella magna; Red- winged Blackbird, Agelaius phoeniceus; Northern Oriole, Icterus galbula; Common Crackle, Quiscalus quiscula; Brown-headed Cowbird, W'eher and Theherge • BREEDING BIRD SURVEYS 561 Molothriis ater; Cardinal, Cardinalis cardinalis; Rose-breasted Grosbeak, Pheucticus ludovicianus; Indigo Bunting, Passerina cyanea; American Goldfinch, Carduelis tristis; Savannah Sparrow, Passerculus sandwichensis; Vesper Sparrow, Pooecetes gramineus; Chipping Sparrow, Spizella passerina; Field Sparrow, Spizella pusilla; Song Sparrow, Melospiza melodia. NEW LIFE MEMBER Robert D. Burns has become a life mem- ber of the Wilson Ornithological Society. Dr. Burns is presently a professor of biology at Kenyon College in Gambier, Ohio. His principal interests in ornithology are pri- marily in the area of population ecology; he has published several studies on the Cardinal. Dr. Burns is also a member of the AOU, The American Society of Mam- malogists, and other natural history organi- zations. He has been very active in the Wilson Society and has served as an elected council member and a member of WOS committees. Dr. Burns is married and has two children. ANALYSIS OF MATERIALS IN CLIFF AND BARN SWALLOW NESTS: RELATIONSHIP BETWEEN MUD SELECTION AND NEST ARCHITECTURE Delbert L. Kilgore, Jr. and Kathy L. Knudsen Studies of the mud nests of Cliff [Petrochelidon pyrrhonota) and Barn swallows {Hirundo rustica) have heretofore emphasized nest-site selection and nest-building activity (e.g., Emlen 1952, 1954, Mayhew 1958, Samuel 1971, Jackson and Burchfield 1975). While it is usually acknowledged in these studies that a “proper consistency of mud” is essential for nest con- struction, there are no specific data available on the actual texture or con- sistency of such materials. Furthermore, it is not known if or how the design of nests constructed by these species may affect the selection of building materials; Cliff Swallows build nests which are enclosed and retort-shaped, whereas Barn Swallows construct simple cup-shaped nests (Samuel 1971). The purposes of this study were to analyze the materials used by these swallows in nest construction and to determine if there are interspecific differences which may be related to nest construction or design. MATERIALS AND METHODS Collecting sites. — Three Cliff and 3 Barn swallow nests were collected at each of 11 sites in western Montana (Fig. 1). At 7 of these sites (1, 3, 4, 8, 9, 10, 11), nests of both species were found within a few meters of each other on the same structure (concrete or wooden bridge, barn, or highway overpass) . At the remaining 4 sites, nests of both species were located on different structures, but were never more than 400 m apart. Analyses. — Samj)les of each nest were analyzed for texture (% sand, % silt, % clay), sand size, organic matter, and water content. Texture refers to the type and relative numbers of particles in the sample ( Baver et al. 1972). Two 50 g samples of dried mud from each nest were analyzed for texture by the hydrometer method described by Bouyoucos (1936). In intact nests (5 Cliff, 7 Barn) 1 sample was taken from the area of attachment (base), while the other was selected from the rim or opening of the nest. In other nests, the samples were not selected from specific regions. Because the amount and kind of organic matter in the samples varied considerably, they were not routinely treated with hydrogen peroxide. Instead, conspicuous pieces of organic matter were removed by hand. The suspension remaining after the textural analysis was washed through 4 brass sieves, sizes 20, 32, 60, and 120. The mesh openings of these sieves are 0.841, 0.557, 0.250, and 0.125 mm, respectively. The portions of the suspension which were retained in the sieves were air dried and weighed on a semi-microbalance. Non-sand materials were eliminated from the residue before weighing. Sand remaining in the size 20 sieve was classified as very coarse sand, that in the size 32 sieve as coarse sand, that in the size 60 sieve as medium sand, and that in the size 120 sieve as fine sand (USDA 1951). An additional 15 g sample from each nest was air dried at 110°C for 24 h and then 562 Kilgore and Knudsen • ANALYSIS OF SWALLOW NESTS 563 ignited at 400° C for 7 h. This procedure yielded information on water content and organic matter in the dried mud sample (Jackson 1958). These values for water content were used in the calculations of textural components. The amount of material in each textural category, the amount of organic matter, and the water content of each sample are expressed as percentages of the total weight of the sample, while the amount of sand in each size category is expressed as a percentage of the total weight of sand in the sample. Statistical treatment of data. — Variation between species, among sites, among nests within each site, and within individual nests in each textural and sand size category was analyzed with a mixed-model 2 factor analysis of variance (ANOVA) with 2 levels of nesting. Variation between species and among sites in water content and organic matter was analyzed with a mixed-model 2 factor ANOVA (Sokal and Rohlf 1969). Differences between samples taken from the rim and base in textural components and sand size were analyzed with f-tests. All percentages were transformed to angles (arcsine transformation) prior to analysis with ANOVA or Mest (Sokal and Rohlf 1969). RESULTS Textural components. — The soil-like material in the nests of both swallows was predominantly sand, with modest amounts of silt and some clay (Fig. 2). Sand particles comprised 61.4 ± 0.8% and 56.4 ± 1.1% of the mud samples from Cliff and Barn swallow nests, respectively, while silt particles accounted 564 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 Fig. 2. Textural components of mud samples from Cliff (left half) and Barn swallow (right half) nests at all 11 collecting sites. Values are mean percentages for each component. for 25.7 ± 0.9% and 31.5 ± 0.9% (values are means ± SE). The mean amount of clay in the mud samples was similar for both species, 12.7 ± 0.7% and 11.9 ± 0.6%, respectively. At most localities (8 of 11), the mud in Cliff Swallow nests contained more sand and fewer silt particles than that from Barn Swallow nests (Fig. 2). These species differences in sand and silt content are statistically significant (P < .05), as are the differences in clay content (P < .05). However, the interspecific differences in clay content are more dependent on the locality, yet there is no recognizable trend among sites (Fig. 2). Cliff Swallows at 4 sites used mud with a greater clay content than that selected by Barn Swallows, while at 4 other localities the reverse was true. The variations in texture among nests within a site and among collecting sites are likewise statistically significant (P < .05). Intra-site variation accounts for 21%, 15.9%, and 17.8% of the total variation in percent sand, silt, and clay, respec- tively, while the variation among sites accounts for 47.3%, 49.9%, and 46%. Sand size. — The sand particles in the mud samples from the nests of both species was mostly of a small size (Fig. 3). In Cliff Swallow nests, 41.0 ± 2.2% of the sand was fine, 27.8 ± 1.3% was of medium size, 9.3 ± 0.6% was coarse, and 21.9 ± 2.0% was of very coarse size (values are means ± SE). Kilgore and Knudsen • ANALYSIS OF SWALLOW NESTS 565 Fig. 3. Distribution of sand sizes in mud samples from Cliff (left half) and Barn swallow (right half) nests at all 11 collecting sites. VC represents the proportion of very coarse sand, C the percentage of coarse sand, M the proportion of medium sand, and F the percentage of fine sand. Values are mean percentages for each component. The corresponding values in Barn Swallow nests were 44.9 ± 2.5%, 23.8 ± 0.9%, 8.0 ± 0.6%, and 21.7 ± 1.9%, respectively. The amount of sand in each of these size categories varied widely among collecting sites and nests at a particular site and between species (Fig. 3). The 2 species appeared to be selecting mud with different amounts of sand in each of these categories, but these species differences were not the same at most sites (Fig. 3). The amount of sand of a specific size occurring in the mud samples was primarily dependent on the locality. Variation in the amount of sand in these size categories among nests within a site and among sites is statistically significant (P < .05). Variation within a site accounts for 25.7 to 36% of the total variation in all size categories, while variation among localities accounts for 35.6 to 47.7%. Organic matter. — There was a moderate amount of organic matter in the mud of all Cliff and Barn swallow nests, although the actual amount and form of the organic matter varied with the species (Table 1). Organic mat- ter accounted for a mean of 6.6 ± 0.5% of the samples from Barn Swallow nests, but only 4.5 ± 0.4% of samples from the nests of Cliff Swallows. This difference is statistically significant, as are the differences among collecting 566 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Table 1 Organic Matter and Moisture Content in Mud Samples from Cliff AND Barn Swallow Nests Collecting sites Organic matter Moisture content Cliff Bam Cliff Bam 1 3.2 ± 0.2= ‘ 7.5 ± 1.7 1.8 ±0.1 1.9 ± 0.3 2 6.7 ± 0.7 8.5 ± 1.1 2.5 ± 0.2 3.5 ± 0.9 3 4.3 ± 0.7 8.0 ± 0.8 1.4 ± 0.1 2.0 ± 0.2 4 1.1 ± 0.2 5.2 ± 0.5 3.4 ± 1.6 1.3 ± 0.1 5 6.2 ± 1.1 5.6 ± 1.2 1.7 ± 0.4 1.8 ± 0.3 6 9.5 ± 1.3 5.2 ± 1.8 3.5 ± 0.6 1.6 ± 0.4 7 2.7 ± 0.2 4.8 ± 0.9 1.2 ± 0.1 1.9 ± 0.3 8 4.3 ± 0.3 6.3 ± 2.0 1.1 ± 0.1 1.7 ± 0.3 9 1.9 ± 0.4 4.7 ± 0.5 1.0 ± 0.2 0.9 ± 0.1 10 3.0 ± 0.6 5.5 ± 3.4 1.1 ±0.1 1.9 ± 0.3 11 5.8 ± 1.2 12.3 ± 2.3 1.3 ± 0.2 2.2 ± 0.5 “ Values are mean percentages ± the standard error of the mean. sites (P < .05). At most localities (9 of 11), mud from Barn Swallow nests contained more organic matter than samples from Cliff Swallow nests (Table 1). The organic matter in Cliff Swallow nests was primarily seeds and other fine particulate matter, while that in the nests of Barn Swallows consisted of coarse items, such as grass stems, horse hair, and feathers. Table 2 Textural Components and Sand Sizes in Mud Samples Selected from the Base and Rim of 5 Cliff and 7 Barn Swallow Nests Cliff Barn Rim Base Rim Base Textural component Sand Silt Clay 61.4 ± 2.8“ 25.0 ± 1.2 13.6 ± 1.8 62.4 ± 2.6 25.9 ± 1.5 11.6 ± 1.1 50.3 ± 1.7 33.7 ± 1.4 16.1 ± 1.3 48.7 ± 1.8 35.4 ± 1.4** 16.0 ± 0.8 Sand size Ver>- coarse Coarse Medium Fine 18.5 ± 7.1 9.8 ± 2.4 28.8 ± 2.7 42.9 ± 8.7 20.3 ± 5.8 10.4 ± 1.5 30.3 ± 2.6 39.0 ± 5.4 31.0 ± 5.7 8.8 ± 1.4 23.4 ± 1.4 36.9 ± 5.3 26.7 ± 4.5 8.9 ± 1.4 24.1 ± 1.4 40.3 ± 5.4'’ “ Values are mean percentages ± the standard error of the mean. ^ Difference between rim and base samples is statistically significant at the 95% probability level. Kilgore and Knudsen • ANALYSIS OF SWALLOW NESTS 567 Moisture content. — The mud samples from nests of both species contained very little water. The mean moisture content of the mud in Cliff Swallow nests was 1.9 ± 0.2%, while that in Barn Swallow nests was 1.8 ± 0.1% (Table 1). These interspecific differences are slight and are not statistically significant (P > .05), but the differences among sites are (P < .05). Locality differences were not correlated with nest placement, ambient humidity condi- tions, or the type of structure on which the nest was located. Intra-nest differences. — Samples taken from specific regions of the nest (i.e., base and rim) were very similar in their textural components and in sand size (Table 2). Intra-nest differences in the percentage of silt and fine sand in Barn Swallow nests are statistically significant at the 95% probability level; all other intra-nest differences are not statistically significant (P > .05). Hence, for all practical purposes the mud in these nests can be considered to be homogeneous. DISCUSSION It is clear from the preceding analyses that the 2 species of swallows select different materials for the construction of their nests. The primary differences among the variables measured are in the texture of mud and the amount of organic matter used in the nest. Cliff Swallows selected mud with a higher sand and lower silt content than that selected by Barn Swallows (Fig. 2). These differences are especially meaningful when one considers that both species presumably had access to the same mud source at each of the collecting sites. Emlen (1954) observed that Cliff Swallows may gather mud from sources 0.8 km or more from the nesting colony, that is, from distances twice the maximum distance separating nests of the 2 species at any of our collecting sites. Buss (1942) also found that Cliff Swallows would carry mud for nest building as far as 1.2 km. Despite these interspecific differences, both species appeared to be using mud within a restricted range of texture. Ninety-five percent of the textural observations are within the bounds of 41-77% sand, 17^3% silt, and 4-28% clay. These textural ranges include sandy loam, sandy clay loam, and loam soil types. Interestingly, Buss (1942) suggested that loam, silt loam, and clay loam might make the best mud for nest-building. The organic matter in the nests of the 2 species differed both in form and in quantity, the former being the more conspicuous difference. Less obvious were the differences in the amount of organic matter incorporated into the nest, which while seemingly slight are statistically significant. Emlen (1954) noted that the quality of mud used by Cliff Swallows varied considerably from locality to locality. Such variation among sites is apparent in the specific parameters of texture, sand size, moisture content, and organic 568 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 matter measured in this study. These differences might be expected to reflect differences in the temporal and geographic availability of particular mud (soil) types. However, the site differences described above cannot be cor- related with any major geographic or geologic features that affect the dis- tribution or abundance of soil types in western Montana. Based on analyses of differences among sites, there is no relationship between the major geo- graphic area of the site and the values of any of the variables. For example, the mean sand content of mud samples from Cliff Swallow nests along the Blackfoot River (sites 5, 6, 7, 8) range from 52% to 66.2%, the lowest and next to the highest values for all localities. The texture of the mud selected for nest-building by these species is un- doubtedly influenced by many factors which are important in the design and construction of a mud nest. The most obvious of these are (1) how well the mud adheres to the supporting substrate (adhesion), (2) how well the particles in the mud hold together (cohesion), (3) how well the dried mud withstands compressive and tensile stresses (strength), (4) how easily the mud is manipulated during construction (workability), and (5) how resistant the mud is to changes in volume (shrinkage and swelling). Interspecific differences in the texture of mud incorporated in the nest, especially where there are also obvious differences in nest shape, would suggest that some of the above factors may be more important than others in the construction of a nest with a particular design. To assess the importance of each of these factors and their relationship to nest design would require a detailed knowledge of the physical properties of the mud both in a plastic and nonplastic stage. Such information is currently unavailable. However, some insight into the importance of these factors in the selection of mud can be derived from the physical properties of adobe, other aggregate building materials (e.g., concrete and stucco), and soils. In soils, the properties of adhesion and cohesion are most affected by moisture content, although texture is also of some importance ( Baver et al. 1972). Cohesion increases with decreasing moisture content and is greater for clays (soils which are largely composed of small particles) than for other soils, while adhesion decreases with declining moisture content. At very low moisture contents, like those measured in the dried mud samples (1 to 6%), the increased cohesion is primarily due to a cementation effect between the dried particles (Baver et al. 1972). In view of the lack of significant inter- specific differences in the moisture content of the mud samples and since the interspecific textural differences, although statistically significant, are rather minor compared with those differences in soils which substantially affect cohesion and adhesion, it would appear that the adhesiveness and cohesiveness of the muds selected by both species are similar. Kilgore and Knudsen • ANALYSIS OF SWALLOW NESTS 569 The compressive and tensile breaking strengths of adobe and aggregate building materials are greatly affected by a number of factors, but most important are differences in texture and the inclusion of impurities (Eyre 1935, Mielenz 1965, Turneaure and Maurer 1908) . Addition of sand to adobe reduces its compressive and tensile strength ; the degree of reduction is directly proportional to the amount of sand added. When compared with “pure” adobe, adobe containing 75% sand (by weight) shows a 56% decline in compressive and a 50% decline in tensile strength, while addition of 25% sand only results in a 20% decline in compression and has no effect on tensile strength. Inclusion of organic matter (straw and manure) in adobe also decreases its strength, both in tension and compression (Eyre 1935). The actual reduction in strength resulting from inclusion of organic matter is largely dependent on the form of the organic matter and the amount added. However, the quantitative relationships between reduction in strength and form and the amount of organic matter is not clear (Eyre 1935). The in- creased sand content in the mud of Cliff Swallow nests effectively reduces its strength relative to that in Barn Swallow nests. Perhaps this reduction in strength accounts for the apparent “fragileness” of Cliff Swallow nests observed by Samuel (1971) and others. Workability of aggregate mixtures (those containing sand or gravel) is a difficult factor to quantify, primarily because it is dependent on many factors. However, the texture or proportion of the mixture is very important ( Mielenz 1965). In concrete and stucco an increased sand content improves the work- ability. Such “lean” mixtures are often used where strength and durability are not of major importance (Mielenz 1965). The higher sand content in the mud of Cliff Swallow nests would undoubtedly improve its workability, therefore facilitating the construction of a nest with a complex shape. Volume changes in the dried mud may be brought about by changes in moisture content (during drying and afterwards) and temperature and are affected by texture (Eyre 1935). Adobe mixed with 50% sand (by weight) has coefficients of expansion and contraction that are % those of pure adobe (Eyre 1935). Addition of straw to adobe also reduces the coefficients of expansion and contraction, but only by about 15%. Because of their relatively high contents of organic matter and sand, and since most nests are sheltered from direct rain, volume changes in the mud are probably not great. The differences in materials (texture of the mud and organic matter) used by the 2 species appear not to be fortuitous, but instead to be related to the factors important in the construction of mud nests and to the differences in the complexity of design. The mud selected by Cliff Swallows, with its high sand content, is more easily manipulated mechanically which assists in the building of a more complex nest. Improved workability of the mud, however, 570 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 occurs at the expense of strength and perhaps cohesion, which may explain why Cliff Swallow nests are noticeably more “fragile.” The inclusion of conspicuous amounts of organic matter (grass, feathers, and hair) in the nest of Barn Swallows may improve its cohesive nature, in that pieces of mud are bound together, which may further explain why these nests appear more durable in comparison to those of Cliff Swallows. The retort-shaped design of Cliff Swallow nests probably prohibits the use of such large pieces of organic matter. SUMMARY There are statistically significant differences in the composition of the mud selected by Barn and Cliff swallows for nest building. Furthermore, these differences appear to be related to the complexity of nest design. Cliff Swallows select mud with a higher sand and lower silt content than that used by Barn Swallows. A high sand content, based on the physical properties of other composite mixtures, probably improves the ease with which the mud may be manipulated and shaped into a retort-shaped nest. However, this increased workability of the mud is accompanied by a reduction in strength. There are also statistically significant differences in the amount and form of organic matter incorporated in the mud of these nests. Mud from Cliff Swallow nests contains small amounts of seeds and other fine particulate matter, while that from the nests of Barn Swallows contains large amounts of coarse items, such as grass stems, horse hair, and feathers. The design of Cliff Swallow nests probably precludes the inclusion of such bulky materials, which in Barn Swallow nests probably serve to bind together pieces of mud. The type and character of the mud used by these species varied from locality to locality, but within relatively narrow limits. The factors affecting the construction and design of mud nests le.g., adhesion, cohesion, etc.) may well place constraints on the type of mud which can be used. ACKNOWLEDGMENTS We thank Gary C. Packard, James D. Rising, Andrew L. Sheldon, and Charles H. Daugherty for their valuable advice in the preparation of the manuscript. LITER-\TURE CITED Bavkr, L. 1)., H. Gardner, and W'. R. Gardner. 1972. Soil physics. John Wiley & Sons, Ine., New York. Bouyoucos, G. J. 1936. Directions for making mechanical analyses of soils by the hydrometer method. Soil Sci. 42:225-229. Buss, I. 0. 1942. A managed Cliff Swallow colony in Southern Wisconsin. Wilson Bull. 54:153-161. Emlen, J. T., Jr. 1952. Social behavior in nesting Cliff Swallows. Condor 54:177-199. . 1954. Territory, nest building, and pair formation in the Cliff Swallow. Auk 71:16-35. Eyre, T. T. 1935. The physical properties of adobe used as a building material. Univ. New Mexico Bull., Engr. Series 1:3-32. Kilgore and Knudsen • ANALYSIS OF SWALLOW NESTS 571 Jackson, J. A. and P. G. Burchheld. 1975. Nest-site selection of Barn Swallows in east-central Mississippi. Am. Midi. Nat. 94:503-509. Jackson, M. L. 1958. Soil chemical analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. Mayhew, W. W. 1958. The biology of the Cliff Swallow in California. Condor 60:7-37. Mielenz, R. C. 1965. Concrete, as a modern material. Pp. 259-363 in Modern Materials, Advances in Development and Applications, Vol. 5 (B. W. Gonser, ed.). Academic Press, New York. Samuel, D. E. 1971. The breeding biology of Barn and Cliff swallows in West Virginia. Wilson Bull. 83:284-301. SoKAL, R. R. AND F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Francisco. Turneaure, F. E. and E. R. Maurer. 1908. Principles of reinforced concrete construc- tion. John Wiley & Sons, New York. U.S. Department of Agriculture. 1951. Soil survey manual. Handbook 18. DEPT. OF ZOOLOGY, UNIV. OF MONTANA, MISSOULA 59801. ACCEPTED 1 AUG. 1976. REQUESTS FOR ASSISTANCE Mexican locality records. — A comprehensive bibliography and gazetter of localities concerning birds in Mexieo is being prepared. We will include all papers dealing with Mexican birds and birds recorded in Mexico. Authors wishing to have material included should send reprints of their materials to Mario and Isabel Ramos, Bell Museum of Natural History, University of Minnesota, 10 Church St. SE, Minneapolis, MN 55455. Artificial nest structure literature. — An international bibliography on the use of artificial nest structures for bird research and management is being compiled. All contributions will be acknowledged in publication. Please send reprints and title lists to Jeffrey B. Froke, National Audubon Society, Box 157, San Juan Capistrano, CA 92675. PRODUCTION AND SURVIVAL OF THE VERDIN George T. Austin A review of avian demography (Ricklefs 1973) demonstrates the dearth of knowledge on this subject. Although certain demographic parameters are relatively well known for a wide variety of species, data are generally lacking for their seasonal, annual, and geographic variability. These, including population densities, nesting season, clutch size, and nesting success, are straight forward and can be obtained with relative ease. Survival and annual recruitment are also of interest, but are difficult to determine under field conditions. If first year individuals (/) can be distinguished from adults (A), the ratio of the 2 can be used to calculate annual mortality and recruitment of first year birds into the breeding population given the assumptions of constant population size and no collecting bias. In many species, the first prebasic molt is incomplete (e.g., Dwight 1900), providing a basis of distinguishing first year birds from adults. Although it has long been recognized that I/A ratios can be used to determine annual survival ( Emlen 1940, Snow 1956), th is method has not been widely used. The Verdin { Auriparus flaviceps) is an ideal species for a demographic study as first year birds can be distinguished from adults through at least March (Austin and Rea 1971), the species is nonmigratory and relatively sedentary, and general aspects of its life history are known (Moore 1965, Taylor 1967, 1971, Austin 1976). In this paper, I will analyze the demo- graphic data available for the Verdin from the literature and my own studies in Arizona and Nevada. Geographical and/or seasonal trends will be exam- ined and estimates of mortality and annual recruitment will be presented. METHODS Clutch size was determined for 144 nests in Pima County, Arizona ( 1970-71 ) , and 87 nests in Clark County, Nevada (1959-70, 1976). Additional data were obtained through the North American Nest Record Card Program, Cornell University (NANRCP). The clutch was considered complete when the number of eggs remained constant for 2 or more successive days. Verdins lay 1 egg per day on successive days until the clutch is complete (Moore 19'i5, Taylor 1971, this study). Nesting success was obtained by periodically (1-3 day intervals) inspecting active nests. Hatching success is percent of eggs to hatch, fledging success is percent of eggs to produce young which fledge, nestling success is percent of hatchlings to fledge, and nest success is percent of breeding nests built to fledge at least one young. I aged museum specimens of Verdins according to the methods of Austin and Rea (1971). Briefly, at fledging, Verdins are grayish-headed and lack the rich chestnut lesser primary coverts characteristic of adults. During the first prebasic molt, the upper greater 572 Austin • VERDIN DEMOGRAPHY 573 primary coverts and the proximal 2-5 primaries are retained. These feathers appear more faded and worn than the newer distal primaries. This difference was reliable through at least late March or April of the following spring after which feather wear was too great for reliable determinations. A bird was considered a first year bird if (1) it was gray-headed (before first prebasic molt), (2) it was undergoing first prebasic molt, or (3) it had retained upper greater primary coverts and proximal primaries after molting. After April, yellow-headed birds which had not begun prebasic molt were considered adult. The latter included some birds which were not as yet one year of age hut had entered one breeding season since the first prebasic molt. RESULTS Relative abundance. — The population density of the Verdin in the United States in winter is greatest in the Big Bend region of Texas, southern Arizona (excluding eastern Arizona), and adjacent southeastern California (Fig. 1). Distribution and relative abundance appears similar during summer (data from U.S. Fish and Wildlife Service breeding bird survey courtesy C. S. Robbins). Population densities during the breeding season in suitable habitat averaged about 8 pairs/40 ha in the San Antonio region of Texas (American Birds breeding bird censuses 1968-1972, 1974), 2-3 pairs/40 ha in the Chihuahuan Desert of western Texas and New" Mexico (Dixon 1959, Raitt and Maze 1968), about 8 pairs/40 ha in southern Arizona and southeastern California (range 3-16; American Birds breeding bird censuses 1941, 1942, 1965, 1969-1971, Hensley 1954, Taylor 1967, Anderson and Anderson 1973, Tomoff 1974, this study ) , and about 4 pairs/40 ha in southern Nevada ( Austin 1970, Miller 1974). Clutch size. — In southern Nevada, modal clutch size remained at 4 through- out the season although mean clutch size decreased (Table ll. In southern Arizona, mean clutch size decreased through the breeding season with a modal size of 4 in March and April and 3 for the remainder of the season. Similarly, Moore (1965) and Taylor (1971) reported a decrease in clutch size through the breeding season in New" Mexico and central Arizona, re- spectively. There also appears to be an east to west and north to south decrease in clutch size (Table 1). Two sets of eggs from Texas contained 6 eggs each (collection at Oregon State Univ., in Taylor 1967), although the usual size in Texas was reported as being 4 ( Attw ater 1892) or 3 to 6 ( Oberholser 1974, NANRCP ) . The average clutch in Texas and New Mexico contained 3.75 eggs, in Arizona 3.41, and in central Baja California, Mexico, 2.41. Only 1 clutch of 100 examined in the latter location contained 4 eggs (Bancroft 1930) . The average southern Nevada and central Arizona clutch was nearly 0.5 egg larger than those in southern Arizona. Breeding season. — The breeding season extends generally from mid-March 574 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 Fig. L Relative density of Verdin populations during winter (values are numbers/10 party hours averaged for the years 1965-1974 from American Birds Christmas bird count). through mid-June over much of the Verdin’s range (Bent 1946, Hensley 1959, Moore 1965, Taylor 1971, this study) with the earliest egg date of 4 March (Bent 1946). In desert grassland in southern Arizona, I have found fresh clutches as late as 23 July with young fledging after mid-August. Near Tucson, a fresh clutch was found in early August (P. Gould, pers. comm.). Table Clutch Size of 1 THE Verdin Mean clutch size ( N ) Locality Early nests Late nests Total Source Texas _ - 4.06 (16) Taylor 1967, NANRCP New Mexico 3.71 (14) 3.00 (6) 3.50 (20) Moore 1965 Southern Arizona - - 4.00 (16) Hensley 1959 Southern Arizona 3.77 (48) 3.02 (96) 3.27 (144) this study Central Arizona 4.13 (16) 3.42 (12) 3.82 (28) Taylor 1971 Southern Nevada 3.83 (53) 3.68 (34) 3.77 (87) this study Baja California - - 2.41 (est.) Bancroft 1930 Austin • VERDIN DEMOGRAPHY 575 Length Table 2 OF Breeding Season in the Verdin Location Number of clutches laid during Index of season length ( months ) Source March April May June July Aug. Texas 1 5 6 7 1 0 3.95 NANRCP Southern Arizona 1 3 5 5 0 0 3.50 NANRCP Saguaro-Palo Verde 22 86 80 23 0 1 3.44 this study Mesquite grassland 0 11 14 18 7 0 3.79 this study Central Arizona 6 10 8 4 0 0 3.80 Taylor 1971 Southern Nevada 4 102 65 11 0 0 2.58 this study Texas birds may occasionally breed into September (Oberholser 1974). Further south in Mexico the breeding season was apparently well underway by late March (Short and Crossin 1967) and extended into August (Amadon and Phillips 1947, Short 1974) or later (Brewster 1902, but see Banks 1963). This longer breeding season in Mexico was further suggested by Bancroft (1930) who found many fresh clutches in mid- June. An estimate of season length adjusted for peaks in breeding activity is derived from the index of “equally probable months for nesting” (Mac- Arthur 1964) : months — e - Spj log Pi where pi = the number of clutches initiated in each month. These data for the Verdin are presented in Table 2. Season length in southern Nevada was nearly a month shorter than in Arizona. This may be due to the lack of a distinct summer rainy season in the Mojave Desert. Here the peak of the breeding season is in April compared to a peak extending over a 2-month period in southern Arizona (Table 2). Number of broods. — In Arizona, most pairs had no more than one successful brood per year although some with successful first nests attempted a second nesting. On the average, % of the pairs built 2 nests. As many as 4 breeding attempts were made in a season and as many as 2 were successful (Taylor 1967, this study) . The length of one successful cycle was approximately 43 days (6 days building, 3 days laying, 16 days incubation, 18 days nestling period). Second and successive cycles were shorter as the male constructs succeeding nests during the previous nest period. After a successful nesting, an average of 2 days elapsed between fledging and the first egg of the next clutch (r^) ; after a failure, an average of 4 days elapsed (r/) (Taylor 1967, this study). 576 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 Table 3 Seasonal Changes in Nesting Success of the Verdin During 1970 and 1971 IN Pima County, Arizona Month Hatching (%) Fledging (%) Nestling (%) Nest (%) N eggs N nests March 77.8 72.2 92.8 88.8 36 9 April 85.6 71.2 83.2 82.9 146 41 May 80.0 70.5 88.2 80.0 95 30 June 48.5 29.4 60.6 36.0 68 25 July 35.3 35.3 100.0 42.9 17 7 Length of an unsuccessful cycle can be calculated (assuming a constant probability of nest loss during the nest period) with Ricklefs’ (1973) equation ^ _Q->nPT ’ mQ where P is the proportion of successful nests, is 1 - P, T is length of egg laying, incubation, and nestling periods and in is the average daily mortality rate (see below). Length of an unsuccessful attempt averaged 19.3 days in southern Arizona (all data combined) and 17.2 in southern Nevada. With these an average nesting attempt can be calculated by T — P{T +rs) Q{Tf Tf) (Ricklefs 1970). This value was 36.9 days in southern Arizona and 31.5 days in Nevada. Thus about 3.7 and 2.1 broods could be attempted, on the average, in the 2 localities given the breeding season length in Table 2. Nesting success. — Several factors affect nesting success in the Verdin. Nest orientation was shown to be an important variable (Austin 1976) as was found for the Cactus Wren iCampylorhynchus brunneicapillus) (Austin 1974) . Success also tended to decrease from beginning to end of the breed- ing season (Table 3). Considerable variation existed in nesting success among the several localities for which data were available (Table 4), possibly a reflection of annual differences, although 2 successive years in southern Arizona were very similar. Overall fledging success for all studies reported in Table 4 was 59% which is within the range given for enclosed nesting species (Nice 1957). Daily mortality rates calculated by -doge P) m T Austin • VERDIN DEMOGRAPHY 577 Table 4 Nesting Success of THE Verdin Location Success ( % ) N e3gs N nests Source Hatching Fledging Nestling Nest Texas 56.3 34.4 61.1 33.3 32 12 NANRCP New Mexico - 56.9 - - 48 18 Moore 1965 Southern Arizona 96.4 82.1 85.5 80.0 56 15 Hensley 1954 Southern Arizona Saguaro-Palo Verde 73.5 65.5 89.2 76.4 264 72 this study Mesquite grassland 61.5 44.1 71.6 54.2 143 48 this study Central Arizona 48.8 29.1 59.5 36.0 86 25 Taylor 1971 Southern Nevada 85.9 72.4 84.2 86.5 199 52 this study Total 72.7 59.1 81.5 67.4 780 224 Average daily mortality rate (%) 1.67 1.42 1.13 1.06 (Ricklefs 1969), averaged 1.42% (0.53-3.33) from laying to fledging for eggs and 1.06% (0.39-2.97) for nests. This indicates that within-nest loss exceeded total losses and was true for all samples except for the Texas sample (Table 4) and Hensley’s (1954) sample. Within-nest losses exceeded whole nest losses during both the egg (0.57%) and nestling (0.26%) stages for all samples combined. Individual losses during the incubation period averaged greater (by 0.54%) than during the nestling period although 2 of the 6 samples (Hensley, southern Nevada) showed the opposite trend (Table 4). Similarly, whole nest losses averaged greater during incubation (0.23%). These low values are typical of enclosed nesting species, and the resultant vectors indicate that hatching failure and possibly desertion are major factors of mortality (Ricklefs 1969). I was able to assign probable factors responsible for mortality of 105 individuals (65 eggs, 40 nestlings). Hatching failure accounted for 24.8% and desertion for 21.9% of all mortality. These factors accounted for only 8.2% and 5.2% of total mortality in 6 species summarized by Ricklefs (1969). Predation accounted for 58.4% of nest mortality in these species compared to 19.0% in the Verdin. Success and productivity comparisons between 3 and 4 egg clutches are presented in Table 5. Four egg clutches produced more young per nest than 3 egg clutches. The difference is greater in early nests than in late nests. In early nests, 4 egg clutches were more efficient in fledging young from eggs which hatched than were 3 egg clutches. In late nests, however, 3 egg clutches were somewhat more efficient in fledging young. The relatively high pro- 1 578 the WILSON BULLETIN • Vo/. 89, No. 4, December 1977 Table 5 Comparison of Production and Nesting Success Between 3 and 4 Egg Clutches OF THE Verdin Early nests clutch size Late nests clutch size 3 4 3 4 Number of nests % Nests with hatched eggs % Nests with fledged young % Nests hatch, young/fledg. young % Eggs hatching % Eggs producing young which fledged % Eggs hatch./producing yg. which fledged Mean no. of fledglings per nest in which eggs hatched Mean no. of fledglings per nest 13 69 81 38 85 94 77 87 69 88 65 74 82 94 85 85 85 85 68 74 62 74 55 56 73 87 81 76 2.2 3.1 2.2 2.6 1.9 3.0 1.7 2.2 portion of 4 egg clutches among late nests may be of value during favorable years. Post- fledging mortality. — Fledgling Verdins remained dependent on the parents for 2 or more weeks post-fledging. In southern Arizona, I was able to follow 21 fledglings of which 17 survived this period. Mortality rates averaged 0.87% per day (1.05% per day during the first 15 days). The immediate post-fledging period has previously been shown to be the time of heaviest mortality of birds out of the nest (Lack 1966, Ricklefs 1973). Immature adult ratios. — Immature to adult ratios for 286 specimens from southwestern United States are presented in Table 6. These were high during and immediately after the breeding season, decreased rapidly thereafter and leveled off in November at a value averaging about 0.67. The change re- flected a greater mortality rate of first year birds than adults until mid-winter. Thus, the breeding population of Verdins was composed approximately of 60% birds in at least their second season. Table 6 Numbers of Immatures (I) and Adults (A) Among Specimens Examined and Ratio OF Immatures to Adults May— June July-Aug. Sept.— Oct. Nov.-Dee. Jan.-Feb. March I 32 40 31 24 17 4 A 18 22 29 39 24 6 I/A 1.78 1.82 1.07 0.62 0.71 0.67 Austin • VERDIN DEMOGRAPHY 579 Numbp:r of Table 7 Young Fledged by Verdins Location Number of pairs X no. young fledged per nest X no. young fledged per individual Source Nevada-Arizona^ - 3.61 1.81 this study Nevada" 9 3.44 1.72 this study Arizona" 20 3.40 1.70 this study Arizona^ 8 2.88 1.44 Taylor 1967 New Mexico^ 14 3.78 1.89 Moore 1955 1 Data for Clark County, Nevada and Pima County, Arizona (1969-1971) assuming 50% of pairs build late nests (Moore 1965, this study) and means of 2.74 and 1.73 young fledged per early and late nest, respectively ( this study ) . ^ Known pairs followed throughout season. ^ Data for 1965 excluding 4 pairs which left study area shortly after laying. * Data for 1965 assuming as stated that 50% of pairs renest and mean of 2.64 young fledged per nest from first nests and 2.27 young fledged per nest from second nests (Moore 1965). Because first year Verdins before the first prebasic molt are distinguishable from adults in the field, a bias by collectors may result. As a check on the accuracy of the I/A ratio during and just after the breeding season, I com- piled data on average production per pair (Table 7). This averaged 1.71 young fledged per adult and was very similar to the I/A ratio (Table 6). Another check was made by multiplying mean clutch size (3.55) by mean fledging success (0.591) and number of broods attempted (1.5) which gave a similar value of 1.57 nestlings per individual (3.15 per pair). Thus, for the Verdin, the I/A ratio (in May-August) appears to be a good indicator of successful production. For a sample of 54 specimens for October to March from Baja California, Mexico, the I/A ratio was 1.16 (29/25) and for 23 specimens from the main part of Mexico a ratio of 0.92 (11/12). These ratios were considerably higher than for southwestern U.S. populations. First year and adult mortality. — From the I/A ratio data, mortality and survival rates can be calculated. Assuming breeding populations are stable and adult mortality rates are constant throughout the year, mortality rates of first year birds can be calculated on a seasonal basis. Because I assumed stable populations and that the proportion of first year birds in the breeding population equaled adult losses, annual adult mortality was 40%. Annual mortality for first year birds was 75%. This complements the 24% first year survival predicated from known production (1.7 fledglings per adult) and adult mortality (40%). Mortality rates of first year birds from 1 July to 1 December averaged 0.81% per day. From 1 December to 1 May, mortality averaged 0.13% per 580 THE WILSON BULLETIN • Vol 89, No. 4, December 1977 day similar to the adult rate of 0.14% per day. The mortality differential between first year birds and adults can be calculated by: loge R{t)/Rio) mA - mi ^ (Ricklefs 1973) where Rio) is the 1/A ratio at some time o and Rit) is the I/A ratio after some period of time iT). This value for the Verdin was nearly 20% per month until 1 December. The mortality differential from 1 July to 1 May was 9.7% per month. The samples from Mexico indicated a greater annual adult mortality (approaching or exceeding 50%) than further north. DISCUSSION Among passerine birds, mortality rate generally decreases at each stage of the life cycle from egg to adult (e.g.. Lack 1966, Ricklefs 1969, 1973). These data for the Verdin are summarized in Table 8. As noted by Ricklefs ( 1973) post-fledging survival rate appears higher in species with long nestling periods reflecting their greater maturity at fledging. In many species post- fledging mortality rate is lower or about equal to nestling mortality rate. Most species in which post-fledging mortality rate approached or exceeded nestling mortality rate, nest in enclosed nests or cavities. Such species are well-known for their high nesting success (Nice 1957). The first week out of the nest may be the most critical for some species ( Snow 1958 1 and newly fledged young may be especially susceptible to severe weather ( Smith 1967). First year survival rate is usually greater than in the nest and lower than annual adult survival rate. As was true for thrushes and tits ( Lack 1946, 1966) survival rates of the Verdin approached adult levels after 6 months. The adult survival rate of 60% per year is at the upper end of the range for passerines given by Lack (1954). In addition, first year survival was about 42% of adult annual survival and considerably higher than the 25% of adult survival suggested for first Table 8 Mortality and Survival Rate of the Verdin Egg staged Nestling stage Post- fledging stage First 6 months- Second 6 months Adult Mortality (% per day) 1.69 1.05 0.87 0.81 0.13 0.14 % Eggs to survive to the end of this period 72.5 59.9 46.9 13.9 11.0 — Number of days 19 18 28 180 180 — 1 Including laying period. - Including post-fledging stage. Austin • VERDIN DEMOGRAPHY 581 year small land birds (Ricklefs 1973) and may reflect the lower level of productivity in arid temperate regions (Ricklefs 1973). The increase in I/A ratio with decreasing latitude and clutch size was counter to the trend for Blue Tits {Parus caeruleus) (Snow 1956) and Rough- winged Swallows {Stelgidopteryx ruficollis) (Ricklefs 1972). The data for the Verdin from Mexico suggest that nesting success or first year survival is greater in Mexico or that the nesting season is substantially longer (as indicated above) and productivity is greater to offset a higher rate of adult mortality. SUMMARY Demographic parameters of the Verdin are discussed. Clutch size displayed a decreasing gradient from east to west and north to south. Breeding season length decreased from south to north. Nesting success was greatest during the first 2 months of the breeding season and varied with locality and year. Based on immature/adult ratios, annual adult survival was 60% in southwestern U.S. and 50% in Mexico. Survival increased during the first year of life and approached adult levels at about 6 months. ACKNOWLEDGMENTS I thank museum curators for access to specimens in their respective collections: A. M. Rea (Prescott College and private collection), J. R. Jehl (San Diego Society of Natural History), J. R. Northern and K. Stager (Los Angeles County Museum), R. J. Raitt (New Mexico State Univ.), E. L. Smith and S. M. Russell (Univ. of Arizona), R. D. Ohmart (Arizona State Univ.), and J. S. Miller and W. G. Bradley (Univ. of Nevada, Las Vegas). The North American Nest Record Card Program supplied data which filled certain gaps. Their assistance, especially that of R. Pantle, is gratefully acknowledged. I thank R. E. Ricklefs for critical comments on one draft of the manuscript. Part of the field work was supported by US/IBP Desert Biome Program under National Science Foundation Grant GB 15886 at the University of Arizona. LITERATURE CITED Amadon, D. and a. R. Phillips. 1947. Notes on Mexican birds. Auk 64:576-581. Anderson, A. H. and A. Anderson, 1973, The Cactus Wren. Univ. of Arizona Press, Tucson. Attwater, H. P. 1892. List of birds observed in the vicinity of San Antonio, Bexar County, Texas. Auk 9:337-345. Austin, G. T. 1970, Breeding birds of desert riparian habitat in southern Nevada. Condor 72:431-436. . 1974. Nesting success of the Cactus Wren in relation to nest orientation. Condor 76:216-217, . 1976. Behavioral adaptations of the Verdin to the desert. Auk 93:245-262. AND A. M. Rea. 1971. Key to age and sex determination of Verdins. Western Bird Bander 46:41. Bancroft, G. 1930. The breeding birds of central lower California. Condor 32:20-49. Banks, R. C. 1963. Birds of Cerralvo Island, Baja California. Condor 65:300-312. Bent, A. C. 1946. Life histories of North American jays, crows, and titmice. U.S. Natl. Mus. Bull. 191. 1 582 the WILSON BULLETIN • VoL 89, No. 4, December 1977 Brewster, W. 1902, Birds of the cape region of Lower California. Bull. Mus. Comp. Zool. 41:1-241. Dixon, K. L. 1959. Ecological and distributional relations of desert scrub birds of western Texas. Condor 61:397-409. Dwight, J., Jr. 1900. The sequence of plumages and moults of the passerine birds of New York. Ann. New York Acad. Sci. 13:73-360. Emlen, J. T., Jr. 1940. Sex and age ratios in survival of the California Quail. J. Wildl. Manage. 4:92-99. Hensley, M, M. 1954. Ecological relationships of the breeding bird population of the desert biome of Arizona. Ecol. Monogr. 24:185-207. . 1959. Notes on the nesting of selected species of birds of the Sonoran Desert. Wilson Bull. 71:86-92. Lack, D. 1946. Do juvenile birds survive less well than adults? Br. Birds 39:258-264. . 1966. Population studies of birds. Clarendon Press, Oxford. MacArtiiur, R. H. 1964. Environmental factors affecting species diversity. Am. Nat. 98:387-397. Miller, J. S. 1974. The avian community structure of Las Vegas Wash, Clark County, Nevada. M.S. thesis, Univ. Nevada, Las Vegas. Moore, D. R. 1965. Breeding biology and nest usage of the Verdin in southern New Mexico. M.S. thesis, New Mexico State Univ. Nice, M. M, 1957. Nesting success of altricial birds. Auk 74:305-321. Oberiiolser, H. C. 1974. The bird life of Texas. Univ. of Texas Press, Austin. Raitt, R. j. and R. L. Maze. 1958. Densities and species composition of breeding birds of a creosote hush community in southern New Mexico. Condor 70:193-205. Ricklefs, R. E. 1969, An analysis of nesting mortality in birds. Smithson. Contrib. Zool. No. 9. . 1970. The estimation of a time function of ecological use. Ecology 51:508-513, 1972. Latitudinal variation in breeding productivity of the Rough-winged Swal- low. Auk 89:826-836. -. 1973. Fecundity, mortality and avian demography. Pp. 366-447, in Breeding Biology of Birds ( D. S. Farner, ed.), Natl, Acad. Sci., Washington, D.C. Short, L. L. 1974. Nesting of southern Sonoran birds during the summer rainy season. Condor 76:21-32. AND R. S. Crossin. 1967. Notes on the avifauna of northwestern Baja California. Trans. San Diego Soc. Nat. Hist. 14:281-300. Smith, S. M. 1967. Seasonal changes in the survival of the Black-capped Chickadee. Condor 69:344-359. Snow, 1). W. 1956. The annual mortality of the Blue Tit in different parts of its range. Br. Birds 49:174-177. . 1958. The breeding of the Blackbird Turdus merula at Oxford. Ibis 100:1-30. Taylor, W. K. 1967. Breeding biology and ecology of the Verdin, Auriparus flaviceps (Sundevall). Ph.D. thesis, Arizona State Univ., Tempe. . 1971. A breeding biology study of the Verdin, Auriparus flaviceps (Sundevall) in Arizona. Am, Midi. Nat. 85:289-328. Tomoff, C. S. 1974. Avian species diversity in desert scrub. Ecology 55:396-403. DEPT. OF BIOLOGICAL SCIENCES, UNIV. OF NEVADA, LAS VEGAS, LAS VEGAS, NV 89154. ACCEPTED 1 AUG. 1976. MALE BEHAVIOR AND EEMALE RECRUITMENT IN THE RED-WINGED BLACKBIRD Patrick J. Weatherhead and Raleigh J. Robertson In most species of birds females devote more energy to reproduction than do males. Consequently natural selection has favored increased discrimination on the part of the females when seeking a mate and thus, in such species, selection of mates is expected to be determined by female choice (Trivers 1972). This would be particularly true in highly polygynous species where the male role in reproduction is very limited. The factors on which female choice can be based are individual differences among males and differences among the territories of males (Orians 1969). When the male contribution to reproduction is restricted to supplying genetic information, or when pair bonds are made away from breeding grounds, female choice must rely solely on differences in individual characteristics (Verner 1964). However, as the male contribution increases to include the maintenance of a territory which influences reproductive success through provision of food or nest sites, then female choice should also be influenced by territory quality (Orians 1969). This latter point has formed the basis for the Orians-Verner model for the evolution of polygynous mating systems (Selander 1972, Wilson 1975). The Orians-Verner model attributes polygynous mating to the existence of major differences in territory qualities. When these differences are great enough, a female will improve her chances of successfully reproducing by choosing to share a high quality territory with another female rather than by choosing to be the sole female in a poor quality territory. Implicit in this model is that the differences in males’ individual characteristics will be reflected by differences in the qualities of their territories, since the fittest males should be best able to defend superior territories. It follows then that female choice will be meditated by territory quality and therefore behav- ioral differences in males should not influence female recruitment. One might further predict that selection would favor a reduction in recruitment and courtship behavior since it should lack any advantage in attracting mates while making the male more vulnerable to predators. From a study of Red-winged Blackbirds {Agelaius phoeniceus) we have shown that female choice as reflected by harem size, did not correlate with territory quality (Weatherhead and Robertson 1977). Some females therefore appeared to be making poor choices in terms of territory quality. We concluded that these choices must result from behavioral influences of territorial males not associated with the quality of their territories. Although 583 5S4 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 such choices lower reproductive success in the Fi generation, we developed a model which showed that as long as those losses did not exceed a critical maximum, the females would ultimately benefit through their male off- springs’ superior ability to recruit a harem (Weatherhead and Robertson, in press). The purpose of the present study was to determine to what extent territorial males differ in their behavior towards females and to establish whether or not such differences reflect success at recruiting females indepen- dent of territory quality. We previously reported (Weatherhead and Robertson 1977) that female reproductive success was negatively correlated with the density of females within male territories and positively correlated with ter- ritory quality with respect to nest site suitability. We predicted, therefore, that if territory quality could be held relatively constant then the density of females in a male’s territory would be an indication of his recruitment ability. This technique is useful in that it allows one to compare recruitment ability relative to territory size. It is also necessary if one is to be able to distinguish between behavioral characteristics related to recruitment and those related to territory acquisition. METHODS The area chosen for this study was Cow' Island Marsh located near the Queen’s University Biology Station, 40 km north of Kingston, Ontario. The vegetation of the marsh is predominantly cattail (Typha latijolia) , bordered by sweet gale {Myrica gale) and alder (Alnus rugosa) . The marsh is approximately 1 ha in area. In order to quantify differences between males, behavioral tests were conducted from the onset of breeding in early May until nesting terminated in late June. Tests were performed between 0900 and 1130 two days a week. A single test consisted of a 5 min presentation of a normally postured, freeze-dried female Red-winged Blackbird to a territorial male. The model was attached to the top of a wooden pole positioned so the model was just above the vegetation, close to the center of a territory to ensure that the beliavior recorded was that of the territory holder. To avoid behavior associated with nest defense the model was never placed within 5 m of an active nest. The ol)server was positioned outside the territory in which the test was conducted, using a portable burlap blind for concealment early in the season and relying on the new growth of vegetation when it became available. The events of the trial were recorded on a portable tape recorder. No male was tested more than once on any given day and the order in which males were tested was varied each day. The scoring system used was similar to that used by Robertson and Norman (1976) in ranking host aggression to cowbirds {Molothrus ater) . It was first necessary to rank tlie behavioral acts in order of increasing intensity of courtship. The basis for the ranking came primarily from the work of Nero (1956a, b) although some intuitive judgments based on field observations had to be made. Such was the case wdien a departure from strict ordinal ranking was made in scoring an act thought to be of much higher intensity than the act ranked below it. The list of acts and their respective scores is presented in Table 1. The distinction between close and distant acts is that the latter occur further than 3 m from the model. Weatherhead and Robertson • RED-WINGED BLACKBIRD BEHAVIOR 585 Table 1 Scoring System for Behavioral Trials Score Act. 1 distant silent observation 2 close silent observation 3 distant observation with “check” 4 close observation with “check” 5 distant display flight (6 sec) 6 distant “song-spread” (3 sec) 7 distant crouch or strutting 8 close display flight (6 sec) 9 close “song-spread” (3 sec) 10 close crouch or strutting 15 aggression to other females* 17 pecking at model 20 attempted copulation * This was considered a recruitment act since it discouraged aggression toward a potential mate by those females aheady recruited. The intensity of courtship is also a function of the length of time various acts were elicited during the 5 min trial. Thus, a duration score was also assigned to acts as follows; 1 for acts elicited for less than 5 sec; 2 for acts lasting between 5 sec and 1 min; 3 for acts lasting between 1 and 3 min; 4 for acts lasting greater than 3 min. For discrete acts which were recorded on the basis of how often they were elicited rather than for how long, their duration score in a trial was the mean time required for that act (given in brackets in Table 1) times the number of elicitations. The score for each test was then computed by multiplying each act score by its duration score and then summing these values. In addition to the model testing, general reproductive information was collected throughout the breeding season. Twice weekly the marsh was thoroughly searched for new nests while the progress of nests found previously was recorded. Territory boundaries were determined as soon as the males became resident. This was accomplished by observing individual movements, use of song posts, and points of conflict between males. Following the breeding season the marsh was surveyed and mapped and territory areas were computed using a polar planimeter. Harem sizes were then determined from nest records as the maximum number of active nests in a territory at any given time during the breeding season. RESULTS Figure 1 is a map of Cow Island Marsh indicating the territorial boundaries of the 11 resident males involved in the study. The range of harem sizes and territory areas (see Table 2) are similar to those found in other Red-wing studies (Holm 1973, Goddard and Board 1967, Case and Hewitt 1963, Orians 1961). We conducted 61 tests, with the number of tests per individual 586 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Fig. 1. Territorial males in the study of area. Solid lines indicate dry land (shaded), wavy lines separate cattail from open water and shrub vegetation. Broken lines indicate territory boundaries. ranging from 4 to 9. Each male was not tested an equal number of times because some males were off their territories on several occasions when they were to be tested. No score was assigned to such trials although a score of zero was given if the male was present at the initiation of a trial and left without eliciting any courtship behavior. Whether such zero scores actually indicated a complete lack of motivation by the male or only a failure to see the model is uncertain. Therefore, the use of such scores in the analysis has been minimized as will be indicated. Before comparisons could be made between the trial scores of individual males and their respective recruitment success, it was first necessary to test the validity of the scoring system. We felt that if the scores obtained in the Weatherhead and Robertson • RED-WINGED BLACKBIRD BEHAVIOR 587 Territory Parameters Table 2 AND Test Scores for all Males Tested Male Territory Area (m-) Harem Size Area Per Female (m^) Score Test Mean Song- Spread Freq. A 1298 3 433 11.5 4.3 B 213 1 213 20.0 2.7 C 833 4 208 42.4 7.3 D 153 2 77 47.5 10.0 E 475 3 158 20.3 13.0 F 453 3 151 35.5 5.3 G 620 4 155 17.5 2.6 H 1013 4 253 8.0 3.0 I 550 1 550 0.0 2.0 J 863 2 432 13.8 3.8 K 2078 2 1039 4.5 0.7 behavior tests were representative of courtship intensity, then the scores should be highest when the receptive females were most abundant. Nero ( 1956a j found that female Red-wings are generally receptive around the time of clutch initiation. Thus, by comparing the distribution of dates of clutch initiation with the mean trial score of all males combined over the breeding season, it is possible to determine if the predicted correlation exists. Figure 2 illustrates the results of this comparison. A highly significant cor- relation (Spearman rank correlation, Ts — 0.85, p < 0.01) was found between the abundance of receptive females and the mean test scores over the breeding season. Only non-zero test scores were included. The results of the model tests for individual males are presented in Table 2. Because of the uncertain meaning of zero test scores, the lowest score was discarded for each male. To prevent any possible biasing due to this drop- ping of the lowest score, the highest score for each male was also dropped. The mean of the remaining scores was then used to determine the overall test score. If males were able to influence female choice by this behavior, we predicted that those males with the highest overall test scores should also be those that were most successful in recruiting females relative to the qualities of their respective territories. The study area was chosen for its consistent nest site quality and therefore the major differences in overall quality among the territories were due to significant differences in area. Thus, the highest scoring males were predicted to be those with the lowest area per female. Using a Spearman rank correlation, a coefficient of -0.87 (p< 0.01) was found for the correlation of female model test score with 588 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 o> w O o (/> o 9 E Fig. 2. Distribution of clutch initiation dates (a) and mean trial scores (b) for all males over the breeding season. area per female. Thus, the males that maintained the highest intensity of courtship to the female model are those that were able to achieve the highest nesting density of females. Since high nesting density is disadvantageous to females, those which chose such a situation must have been influenced to do so by the behavior of the territorial male. Figure 3 clearly demonstrates this finding. As the area per female drops below 300 m^ the intensity of courtship required by males to increase density further rises sharply. There is also a significant negative correlation ( Spearman rank correlation, r = -0.65, p < 0.05) between the model test score and territory area (Table 2l. This indicates that male behavior associated with recruitment ability is not synonymous with the ability to establish a large territory. In fact, it appears that the 2 are in some way mutually exclusive since the males with the best territories scored the lowest in the behavioral tests. Of the 13 acts that were recorded in the model tests, the song-spread display was most frequently observed and is perhaps the behavior most commonly associated with male Red-winged Blackbirds. Nero (1956a) considers the song-spread to be a warning display to other males although he does state that it is given more frequently in the presence of females. The mean fre- quency of song-spread displays per trial where the male was present for the Weatherhead and Robertson • RED-WINGED BLACKBIRD BEHAVIOR 589 full 5 min was determined for each male. The results of this investigation are included in Table 2. Using Spearman rank correlation, a significant positive correlation (p < 0.05) was found between song-spread frequency and mean trial score, and a significant negative correlation (p < 0.05) be- tween song-spread frequency and area per female. Thus, it appears that during the time of model testing, this display was of significant importance in female recruitment and did not function solely as a warning display. DISCUSSION The notion that males are able to influence the choice of mates by females is not a new one. In his treatment of the topic of sexual selection, Darwin (1871) considered this ability of males to be a major driving force in the evolution of secondary sexual characteristics. The importance of individual differences between males has not been ignored in the Orians-Verner model of polygyny. However, in defining the polygyny threshold in terms of ter- ritory quality, the assumption has been made that the difference in qualities of males is reflected in the difference in the qualities of their respective territories. This assumption has not allowed for the possibility that the ability to maintain a good quality territory may be derived from attributes unrelated to those involved in female recruitment. The results of this study suggest that this is in fact the case since recruitment success, as measured by female density, was generally higher for those males with the smaller territories. Results of other studies of the Red-winged Blackbird support the conclu- 590 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 sions drawn above. Smith (1972) investigated the role of the male’s red epaulets and found them to be important in territory maintenance but not female recruitment. This indicated that if males were actively recruiting females the epaulets are not important. However, it did not, as Smith suggests, prove that males exert no influence over female choice. Peek (1972) performed similar experiments to those of Smith in which male epaulets were blackened. His results supported Smith’s in that the loss of red epaulets resulted in a reduced ability to maintain a territory. However, Peek also performed experiments in which male Red-wings were muted and his results suggest that muted males were unable to obtain mates. In a study of 97 territorial male Red-wings, Weatherhead ( 1976 ) reported that only one male remained unmated. This male was also unique in its inability to perform the normal vocalization accompanying the song-spread display, uttering only a high-pitched squeak in its place. The results of the studies cited above suggest that the ability to maintain a territory is related more to visually directed displays while the ability to recruit females has its basis in vocaliza- tions. However, a more recent study by Smith (1976) produced contradictory results. Males that were vocally altered (= muting by Peek) did not appear to suffer any loss of ability to maintain a territory or attract females. Smith interprets the contradictory results as a possible consequence of habitat differences where the 2 studies were conducted. He suggests that only in the best habitats where competition between males is most intense would a male perform poorly if his vocal or visual displays were altered. It has been demonstrated that males differ behaviorally and that these differences affect recruitment success. However, what remains to be explained is why those males most successful at recruitment are least successful at the establishment of good quality territories. A possible explanation might be that, given a fixed amount of energy available for reproduction there are 2 strategies available. One would be to expend a great deal of energy in establishing a large, good quality territory. In addition to the high energetic costs of the acquisition of such a territory would be the accompanying high maintenance costs throughout the breeding season. This would leave only limited time and energy available for recruitment. The consequence of this strategy would be that fewer females would nest in the territory than would be expected, but because of low nesting density, individual success should be high. The alternative strategy would be to use very little energy in territory establishment, thereby securing a small territory with low maintenance costs. This would allow much more energy to be devoted to attracting females. The consequence here would be that more females would nest in the territory than would be predicted from its quality. Although female success would be W'eatherhead and Robertson • RED-WINGED BLACKBIRD BEHAVIOR 591 reduced, the male would still benefit by virtue of the number of females recruited. As a consequence of the 2 male strategies, females could choose either a high quality territory with low female density and therefore higher chances of nest success or a low quality, high density territory in which the chances of success were reduced but any male offspring produced would be expected to have superior ability to recruit mates. It is not expected that only 2 “pure” strategies would be observed in nature since males should differ with respect to their total energy available for reproduction. Differences in past experience may also affect how effectively this energy is partitioned. The consequence therefore would be that a range of males might exist such that the most fit obtained many females and high quality territories while the least fit obtained few females and poor quality territories. Between the 2 extremes would be a range of combinations of harem sizes and territory qualities. This would account for the consistent finding of many Red-winged Blackbird studies that harem size and territory area are not correlated (Holm 1973, Case and Hewitt 1963, Orians 1961, Nero 1956b) . SUMMARY Behavioral tests using freeze-dried female Red-winged Blackbird models were conducted on 11 territorial males through one breeding season. The intensity of courtship in the tests reflected recruitment success but not territory quality, indicating that those male attributes associated with territory establishment differ from those related to female recruitment. A negative correlation was found between recruitment success and territory quality and a possible explanation is presented. ACKNOWLEDGMENTS For her assistance with all aspects of the field work we are grateful to K. Clark. We wish to thank R. Norman and F. Phelan for their help in surveying the study area. For their valuable comments, criticisms, and discussion we thank the ecology group at Queen’s, and in particular P. Colgan for reading the manuscript. The use of facilities at the Queen’s University Biology Station greatly aided this study. LITERATURE CITED Case, N. A. and 0. H. Hewitt. 1963. Nesting and productivity of the Red-winged Blackbird in relation to habitat. Living Bird 2:7-20. Darwin, C. R. 1871. The descent of man and selection in relation to sex. Appleton, New York. Goddard, S. V. and V. V. Board. 1967. Reproductive success of Red-winged Black- birds in North Central Oklahoma. Wilson Bull. 79:283-289. Holm, C. H. 1973. Breeding sex ratios, territoriality, and reproductive success in the Red-winged Blackbird (Agelaius phoeniceus) . Ecology 54:356-365. 592 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Nero, R. W. 1956a. A behavior study of the Red-winged Blackbird. I. Mating and nesting activities. Wilson Bull. 68:4-37. ■ — . 1956b. A behavior study of the Red-winged Blackbird. II. Territoriality. Wilson Bull. 68:129-150. Orians, G. H. 1961. The ecology of blackbird (Agelaius) social systems. Ecol. Monogr. 31:285-312. 1969. On the evolution of mating systems in birds and mammals. Am. Nat. 103:589-603. Peek, F. W. 1972. An experimental study of the territorial function of vocal and visual display in the male Red-winged Blackbird {Agelaius phoeniceus) . Anim. Behav. 20:112-118. Robertson, R. J. and R. F. Norman. 1976. Behavioral defenses of potential hosts to brood parasitism by the Brown-headed Cowbird. Condor 78:166-173. Selander, R. K. 1972. Sexual selection and dimorphism in birds. Pp. 180-230, in Sexual selection and the descent of man 1871-1971 (B. Campbell, ed.). Aldine, Chicago. Smith, D. G. 1972. The role of the epaulets in the Red-winged Blackbird {Agelaius phoeniceus) social system. Behavior 41:251-268. . 1976. An experimental analysis of the function of Red-winged Blackbird song. Behavior 56:136-156. Trivers, R. L. 1972. Parental investment and sexual selection. Pp. 136-179, in Sexual selection and the descent of man 1871-1971 (B. Campbell ed.). Verner, J. 1964. Evolution of polygamy in the Long-billed Marsh Wren. Evolution 18:252-261. Weatheriiead, P. j. 1976. Polygyny in the Red-winged Blackbird. M. Sc. thesis. Queen’s Univ., Kingston Ontario. Weatheriiead, P. J. and R. J. Robertson. 1977. Harem size, territory quality and reproductive success in the Rew-winged Blackbird {Agelaius phoeniceus). Can. J. Zool. AND . In press. Offspring quality and the polygyny threshold: “The sexy son hypothesis.” Am. Nat. Wilson, E. 0. 1975. Sociohiology. Belknap, Harvard. DEPT. OF BIOLOGY, QUEEN’s UNIV., KINGSTON, ONTARIO, CANADA. ACCEPTED 9 NOV. 1976. REDUCTION OF COURTSHIP BEHAVIOR INDUCED BY DDE IN MALE RINGED TURTLE DOVES M. A. Haegele and Rick H. Hudson Some wildlife investigators who have studied declining bird populations have reported what they consider to be aberrant reproductive behavior. For instance, Gress (1970) reported abnormal behavior in the California Brown Pelicans (Pelecanus occidentalis) nesting on Anacapa Island, and Snyder et al. (1973) observed abnormal reproductive behavior in Cooper’s Hawks {Accipiter cooperii) . These investigators suggested that the behavior prob- lems were caused by DDE residues. Effects of DDE on behavior have been observed in laboratory studies concerned with reproduction of birds. Lincer (1972) observed possible abnormal parental behavior in American Kestrels (Falco sparverius) treated with dietary DDE and polychlorinated biphenyls (PCB’s). In our own studies with Ringed Turtle Doves {Streptopelia risoria) we found that DDE-treated birds in undisturbed reproductive cycles took an average of 2.5 times longer to renest than did control birds (Haegele and Hudson 1973). In the laboratory and field studies cited, few quantitative data were avail- able to test if DDE does affect reproductive behavior of birds. The purpose of the present study was to measure the quantitative effects of dietary DDE on the initial courtship behavior of male Ringed Turtle Doves. METHODS The 18 pairs of Ringed Turtle Doves used in this study were hatched and raised at the Denver Wildlife Research Center. Each pair had raised at least 1 young to 21 days of age prior to use in this study. After all 18 pairs of birds had successfully completed 1 breeding experience, each bird was isolated from viewing others (sight isolation) for 18 days. At the start of isolation, the outside of each wing of all males was marked with red ink so that we could easily distinguish the male birds when paired for courtship obser- vations. Lights were clock-controlled, turned on at 06:30 and off at 20:00 MST and temperature in all rooms was maintained between 22° and 24°C. Water, food, and min- eralized grit were provided ad libitum. After the 18th day of isolation, each male dove was randomly paired with a female dove with which he had never previously mated, and the pairs were placed in observation cages for 12.5 min each day for 5 consecutive days. Courtship behavior displayed during this period was recorded on video tape. This procedure provided a base line for normal courtship behavior. In order to keep assigned pairs separate until the video tape equipment was in operation, each observation cage was divided by an opaque partition. These parti- tions were removed after our recording equipment was started. To quantify courtship behavior, the video tapes were later played back and the number of bow-coos performed by each male bird was counted for each observation period. A 593 594 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 stopwatch was used to measure total time each male bird spent in performing sexual behavior. The types of behavior which we timed included driving, bow-coo, wing-flip, hetero-preening, billing, sex-mount, and stick-nest building (Miller and Miller 1958). The first 30 sec of video-taped behavior for all 12.5-min observation periods were not used. This allowed the investigator to leave the room and gave the doves time to adjust to the experimental conditions and their assigned mates. The video tape recorder and TV monitor were kept in an adjacent room so that no one was present in the room with the doves during a taping session. At the end of each observation period, all birds were returned to sight isolation until the next period. During all recording periods, each male dove was always paired with the same female. After the pretreatment data were gathered, the 18 pairs of doves were randomly assigned to 3 treatment groups each consisting of 6 pairs. One group was given a 10-ppm p,p'-DDE diet and one a 50-ppm diet dry weight basis. The remaining group served as a control and received an uncontaminated pigeon checkers diet. Initiation of treated diets was designated as day 0. Treated diets were made by adding proper amounts of DDE to ground Purina pigeon checkers. (Trade names are provided for identification only. Their mention does not imply endorsement of commercial products by the Federal Government.) All birds were fed ad libitum for 63 days at which time the study was terminated. The 10-ppm dietary treatment was chosen to simulate what was considered a possible field exposure level and the 50-ppm dietary treatment was selected to approximate the exposure level which adversely affected reproduction in our earlier study ( Haegele and Hudson 1973). To minimize possible intoxication effects, both treatment levels were selected to be sufficiently below the 20-day LC50 value of 250-300 ppm p.p'-DDE for Kinged Turtle Doves, which we determined before initiating this study. On days 31-35 and days 59-63, all birds were again paired with the same female and put into observation cages for video taping of courtship behavior. All doves were kept in sight isolation between observation periods. After completion of the observations, on day 63, all birds were sacrificed and the males plucked, eviscerated, and analyzed for residues. All birds were examined internally to confirm that they had been correctly sexed. One pair on the 50-ppm diet consisted of 2 males; therefore, all results for the 50-ppm dietary treatment are based on a sample size of 5 pairs rather than 6. A 3-factor analysis of variance with repeated measures on the last 2 factors (Winer 1971) was used to determine significant differences in bow-coo frequency and total activity time. Factor 1 was the 3 treatments, factor 2 was the 2 time periods (days 31-35 and 59-63), and factor 3 was the 10 1-day periods. Analysis of variance and Duncan's new multiple range test were used to determine significant differences between percent lipids found in whole body carcasses. Residue values were determined by the method of Peterson et al. (1976). RESULTS The mean number of seconds of total courtship activity time displayed hy male Ringed Turtle Doves was reduced by the DDE treatment (Fig. 1). Control birds increased their average courtship activity by 25% and 23%, respectively, for the 31-35 and 59-63 day posttreatment observation periods. The 10-ppm-treated birds showed no courtship activity time behavior effects at 31-35 days, but activity time decreased 55% from pretreatment values at 59-63 days. The birds on the 50-ppm DDE-contaminated diet had a 30% Haegele and Hudson • DDE AND DOVE BEHAVIOR 595 Fig. 1. Mean total courtship activity time displayed by male Ringed Turtle Doves per 12.5-min observation period. The vertical lines indicate ± S.E. of the mean. reduction in activity time at 31-35 days, and a 67% reduction at days 59-63, when compared to pretreatment activity. The analysis of variance showed: (1) a significant difference among the 3 treatments (F[2, 14] = 6.58, p < 0.01), (2) a significant difference between the 2 time periods (F[i,i4] = 16.51, p < 0.005) , (3) a significant difference among the 6 treatment X time period inter- action means (F[2,i4i = 4.08, p < 0.05), (4) a significant difference among the five 1-day periods (F[4,s6] = 2.72, p < 0.05), and (5) a significant differ- ence among the 10 time period X 1-day period interaction means (F[4,56] = 3.81, p < 0.01). In summary. Figure 1 shows that the controls maintained a high level of activity time for both observation periods while the treated birds showed a marked reduction in activity time at the 59-63-day period. These results indicate that the difference in activity time between the 2 observation periods was due to the DDE treatment over time. The mean bow-coo frequency by male Ringed Turtle Doves treated with the DDE diet was also reduced (Fig. 2). When compared with their pre- treatment values, the bow-coo frequency of control males increased 19% at 596 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Fig. 2. Mean number of bow-coos performed by male Ringed Turtle Doves per 12.5-min observation period. The vertical lines indicate ± S.E. of the mean. 31-35 days and then decreased 1% at 59-63 days. The 10-ppm group also increased (26%) at 31-35 days; however, after 59-63 days, the 10-ppm treatment had begun to affect bow-coo frequency causing a reduction to 53% fewer bow-coos than were observed during pretreatment observations. The 50-ppm group had a reduced how-coo frequency of approximately 43% during the first observation period and 84% during the second. The analysis of variance showed a significant difference among the 3 treatments (F[2,i4] = 3.58, p < 0.06) and a significant difference between the 2 time periods (F[i,i4i = 11.05, p < 0.01). The difference between the 2 observation periods indi- cates that, as DDE residues were increasing in treated birds, bow-coo frequency was decreasing. Figure 2 shows that all groups performed fewer bow-coos at 59-63 days than they did at 31-35 days, but the greatest changes in bow-coo frequency were shown by the DDE-treated groups. For this study, total activity time, rather than the frequency of bow-coos, was probably a more sensitive index to the effects of the treated diet. Whereas bow-coos are only an early segment of courtship behavior displayed by male Haegele and Hudson • DDE AND DOVE BEHAVIOR 597 Percent Body Weight Loss of Table 1 Ringed Turtle Doves During the Study (0-63 Days) Dietary DDE treatment N Males Mean percent body weightless (S.E.) Females Mean percent body weight loss ( S.E. ) 0 ppm 6 6.0 (3.4) 5.3 (3.7) 10 ppm 6 7.7 (2.7) 11.7 (6.2) 50 ppm 5 12.5 (10.6) 14.1 (7.9) Ringed Turtle Doves, total activity time took into account all types of court- ship behavior occurring during the observation period, such as nest site selection, wing-flipping, hetero-preening, billing, driving, bow-cooing, and copulation. The fact that male Ringed Turtle Doves could bow-coo infrequently during the observation period but still score very high in total time spent in courtship activity explains why our control birds showed a 1% decrease in bow-coo frequency at 59-63 days and yet had an increase of 23% in activity time during the same period (Figs. 1 and 2). Body weight losses for both the male and female Ringed Turtle Doves are given in Table 1. All birds tended to lose weight during the study. A small amount of weight loss during isolation seems to be normal. We have usually observed this effect in our laboratory when we have isolated doves for any extended period of time. Although the doves fed 50 ppm DDE weighed somewhat less than others at the termination of the experiment, group differ- ences were not statistically significant. Brain and whole-body DDE residues, along with lipid values, are given in Table 2. Small amounts of Aroclor 1254-like residues, which could have accumulated from small amounts of background contamination in the feed, were also found in all birds. Although there may be a possibility of PCB’s acting synergistically with DDE, the levels found in the different treatment groups, including controls, were probably too low (0.80 to 0.86 ppm wet weight) to affect the results. The percentage whole-body lipid in the male Ringed Turtle Doves was significantly (p ^ 0.05) reduced by the diets con- taminated with DDE. The standard errors for whole-body lipids were 0.5, 1.2, and 1.3 for the controls, 10-ppm DDE and 50-ppm treatment groups, respectively. Thus, DDE treatment not only reduced the percentage whole- body lipid found in male Ringed Turtle Doves, but it also increased the within-treatment variance of percentage whole-body lipids. The percentage lipid in brain tissue, not included in the whole-body lipid determination, was constant regardless of treatment (Table 2). 598 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Table 2 DDE Residues Found in Male Ringed Turtle Doves After 63 Days of Dietary Treatment Brain Dietary DDE treatment N ppm wet weight (S.E.) ppm lipid weight (S.E.) Percent lipid (S.E.) 0 ppm 6 0.09 (0.02) 1.6 (0.3) 6.7 (0.2) 10 ppm 6 2.9 (0.5) 40.9 (7.1) 7.0 (0.0) 50 ppm 5 7.6 (0.9) 116 (14.6) 6.6 (0.2) Whole body ppm wet weight (S.E.) ppm lipid weight (S.E.) Percent lipid* (S.E.) 0 ppm 6 0.39 (0.16) 1.8 (1.8) 13.4M0.5) 10 ppm 6 37.8 (4.4) 446 (108) 9.9M.2) 50 ppm 5 153 (25.2) 2477 (1107) 8.4” (1.3) * Those % lipid values with different letters as superscripts are significantly different from each other (p ^ 0.05) — Duncan’s New Multiple Range Test. Discussion Throughout this study, we continued the exposure to the DDE-treated diet and did not assess whether or not or how long it would take for the birds to recover from the treatment. It is not common for a population of wild birds to be continually exposed to levels of DDE of the same magnitude as were used in our experiment. Yet, it is not unusual for wild birds to be exposed to levels of DDE equal to our 10-ppm (dry weight) contaminated diet for fairly long periods of time during some part of the year and build up body residues equal to or greater than those found in our 10-ppm birds (Table 2). Some examples are Brown Pelicans (Keith et al. 1970), Prairie Falcons [Falco mexicanus) (Enderson and Berger 1970), and Peregrines {Falco peregrinus) (Lincer et al. 1970, Cade et al. 1968, Enderson and Berger 1968). This study supplements our earlier work ( Haegele and Hudson 1973) by showing that dietary DDE at low levels can adversely affect sexual behavior and performance in Ringed Turtle Doves. The 40-ppm DDE-contaminated diet in our earlier study caused a significant delay in renesting and an abundance of single-egg clutches. This delay in renesting could have been caused by a reduction in courtship behavior displayed by the male Ringed Turtle Doves treated with DDE. The 1-egg clutches observed in our 1973 study could also have been caused Haegele and Hudson • DDE AND DOVE BEHAVIOR 599 by behavioral effects of DDE. Courtship and nest building behavior are very important in stimulating the female Ringed Turtle Dove to ovulate and lay eggs (Lehrman et al. 1961, Lehrman 1965). The DDE diet could have reduced male and female Ringed Turtle Dove courtship behavior to a level that was insufficient to cause normal egg laying. Embryo development was absent in the eggs from single-egg clutches after 18 days of incubation ( 14 days is the normal incubation period) ; this suggests that the pairs of doves with infertile single-egg clutches never copulated before egg laying. The fact that these birds may not have copulated could be related to DDE effects on reproductive behavior. The reduced intensity of male Ringed Turtle Dove courtship behavior observed in this study supports the suggestion that DDE accumulation in wild birds is causing aberrant reproductive behavior and thereby reducing fecundity. Cress (1970) reported what he felt was aberrant reproductive behavior in Brown Pelicans, and he believed that the presence of large amounts of chlorinated hydrocarbon residues found in pelican tissue should be con- sidered as a potential cause of this erratic behavior. Snyder et al. (1973) suggested that the disturbed behavior they observed in Cooper’s Hawks might also have been linked to DDE residues present. Switzer et al. (1971) felt that the cause for the low reproductive success of their study population of Common Terns {Sterna hirundo) was attributable to aberrant reproductive behavior caused by DDE residues. Koeman et al. (1972) state that in their study 60% and probably 80% of the reproductive failures were caused by some intrinsic derailing factor in the Sparrow Hawk’s (Accipiter nisus) breeding process. They felt this might be caused by DDE residues the birds were carrying. All of the above examples suggest that some wild populations of birds contaminated with DDE are showing altered reproductive behavior and con- sequent reproductive failure. Judging from the effects dietary DDE had on courtship behavior of male Ringed Turtle Doves in our study, it is likely that DDE is affecting behavior in wild birds such as those mentioned above. The subsequent degree of reproductive inhibition may depend on the impor- tance and complexity of courtship behavior for different species. Whether other pollutants may contribute to these reproductive failures cannot be answered without further study, but DDE may be the most important factor. A possible mechanism for DDE effects on reproductive behavior has been suggested by Peakall (1970). He found that Ringed Turtle Doves treated daily with DDT had significantly lower estradiol levels in the blood associated with increased hepatic enzyme activities. Since DDE is also a microsomal enzyme inducer (Conney et al. 1967), it could be influencing reproductive behavior in the same manner as DDT. However, no study to date has demon- 600 THE WILSON BULLETIN • Vol. 89, No. 4, December 1977 strated a sequential occurrence of elevated hepatic microsomal enzyme activity, reduced circulating levels of sex hormones, and altered reproductive behavior in the same birds. The results of this study demonstrated that environmental levels of dietary DDE can lower the intensity of courtship behavior in male Ringed Turtle Doves. Therefore, we feel that DDE might indeed be a significant factor contributing to reproductive failure in wild birds. SUMMARY The effects of p,p'-DDE on the intensity of male Ringed Turtle Doves’ courtship behavior were determined for dietary levels of 10 ppm and 50 ppm (dry weight). Pairs of doves were placed in cages for 12.5 min on 5 consecutive days for behavioral observation before dietary treatment and for periods 31-35 and 59-63 days after initiation of the treated diet. Total amount of time spent displaying courtship behavior and bow-coo frequency were analyzed through video tape recording. The 50-ppm diet caused a reduction in total courtship activity time and in bow-coo frequency for both posttreatment observation periods. The 10-ppm diet did not affect bow-coo frequency and total activity time at 31-35 days but did cause a significant reduction in courtship behavior during the 59-63-day observation period. The DDE residues found in the male Ringed Turtle Doves showing reduced courtship behavior were lower than residues found in many species of birds that have shown reproductive failures in the wild. ACKNOWLEDGMENTS This work was conducted at the Denver Wildlife Research Center prior to a program consolidation of 28 March 1976. We thank co-workers at the Center, especially Kristine M. Stahl, for DDE residue analyses. LITERATURE CITED Cade, T. J., C. M. White, and J. R. Haugii. 1968. Peregrines and pesticides in Alaska. Condor 70:170-178. CoNNEY, A. H., R. M. Welch, R. Kuntzman, and J. J. Burns. 1967. Effects of pesticides on drug and steroid metabolism. Clin. Pharmacol. Ther. 8:2-10. Enderson, j. H., and D. D. Berger. 1968. Chlorinated hydrocarbon residues in Pere- grines and their prey species from northern Canada. Condor 70:149-153. AND . 1970. Pesticides: eggshell thinning and lowered production of young in Prairie Falcons. BioScience 20:355-356. Cress, F. 1970. Reproductive status of the California Brown Pelican in 1970, with notes on breeding biology and natural histoiy. Wildl. Manage. Branch Admin. Rep. No. 70-6. Calif. Dep. of Fish and Game. Haegele, M. a. AND R. H. Hudson. 1973. DDE effects on reproduction of Ring Doves. Environ. Pollut. 4:53-57. Keith, J. 0., L. A. Woods, and E. G. Hunt. 1970. Reproductive failure in Brown Pelicans on the Pacific coast. Trans. 35th N. Am. Wildl. Nat. Resour. Conf. 35: 56-64. Haegele and Hudson • DDE AND DOVE BEHAVIOR 601 Koeman, J. H., C. F. Van Beusekom, and J. J. M. De Coeij. 1972. Eggshell and popu- lation changes in the Sparrow Hawk {Accipiter nisus) . TNO-News 27:542-550. Lehrman, D. S. 1965. Interaction between internal and external environments in the regulation of the reproductive cycle of the Ring Dove. Pp. 355-380 in Sex and Be- havior (F. A. Beach, ed.), Wiley, N.Y. , P. N. Brody, and R. P. Wortis. 1961. The presence of the mate and of nesting material as stimuli for the development for incubation behavior and for gonadotrophin secretion in the Ring Dove (Streptopelia risoria) . Endocrinology 68:507-516. Linger, J. L. 1972. The effects of organochlorines on the American Kestrel {Falco spar- verius) . Ph.D. thesis, Cornell Univ., Ithaca, N.Y. , T. J. Cade, and J. M. Devine. 1970. Organochlorine residues in Alaskan Peregrine Falcons {Falco peregrinus Tunstall), Rough-legged Hawks {Buteo lagopus Pontoppidon) and their prey. Can. Field-Nat. 84:255-263. Miller, W. J., and L. S. Miller. 1958. Synopsis of behavior traits of the Ring Neck Dove. Anim. Behav. 6:3-8. Peakall, D. B. 1970. p,p'-DDT: effect on calcium metabolism and concentration of estradiol in the blood. Science 168:592-594. Peterson, J. E., K. M. Stahl, and D. L. Meeker. 1976. Simplified extraction and cleanup for determining organochlorine pesticides in small biological samples. Bull. Environ. Contam. Toxicol. 15:135-139. Snyder, N. F. R., H. A. Snyder, J. L. Linger, and R. T. Reynolds. 1973. Organo- chlorines, heavy metals, and the biology of North American accipiters. BioScience 23:300-305. Switzer, B., V. Lewin, and F. H. Wolfe. 1971. Shell thickness, DDE levels in eggs, and reproductive success in Common Terns {Sterna hirundo) , in Alberta. Can. J. Zool. 49:69-73. Winer, B. J. 1971. Statistical principles in experimental design. McGraw-Hill Book Co., N.Y. U.S. FISH AND WILDLIFE SERVICE, PATUXENT WILDLIFE RESEARCH CENTER, LAUREL, MD 20811. (PRESENT ADDRESS RHH: DEPT. OF ZOOLOGY, UNIV. OF WASHINGTON, SEATTLE 98195). ACCEPTED 1 AUG. 1976. MOVEMENTS OE THE GREAT-TAILED CRACKLE IN TEXAS Keith A. Arnold and Leon J. Folse, Jr. The Great-tailed Crackle [Quiscalus mexicanus) is an interesting species in that it has expanded its distribution rather dramatically in the 20th century (Selander and Giller 1961, Oberholser 1974). The species has received much attention, including systematics (Selander and Giller 1961), vocal behavior (Kok 1971), food habits (Davis and Arnold 1972), and growth rate and thermoregulation (Gotie and Kroll 1973). Little information is available on the movements of this grackle, especially in the area of its range expansion. This paper is intended to present such information for the Great-tail in the central parts of Texas. METHODS AND STUDY AREA Arnold initiated a banding and color-marking program in April 1967, as part of a study on the population dynamics and social structure of the Great-tailed Grackle. Except for minor interruptions, this banding program has been continued by Arnold and his graduate students to the present. Emphasis in mode of capture has varied from one year to the next. Thus most bandings of nestlings took place in the breeding seasons of 1967, 1968, 1971, and 1972, while mass banding at roosts with a light trap was limited to 1969, 1970, and 1971. Decoy traps were used every year, generally from September through March since 1969, but were used through July in 1973 and used continuously since September 1973. This latter method caught relatively few Great-tails until November 1973, when the permanent trap began to catch this species almost exclusively. We handed most birds in the area of Br>an-College Station, Texas. However decoy traps have been operated up to 15 km from these cities, and a light trap was used at a roost approximately 17 km west of Bryan. The area encompassed by our handing operations included the Bryan-College Station metropolitan area, flood plains of the Brazos and Navasota rivers, plus many hectares that were originally post oak {Quercus stellata) savannah and blackland prairie (see Coon 1974, for detailed description of the study area). The Brazos River lowlands w'ere used primarily for production of cotton (Gossypium hirsutum) , grain sorghum (Sorghur?! bicolor), and soybeans ^Glycine max), while much of the uplands and the Navasota River lowlands was used for pasturing livestock (Coon 1974). We grouped our recoveries for analyses by location of recovery, sex, and age at handing. The age groups differ between males and females, but these differences reflect breeding characteristics of the birds. For males in the spring, second year birds can he distinguished from older birds by plumage characteristics, and these second year males do not breed. Among females in the spring, second year birds are difficult to distinguish from older birds. However, these second year females do nest along with the older females. Consequently, we class second year males as immatures, but second year females as adults. Kruskal-Wallis statistical tests were from Conover (1971) and are based on ungrouped 602 Arnold and Folse • GREAT-TAILED CRACKLE MOVEMENTS 603 Number Table 1 OF Birds Banded by Year in each Sex-age Category 1967 1968 1969 1970 1971 1972 1973 1974 Total Males Adults 97 78 114 288 131 8 71 239 1026 Immatures 30 19 37 25 76 1 28 894 1110 Nestlings 42 51 18 6 61 26 - - 204 Females Adults 109 58 106 286 542^ 113 121 507 1842 Immatures 122 23 22 49 2 249 1207 1674 Nestlings 50 34 12 2 53 32 - - 183 Totals 450 263 309 656 863 182 469 2847 6039 * No attempt made this year to age females in roost-trapping. data, and other statistical tests were from Steel and Torrie (1960). We used a .05 significance level in all statistical tests. RESULTS Between April 1967 and December 1974, we banded over 6000 Great- tailed Crackles within the study area (Table 1). We received 117 recoveries, of which 60 were from outside the study area (Fig. 1). All recoveries but one are from Texas. The exception is a winter-banded adult male recovered the second spring after banding in Oklahoma City, Oklahoma. These recoveries may be roughly grouped into 3 categories: (1) those recovered south (S) of the study area in the drainage of the Brazos River and adjacent portions of the coastal plain; (2) those recovered to the west and northwest (WNW) of the study area, generally within the drainage of the Brazos River; and (3) those recovered to the north (N) of the study area and out- side the drainage area of the Brazos River. The remaining 57 recoveries were within the Bryan-College Station area (BCS). The recoveries from the different areas did not occur at the same times of year (Fig. 2; Kruskal- Wallis test, based on days to recovery after 1 December). Basically, those recoveries from the south (S) occurred in late winter or spring migration, while those from the remaining areas were made at other times of the year. There were no significant differences among areas, between the sexes or between birds banded as adults and fledglings in mean time to recovery after banding. Mean time to recovery of birds banded as nestlings was less than 604 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Fig. 1. Locations of Great-tailed Crackles recovered in Texas outside the study area (stippled). Counties as indicated by letters: A, Montague; B, Dallas; C, Kaufman; D, Ellis; E, Navarro; F, Freestone; G, McLennan; H, Falls; J, Bell; K, Milam; L, Robertson; M, Williamson; N, Burleson; 0, Bastrop; P, Washington; R, Austin; S, Waller; T, Harris; U, Fort Bend; V, Colorado. Insert shows relative portion of Texas depicted on map. that for birds banded when older (Kruskal-Wallis tests, based on days to recovery after banding). Recoveries were classified by location, sex, and age at banding (Table 2). An analysis of variance with these 3 factors revealed no significant variation with respect to sex and age upon percent of banded birds recovered. There was significant variation in percent recaptures among the recovery areas. Single degrees of freedom comparisons showed that percent recoveries were Arnold and Folse • GREAT-TAILED CRACKLE MOVEMENTS 605 DJ FMAMJ JASON DJ FMAMJ JASON Fig. 2. Recoveries of Great-tailed Crackles by month and location. (See text for locations.) 606 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 Number and Percent i Table 2 OF Texas Recoveries by Regions Northwest Bryan-College North* & West South Station Total Males Adults 3 (.29)-* 6 (.58) 2 (.20) 13 (1.27 ) 24 (2.34) Immatures 1 (.09) 8 (.72) 6 (.54) 12 (1.17) 28 (2.52) Nestlings 0 1 (.49) 0 4 (1.96) 5 (2.45) Females Adults 5 (.27) 13 (.71) 6 (.33) 16 (.87) 40 (2.18) Immatures 1 (.06) 4 (.24) 2 (.12) 7 (.42) 14 (.84) Nestlings 0 2 (1.09) 0 4 (2.19) 6 (3.28) * Includes the Oklahoma recovery. ** Percentages are based upon numbers originally banded in each sex-age category from 1967 through 1974. greatest in Bryan-College Station with WNW next, followed by S and N which were indistinguishable from each other. DISCUSSION The Great-tailed Crackle has expanded its range northward into Kansas in the past few years and has become established there as a summer resident (Schwilling 1971). In this northward expansion, the status of the species has changed dramatically in central and northern Texas (Davis 1940, Arnold 1973). These northern Texas populations are apparently partially migratory as evidenced by recoveries in this study and by the dramatic increase in numbers of this species that occur in the local fall and winter roosts. Our banding studies demonstrate southward shifts in blackbird populations of the Bryan-College Station, Texas area in mid-winter with replacement by populations from the north; this is indirectly indicated by the lack of recaptures for Great-tails banded in the earlier part of the winter even though large numbers of this species continue to be captured. These popula- tion shifts, which generally coincide with the onset of the severe portion of our winter, do not always occur in mild winters. We know of several winter roosts in the Dallas-Ellis county area that have Great-tails associated with them, so not all Great-tailed Crackles migrate south from that area. We have a number of recaptures and sightings in the Bryan-College Station area of color-marked birds or birds marked during the breeding season or fall and early winter that were made in the severe portion of the winter. Further, at least 2 Great-tails have been recovered in the Dallas-Ellis county area Arnold and Folse • GREAT-TAILED CRACKLE MOVEMENTS 607 during the winter months subsequent to their banding on the study area in a preceeding October. It is not clear whether these birds represent an age class with migratory patterns different from other age classes or whether this represents a failure to migrate due to mild climatic conditions; only further recoveries will resolve this point. The latter possiblility is likely since 2 winters occurred (1969-70 and 1973-74) without a mid-season population shift for Great-tailed Crackles in our study area. We believe that the Brazos River valley serves as a natural migration route, based on recoveries south and west-northwest of our area. The grackles recovered to the north, however, are in the Trinity River drainage. Area N (Fig. 1) lies at the eastern edge of the post oak savannah and black land prairie regions and at the western edge of the East Texas pinelands. As the Great-tail is essentially a bird of the open country, it appears to us that the southward migration from the Dallas-Ellis county area represents avoidance by this grackle of unsuitable vegetation zones. Recoveries from Navarro and Freestone counties, intermediate in geographic position, support this idea. The presence of livestock operations and extensive growing of grain sorghum in the Brazos River valley may offer attractive food sources to flocks of migrating Great-tails. Two female grackles recovered to the south of Bryan-College Station rep- resent another problem; both were recovered in May (one each in Waller and Washington counties). The date is late for migrant grackles, but the locations are geographically intermediate between wintering areas on the coastal plain and the study area. As one of the females was banded as an adult the previous May, the possibility exists that the bird had changed breeding areas. Many of the birds recovered from the north and west-northwest were banded in September and October as immatures. We do not know whether these birds were hatched locally or were migrants. Some of these recoveries must represent dispersal of young birds in a species that is rapidly expanding its range. This is supported by recoveries of 3 birds banded as nestlings on the study area: a male and a female recovered in Robertson County, and a female recovered in McLennan County, both northwest of our study area. It is possible that birds entering the local roosts in late summer and early fall (August to October) may represent the coalescing of adjacent breeding populations with those of the Bryan-College Station area; however additional studies are needed to clarify this. We know that many breeding colonies of this species exist in surrounding communities, but Great-tails are difficult to observe in those areas during the winter months. Further recoveries will clarify our knowledge of the migration of Great-tailed Grackles. 608 THE WILSON BULLETIN • VoL 89, No. 4, December 1977 SUMMARY Between April 1967 and December 1974, over 6000 Great-tailed Crackles were banded in Brazos County, Texas. Through September 1976, 117 recoveries were reported, with 60 from outside of Brazos County. These outside recoveries were grouped into 3 geographic areas: birds recovered to the south that represent a mid-winter population shift; those recovered to the west and northwest, representing migration along the Brazos River valley; and birds from the north that may represent migration along the edge of the East Texas pine lands. These latter 2 groups probably also include young birds dispersing away from breeding colonies in the study area. ACKNOWLEDGMENTS This study is a part of a continuing project of the Texas Agricultural Experiment Station (TAES contribution number TA 12114) . We thank the following persons for their help in banding the grackles: W. R. Davis II, D. W. Coon, R. F. Gotie, R. Klein, R. Summers, J. C. Newman, A. L. Barr, M. T. Hanson, and C. E. Grue. R. D. Slack and B. W. Cain critically reviewed the manuscript. C. S. Robbins provided many useful suggestions. LITERATURE CITED Arnold, K. A. 1973. The birds of Brazos county; Thirty years in retrospect. Bull. Texas Ornithol. Soc. 6:4-6. Conover, W. J. 1971. Practical nonparametric statistics. John Wiley, New York. Coon, D. W. 1974. Daily movements of the Brown-headed Cowbird in south-central Texas during winter. Ph.D. thesis, Texas A&M Univ., College Station. Davis, W. B. 1940. The birds of Brazos county, Texas. Condor 42:81-85. Davis, W. R., II and K. A. Arnold. 1972. Food habits of the Great-tailed Crackle in Brazos County, Texas. Condor 74:439-446. Cotie, R. E. and J. C. Kroll. 1973. Growth rate and ontogeny of thermoregulation in nestling Great-tailed Grackles, Cassidix mexicanus prosopidicola (Icteridae). Condor 75:190-199. Kok, O. B. 1971. Vocal behavior of the Great-tailed Crackle {Quiscalus mexicanus prosopidicola) . Condor 73:348-363. Oberiiolser, H. C. 1974. The bird life of Texas. 2 vols. Univ. Texas Press, Austin. ScHWiLLiNG, M. D. 1971. Rapid increase and dispersal of Boat-tailed Grackles in Kan- sas. Kansas Ornithol. Soc. Bull. 22:15-16. Selander, R. K. and D. R. Giller. 1961. Analysis of sympatry of Great-tailed and Boat-tailed Grackles. Condor 63:29-68. Steel, R. G. D. and J. H. Torrie. 1960. Principles and procedures of statistics. Mc- Graw-Hill, New York. DEPT. OF WILDLIFE AND FISHERIES SCIENCES, TEXAS A&M UNIV., COLLEGE STA- TION 77843. ACCEPTED 3 NOV. 1976. GENERAL NOTES Tropical Screech Owl nest defense behavior and nestling growth rate. — The Tropical Screech Owl {Otus choliba) is a common and wide spread resident in the Neotropics from Costa Rica to Argentina. General accounts of it are given by ffrench (A Guide to the Birds of Trinidad and Tobago, Livingston Publ. Co., Wynnewood, Pa., 1973), Haverschmidt (Birds of Surinam, Oliver and Boyd, Edinburgh, 1968) and Wetmore (The Birds of the Republic of Panama, Part 1, Smithsonian Institute Press, Washington, D.C., 1965). Although there are more than 30 owls, world wide, in the genus Otus, little has been published on their nestling behavior and growth, except for Otus asio, O. trichopsis, and 0. flammeolus by Bent (U.S. Natl. Mus. Bull. 170, 1938). There is also a note on nestling screech owl behavior and growth by Sumner (Condor 30:333-338, 1928). References indicated below for soft part colors are from Smithe (Naturalist’s Color Guide, The American Museum of Natural History, New York, 1975) . This note presents information on the defense behavior of adults and nestlings, and the growth rate of the latter, from observations made at a nest in a rural residential area 30 km south of Caracas, Venezuela. The nest was located in an abandoned wooden box, a cube of about 40 cm wedged into the lowest branches of a pardillo tree {Cordia alliodora) , 3 m from the ground. The box was in an angled position and about half of one side was open. The owl’s choice of such a nest site probably reflected the lack of suitable natural cavities in the area, a region of fallow, grassy hillsides with only a few scattered trees. On 25 February 1975 the nest contained 2 eggs (smooth, white, and nearly spherical; 33.3 X 30.5, 32.7 X 29.9 mm) which rested in the lowest corner of the box on some weathered fragments of cloth. The cloth was probably left over from the former tenant, a pet monkey. To stop tree climbing predators I fastened aim wide piece of aluminum around the tree trunk. Nestling period. — By the evening of 12 March both nestlings had hatched and their natal down was completely dry. They weighed 17.0 and 13.5 g. I believe the larger one had hatched on the previous day. They were covered with white down, sparser on the dorsal area. The legs and feet were pale pink (color 7) and the bills pearl gray (color 81) with a white egg tooth. The nestlings’ eyelids were closed, but on the 6th day of obser- vation the smaller bird’s eyelids were slightly open, the larger nestling’s eyelids were not fully open until the 10th day. At first the irides were cream color (color 54), they did not become nearly spectrum yellow (color 55) until about the 19th day. The nestlings’ bills had changed by the 4th day to pale neutral gray (color 86) and the egg tooth was greatly diminished; the legs and feet were now almost translucent white. Feather sheaths were clearly visible on the larger nestling on the 6th day on its spinal, alar, and ventral tracts. Rictal bristles appeared on both birds in 2 weeks. By the 16th day their white natal down (protoptile) had generally changed to pale neutral gray (color 86) and white banded down (mesoptile) and their primaries and secondaries were beginning to expand. The central rectrices started to emerge from their sheaths on the 26th day. A pesola scale was used to weigh the nestlings daily (except days 21 and 28) at about 18:00, before their first nightly feedings. The fairly constant increase of weights is shown in Fig. 1. The decreased weights recorded on the evenings of observation days 14 and 16 might be indicative of a lesser quantity of food brought to the nestlings on the previous nights. Weight losses on days 23, 25, and 27 possibly signal the approaching end of the nestling period. 609 610 THE WILSON BULLETIN • Vol. 89, No. 4 Fig. 1. Daily weights of Tropical Screech Owl nestlings. The nestling owls never regurgitated food on being handled; therefore the only food seen was that encountered in the nest. At different times I found all or parts of 2 mice, 1 rice rat (Oryzomys) , 2 cockroaches ( Penplaneta and Blaberus) , 1 locust (Schistocerca) , 1 very small snake {Helminthophis) and 1 earthworm (Lumbricus). On the 27th day I collected the only pellet found in the nest. It was moist, measured 2 X L25 cm, and contained insect and bone fragments too small for identification. Nest sanitation was poor only on the first 2 days; thereafter the nest was kept clean. Ectoparasites found on the nestlings were 1 louse fly Icosta (= Lynchia) americana and at various times 4 mag- gots were found attached to the nestlings’ ventral sides. One maggot has been tentatively identified as Neomusco ( r= Philornis) falsijicus. An adult owl brooded the nestlings all day through the 6th day, leaving only after dark. However after a cessation of brooding for 3 days, an adult stayed all day in the nest for a single day, the 10th. I was unable to establish whether 1 or 2 owls attended the nest. Presumably the nestlings left the nest normally during the night of 8 April at the ages of about 28 and 29 days. That night around 02:00 there was much low calling by an adult owl. The next morning the nest box was empty and I could not locate the young owls in the area. Defense behavior by the adult. — The first time that the nest box was examined, during the day, an adult owl was incubating the eggs and it puffed up its feathers and erected its “ear tufts.” Then it flew out of the box onto the lowest limb of a tree 5 m distant December 1977 • GENERAL NOTES 611 Fig. 2. Tropical Screech Owl nestling on the 26th day showing black feathers on eyelids. from where it watched the nest. On subsequent nest examinations, in addition to ruffling out its plumage and erecting its “ear tufts'’ the adult owl snapped its hill, swayed back and forth sideways, and raised and lowered its body by flexing its legs. During the first examination after the nestlings had hatched, the brooding owl threw itself flat on its back over the nestlings and raised its legs with claws extended toward me. Gradually, however, the attending adult owl appeared to become accustomed to the daily visits and soon I was able to easily remove and replace the nestlings from under it without causing more than a token feather ruffling. On the 16th day I examined the nest for food every hour on the hour from 18:00 to 06:00. At 22:00 while I was at the nest, an owl struck me quite hard on the arm, leaving 3 small bloody marks from the claws’ impact in a triangle roughly 3.5 to 5 cm. In no way was this a serious wound, it is mentioned only as a measure of the probable size of grasp of an adult Otus choliba. An owl struck me again, twice at 01:00 and once more at 03:00, as well as on another night at about 21:00. I heard no sound of the bird’s approach before the strikes nor was I able to see an owl in the area before climbing to the nest. Defense behavior of the nestlings. — Nestling defense behavior may have started on the first day of examination when they made tiny squeaking sounds or distress calls on being weighed. They both continued their vocal protests with increasing vigor as they grew. On the 6th day, the smaller nestling gripped my glove with its bill and hung on tenaciously. The larger nestling began bill snapping on the 11th day; the smaller bird 2 days later. The more active nestling, the smaller one, dug its claws into my exposed finger on the 16th day of examination. That same day, for the first time, both nestlings fluffed up their downy feathers and crouched down deeply on their tarsi at my approach. The next 612 THE WILSON BULLETIN • VoL 89, No. 4 day they both began flapping their wings on being handled. On the 23rd day the larger nestling threw itself on its back with its legs raised and claws extended, exactly as the previously described adult behavior. Most of the nestling owls’ day was spent sleeping pressed tightly together. At 3 weeks they began sleeping facing the open entrance of the nest box. It was then that I first saw, on looking into the box, what appeared to be 2 pairs of large eyes staring at me, when, in fact, the nestlings had their eyes closed in sleep. Careful examination revealed that the eyelid skin was light pearl gray (color 81) with a tuft of sparce white feathers near the outer corner, and that centrally on the lid there was a prominent area of short black feathers. The effect of this pattern was to make the eyelids, when closed and viewed in the weak light of the nest, excellent mimic eyes (Fig. 2) . This appears to be an adaptation for startling a predator who might look into the nest hole in daylight, after the nestlings are no longer brooded by an adult owl. Summary. — This paper presents observations of the nesting of the Tropical Screech Owl {Otus choliba) and discusses development of the nestlings, their food, defense behavior of the adult and the nestlings, as well as a description of a probable warning color pattern development of eyelid feathers as a defense strategy. Acknowledgments. — I wish to thank Paul Schwartz for reading the first draft of this paper and for making a number of helpful criticisms of it. I am also indebted to Dr. Francisco Fernandez Yepez for identification of the insect food items, and to Dr. Andrew J. Main, Jr. and Dr. T. H. G. Aitken for identification of the ectoparasites. — Betsy Trent Thomas, Apartado 80450, Caracas 108, Venezuela. Accepted 1 Aug. 1976. Three more new specimen records for Guatemala. — During continuing studies of the avifauna of the Pacific lowlands of Guatemala near La Avellana, Santa Rosa Depart- ment (Dickerman, Wilson Bull. 87:412-413, 1975), I collected 3 more species of birds not previously taken in the country. During April and May 1975 Richard R. Viet and in April 1976 Alexander Brash and Thomas Will participated in the field work. Collecting was done under permit from the Departmento de Vida Silvestre, Institute Nacional Forestal •of Guatemala and specimens are deposited at the American Museum of Natural History. Ca/idris bairdii, Baird’s Sandpiper. — Although Baird’s Sandpiper is a regular migrant at least in interior areas of Mexico, and winters in South America, there are few specimen records from Central America, and to date none from Guatemala. On 21 April 1976 Thomas Will identified 2 Baird’s Sandpipers and I collected one, on mud flats between the villages of La Avellana and Monterrico, Santa Rosa Department. The birds were found in the afternoon following a major rainstorm the night of 20-21 April that continued to mid-morning of the 21st. The specimen collected, an immature female, weighed 33.3 g and was moderately fat. Calidris alpina. Dunlin. — An adult male Dunlin, well advanced in prealternate molt, was taken about 21:30, 6 April 1976 from nets set on the same mud flats mentioned above. This is apparently the southernmost specimen record for the species in the New World. Its wing chord measures 115 and the culmen 36.5 mm; the specimen was identified by John Farrand, Jr. as E. a. pacifica based both on measurements and on the deep coloration ■of the fresh alternate plumage. Sterna albifrons, Least Tern. — Although the AOU Check-list (1957) records 5. a. browni, the race that nests along the coast of California and Baja California, as probably ranging south to Guatemala in winter. Land (Birds of Guatemala, Livingston Publ. Co., Wynne- wood, Pa., 1972:112) knew of no specimens of the species from Guatemala. Accordingly December 1977 • GENERAL NOTES 613 he placed the species name in brackets indicating a hypothetical status. He mentioned August and September sight records for both coasts. We have seen 1 to 3 Least Terns on the following dates at the mouth of the Rio Los Esclavos, Santa Rosa Department: 2 and 4 May 1974, 24 April 1975 and 29 April 1976. An adult collected 24 April 1974 was identified by John Farrand, Jr. as S. a. antillarum, the race of central and eastern United States and of the Caribbean region. — Robert W. Dickerman, Dept. Microbiology, Cornell Univ. Medical College, 1300 York Avenue, New York 10021. Accepted 20 Oct. 1976. Feeding behavior of two hummingbirds in a Costa Rican montane forest. — Between the end of April and mid-June 1974, I made observations on hummingbirds feeding in primary forest between 1480 and 1680 m at Monteverde on the Pacific slope of the Cordillera de Tilaran of Costa Rica. At this level the forest is transitional between the Lower Montane Rain Forest and the Lower Montane Wet Forest (Holdridge, Life Zone Ecology, Tropical Science Center, San Jose, Costa Rica, 1967). Within the shade of this forest the 2 most abundant resident hummingbirds were the Purple-throated Mountain Gem (Lampornis calolaema) and the Guy’s Hermit (Phaethornis guy). At Monteverde these 2 species largely overlap in habitat and altitudinal range, as they do elsewhere in their geographical range (Slud, Bull. Am. Mus. Nat. Hist. 128, 1964), but I observed no overlap in the flowers visited for nectar (Table 1). These hummingbirds also differ in size and proportion; the mean measurements, sexes combined, of Guy’s Hermit are: weight 5.8 g (N = 3), wing 61 mm (N = 20), culmen 42.5 mm (N 20) (Wetmore, The Birds of the Republic of Panama, Part 2, Smiths. Misc. Coll. 150, 1968) . Those for the Mountain Gem, treating the sexes separately, are: female — weight 4.2 g (N = 2), wing 56 mm (N = 6), culmen 22.1 mm (N = 6) ; male — weight 5.6 g (N = 7), wing 63 mm (N =i 7), culmen 21.4 mm (N = 6) (Feinsinger, Organization of a Tropical Guild of Nectarivorous Birds, Ph.D. thesis, Cornell Univ., Ithaca, N.Y., 1974 for weights quoted). In addition the Guy’s Hermit has the typical curved bill of the hermits and the Mountain Gem a straight bill. Except for 1 tree (Quararibea sp.) the feeding records were from vines, shrubs, her- baceous plants, or epiphytes (Table 1). With the exception of the epiphytes, the other plants appeared to be shade tolerant species growing and flowering under unbroken canopy or in the partial shade of mountain paths or very steep slopes. The herbaceous plants flowered at heights of 30-90 cm and the shrubs and vines at 0.6-6 m. Of the 2 epiphytes, Columnea magnifica grew sparsely on trees just below canopy level and continued to grow and flower in partial shade on fallen trees; whereas the epiphytic heath (Thibaudiae) grew in large clumps locally dominating its host tree and enjoying full sunlight, but was not seen flowering on fallen trees in partial or heavy shade. Guy’s Hermits were feeding at 3 levels, between 30 and 90 cm when feeding at her- baceous plants, between 1.5 and 2 m when feeding at vines, and just below the canopy at 12-18 m when feeding at the epiphyte. The Mountain Gem fed at shrubs between 60 cm and 6 m, at the epiphyte between 6 and 10.5 m, and at the Quararibea tree between 12 and 15 m. I defined a feeding record as a bout of feeding by an individual hummingbird at 1 plant species. The actual number of flowers visited in a feeding bout varied greatly depending on flower size; thus a record of Guy’s Hermit feeding on Drymonia concho- calyx would consist of probes into 2 or 3 flowers, while a record of a Mountain Gem at Palicourea typically consisted of probes into 20 or more flowers. To attempt to adjust the data to number of flowers visited would distort the results because some nectar 614 THE WILSOxN BULLETIN • Vo/. 89, No. 4 Table 1 Flower Characteristics and Hummingbird Feeding Records Corolla length^ ( mm ) \\’idth ( mm ) Lam] calol oornis aema Phae- thornis gay Plant fonn Color d 9 Gesneriaceae Drymonia conchocalyx vine dark pink 56 10 1* 4 Drymonia sp. vine orange 38 8 1 Alloplectus tetragonus herb red 44 6 1 Besleria formosa shrub -orange 16 6 2 15 Campanaea huml)oldtii vine green - 6 3 Columnea magnifica epiphyte orange 28 4.5 1 Rubiaceae Cephaelis data shrub whitet 17 2.0 7 4 Palicourea lasiorrhachis shrub yellow 14 2.5 13 9 Lobeliaceae Centropogon solanifolius herb red 43 7.5 2 Musaceae Heliconia tortuosaX herb yellowt 34 4.0 2 Bombacaceae Quararibea sp. tree white 23 2.0 11 Ericaceae (Thibaudiae) epiphyte pink 19 3.0 3 2 Total 36 34 11 * Corolla pierced, t Red bracts. + Ta.xon H-5 (Stiles 1975). 1 Corolla lengths measured are the distance from the opening of the corolla tube to the nectar cham- ber. One t>pical corolla measured from each species. sources were more scattered than others and a feeding hummingbird in the forest is more quickly lost to view when feeding on scattered flowers than on the more concentrated ones. Except for the record of the Mountain Gem piercing the corolla of Drymonia concho- calyx, no other hummingbirds were seen feeding at the flowers exploited by the Guy’s Hermit within the 1480-1680 m altitudinal limits. Below these altitudes the Violet Sabrewing {Campylopterus hemileucurus) commonly fed at Heliconia tortuosa and tbe Stripe-tailed Hummingbird {Eupherusa eximia) occasionally did so. There was more competition for the flowers exploited by the Mountain Gem. The larger clumps of the epiphytic heath were dominated by the Fiery-throated Hummingbird iPanterpe insignis) , and the Slaty Flower-piercer {Diglossa plumbea) also fed at it. The Stripe-tailed Hummingbird was the only other hummingbird seen exploiting any of the nectar sources of the Mountain Gem listed in Table 1. I have 3 records of it feeding at Palicourea lasiorrhachis; 2 were below 1480 m. Feinsinger (op. cit.) frequently recorded December 1977 • GENERAL NOTES 615 the Mountain Gem below 1480 m where it defended high density nectar sources in more open habitats of secondary growth and forest edge and competed with several other hummingbird species. The flowers it visited were all different from those listed in Table 1. There are several differences between the flowers exploited by Guy’s Hermit and the Mountain Gem. The 6 hermit flowers are larger, with an average corolla length of 44 mm and a width at the base of the corolla of 7.2 mm. The corollas are curved and colored either orange, red, or dark pink (in Heliconia tortuosa the corolla is yellow but the flower is embedded in a red bract). The color of the 6 species at which the Mountain Gem fed are more varied and include pink, orange, yellow, white, and green. The corollas are all straight rather narrow tubes with an average length of 18 mm and a basal width of 3.1 mm. The vine Campanea humboldtii is an exception and not included in the above average; its corolla is a large open bell with a width of 30 mm at the mouth. The Mountain Gem feeds at this flower with its whole head inside the bell. These differences in feeding niche between the Guy’s Hermit and the Mountain Gem are generally similar to the differences in Trinidad between the Guy’s Hermit and the Blue-chinned Sapphire ( Chlorestes notatus) , a straight-billed forest hummingbird slightly smaller than the Mountain Gem (Snow and Snow, J. Anim. Ecol. 41:471^85, 1972). Guy’s Hermits were not seen defending nectar resources and are evidently “trapline” feeders as are other hermit hummingbirds that have been studied (Stiles, Ecology 56: 285-301, 1975) . Mountain Gems, on the other hand, were frequently seen defending their nectar resources against conspecifics. Since the sexual difference in plumage is apparent in this species before the young leave the nest (Skutch, Publ. Nuttall Ornithol. Club 7, 1967) it was possible to separate with certainty the feeding records of the sexes. The differences in the feeding niches of the sexes (Table 1) reflect the male’s dominant behavior over the richer resources. Quararibea sp., a tree reaching canopy level, was a con- centrated source over which males held feeding territories; they also held territories at the smaller patches of the epiphytic heath, the larger patches being dominatd by the Fiery- throated Hummingbird, and at the shrub Palicourea lasiorrhachis. Palicourea, growing to 6 m with an abundance of small flowers, is the biggest of the 3 shrubs at which I recorded the Mountain Gem feeding and the one over which males most frequently held territories. Palicourea lasiorrhachis grows in 2 forms; both have similar yellow corollas, but 1 form, common between 1400 and 1540 m, has a red calyx and pedicel; the other form, not noted growing below 1530 m and generally a smaller plant, has a green calyx and pedicel. Nine of 13 records of male Mountain Gems feeding at Palicourea were from the red-calyxed form, but only 4 of 9 records of females feeding at Palicourea were from this form and 2 of these appeared to be permitted intrusions by a male into his territory (see below). Besleria formosa is a much smaller shrub than Palicourea, growing to only 90 to 120 cm, and is thinly scattered through deeply shaded forest; typically each shrub has between 4 and 8 open flowers at one time, but more where it grows at path edges. Exploited largely by females, individual shrubs were re-visited on an average of every 10 min. Be- tween feeding circuits females usually perched near one of the larger clumps of Besleria from which conspecifics were driven off. Periodically during a bout of aerial nectar extraction, both male and female Mountain Gems perch to feed at one particular flower, and re-perch at the same flower at each subsequent visit. This was noted when they were feeding at Besleria, Quararibea, and the epiphytic heath. Skutch (op. cit.) also noted this behavior of Mountain Gems feeding at epiphytic heaths. Observations on the insect searching strategy of Mountain Gems produced 2 records of males hawking for aerial insects from their territorial perches 6 to 12 m up, and 2 records of females searching amongst very dense vegetation, presumably for resting insects, once in the herbaceous layer at 60 to 120 cm and once in the foliage 616 THE WILSON BULLETIN • VoL 89, No. 4 of a small tree at 3 to 6 m. On each occasion the female’s wings were audibly hitting the leaves as she hovered amongst them. While Skutch (op. cit.) has described the nesting of the Mountain Gem and once observed a young male, still being fed by his mother, who was persistently singing a very faint song, he has never heard song from adult males or observed any other courtship activity. During my observations both male and female Mountain Gems were usually silent except for occasional flight notes uttered during longer flights between nectar sources. But on 11 June I observed a male briefly uttering an insect-like song from a perch beside a Palicourea shrub at which it was periodically feeding. Another observation suggested that there may at times be a sharing of nectar resources between the sexes. Between 11:40 and 12:00 on 13 June I watched a male Mountain Gem which held a feeding territory over 3 flowering Palicourea shrubs. During this time he was observed both feeding at the shrubs and chasing off a female from them; then at 12:00 a female came to one of the Palicourea shrubs and began to feed, and between each probe she uttered a short call which I transcribed as trrrt. While she fed, the male was perched immediately below her on the same perch he had been using the previous 20 min. He remained perched there throughout the female’s feed and once uttered an answering trrt. The only other occasion when this call was heard was earlier on the same day when a female, feeding at the same Palicourea, was noted as uttering the call between each feeding probe. Interpretation of this behavior on a single observation would be premature, but it suggests that males may have a special relationship with particular females, and may allow them to share the nectar in their feeding territories. Wolf and Stiles (Evolution 24:759-773, 1970) found that male Fiery-throated Hummingbirds allowed females with whom they mated to feed within their defended territory. I acknowledge with thanks financial assistance from the Frank M. Chapman Memorial Fund of the American Museum of Natural History. I should also like to thank Dr. Luis Poveda and Dr. Richard Baker for botanical identifications. — Barbara K. Snow, Old Forge, W ingrave, Aylesbury, Bucks, England. Accepted 12 Oct. 1976. Black-legged Kittiwakes nesting on snowbank. — On 4 July 1975 we found 20 nests of the Black-legged Kittiwake ( Rissa tridactyla) being built on a snowbank at St. Paul Island, Pribilof Islands, Alaska < Fig. 1). The snowbank, approximately 10 m high, 100 m long and sloping at an angle of 75°, was blocking access to an area of south-facing cliff just east of Southwest Point. More Black-legged Kittiwakes and several other species of seabirds were nesting on the cliffs on either end of the snowbank. The nests on the snowbank were not noted on 28 June, the date of the previous visit to the area. During the next 10 days after 4 July, the nests disintegrated and fell as the snow melted. No eggs were seen nor were the adults noted incubating. These nests were built relatively late in the breeding season, as the first eggs of this species on the island were seen on 27 June. On 7 July 85% of the Black-legged Kittiwake nests in a nearby study area were being incubated. It is unclear whether this use of a snowbank as a nest substrate was the result of site tenacity on the part of the kittiwakes or of the lack of suitable alternative nest sites. Sealy (Auk 92:528-538, 1975) discusses a similar situation in which Least Auklets {Aethia pusilla) and Crested Auklets (A. cristatella) on St. Lawrence Island laid eggs on snow. Snow nesting of the auklets was restricted to those birds faithful to nesting habitat that remained snow covered until mid-July. Belopol’skii (Translated from Russian book Akademiya Nauk SSSR, Karel’skii filial. U.S. Dept, of Commerce 61-11487, p. 118, 1957) states that Herring and Great Black-backed gulls (Larus argentatus and L. marinus) December 1977 • GENERAL NOTES 617 Fig. 1. Black-legged Kittiwakes building nests on a snowbank, St. Paul Island, Alaska, 4 July 1975. 618 THE WILSON BULLETIN • VoL 89, No. 4 nesting on Kharlov Island in the Barents Sea occasionally build nests on snow, although the majority of pairs wait until the snow has melted. The support of contract number 03-5-022-72 from the National Oceanic and Atmospheric Administration to the senior author is gratefully acknowledged. We thank the National Marine Fisheries Service, St. Paul Island Project, for logistical support. — George L. Hunt, Jr., Dept, of Ecology and Evolutionary Biology, JJniv. of California, Irvine 92717 and Max C. Thompson, Dept, of Biology, Southwestern College, Winfield, KS 67156. Accepted 3 May 1976. Evidence of double brooding by American Kestrels in the Colorado high plains. — Double brooding, although considered uncommon in the Falconiformes, has been reported in the Harris Hawk (Parabuteo unicinctus) in southern Arizona ( Mader, Living Bird 14: 59-85, 1975), Caracara iCaracara cheriway; Bent, U.S. Natl. Mus. Bull. 170, 1937), and the American Kestrel (Falco sparverius) in Florida (Howell, Florida Bird Life, Coward- McMann, New York, 1932). Captive American Kestrels have also produced second clutches after fledging the first brood (Porter and Wiemeyer, Condor 74:46-53, 1972). Observations at 2 nest boxes in southeastern Colorado during 1975 and 1976 breeding seasons suggest that double brooding also occurs in American Kestrels under natural conditions in a temperate climate. The boxes, 2 of 25 attached to wooden H-frame towers of a 230 kV transmission line, were approximately 13 km SSE of Ellicott, El Paso County, Colorado. The terrain is rolling sandhills vegetated with yucca (Yucca glauca) , sand sagebrush (Artemesia fili- folia) , and a variety of herbs and grasses. Insects (Orthoptera, Coleoptera), small lizards, and Horned Larks ( Eremophila alpestris) were available and used as prey items. An adult female American Kestrel was flushed from 5 eggs in Box A on 19 April 1975. She was brooding 3 recently hatched chicks on 14 May, and on 5 June, three 3^-week-old young were banded. Two infertile eggs were also removed. The empty box and its heavily muted top on 17 June suggested successful fledging; 1 kestrel was heard but not seen. On 8 August, 4 infertile eggs and a 2-week-old nestling were found in the box. Four 2V^-week-old young were banded on 25 May 1976 at Box B, 7 km northeast of Box A. On 24 June an ASY female, aged according to Parkes (Wilson Bull. 67:194-199, 1955), was captured on 5 warm eggs. A male escaped from the box while the female was being removed for banding. Another male and female were perched on the tower above Box B. The latter male aggressively defended the nest box suggesting that it was the mate of the incubating female. The other 2 kestrels were passive and less wary than the occupying adults, remaining perched throughout the nest visit. Young kestrels tend to remain in the breeding territory of their parents until fall migration (Balgooyen, Univ. Cal. Publ. Zool. 103, 1976). Since the nesting pair tolerated the extra kestrels in and near the nest, we believe that they were progeny of the first nesting attempt by the occupying pair. The 5 eggs were warm on 16 July but cool on 7 August and showed no development when opened. Handling the female during early incubation may have caused a temporary abandonment, killing the embryos, or all eggs may have been infertile. We realize that our evidence is circumstantial. The 2 passive kestrels at Box B could have been members of an adjacent breeding pair. Nests of American Kestrels have been reported within 34 m of each other (Nagy, Wilson Bull. 75:93, 1963) and no territorial defense was observed between pairs nesting within 60 m ( Smith et al.. Southwestern Nat. 17:73-83, 1972). However, Balgooyen (op. cit.) found that Kestrels defended their territories from other Kestrels primarily by mutual avoidance rather than repeated defense December 1977 • GENERAL NOTES 619 of borders. It is also possible that the 2 extra birds at Box B were nest helpers (Wegner, Wilson Bull. 88:670, 1976). However, nest helpers at Red-tailed Hawk (Buteo jamaicensis; Wiley, Condor 77:480-482, 1972) and Harris Hawk ( Mader op. cit.) nests were as aggres- sive as the nesting pair. The combined evidence from the 2 breeding seasons suggests that double brooding may occur in American Kestrels in southeastern Colorado. Data for this note were collected as part of a study of impacts of transmission lines on wildlife and funded by Tri-State Generation and Transmission, Denver, Colorado, Colorado Division of Wildlife, and Colorado State University. We thank Clait E. Braun and Ronald A. Ryder for their critical review of this manuscript.- — Dale W. Stahlecker and Herman J. Griese, Dept, of Fishery and Wildlife Biology, Colorado State Univ., Ft. Collins, 80523. Accepted 22 Feb. 1977. Further comments on sexual size dimorphism in birds. — Ralls (Wilson Bull. 88:149-150, 1976) published a note on extremes of sexual size dimorphism among birds. We corresponded on the subject and this was useful to me as I was then writing on the selective basis for the “reversed” dimorphism that exists in birds of prey ( Amadon, Raptor Research 9:1-11, 1975). I did not, however, see her manuscript and here offer a few additional comments on the subject. Ms. Ralls observed that Lack (Ecological Adaptations for Breeding in Birds, Methuen, London, 1964:161) quoted me (Amadon, Proc. Am. Phil. Soc. 103:531-536, 1959) as source that the Australian Brown Songlark Cinclorhamphus cruralis is an extreme example of sexual size dimorphism in passerine birds, while the hawk Accipiter fasciatus vigilax shows the extreme of size divergence among those birds in which the female is the larger size. She adds for both examples: “the figures Lack gives are not in the paper by Amadon he cites.” Of the Accipiter this is literally true; they are from an earlier paper (Amadon, Wilson Bull. 55:164^-177, 1943) and are for weight, not wing length as Lack has it. The measurements of Cinclorhamphus are in the 1959 paper but again are for weight not wing length! This confusion does raise the question: What is the best general measure of difference in body size? In my 1959 paper, cited above, which is a review of the subject, I noted that in most species of birds and mammals males compete for females and are the larger sex; while in the few groups in which females compete for males (phalaropes, for example) females are larger. Furthermore, dimorphism is usually greater in polygamous or poly- gynous species in which a few males do most of the mating and individuals of that sex are hence to a degree, expendable. Competition for mates often consists largely of display and threats but actual physical conflict is always latent, and overall body size and prowess, of which gross weight is the best available measure, provide the basis for the selection. Considering now the extremes of sexual size dimorphism in the Class Aves, in a few polygynous or promiscuous species such as the Capercaillie (Tetrao urogallus) the weight of males averages a trifle more than twice that of females. Cinclorhamphus cruralis, mentioned above, is uniquely dimorphic for a passerine bird (and perhaps for all birds). Combining weights of several individuals from South Australia for which I am indebted to Dr. L. L. Short, with those I (1959:533) published earlier, we have the following: 6 ^ 65-83 (69) g; 4 $ $, 28-32 (31) g; the males thus average 2.2 times heavier than the females. A few weights of the smaller and only other species of the genus, C. mathewsi, also supplied by Dr. Short, indicate that it is less strikingly dimorphic: 6 $ $ average 39 g and 3 9 9 23.5 g; thus the males average 1.6 times heavier than the females. Dr. Short informs me that the flight displays of the large males of Cinclorhamphus cruralis cover a wide area and he thinks it highly likely that the species is polygynous. 620 THE WILSON BULLETIN • VoL 89, No. 4 In birds in which secondary sexual behavior is “reversed,” such as phalaropes, females, as noted above, are larger, but not greatly so; perhaps they weigh 10 or 15% more than the males. In hawks and owls also females are larger than males, even though courtship and parental care roles are not reversed. The selective basis for this dimorphism is debated; I (Amadon 1975) think it is because in these aggressive, taloned, predatory species females would be in peril from males at pairing time, were they not larger and able to be the dominant partner. In various species of Accipiter, females average about 1.7 times as heavy as males, and if pairing is random some males will have mates of twice their own weight. For example Piechocki (in Glutz, Bezzel and Bauer, Handbuch Vogel Mitteleuropas, 4, Akademische Verlag, Frankfurt, 1971:420) gave the weights of a series of Accipiter nisus as 13 $ $ 134-162 (149) g; 58 $ $ 220-310 (258) g. Dimorphism in some of the other species of the genus Accipiter, for example the American A. striatus, is similar; I (1943 op. cit.) used A. fasciatus vigilax of New Cale- donia merely because a series of weights was available. Judging from skins, I suspect there may be a few species of raptors in which females will be found to average twice as heavy as males: for example Erythrotriorchis radiatus, or Hieraaetus kienerii. Some of the data presented by Ralls are in the form of cube root of weight. This statistic is useful when it is necessary to compare weights with linear measurements — it reduces the w^eight to a linear equivalent (Amadon 1943) ; but, as stated above, it would appear to be the weight, mass, or bulk of a bird per se that provides the selective material for sexual size dimorphism, and the raw weights themselves should be used. Ralls used 2 other measurements in her comparisons: wing length and total length. The latter is rarely employed because differences in the make of museum specimens, and other factors, render it somewhat unreliable. Also, total length in birds as usually defined includes the tail (feathers), which are not really part of the body. This measurement does, however, suggest a different category of sexual dimorphism in birds; one that reaches its extreme in such species as the African Long-tailed Whydah, Euplectes progne delamerei. In breeding plumage males of this weaver-finch are 50 to 60 cm long, of which 75%, roughly, is represented by the very long tail feathers. Females are only about 15 cm long, so measured this way the males are about 4 times as long as the females. Such dimorphism is based largely upon insubstantial feathers and does not reflect true body size. But in another sense it is size dimorphism: the male whydah looks very much larger than his mates, and does fill more space. Some pheasants (Rheinardia. Pavo) and a few other birds equal or approach the dimorphism in display characters found in Euplectes. This kind of dimorphism, unlike that in weight, is presumably based not upon physical competition for mates but rather upon sexual selection in a Darwinian sense. During the evolution of such species, females have tended to prefer males that were superior in display and the “ornaments” that go with it. Not unexpectedly this type of sexual dimorphism also has been able to proceed further in polygynous or promiscuous species such as those cited. Still, some monogamous species such as the Quetzal, Pharomachrus mocinno, are strikingly dimorphic. As to comparative sexual size dimorphism in birds and mammals, no birds, as Ralls noted, approach the degree of disparity in weights found in a few mammals ( elephant- seals, Mirounga, etc.) in which the male weighs several times as much as the female. Mammals are also far ahead in the development of bulky weapons (antlers), though a few birds are spurred. On the other hand, as befits their visual orientation and the structural plasticity yet light weight of feathers, some birds far exceed mammals and perhaps any other group of animals in the size of display ornaments.- — D. Amadon, American Museum of Natural History, New York, NY 10024. Accepted 1 Sept. 1976. December 1977 • GENERAL NOTES 621 Response of incubating Black-bellied Whistling-ducks to loss of mates. — Earlier work reported that Black-bellied Whistling-ducks (Dendrocygna autumnalis) mate for life and that both sexes in this species share incubation duties (Bolen, J. Wildl. Manage. 35:385-388, 1971). There is no verification from field studies of shared incubation among the other seven species of whistling-ducks except for the single observation of Flickinger (Wilson Bull. 87:106-107, 1975) for the Fulvous Whistling-duck iD. bicolor). This poses the question as to whether the loss of one member of a pair of Black-bellied Whistling-ducks might cause nest failure during incubation. Experimental evidence has thus far been lacking, although our field records include an instance when the death of a male led to nest abandonment by the female. We attempted to experimentally examine this question by removing 1 mate of 2 pairs of incubating whistling ducks and then observing the nest and the remaining mate to determine whether incubation would continue or whether the survivor would remate and renest during the current season. On 28 June 1975 we removed 2 Black-bellied Whistling-ducks from separate nests in boxes designed for their use (Bolen, J. Wildl. Manage. 31:794-797, 1967) ; a male was taken from one nest and a female from the other. We held these birds in captivity for 4 days, then released them 60 km distant from the nesting site. The respective mates of each bird had been previously banded and marked prior to our experiment. In each case the nest was abandoned following the removal of the incubating bird. In one instance, we know that the surviving mate (female) was immediately available to assume incubation as she was repeatedly seen loafing on the pond near the nest box; she was seen in virtually the identical spot the day following the removal of her mate from the nest. Furthermore, a small string placed atop the eggs at the time the male was removed remained undisturbed for 24 h, indicating that the hen had not entered the nest box unseen. The male was not seen again following his release, and we likewise have no further history of the hen following 29 June. In the second case, the nest also failed although the female returned to the nesting area on 2 July following her release; this hen’s mate was noted in the company of a banded bird on 26 July (presumably the pair was then reunited), and on 16 August the male was captured incubating a clutch apparently begun about 24 July. On 27 August the female was captured on this nest and confirmed by her band number as the bird captured and released earlier. The nest successfully hatched by 4 September, and represented a successful renesting attempt on the part of this pair ( cf. Delnicki and Bolen, Auk 93:535-542, 1976). These observations support the earlier observations of Bolen ( Ph.D. thesis, Utah State Univ., 1967) and Delnicki (M.S. thesis, Texas Tech Univ., 1973) that the exchange of incubation duties is initiated by the bird (of either sex) on the nest; this is accomplished simply by the incubating bird leaving the nest and flying to a loafing area where it joins the waiting mate. The loafing mate thereafter returns to the nest to continue incubation without further behavioral interaction or nest exchange ritual. Thus, the simulated or actual death of the bird on the nest interrupts the sequence and the nest is abandoned when the incubating bird fails to join its mate at a loafing site. This study was part of a M.S. thesis accepted by the faculty of Corpus Christi State University. — Richard E. McCamant and Eric G. Bolen, Rob and Bessie Welder Wildlife Foundation, P.O. Drawer 1400, Sinton, TX 78387. (Present address REM: Buffalo National River, P.O. Box 1173, Harrison, AK 72601.) Accepted 5 May 1977. 622 THE WILSON BULLETIN • VoL 89, No. 4 Late Pleistocene Williamson’s Sapsucker from Wyoming. — Archaeological ex- cavations at the Casper Site, a bison kill site in Casper, Natrona Co., W'yoming, yielded a single left humerus referable to Sphyrapicus thyroideus, Williamson’s Sapsucker. The site, described by G. C. Frison (The Casper Site: a Hell Gap Bison Kill on the High Plains, Academic Press, N.Y., 1974) , has been radiocarbon dated to 10,060 ± 170 years B.P. (8110 B.C.: RL-208) and 9830 ± 350 years B.P. (7880 B.C.: RL-125), and there- fore lies on the Late Pleistocene-Holocene boundary. The extinct camelid Camelops is also present in the fauna, lending it a Late Pleistocene aspect. Dental eruption and attrition in the bison population, referred by M. Wilson iin G. C. Frison. op. cit.. p. 132) to Bison bison antiquus, suggest a late autumn kill event. Shed coyote {Canis latrans) deciduous premolars also suggest a late summer to autumn occurrence. If a natural occurrence, the sapsucker could have been a migrating individual, as the sand-dune setting cf the site is at variance with modern habitat preferences of this species. However, its emplacement in the bone bed may have come through the action of human or other predators. Table 1 Maximum Measurements of Sapsucker Humeri UWA27269 S. thryoideus^ S. varius- Mid-shaft diameter 2.85 2.85 2.70 Breadth distal end 6.70 6.55 6.45 Ectepicondylar prominence to external trochlear condyle 4.00 3.75 3.70 CD II ;i CO II Asyndesmus lewis and species of Melanerpes were eliminated on the basis of size as well as characters of the distal end (the proximal head of the fossil is missing). The similar humerus of Picoides villosus differs in having (1) a larger, more deeply excavated olecranal fossa, (2) a larger depression of brachialis anticus, and (3) a shorter ectepi- condylar prominence. The external trochlear condyle of Sphyrapicus thyroideus appears more bulbous than that of S. v. varius and 5. v. nuchalis. In addition, the larger size of the fossil indicates S. thyroideus rather than S. varius (Table 1). The specimen (NC2559, recatalogued UWA27269) is in the University of W'yoming Anthropology collections. P. Brodkorb (Catalogue of Fossil Birds, Part 4, Bull. Florida State Mus. 15:162-266, 1971) lists 2 Pleistocene records for S. varius (1 of them uncertain), but none for S. thyroideus. The Casper Site specimen therefore appears to be the first Pleistocene record of the species. We are grateful to Dr. George C. Frison for his loan of the specimen and for financial assistance to the senior author through the Wyoming Recreation Commission and the University of Wyoming, for the analysis of the Casper Local Fauna. — Michael Wilson, Dept, of Archaeology, Univ. of Calgary, Calgary, Alberta, Canada T2N IN 4; and Amadeo M. Rea, Dept, of Ecology and Evolutionary Biology, Univ. of Arizona, Tucson 85721. Accepted 14 April 1976. December 1977 • GENERAL NOTES 623 American Kestrel rejects captured spadefoot toad. — Although the diet of the American Kestrel iFalco sparverius) includes a wide range of prey items ( Heintzelman, Wilson Bull. 76:323-330, 1964), I could find no records of American Kestrels preying on toads. Therefore, the following observation of a kestrel capturing hut not eating a toad is of interest. On 17 October 1975 at 09:40, about 5 km west of Elgin, Arizona, I observed a female American Kestrel fly about 50 m from a utility pole to the ground in an open short-grass field. Shortly she returned to the pole carrying a toad in one foot. The kestrel picked at the head of the toad sporadically and occasionally shook her head from side to side. After 2 min 55 sec, when she was frightened by a passing vehicle, she carried the toad about 150 m to another pole. She held the toad 3 min 45 sec on this perch before making an attempt to eat it, then began biting the head again but shook her head violently after each bite. After 5 min 55 sec of intermittent bites and head shakes she carried the toad about 200 m to a fence post. Soon she flew a short distance to the ground and returned to the post without the toad. She sat on the post with her feathers ruffled, constantly changed foot positions, and continued the head shaking. After 2 min she flew to the ground and captured a grasshopper which was carried to a utility pole farther out in the field. After eating the grasshopper she still occasionally shook her head. I found the toad on the ground near a small bush. It was crawling feebly and the rostrum was covered with blood but it had no other injuries. The toad, a western spade- foot {Scaphiopus hammondi) , measured 44 mm SVL and weighed 14 g. It exhibited normal locomotion and behavior within 24 h and lived for 22 days before being released. Bent ( U.S. Natl. Bull. 170, 1937) lists “toads” in the diets of 4 species of Accipitridae and Sexton and Marion (Wilson Bull. 86:167-168, 1974) report evidence of Swainson’s Hawks iButeo swainsoni) feeding on plains spadefoot toads i Scaphiopus bombifrons) . Perhaps there are differences in the tolerances of different hawks to the distastefulness of toads and differences in the distastefulness of different species of toads. That the toad was carried to the ground and released rather than dropped from a perch is probably explained by the food storing behavior of American Kestrels. Tordoff (Wilson Bull. 67:139-140, 1955) and Stendell and Waian (Condor 70:187, 1968) reported food storing by American Kestrels, and I have observed it in the Elgin area on 5 occasions; 3 times prey was stored in a small bush. — G. Scott Mills, Dept, of Ecology and Evolu- tionary Biology, Univ. of Arizona, Tucson 85721. Accepted 20 July 1976. Winter distribution of Red-tailed Hawks in central New’ York state. — The winter distribution of raptors in relation to their prey has seldom been investigated systematically. Several authors (e.g. Snyder and Hope, Wilson Bull. 50:110-112; Weller et ah, Wilson Bull. 67:189-193) have noted concentrations of raptors where meadow voles (Microtus sp.) were abundant and Craighead and Craighead ( Hawks, Owls, and Wildlife, Dover, N.Y. 1969:144) concluded that in a 90 km^ study area in Michigan raptor density in winter was highest where vole density was highest. While driving between Ithaca and Albany, New York I noticed on several occasions that Red-tailed Hawk iButeo famaicensis) density along the route varied greatly. This study was undertaken to determine whether the differences in hawk density were cor- related with density of meadow voles (Microtus pennsylvanicus) , one of their principal prey species (Craighead et al., USDA Circ. 370, 1935). Methods. — Five surveys were made on clear days between 1 February and 1 March 1974 on the 241 km route which followed US Rt. 13, NY Rt. 26, and US Rt. 20. 624 THE WILSON BULLETIN • Vol. 89, No. 4 Table 1 Relative Density of Red-tailed Hawks, Frequency of Vole Habitat, and Vole Run- ways Between Ithaca and Albany, New York in Winter, 1974 Hawk density Hawks/km Frequency of good habitat Runways in good habitat Mean* (SD) Frequency of fair habitat Runways in fair habitat Mean* (SD) 0.016 (low) .006 54.8 (15.0) .458 4.0 (4.9) 0.057 (medium) .050 44.7 (7.4) .372 1.0 (1.7) 2.190 (high) .440 93.5 (10.7)** .125 1.0 (1.5) * N = 6 fields in each case (20 samples/field). ** Use of the Student-Newman-Keuls test (Sokal and Rohlf, Biometry, Freeman, San Francisco, 1969 ) showed that in good habitat there were more (p ^ ,01) runways in high hawk density areas than in medium or low hawk density areas. No other differences in runway density were significant. I measured Microtus habitat by driving the route slowly and visually classifying the habitat every .32 km on both sides of the road as “good,” “poor,” or “unsuitable.” Good habitat consisted of recently abandoned fields with a matted, grassy cover. Grass shoots, on which voles feed, were common under this cover. Poor habitat included fields without the distinctive matted cover and had few grass shoots for voles to feed on. Unsuitable habitat included all areas such as plowed fields or woodlots where voles would not be found or where Red-tails would not hunt. Hawk density was particularly high in a short 4.2 km section at the east end of the route. To measure Microtus habitat frequency more accurately in this section, I used aerial photographs and ground surveys to prepare a habitat map of the entire area within 400 m of the read (the average maximum distance at which I recorded hawks). Using the results of the roadside survey (1480 samples) and the habitat map prepared for the short section at the east end of the route, I calculated habitat frequency for the entire route (Table 1). Microtus population levels were then measured in good and poor habitats by counting the number of runways crossing the perimeter of a randomly placed .25 m“ wire frame. Twenty samples were obtained in each of 18 good and 18 fair habitat fields regularly spaced along the route (Table 1). Results. — Starting at the east end of the route, hawk density per kilometer varied from 2.19 in the first 4 km to .063 in the next 51 km to .016 in the final 186 km. Paralleling the change in hawk density, the fre(}uency of good vole habitat dropped from .44 to .05 to .006 in the high, medium, and low hawk density sections respectively (Table 1). Good vole habitat thus varied about as much, and in the same direction, as hawk density. There were few vole runways in any of the poor habitats sampled ( 0-4 runways per field). The number of runways in good habitat was about equal in low and medium hawk density areas hut significantly higher (p ^ .01) in high hawk density areas (Table 1). The number of Red-tailed Hawks observed was thus correlated with the frequency of good Microtus habitat and with high Microtus population indices within good habitat. These results support the hypothesis that, in the study area, Microtus distribution is a major factor determining the distribution of Red-tailed Hawks in winter. The Microtus sampling method was suggested by M. Richmond who also greatly im- December 1977 • GENERAL NOTES 625 proved the manuscript. T. Cade and L. Oring offered additional helpful suggestions on the manuscript. — Jonathan Bart, New York Cooperative Wildlife Research Unit, Dept, of Natural Resources, Cornell Univ., Ithaca, NY 14853. Accepted 14 Sept. 1976. Osprey catches vole.- — On 3 October 1975 at Lighthouse Point Park, New Haven Co., Connecticut, I observed an Osprey (Pandion haliaetus) circle low over a salt marsh, rise slightly, hover in the same pattern it would in catching a fish and then plunge to the ground. It sat motionless for a moment in the short Spartina patens grass looking at its feet then took flight clutching a small rodent. It flew to the ridgepole of a nearby cottage and through a 20X spotting scope I watched it tear its prey apart. When it had finished and left, I retrieved all that remained: the skin from the sides, feet and some entrails of a meadow vole ( Micro tus pennsylvanicus) . Brown and Amadon (1968. Eagles, Hawks, and Falcons of the World. McCraw Hill, New York) list numerous vertebrates as acceptable Osprey prey including birds, frogs, and crustaceans in addition to its normal diet of fish. Wiley and Loher (Wilson Bull. 85:468-470, 1973) give detailed lists of Osprey prey including 12 species of birds, several reptiles and amphibians, and 8 species of mammals, but not M. pennsylvannicus. Spitzer (pers. comm.) found what he believed to be M. pennsylvannicus remains in at least 1 Osprey nest. The literature is lacking in actual sightings of how these mammals are taken. — Noble S. Proctor, Biology Dept., Southern Connecticut State College, 501 Cres- cent St., New Haven 06515. Accepted 6 Aug. 1976. Patterns of feeding Field Sparrow young. — As part of a study of Field Sparrow (Spizella pusilla) breeding ecology (Best, Ph.D. thesis, Univ. of Illinois, Urbana, 1974), I recorded the activities of parents feeding nestlings on the 6th day after the first young hatched. Observations were made from a blind and covered the periods: dawn-08: 00, 09:00-12:00, 13:00-16:00, and 17:00-dusk. A mirror positioned above the nest permitted observation of its contents. Airplane paint was applied to each nestling’s bill for individual recognition (this had no noticeable effect on parental feeding behavior) and adults were marked with colored leg bands. Besides documenting the frequency and temporal dis- tribution of feeding visits (Best, Auk, 94:308-319, 1977), the pattern of food delivery to individual nestlings was also recorded for 6 broods. The pattern of food delivery, which is rarely reported, is the subject of this note. To determine if the sequence of feeding nestlings was random, an interval-distribution test (Ghent and Hanna, Am. Midi. Nat. 85:188-195, 1971) was employed. In only 2 of the 16 nestlings tested (representing 2 of 6 broods), were the intervals between feedings significantly different from a random sequence iP < 0.05). Although this implies no sequential pattern in feeding most nestlings, certain nonsignificant trends were evident. In all 16 nestlings the “observed” frequency of consecutive feedings ( the same nestling being fed twice in immediate succession) was less than the “expected” frequency, while the observed frequency of alternate feedings ( another nestling being fed between suc- cessive feedings of the nestling in question) was greater than the expected frequency in all but 3 nestlings (representing 2 broods). These trends indicate that on the basis of 626 THE WILSON BULLETIN • VoL 89, No. 4 Table 1 The Distribution of Feeding Trips (Male/Female/Both) Among Nestlings Within Each Brood Brood“ Individual nestlings 20 June 15/11/26 (7.8/1.8)‘’ 6/13/19 (8.3/1.8) 9/17/26 (6.0/1.6) 10 14 24 (7.9/1.8) 24 June‘S 8/15/23 (8.0/1.8) 17/ 7/24 (8.7/1.9) 9/15/24 (9.1/1.8) 5 August 20/28/48 (7.0/1.6) 25/21/46 (6.8/1.7) 30/26/56 (8.4/1.8) 7 August*" 44/22/66 (8.3 /1. 8) 32/26/58 (8.0/1.8) 33/12/45 (6.7/1.7) 3/10, 13*" (3.5/1.3) 28 August 15/28/43 (8.5/1.8) 12/27/39 (7.9/T.8) 1 September 17/21/38 (7.4/1.8) 13/25/38 (6.8/ 1.8) ® Date when brood was observed being fed. ^ Nestling weight (g) and tarsal length (cm) measured the day before the feedings were recorded. ^ Significant differences in the proportion of feedings by the male and female to each brood member. ** This nestling had a broken leg. cliance alone, nestlings are fed less often than expected on consecutive feedings but more often than expected on alternate feedings. There were no consistent departures from expected feeding frequencies for intervals greater than 1. The distribution of the total feeding trips among nestlings of a given brood was not significantly different from uniformity in 5 of the 6 broods observed (Chi-square good- ness of fit test) (Table 1). In the brood fed differentially, 1 nestling had a broken leg. This nestling was fed much less frecjuently than the others and could not reach as high when begging for food. When the analysis was restricted to the remaining nestlings, the difference was not significant. The above results suggest that size differences among nestlings did not significantly influence the number of feedings each received, although there was a tendency in most cases for the larger nestlings to be fed more frequently (Table 1). Brood reduction resulting from starvation was not observed during the entire study and the only nestling exhibiting abnormally slow growth was the one with a broken leg. Availability of nestling food did not appear to limit breeding success on the study area (Best, Auk, op. cit.). The proportion of feeding trips by the male and female to each nestling of a brood generally differed (Table 1), although in only 2 of 6 instances (both broods of the same pair) was the difference statistically significant (Chi-square contingency analysis). In most cases differences were complementary, tending to balance the frequency of feeding each nestling. The influence of spatial arrangement in the nest on how frequently each nestling received food was determined for 4 broods by comparing the positions of all nestlings December 1977 • GENERAL NOTES 627 during each visit with the position (s) of the nestling (s) receiving food (occasionally 2 nestlings were fed during a visit, but usually only 1). Twelve positions were selected reflecting the hours on a clock face (e.g. during a visit the 3 brood members were at 01:00, 05:00, and 10:00 with the nestling at 05:00 receiving food). The adult’s position on the nest rim was also recorded during each visit. A Chi-square test for goodness of fit was used to determine if the frequency of feeding nestlings at various positions departed significantly from the frequency nestlings occupied those positions during feedings. Adult male and female feedings were considered separately as well as combined. In only one instance were the results statistically significant (Fig. 1) and then only for the spatial feeding pattern of the male (P < 0.005). Apparently the frequency of feeding nestlings in various regions of the nest is usually determined by how frequently those positions are occupied by young, and not by the adults’ preference to feed in particular areas. All adults did, however, show strong preferences to feed from specific areas on the nest rim (see Fig. 1 for example) . In some instances both members of the pair used the same feeding position while in other cases they did not. Fig. 1. The positions of adults and nestlings during feedings of the 7 August brood. Bar lengths indicate frequencies. Diagram A shows the feeding positions of the adults from the nest rim (bars outside circle), the positions of all nestlings during feeding visits (bars inside circle), and the positions of the nestlings actually receiving food (black portion of bars). Diagram B illustrates the positions of the 4 individual nestlings during feedings. Although the young rearranged their positions in the nest frequently throughout the day, brood members showed a strong propensity to occupy different regions of the nest in all 4 broods considered (P < 0.005, Chi-square contingency analysis) (see Fig. 1 for example). When parents feed the young preferentially in different regions of the nest (which generally appeared not to be the case in this study), the nestlings’ spatial arrange- ment in the nest could result in differential feeding. The referees’ suggestions for revising the manuscript were appreciated. — Louis B. Best, Dept, of Animal Ecology, Iowa State Univ., Ames 50011. Accepted 3 Nov. 1976. 628 THE WILSON BULLETIN • Vol 89, No. 4 Avian bone pathologies from Arikara sites in South Dakota. — Vertebrate remains recovered in aboriginal sites often provide the archaeologist with valuable data pertaining to the economic and social use of animals by various Indian groups. Certain groups of birds such as the Anatidae comprised a significant part of their diet ( Howard, Univ. Calif, Publ. Zool. 32:301-387, 1929; Parmalee, Bull. 111. Arch. Surv. 10:137-155, 1976) while others, for example representatives of the Accipitridae and Corvidae, played a major role in ceremonial and related customs (Ubelaker and Wedel, Am. Antiquity 40: 444-452, 1975t , The identification of avian bones found during archaeological excavations may provide the zoologist with noteworthy prehistoric or early historic species’ distribution and abundance records. Data pertaining to sex and age ratios of certain species, season of death (collection by the Indian), significant taxonomic characters, and osteological anomalies may also be obtained from archaeologically derived faunal samples. The occurrence of mammalian bones which exhibit some form of anomaly or pathology are recovered occasionally in Indian refuse deposits, but similar elements of birds are extremely rare. This may be due to the fact that birds were seldom taken in numbers comparable to those of mammals and, consequently, fewer elements were preserved. How- ever, the chances of a bird with a broken leg or especially a wing surviving during the period necessary for healing are probably small. Even if such a break would heal in a manner that would allow the individual to again function normally, the possibility that the Indian would later kill such a bird and its bone be preserved in the midden debris is even more remote. Therefore, the recover>- of 3 extreme cases of bird bone pathology encountered during identification and analysis of an avifauna from 51 Arikara sites in South Dakota are worthy of description. Archaeological salvage work was carried out along the Missouri River in North and South Dakota from about 1950 to 1965 prior to the construction of 5 major dams by the U.S. Army Corps of Engineers. During this period considerable quantities of animal refuse were salvaged from several Plains Indian village sites which were to be inundated. I have examined nearly 3100 bird bones from 51 of these South Dakota sites which were formerly occupied by groups of Arikara. This Plains tribe was originally a part of the Pawnee confederacy; these Indians migrated to the Middle Missouri River area from Nebraska and established numerous villages along the river. Approximately 68 species of birds, representing 22 families, were identified. Remains of waterfowl, hawks and eagles, grouse, and corvids comprised about 84% of the total. A minimum of approximately 870 individuals were represented in these 51 sites and of this total, elements of only 3 birds. 2 hawks and 1 duck, exhibited a pathological condition. Four such elements, occurring in the same archaeological excavation unit and undoubtedly from the same individual, were recovered at the Crow Creek site (39BF11: occupied ca. AD 1400-1550). The bones consisted of a right tarsometatarsus and radius/ ulna and a left humerus < Fig. 1) and were those of a hawk iButeo sp.). Several of the hroad-winged hawks are extremely difficult to separate osteologically even when the elements are normal, so these distorted bones make a specific determination more un- certain; however, these compare most closely with the Rough-legged Hawk iB. lagopus) . The most interesting of the 4 elements are the radius and ulna which had been broken about 1/4 the distance from the distal end. During the healing process these bones became fused at or near the point of fracture by a bridge of callus bone. The distal % to % of these elements was greatly swollen with porous new bone which resembles a tumor. This swollen appearance is apparently due, in part, to the overlapping broken ends of the bone which double the normal diameter of each; healing in such a manner would tend to shorten the length of the wing. The distal articulating surfaces of the radius and ulna December 1977 • GENERAL NOTES 629 appeared only minimally affected and there was probably little or no disruption in free articular movement with the cuneiform, scapholunar, and proximal end of the carpo- metacarpus. Once complete healing had occurred, the bird was probably able to again fly. Neither the humerus nor the tarsometatarsus recovered in the same unit with the pathological radius/ulna had been broken, but the deformities and surface irregularities apparent in both suggest they may well have been from the same hawk. It is possible that these 2 elements, as well as others, could have been affected as a result of the traumatic condition of the broken wing and/or possibly inadequate diet during the healing process. The proximal end of the humerus is bent upward (anconal view) and the area of attachment of the latissimus dorsi, pneumatic fossa, bicipital surface, and bicipital furrow appear granular or roughened. The proximal end of the tarsometatarsus is also bent upward (anterior view) and slightly twisted (Fig. 1). Although the break in the wing of this hawk healed completely, the total effect osteologically may have been detrimental to a normal existence. The second incidence of pathology involved the left tarsometatarsus of a hawk, possibly Rough-legged Hawk or Red-tailed Hawk (B. jamaicensis) found at the Black Partizan site (39LM218: occupied ca. AD 1550-1675). The element had been broken near the distal end (Fig. 2, A) and, as evidenced by the deposition of porous new bone, a certain amount of healing had taken place. Whether or not the fracture had completely healed at the time of the bird’s death is uncertain since the proximal section of the tarsometa- tarsus was not recovered. The callus bone was extremely irregular, however, and the use of the tarsometatarsus was probably considerably reduced as evidenced by the thin areas of ossification. In any event, this hawk had survived long enough before being killed to permit a considerable amount of healing at the fracture. In his study of bone injuries in birds, Tiemeier (Auk 58:350-359, 1941) found that nearly 13% of 256 skeletons of the ducks he examined possessed one or more broken or otherwise damaged and repaired elements. Generally, for all groups of birds he exam- ined, wing and leg elements were the ones most subject to injury, although damage to skulls, sterna, and furcula, especially in the passerines, was not uncommon. Of 6212 bird skeletons (59 families) examined by Tiemeier (op. cit.) from the collections of the Museum of Natural History, University of Kansas, Lawrence, no injury to the coracoid was mentioned. Therefore, the apparent rarity of injury to this element, coupled with the extreme pathological condition of a left coracoid from the Hosterman site (39P07: occupied ca. AD 1550-1675) is of particular interest. This duck coracoid (Fig. 2, B-D) appears to have been broken or at least severely damaged in the area of the glenoid and/or scapular facets. Unlike the hawk elements described above, the callus was small and very compact, although the bone exhibited an extreme degree of deformity. The shaft was divided into 2 sections, producing 1 large and 1 small hole (viewed laterally) between them; the scapular facet appears as a deep U-shaped groove; the glenoid facet is reduced, or divided, and bent at about a 45° angle; and the surface of the ventral “half” of the shaft possesses an oblong depression in which the head of the humerus may have articulated. Because of its deformed condition, this element could not be specifically identified; the overall length and general configuration of the head and sternal facet compare with that of the Mallard {Anas platyrhynchos) . There is little doubt that such an injury to the coracoid would have inhibited the wing from functioning normally and would have limited or entirely prohibited its use for flight until healed. At least 1 tribe of Plains Indians, the Mandan of North Dakota, are known to have kept live owls as soothsayers (Thwaites, ed., Early Western Travels, Vol. 23, Arthur H. Clark 630 THE WILSON BULLETIN • VoL 89, No. 4 Fig. 1. Pathological conditions of hawk {Buteo sp.) elements from the Crow Creek site, South Dakota. A: left humerus, anconal view; B: fused right radius and ulna, palmar view; C: tarsometatarsus, anterior view. Co., Cleveland, OH, 1906) , hut there is no evidence suggesting that hawks were kept alive for any reason. In discussing animal ceremonialism of the Miwok Indians of California, however, Heizer and Hewes (Am. Anthropologist 42:587-603, 1940) state that “Eagles, condors and falcons were kept captive, and might he traded from tribe to tribe,” but apparently such birds were held only temporarily. Eagle trapping was a significant trait of all Plains tribes and a ritual which involved a considerable amount of preparation and December 1977 • GENERAL NOTES 631 Fig. 2. A: anterior view of a fractured left tarsometatarsus of a hawk (Buteo sp.) from the Black Partizan site, South Dakota. Three views (B, lateral; C, dorsal; D, external) of a fractured left duck coracoid from the Hosterman site, South Dakota. organization (Wilson, Am. Mus. Nat. Hist. Anthro. Papers 30:99-245, 1928). Large numbers of hawks and eagles were often taken during these hunts and, although the first few eagles caught were sometimes kept as decoys, most were killed at the end of the hunt. There is no mention of hawks being used in a like manner. Hargrave (Univ. Ariz. Anthro. Papers No. 20:1-67, 1970), in his osteological study of macaws (Ara spp.) from prehistoric Pueblo sites in Arizona and New Mexico, found that 47% of the 145 individuals exam- 632 THE WILSON BULLETIN • Vol 89, No. 4 ined displayed pathological bones; he concluded these conditions reflected normal accidents or dietary deficiencies. The high percentage of bone pathologies evident in the maca’/s are indicative of a captive state with inadequate diet and generally poor treatment contributing to this condition. With no evidence of the Plains Indian keeping captive hawks, it may be reasonably assumed that the 2 birds exhibiting fractured bones were injured in “natural” accidents. Regardless of the cause, the fact remains that many wild birds which suffer severe fractures are able to survive adverse conditions during the healing period and eventually return to their natural mode of life. The broken and repaired humeri of ducks described and illustrated by Abbott (Auk 60:447, 1943) and Dillon (Auk 78:273-274, 1951) exemplify the fact that a healed break of a major wing element, even though distorted, may enable the bird to regain the power of flight. The extremely deformed coracoid from the Hosterman site is of special interest, however, both from the standpoint of how such an injury occurred and its possible effect on normal wing movement. — Paul W. Parmalee, Dept, of Anthropology, JJniv. of Tennessee, Knoxville 37916. Accepted 1 Nov. 1976. Nest reciprocity in Eastern Phoebes and Barn Swallows. — A few investigations of the Eastern Phoebe (Sayornis phoebe) and Barn Swallow (Hirundo rustica) have revealed use, with modification, of old nests of 1 species by the other (Stoner, New York State Mus. Circ. 22:1-42, 1939; Graber et ah. 111. Nat. Hist. Surv. Biol. Note No. 86, 1974; Jackson and Weeks, Alabama Birdlife 24:7-9, 1976). In March 1970 on Crane Naval Ammunition Depot (NAD Crane), Indiana, I examined and marked all Eastern Phoebe and Barn Swallow nests that remained under bridges and culverts from previous nesting seasons. Of the 242 old nests examined, I found 7 instances of reciprocal use — 3 former Barn Swallow nests modified by phoebes, 1 phoebe nest converted by a Barn Swallow, and 3 nests illustrating multiple reciprocity. In these and subsequent modifications, the pattern was similar, with Barn Swallows adding mud and dried grass, and phoebes adding moss to the nests’ rims; each species also lined the nest with the appropriate material. The multiple reciprocity nests were all found beneath bridges. The largest nest, 32 cm in height, was composed of 7 alternating Barn Swallow and phoebe nests beginning with a Barn Swallow base (Fig. 1). Another nest was of similar construction but com- posed of 4 tiers of alternating nests built on a phoebe base and 24 cm in height. A third nest appeared to be a single, excessively large (21 cm in height) Barn Swallow nest modified by phoebes, and only at removal at the end of the season did the nest separate to reveal a second, intermediate layer of phoebe nesting material. Of the 235 normal nests marked prior to the 1970 nesting season, 8 were modified and used during the season by the other species. Six old Barn Swallow nests were adapted by phoebes, and 2 old phoebe nests converted by Barn Swallows. One of these latter nests and another phoebe nest, built early in the 1970 season, fledged broods of phoebes prior to their modification and successful late season use by Barn Swallows. Since old nests were removed at the end of the 1970 season, no nests were available for reuse in spring 1971. However, 1 phoebe and 1 Barn Swallow nest built early in the 1971 season were subsequently converted and used by the alternate species. All nests December 1977 • GENERAL NOTES 633 Fig. 1. An extreme example of successive nest use by Barn Swallows and Eastern Phoebes. were removed at the end of the 1971 season with only sporadic examinations made from 1972 through 1975. At the close of the 1976 breeding season, 1 found 6 Barn Swallow nests altered by phoebes and 3 phoebe nests remodeled by Barn Swallows; all were under bridges. Both species regularly nest under bridges and add material to and use old nests of conspecifics, so it is perhaps not surprising that these birds occasionally demonstrate reciprocal use. It seems that in most instances both species recognize and prefer old nests of their own species. The question then arises as to why a bird chooses an old 634 THE WILSON BULLETIN • VoL 89, No. 4 nest of another species while one or more conspecific nests are available. And further- more, why is a particular nest chosen from among all those available? The possibility exists that certain sites are optimum, representing the ultimate com- bination of location parameters for both species. However, I compared empirically the site parameters for all modified nests and found no consistent relationship between nests in such features as height above ground or water, location over land or water, direction faced, distance from nearest side of bridge, and distance from nearest end of bridge or culvert. There was, however, a definite tendency for birds to choose nests that were adherent, as opposed to statant (Samuel, Wilson Bull. 83:284-301, 1971), and in which the distance to overhead cover was relatively great. This was especially evident in the selection of Barn Swallow nests by phoebes. In 5 instances phoebes chose ( from among 5-30 available nests under a single bridge) the one with the greatest rim-to-overhead cover distance. Phoebes generally prefer a greater distance to overhead cover than do Barn Swallows; significantly (P < 0.05; %^) more phoebe than Barn Swallow nests had rim-to-overhead distances ^6 cm. Of adherent Barn Swallow nests, 35.6% had <5 cm rim-to-overhead distance. In addition to giving the incubating phoebe abnormally little head space, the typical Barn Swallow nest would prevent the addition of any substantial amount of moss to the nest rim. Barn Swallow nests that phoebes modified were usually built lower on the supporting beam to take advantage of some feature, e.g., mud dauber ( Trypoxylon poll turn or Sceliphron caementarium) nests, wood splinters, or rough concrete, that facilitated initial attachment of nesting material. Although the top rim of the multiple nest in Fig. 1 was within 7 cm of overhead cover when discovered, the initial nest had 28 cm rim-to-overhead cover distance. There seems to be no consistency with respect to distance to overhead cover or other location parameters in the few phoebe nests modified by Barn Swallows. Of the 6 nests for which complete data are available, the distance to overhead cover ranged from 6 to 25 cm and height from 1.1 to 3.4 m; nests faced 3 cardinal directions and both upstream and downstream. The fact that reciprocal nests are usually adherent is not surprising since both species prefer these types of locations. Over 92% of all Barn Swallow nests and 54% of all phoebe nests were adherent. The size of the prospective nest may also play a role in its selection for modification. Once a nest is modified, it appears to he used year-after-year with greater regularity than nests without reciprocal alteration. Six of 7 nests (86%) built and modified prior to 1970 were used in 1970, while only 54% of all other nests present before 1970 were used during that year. Converted nests don’t seem to accumulate, e.g., about as many nests were adapted in 1970 (8 nests) as had accumulated from all seasons prior to the 1970 (7 nests) season and in the years from 1972 to 1976 (9 nests). Modifications are likely more frequent than observed from periodic counts, with the intervening rim- tailorings being masked by material added in the successive species shift. The result of these annual additions, whether reciprocal or conspecific, is a larger nest. Birds frequently selected one of the largest available nests for repair or modification and use. The selective advantage to birds in choosing large nests may he that such nests are more durable. While nest turnover is not excessively high, 5 to 10% are lost each season. Those nests that survive many years undoubtedly have firm initial attachments, resulting in less likelihood of destruction during the nesting cycle. As did Jackson and Burchfield (Am. Midi. Nat. 94:503-509, 1975) in Mississippi, I found mud dauber nests of utmost importance in facilitating a firm initial attachment, with 38.0% of adherent Barn Swallow and 51.5% of adherent phoebe nests receiving some support from these insect nests. December 1977 • GENERAL NOTES 635 I wish to thank C. M. Kirkpatrick and F. H. Montague for critical review of this manuscript. This is Journal Paper No. 6464 from Purdue Agricultural Experiment Station. — Harmon P. Weeks, Jr., Dept, of Forestry and Natural Resources, Purdue Univ., West Lafayette, IN 47907. Accepted 12 Jan. 1977. REQUESTS FOR ASSISTANCE Colored-marked Sandhill Cranes. — During summers of 1975 through 1977, researchers at Clarence Rhode National Wildlife Range on the Yukon-Kuskokwim Delta, Alaska, banded and color-marked Lesser Sandhill Cranes. In July and August of each year, chicks were marked with black-numbered yellow collars and leg bands as well as standard Fish and Wildlife Service aluminum bands. To date, 10 resightings have been reported, yielding valuable information on migration routes and wintering areas. More information is needed on timing and pathways of migration, however. Observers are asked to report the date and location of sightings, size of the flocks with which marked birds were seen, whether color bands were on the birds’ right or left legs and, if possible, the numbers on the collars and leg bands. Report sightings to Cheryl Boise, Wildlife Research Unit, Irving Building, Univ. of Alaska, Fairbanks, AK 99701. Sightings of Sandhill Cranes in northwestern Ontario. — The Greater Sandhill Crane breeds throughout the Great Lakes states and parts of Saskatchewan and Manitoba. However, in northwestern Ontario (area south of 51° latitude and west of Sault Ste. Marie) they are considered rare transients. Over the past decade, cranes believed to be Greater Sandhills have been sighted with increasing frequency throughout this area. Occurrences of immatures in the last year suggest that there is a population of Greater Sandhills breeding in N.W. Ontario. Information from sightings will be used to estimate the popu- lation distribution and to determine an area for an intensive study of biology, migration, and taxonomic verification. Please include date, location, and number of birds. Dr. C. D. Ankney, Department of Zoology, University of Western Ontario, London, Ontario, Canada, N6A 5B7. ORNITHOLOGICAL LITERATURE The Birds of the Bahamas. By P. G. C. Brudenell-Bruce. Taplinger, New York, 1975: 142 pp., 4 color and 9 black-and-white plates, 2 maps. 110.95. — The first thing that confronts a reviewer of this book is the title. Apparently the American publisher decided that a field guide to “The Birds of New Providence and the Bahama Islands,” the title of the British edition, would sell better under the broader designation. Accord- ingly, although this book is printed in England and to all intents is the British edition (complete with British spellings), the dust jacket cover, end flaps, and title page of the “American edition” have been changed. Taplinger did not go to the expense of altering the title on the book spine (under the dust jacket, from a purchaser’s viewpoint) or on the half title page; these still declare it to be “The Birds of New Providence and the Bahama Islands.” And so it is. A field guide to Bahamian birds has long been needed. The only other recent book on the subject is so poor as to be useless. Therefore, until the present guide appeared, one had to carry both North American and Bond’s West Indies guides to identify birds seen in the Bahamas. Now one still has to carry a North American guide, but Brudenell- Bruce’s book weighs less than Bond’s and is limited to the Bahamian fauna. It includes all the species recorded in the Bahamas (206, plus 69 accidentals), but illustrates only 31 native land birds in color and another 61 species in black-and-white. The remaining 111 species are referenced by plate number to Peterson’s eastern guide. Apparently by error, illustration references to 3 species are omitted; all 3 are available in Peterson. The hook focuses on the birds of New Providence both because that island is by far the best known ornithologically, and because the author could draw on his own expe- riences during the 4^/4 years he lived there. New Providence is a small island, with only 1.3% of the land area of the country. Nevertheless, it has over 60% of the human population, the capital city of Nassau, and a busy tourist trade. From the Bahamian point of view, Nassau New Providence are of paramount importance — hence the custom of collectively calling all the other islands the “Out Islands” (or, more recently, the “Family Islands”) even though they make up 98.7% of the total land area. The Bahamas contains approximately 17 major islands and some 700 smaller ones. Ornithologically, New Providence is both typical and atypical of the country. It is one of the 4 northern islands to contain pine forest. It also receives more rainfall than do the southern islands, and is not subject to the extreme drying effects of the southern trade winds. New Providence is by far the most disturbed island in the Bahamas, hut it may also he the most diverse in terms of numbers of habitat types. It has more cleared land, flowering gardens, and large trees (other than pines) than do the others. Approximately 222 species of birds have been recorded there — with only an additional 53 known from all the rest of the country. Of the New Providence species, 40 are year-round residents, 9 are summer breeders, 70 are winter visitors, 71 are passage migrants only, and 32 have been listed as vagrants. The text arrangement and nomenclature follow the AOU Check-list and Bond. Each species account begins with the English language name ( occasionally supplemented wdth older, British, or local synonyms) and the scientific name. The bcdy of most accounts is divided into: status (first on New Providence, then in the Out Islands) ; description and habits; voice; and nest. The thoroughness of each account is heavily influenced by Brudenell-Bruce’s experience with the species. The treatment of birds he knows well is generally excellent, filled with useful, often new, information. If a species does not 636 December 1977 • ORNITHOLOGICAL LITERATURE 637 occur on New Providence, however, it is too often given short shrift. For example, the interesting and rare endemics that are found only on other islands: the Cuban Parrot, Amazona leucocephala bahamensis, rates only 4 lines (and those partly inaccurate) ; the Black-cowled Oriole, Icterus dominicensis northropi, another 4 lines (vs. the New Providence passage migrant Icterus galbula with 14 lines) ; and the West Indian Red- bellied Woodpecker, Melanerpes superciliaris (3 endemic subspecies), only 6 lines. Even the national bird, the American Flamingo i Phoenicopterus ruber), is granted only 9 lines, heavily devoted to explaining that it can be seen on New Providence only in captivity. Thus the author not only skimps on Bahamian specialties, but also generally ignores the fact that many are conservation/protection problems. Of the endemics listed above, only the oriole is not a “Red Data Book” bird. As Brudenell-Bruce has visited only 4 other islands for brief periods, most of the information on Out Islands species is drawn from the experience of others, notably from James Bond’s writings and records obtained from C. Russell Mason. Granted that many of these islands are poorly known ornithologically, more information is available than the author found. Even in the references he used, he sometimes missed pertinent information. For instance, he states that the Bobwhite is not found in the Out Islands, yet Bond (at least as early as 1961) gives it as also introduced on Andros and Eleuthera. From personal experience I know it is doing well on Andros. The author curiously ignores Rock Doves (which I have seen at least on New Providence, Eleuthera, and Great Exuma), but he includes Starlings and House Sparrows as well as more exotic introductions such as White-bellied Doves {Leptodla jamaicensis) and 2 species of grassquit (Tiaris) . In a field guide it is never possible to cite references for all records, but I wonder on what authority the author suggests that the Northern Mockingbird (Mimus polyglottos) might not occur on Crooked Island (I saw it there in March 1976). The author’s lack of experience outside New Providence also shows in some of his species descriptions. No one familiar with the Melanerpes in the field would italicize red belly as an important field mark. The upper border of the face mask on Bahama Yellowthroats (Geothlypis rostrata) is only lightly washed with yellow in some individuals, and is a poor field character. The wintering North American race of Yellow-throated Warbler ( Dendroica dominica) is described (as having white underparts), but not the distinctive endemic race (with yellow underparts) on Grand Bahama and Abaco; nor is the reader aided by the black-and-white illustration. The plates by Hermann Heinzel range from good to excellent. Some of the species that apparently were unfamiliar to the artist look a bit wooden, but most are good field guide renditions. I noticed, however, a few errors: for example, the Melanerpes wood- pecker is far too white on the head and underparts; and the Bananaquit (Coereba flaveola) lacks its distinctive red mouth corners. Interestingly, the Bahama Yellowthroat male was painted with a gray-blue-white mask border, while the plate caption states it is “Distinguished from Common Yellowthroat by . . . yellowish upper border to mask . . . ,” and the Greater Antillean Bullfinch (Loxigilla violacea) plumages are marked $ and 9 on the plate, but adult and immature (correct) in the caption. This book clearly has drawbacks, especially as a guide to the Bahamas as a whole. Many of the descriptions are too brief, especially of birds that are not illustrated, absolutely requiring the supplementation of a North American guide. My main objection, however, lies with the too-brief accounts of many of the native species — the sort of material not easily found elsewhere and surely to be expected in a guide to a limited avifauna. Certainly the resulting book is slim enough that much more text could have been included. 638 THE WILSON BULLETIN • Vol. 89, No. 4 On the positive side, the author has done a generally competent job on New Providence birds, and has included several useful features such as local common names, appendices with details of accidental records and song periods of 9 breeding species, and a good index. The book is well bound — my copy has had hard use in the field and is still in fine condition — especially for an English binding. But mostly this guide is, at last, a field book limited to Bahama birds — a relief after years of picking through Bond’s guide which, excellent as it is, covers a much larger fauna. — Mary H. Clench. Ornithological Gazetteer of Ecuador. By Raymond A. Paynter, Jr., and Melvin A. Traylor, Jr. Museum of Comparative Zoology, Cambridge, Mass, and Field Museum of Natural History, Chicago, 111., 1977: viii -|- 151 pp., 2 maps, paperbound. $5.00. Order from Bird Dept., Mus. Comp. Zook, Harvard Univ., Cambridge, Mass. 02138 or Bird Division, Field Mus. Nat. Hist., Chicago, 111. 60605. — This is the second in a series of gazetteers of Neotropical localities where birds have been collected or observed. With a few exceptions, the format of the first (Bolivia) gazetteer has been maintained (for review, see Wilson Bulletin 88:679-680, 1976). The bibliography of the Bolivian volume listed only publications cited in the text, but nevertheless amounted to what the authors had described as “virtually a complete bibliography of Bolivian ornithology.” Paynter and Traylor, in their new volume, have tried to assemble “a complete list of all publications concerned primarily with Ecuadorian birds (exclusive of the Galapagos), not merely those that have been cited in the gazetteer.” I need hardly say that this bibliography of 198 titles will be invaluable to workers on South American birds, and the authors are to be congratulated for their decision to augment the list of references cited. Paynter and Traylor explain that preparation of a gazetteer for Ecuador was vastly more difficult than was true for Bolivia. The literature pertaining speeifically to Bolivia is relatively small, and only a few museums have significant collections from that country. Ecuador, on the other hand, has a long and complicated history of ornithological explora- tion, much of it by professional collectors who sent birds by the tens of thousands to museums and private collections around the world. Label data from many of these specimens are inadequate or misleading. The authors’ introduction goes into some detail as to specific problems encountered in trying to verify itineraries and localities. An innovation is a second map ( in addition to that showing major political subdivisions and rivers) of Ecuador on which all collecting sites (and a few observation sites) have been dotted — the only other information on the map is the 1000 meter contour. “This should help the zoogeographer to decide whether a gap in a distribution may be of biological significance or whether it may merely reflect the absence of collectors.” The map certainly calls attention to those portions of Ecuador that have been relatively neglected, notably the Amazonian east and the northwest, in contrast to the heavily collected high Andes in the vicinity of Quito. The authors have done such an important service to Neotropical ornithology that a reviewer must seem to be a cranky ingrate for wanting even more. Yet the gazetteer could have been substantially more complete and its accuracy improved had the authors followed the suggestion made in my earlier review, and circulated a preliminary copy among museums with major holdings in Ecuadorian birds. The amount of additional work that would have fallen on the shoulders of the authors themselves would have been minimal. For example, there are nine localities listed as “not located” that are cited only from publications of J. T. Zimmer of the American Museum of Natural History. During a visit to that museum, I was able to solve six of these problems in less than an December 1977 • ORNITHOLOGICAL LITERATURE 639 hour. As Paynter and Traylor suspected, some of these were typographical errors or were mistranscribed from labels by Zimmer: “Guapiles” and “Guapilo” for Guapulo, and “Lonambo” for Conambo, for example. “Rio Yamisa,” which they could not locate, is “Rio Yamasa” on the specimen label. This spelling is not in the gazetteer, but the label data also include the altitude, date, and collector’s name, which might have helped to identify this locality. The authors were unable to find “Chitoque,” a locality listed by Zimmer for Tangara nigroviridis consohrina. By looking up the cited specimen, I was able to determine from the label that Chitoque is or was on the Alamor-Guachanama trail, Prov. de Loja, at an altitude of 57 feet, and that the specimen had been taken by Cherrie and Gill on 13 September 1921. I have been assured by ornithologists at the American Museum that they would have been delighted to provide this kind of information had the authors so requested. Carnegie Museum of Natural History has relatively few birds from Ecuador. Most of these were obtained by exchange from the Moore Laboratory of Zoology, Occidental College. In checking the collecting localities represented in our Moore material, I found ten that are not listed at all by Paynter and Traylor, and supplementary information (such as alternative spellings and extensions of collectors’ itineraries) for 19 more. Although I have not discussed this with staff members of the Moore Laboratory, I have no doubt that they would have been happy to cooperate in this important project by consulting the itineraries and maps of Robert T. Moore that are housed at Occidental, thus filling in some significant gaps in the gazetteer. The Bolivian and Ecuadorian gazetteers will be extremely valuable assets to the working libraries of students of Neotropical birds, as will, I am sure, additional volumes in this series. Again I urge the authors to take the time (I see no reason why there should be critical publication deadlines for works like these) to circulate copies of future manu- scripts among those of us who might be in a position to help make the gazetteers even more nearly complete. Few ornithologists of my acquaintance would begrudge the time needed to assist such a worthwhile project in view of the tremendous effort already made by the authors. — Kenneth C. Parkes. Maintenance Behavior and Communication in the Brown Pelican. By Ralph W. Schreiber. Ornithological Monographs No. 22, 1977: 78 pp., 38 figures. $6.00 ($5.00 to A.O.U. members). — This highly detailed description of Brown Pelican colony behavior is the first full ethogram for any pelican, and will doubtlessly become the basis for comparison with other species in the family. Schreiber’s presentation is well organized, thorough, and usually very clear. The rather scanty extant literature on other pelicans is integrated throughout. A major drawback for anyone other than a pelican worker is that the monograph is boring to read. Few readers will care to plod through 34 pages of raw descriptions of comfort activities, and the “communication” half is not much more exciting. Schreiber’s carefully worded descriptions of motor patterns are frequently interspersed with lengthy quotations from his own field notes, as if they were a special source that must be quoted verbatim. Only rarely is the reader’s perserverance rewarded with an eyebrow-raising anecdote (for example, if a male pelican drops a stick while flying back to the nest he usually completes the elaborate Nest Material Presentation anyway). The monograph contains little quantification, mostly just the statement of how many times Schreiber observed a particular behavior in the field. 640 THE WILSON BULLETIN • VoL 89, No. 4 Considering that Schreiber has spent several years doing the field work (his diligence and care are apparent on every page), I was disappointed in the monograph on two counts. First I felt that Ornithological Monographs got stuck with the driest material from a very holistic field study. (Schreiber notes on page 1 that his data on nesting cycles, population fluctuations, age-class composition, reproductive success, and plumage characteristics will be published elsewhere. His data on chick growth patterns, feeding behavior, pesticides, and population history have already been published.) I wonder if the monograph series might not have rated something beyond raw ethological description. Secondly I felt that Schreiber could have posed some more fundamental (and interesting) evolutionary questions about the behavior he described. For example, he explains that Brown Pelicans have remarkably sluggish social lives. They use only 5 displays and generally keep things very simple. By contrast, many other colonial birds, some of which nested in the study colonies near the pelicans, have highly complex social interactions. Why are the pelicans so simple? How can they accomplish the many tasks of pair- formation and successful reproduction with only 5 signals? Schreiber is probably the only person in the world sufficiently knowledgeable about pelicans to intelligently ponder such matters and his speculations could have been provocative. The paper begins with maintenance behavior (shaking, stretching, scratching, preening, bathing, locomotion, and thermoregulation) because some of these motor patterns have been modified by evolution as social signals. Then comes a brief transitional section on “Attack and Escape” before the concluding “Social Behavior.” The generally lucid descriptions are profusely illustrated with 270 mediocre line drawings, including 30 of preening and 10 more of a comfort activity called “glottis exposure.” Do we need 10 illustrations of “glottis exposure?” There are no photographs, though many of the drawings were derived from original photos and a few are even called “photos” in the text ( p. 61 ) . The volume is reasonably free of editing errors though “flys” and “uropygeal” were found. Display names are capitalized in accordance with Moynihan’s established con- vention and then the word display is redundantly tacked on (e.g., Bowing display). The caption for figure 25 is nonsensical. On page 36 an apparent error in paragraph organiza- tion implies that courting male pelicans may go as long as 3 weeks without food! (If this is not an error then it deserves considerably more explanation! ) Elsewhere the description slips occasionally into vagueness (e.g., we are informed that the Upright is often main- tained “. . . for some time" ) . One popular but unfortunate descriptive trick that Schreiber uses repeatedly is the concept of behavioral “intensity.” This is a nonterm referring to variability in the broadest sense. We read about high- and low-intensity display per- formances, high- and low-intensity courtship activity, etc. In “high-intensity Bowing,” for example, the pelican’s head is held below foot level while in “low-intensity Bowing” the head is above foot level. There is also some suggestion that the bird is somehow more excited when it is involved in “high-intensity” behavior. Schreiber also uses the word in its general sense (“intense eye contact”) which makes things even more confusing. It seems paradoxical that the concept of “intensity” should have such sustained popularity among ethologists who generally credit themselves with using only descriptive terminology. Schreiber could as easily have subdivided the variability observed in Bowing into “high- neck” and “low-neck" descriptive categories. Perhaps a more crucial problem is that Schreiber frequently overstates his evidence. At times this could have been avoided by the editorial insertion of a qualifying adverb (e.g., “probably” or “apparently”) that would have softened the sentence’s tone. Else- where he seems to have overstepped the limits of scientific prudence, as in the following generalization: “. . . the subtle and highly variable differences in context in which the December 1977 • ORNITHOLOGICAL LITERATURE 641 displays are performed and received modifies these messages, and thus each encodes different precise probabilities of further action” ( p. 72). This is an attractive working hypothesis — being the basic premise of W. John Smith’s message-meaning approach — but it is not fact. Schreiber gives no evidence to support even the broadest outlines of the claim, much less to demonstrate “precise probabilities.” My last complaint concerns the widespread opinion that pair-formation in many monogamous species can be justly characterized as a “female-choice” system. Schreiber shares this view as shown by his statement that “. . . mate selection is accomplished by the female, who selects the male” ( p. 37) . It is true that in many colonial birds, including pelicans and berons, the males take fixed positions and display while females move about as if “shopping” among them. Outwardly it looks as if females do the choosing and the males are merely passive merchandise. The theoretical implication is that natural selection acts only on the females despite the fact that these males make enormous parental investments. Actually, of course, males are very choosy about their mates; a preference they show by driving away (or ignoring) any females they wish to reject. Schreiber’s own observations confirm this view just 3 sentences after the female-choice statement: “. . . some females are not allowed on the perch, and frequently a female will be allowed onto the perch but then is kept off the nest site. In these cases pair bonding does not occur.” As evolutionary theory would predict, both sexes choose care* fully. The lengthy courtship is clearly a period of mutual assessment. In conclusion I should like to stress the strengths of this paper because it does provide remarkably detailed information on Brown Pelican behavior. I expect Schreiber’s collective works to become the starting place for an exciting comparative literature on the ecology and behavior of pelicans. His work may become to the Pelecanidae what Bryan Nelson’s Gannet study was to the Sulidae. Its tedium does not detract from its value as a reference work, though it will probably not attract a wide readership. — Douglas W. Mock. Alberta Birds, 1961-1970, with Particular Reference to Migration. By Thomas S. Sadler and M. Timothy Myres. Occasional Paper No. 1, Provincial Museum of Alberta, Natural History Section, Edmonton, Alberta, Canada, 1976: 314 pp., 1 map. $3.25. — This work summarizes many bird observations by 225 observers in various parts of Alberta and taken from the published literature. Major justification for the book is to “provide basic data for subsequent analysis by . . . students of the migratory behaviour of particular groups of Alberta birds.” The main part of the work is 233 pages of “species records and summaries, 1961-1970.” Each entry begins with a brief summary of the decade’s records on migration, distribution, breeding records and population status and trends. Records are then listed by year in a telegraphic style, with each entry prefaced by an alphabetical code identifying the kind of observation it is, i.e. spring arrival, peak numbeis recorded in migration, and so on. The observations usually include the number of birds seen, date, locality, and the observer or source of the data. Occasional comments show, for example, that the spring of 1967 was particularly harsh for migrants as indicated by the following observations: a Pygmy Owl was found “frozen to death in a blizzard” in April, the Tree Sparrow migration was “interrupted by a mid- April snow-storm” and in late April about 1(X)0 Lapland Longspurs were seen “in the shelter of a barn during a snow-storm.” Some observations are also of ecological interest as, for example, the observation of very high nesting densities of Short-eared Owl in the Calgary area in 1969 because of a high spring vole population. 642 THE WILSON BULLETIN • VoL 89, No. 4 Several short sections precede the species list. There is an Introduction, a tabulation of the bird highlights of the decade, and a brief section on birds and main weather events of the decade. The latter section and the Introduction are attributed to the junior author. There is also a section listing the place names mentioned in the text with references to a map of Alberta by which they can be located. Another section lists the dates of the Sundays in the decade, and finally a section explains the symbols used in the species accounts. The contributors to the volume are listed immediately after the species accounts. Then follows a bibliography of 259 publications concerned with Alberta birds in the decade covered by the book. Separate indexes to scientific names and English common names conclude the work. The book discusses 341 species. These include confirmation of the historic presence of the Passenger Pigeon in Alberta, the recent so-far-successful establishment of Wild Turkey, and a recent unsuccessful release of Chukar. Other first occurrences in the decade were Cattle Egret, Ruff, Black-necked Stilt, Band-tailed Pigeon, Yellow-billed Cuckoo, and Scarlet Tanager. Of more significance are changes in the status of some species, such as the continued decline of the White Pelican and the Peregrine, and the apparent decline of the Burrowing Owl. The Black-crowned Night Heron, Cinnamon Teal, White-breasted Nuthatch, and Yellow-headed Blackbird are regarded as increasing, while such species as the Cooper Hawk, Blue Jay, and Herring Gull are reported to be more widespread than previously. Considering the authors’ stated objectives this work is fairly successful. It does provide many records of arrivals, departures, flock sizes, breeding records, and distribution data. Students of Alberta bird populations will find this a very useful source of such information. However, I would have liked to see more analysis of the data, more effective presentation of climate data, and more discussion of the interactions of birds and climate. Had this been done the work would he a more important contribution than it is. The hook seems free of minor typographical errors, hut there is a major problem with the indexes. The page references in the scientific and vernacular indexes are all incorrect although they agree with each other. The publishers should correct this error as soon as possible. The last four items in the Table of Contents are also incorrect. I recommend this hook to anyone interested in the birds of Western Canada. It may he purchased from: The Bookshop, Provincial Museum of Alberta, Edmonton, Alberta, Canada, T5N 0M6.— William J. Maher. Bird Populations of Aspen Forests in Western North America. By J. A. Douglas Flack. Ornithological Monographs No. 19, 1976: viii + 97 pp. $7.50 ($6.00 to A.O.U. members). (Obtainable from Glen E. Woolfenden, Department of Biology, University of South Florida, Tampa, Florida 33620.) — Aspen forests in western North America con- stitute a physiognomically distinctive community, somewhat isolated from other broad- leaved vegetation types. In this important study the avifauna of this community is examined through the strip censusing of 41 plots, each in a homogeneous aspen forest. Two visits were made to each plot, the area censused in most cases being 12.5 acres. The stands censused fall into 2 regional groupings, western montane (27 plots) and prairie parkland at lower elevations in Canada. Aspen forests of the 2 regions, which are disjunct in Montana, differ in that summer nights in the parkland are warmer, and summer moisture greater, contributing to faster growth, greater incidence of diseases and shorter life spans December 1977 • ORNITHOLOGICAL LITERATURE of the trees there. Undergrowth usually is more dense in the parkland forests which have experienced more disturbance in pre-settlement time. In contrast, the montane stands tend to be mixed in their age composition. The population data are minimal but the relative values are used to probe significant ecological and evolutionary questions. Species composition and total numbers are con- sidered in relation to many vegetational parameters that relate to the birds’ habitat requirements. Such responses are analyzed further by the grouping of species into 5 nesting guilds to which species are assigned by the positions of their nests. This cate- gorization points to regional differences in the representation of certain groupings, such as a paucity of cavity-dependent nesters in the parkland. Although there are fewer species and individuals in the montane stands, bird species diversity (Shannon-Wiener function) in both regions appears greater than expected on the basis of foliage profile features. Size, spacing and health of the trees emerge as other factors contributing to bird species diversity. Species compositional differences between the two regions receive considerable attention. Twenty aspen-dwelling species are considered restrictedly montane, 24 occur only in the parkland and 24 are shared. The greater number of species in the parklands is attributed to the proximity of a larger pool of prospective colonizers of aspen stands in the eastern deciduous forests. Representation of species derived from that source decreases progres- sively southward in montane aspens. In the absence of 10 species found in such timber farther north, species diversity in aspen forests in Arizona is maintained by such charac- teristic conifer forest species as the Evening Grosbeak and Western Bluebird, The last example points to a considerable contrast in the avifaunas of the 2 regions that is only partially borne out in a schematic summarization of “Geographic Replacement of Morphologically Similar Species” (Table 5), Here, species of minor occurrence ( = frequency), such as the Western Tanager, are not distinguished from those that are widespread and/or numerous. In a succeeding table tbe species are ranked according to their importance regionally. Table 5 would have been more meaningful had the relative importance of the species included been denoted by size or boldness of type. The contrast between the avifaunas of the 2 regions is diminished further by the omission of other species (such as the Broad-winged Hawk, which lacks a counterpart in montane aspen). With reference to ecological equivalence, I question the author’s view (p. 64) that the more rigorous montane climate does not limit species composition. Several western forms equivalent to or conspecific with parkland inhabitants appear confined to levels below the aspen belt in the central Rocky Mountains (Bullock’s Oriole, Gray Catbird) or they reach greatest abundance at lower elevations (Brown-headed Cowbird). The role of the elevational difference is conceded indirectly by Flack in his discussion of foot- hill aspen stands in Alberta as being faunistically and climatically intermediate (p. 77). The topics discussed above demonstrate the emphasis placed upon historical factors, many of which (e.g., routes of colonization) are elusive. Ecological questions, however, receive at least equal attention. Attempts are made to explain abundance or absences on the basis of vegetational features, and the question of saturation in this community is weighed carefully. The success of Flack’s study derives from the application of relative abundance values based upon standard censuses by one investigator to comparisons of broad geographic scope, and from a thorough integration of his findings with a diverse literature. Many of the questions raised, such as fluctuations in populations of widely distributed species, can be answered only by long-term studies that should be undertaken by resident naturalists. The reader’s task would have been aided by the provision of a map showing the extent 644 THE WILSON BULLETIN • Vol. 89, No. 4 of the parkland in relation to other vegetation types. Membership in the seven somewhat subsidiary foraging guilds might have been coded in the nesting guild lists. The few proofreading lapses (even “Red-bellied Sapsucker”) do not hinder the reader’s under- standing. However, these criticisms detract but little from an informative and thought- provoking monograph. — Keith L. Dixon. Watching Birds. By Roger F. Pasquier. Houghton-Mifflin Co., Boston, 1977: 301 pp., 100+ black and white drawings by Margaret La Farge. ^10.00. — ^^Watching Birds is . . . intended to unite the bird watcher’s perception of specific details with the environ- mentalist’s awareness of general truths” ( p. viii). This handsome elementary ornithology book is dedicated to the amateur birdwatcher who wishes to advance beyond the mere “life list” stage of his hobby and expand his understanding of most aspects of bird biology, including the relationships of birds to each other and to their environment. Any non-professional ornithologist could profit from reading this book, which is mercifully free of jargon and which sticks to basic facts about birds. This volume would serve as a high school ornithology text, or for a reference to supplement the library of a bird aficionado. Chapters cover such aspects of avian biology as Origin, Evolution and Speciation (Chap. 3), various aspects of anatomy and locomotion 1 Chaps. 4 and 6), Behavior (Chaps. 5, 7, 8, 10, 11), and Zoogeography (Chaps. 10 and 12). Also included are discussions on conservation, general birdwatching, and an overview of current progress in ornithology (Chaps. 1, 13, 14, 15). By and large the text is quite enjoyable and complete. Occasionally, some statements smack of dogma (e.g. “In every case the individuals that are going to survive and reproduce will be those best adapted to their niche . . .” p. 27), and the discussion of speciation ( p. 28) is superficial. But there is little sense in cluttering up a very basic and readable text with current and sometimes confusing problem areas of biology. Mr. Pastjuier has a feel for birds, and his text imparts this delight of the subject matter to the reader. Reading it could mark a turning point in the life of those birders whose major thrill is hunting down and checking off a new trophy on their lists. New observations of behavior of familiar species can be as exciting as, and certainly more enlightening than, the hunt involving pad and pencil. The illustrations by Margaret La Farge are superb, both scientifically and aesthetically. They are well chosen and delightful. Both the author and artist would likely agree with Thoreau who said, “The wood thrush is a more modern philosopher than Plato and Aristotle. They are now dogma, hut he preaches the doctrine of this hour.” Their book reflects such thinking and I recommend it to anyone who loves birdwatching, and, more importantly, birds. — Michael A. Mares. A Guide to Bird Finding East of the Mississippi, Second Edition. By Olin Sewall Pettingill, Jr., illus. by George Miksch Sutton. Oxford University Press, New York. 1977: xxvii + 689 pp. S15.95.- — The standard guide to finding birds in the eastern United States has been revised to take into account changes in distribution, changes in habitat, and to provide new travel directions resulting from an expanded road network, especially the interstate highway system. For each of the 26 eastern states the major birding localities are listed, and for each there are instructions for reaching it, a brief description of the December 1977 • ORNITHOLOGICAL LITERATURE 645 habitats, and comments on the kinds of birds to be found there at various times of the year. — R.J.R. Summer Birds of the San Juan Valley, New Mexico, By Carl Gregory Schmitt. New Mexico Ornithological Society Publication No. 4, 1976: 22 pp., no price given. — An annotated list of about 147 species observed during the summers of 1971 and 1972, with comments on numbers, breeding status, and habitats. — R.J.R. Pluvianellus Socialis: Biology, Ecology and Relationships of an Enigmatic Patagonian Shorebird. By Joseph R. Jehl, Jr. Trans. San Diego Soc. Nat, Hist,, 18 (3) : 25-74, 1975. 28 figs., 3 tables. — In recent years there have been a number of studies on the ecology and behavior of shorebirds, many dealing with social organization. However, a clear picture of the evolution of social systems within a group is only possible once the behavior, ecology, and taxonomy of many species is described. Although not com- plete, such baseline information exists for shorebirds and the task at hand is to add basic information on key species. Pluvianellus socialis is such a species and Jehl has provided an excellent description of its behavior and ecology. This paper concerns such topics as habitat and distribution, vocalizations, pre-nesting and nesting behavior, growth and care of the young, foraging and feeding behavior, molts and plumages, and systematic relationships. Jehl studied Pluvianellus on its wintering grounds in 1971 and 1972 and on its breeding grounds in 1973. The total population, which may not exceed 1000 individuals, breeds along lakes from the Rio Grande (Tierra del Fuego) north along the south- eastern coast of Patagonia to Puerto Deseado. It winters along the coast from the Strait of Magellan to the Valdes Peninsula. Pluvianellus has been considered a plover, but Jehl, whose knowledge of shorebird taxonomy is considerable, suggests from his observations that “its relationships are far less obvious.” Pluvianellus returns to the breeding grounds in early September and begins breeding up to several weeks before other Fuegian shorebirds. The chicks hatch before most North American migrants arrive. The birds nest on the shores of shallow ponds, lagoons, and lakes in the steppe region of northern Tierra del Fuego and southern Patagonia. They nest along brackish and fresh water lakes, but not along streams, rivers, or the ocean. Obvious requisites are beaches of intermixed small stones and mud with adjacent stretches of open shoreline. The number of pairs at a lake was limited by the amount of suitable habitat and the presence of other shorebirds, but not by the size of the lake. Due to the arrival time of the investigator, information on pair formation and territory acquisition are omitted. However, renesting by one pair allowed Jehl to describe this phase qualitatively. Pluvianellus defends linear territories of 300 to 500 m along the shoreline. The complicated territorial defense displays, described with diagrams and photographs, involve the members of a pair acting as a unit as is typical of oyster- catchers and plovers. Territorial clashes increase in frequency and intensity as the chicks become more mobile. Although Jehl’s descriptions of these clashes provide an excellent qualitative picture of the behaviors involved, quantitative data are necessarily lacking on daily and seasonal variations since the rarity of the species in general, and the small number of pairs on any one lake make it difficult to quantify the behavior described. Nests, located 0.7 to 25 m from water in fully exposed situations, were excavated by digging and lined with small bits of gravel. Pluvianellus lays 2 eggs, but only one chick 646 THE WILSON BULLETIN • Vol 89, No. 4 is actually raised. Egg laying occurred from 4 September to 17 November. Both sexes incubate. Jehl observed no exchange displays and no distraction displays by adults incubating or caring for chicks. Jehl notes that the young chicks are less agile than plover or sandpiper chicks of the same age. The 2 eggs hatch 8-14 h apart, and the chicks leave the nest the day after hatching. The slight age difference results in the success of only the older chick; the 5 families over 3 days of age observed by Jehl included only 1 chick. Both sexes feed the young by regurgitation as -well as with food carried in the bill. Parental feeding is “extremely unusual in the Charadrii, and the use of the crop to regurgitate differen- tiates it from all other shorebirds.” Chicks obtain all their food from their parents for the first 2 weeks, then begin to forage for themselves. During this dependency period chicks remain concealed and depend on their coloration for protection. Jehl concludes from the growth pattern that Pluvianellus ehicks fledge at a much higher weight than that of similarly sized shorebirds that forage for their own food. Pluvianellus’ foraging and feeding behavior parallels that of turnstones. It is apparently the only Charadriiform bird that digs for food. Wintering behavior is also described. The species’ winter requirements are intertidal rocky areas and debris covered sandy beaches on which to feed. The species is uncommon and irregularly distributed along the Patagonian coast. In contrast to its breeding season behavior, the species avoided the water’s edge in winter and fed in flocks with other shorebirds. The concluding section on systematic relationships lists the aspects of Pluvianellus’ morphology and behavior that differ from the usual plover condition: turnstone-like body, short stout legs and blunt elaws, foraging pattern which includes digging, territo- rial defense behavior involving the pair acting as a unit, complex pre-copulatory and scrape displays, courtship feeding, small eggs, clutch size of 2, rearing of only 1 chick, absence of distraction displays, semi-precocial chicks, slow chick growth, prolonged dependence of chicks, unique natal down color, parental feeding of chick, and apparent dove-like drinking behavior. After an exeellent discussion of the problem, Jehl concludes that the species should be in a new monotypic family Pluvianellidae. In general, the problem is clearly defined, the paper is succinctly written, and the diagrams and photographs are clear and contribute to the descriptions. This first de- tailed study on Pluvianellus provides excellent descriptive data on breeding and non- breeding behavior so necessary for the comparative approach to shorebird behavior. — Joanna Burger. Birds of the Antarctic and Sub-Antarctic. By George E. Watson, in collaboration with J. Phillip Angle and Peter C. Harper, illus. by Bob Hines. American Geophysical Union, Washington, D.C., 1975: 350 pp., 11 color plates, 55 black-and-white illustrations, 51 range maps, 11 numbered text figures, 7 tables, hardbound. $15.00. — This is the most unusual book review that I have written, because long before the book’s release, I was asked by George A. Llano of the National Science Foundation to field test the original manuscript in the Weddell Sea off Antarctica. In late December of 1972 I boarded the Coast Guard Icebreaker USCGC Glacier at the American base McMurdo by the Ross Sea. The voyage that followed took a course nearly 180 degrees around the Antarctic continent, including a northward thrust to southern South America and a southward one that penetrated pack ice deep within the Weddell Sea. My travelling companion was S. D. MacDonald of the National Museum of Canada. He was especially December 1977 • ORNITHOLOGICAL LITERATURE 647 valuable to the project because of his keen eyes and exceptional ability at identifying objects far off. MacDonald and I are experienced birders, and have often worked as a team in the High Arctic, but we were totally inexperienced at identifying southern sea birds. Up to the time the Glacier broke free of the pack ice in the Ross Sea and entered the open ocean, neither of us had seen an albatross or the many petrels and storm-petrels that cover these southern waters. We were truly in a good position to test Watson’s descrip- tions and Bob Hines’ illustrations. Many people have had important input in the production of the handbook. Foremost among these was George Llano, who thought not only in terms of a handbook of birds to serve the growing number of people visiting far southern places, but of a series of handbooks covering various biological disciplines for these areas. With the financial backing of the National Science Foundation and continued assistance by Llano, the handbook on birds became a reality. The selection of George Watson as author was a very good choice. He, J. Phillip Angle and others had earlier finished an important scientific work entitled “Birds of the Antarctic and Subantarctic,” edited by Vivian Bushnell and published in 1971 by the American Geographical Society as Antarctic Map Folio Series 14. No doubt this work formed the skeletal structure of the present hand- book. At the beginning of our voyage, MacDonald and I experienced difficulty in identifying southern sea birds, especially prions and immature albatrosses, but by and large things went well. The manuscript and illustrations were indispensable, reaUy a tremendous aid, and there is no question in our minds that we would have been severely handicapped without them. We caught some inconsistencies, a few errors or oversights, and pointed out a number of troublesome areas dealing mostly with at-sea identifications. Hopefully, our efforts produced a better handbook. As good as the text and illustrations are, there still remain shortcomings that only too soon became apparent to the user. There is to my knowledge no easy method for separating at a distance Antarctic Terns in first-year plumage from either young or old Arctic Terns in winter plumage, especially in areas where both species occur. This point really struck home as I recently observed a number of experienced birders aboard the Lindblad Explorer chalk up Arctic Terns when in fact they were seeing immature Antarctic Terns. I knew this to be the case for I had learned by experience that immature Antarctic Terns often associate with adults at or near the breeding colonies, as were the birds observed by the Lindblad birders. The void between research and publication is always a hopeless matter. Months before publication of the handbook we had new information on terns and skuas that should have been included, but there is no stopping the publication machinery once set in motion. How I wanted to tell artist Bob Hines not to use large pupils in the eyes of his penguins, for even in fairly poor light the pupils of most penguins seen by us appeared as pinpoints, giving the penguin its colorful hut blank, pupilless appearing eye. Be as it may, the handbook contains a wealth of useful information. Included in this durable pocket-size book, in addition to species descriptions and much life history information, are accounts of the geography and environments of the southern lands and seas, including climate, vegetation, record taking, preserving and shipping specimens, conservation, and specially protected areas. The distribution maps and tables are very' useful. The references are extensive and the index complete. All these many facets, really an amazing assemblage of material for a small size book, are logically arranged and written in a clear, succinct manner. 648 THE WILSON BULLETIN • Vol 89, No. 4 The black-and-white illustrations play their role well, but it is the color plates that one will return to time and again throughout one’s voyage. Composites showing many birds in flight are a difficult, dreary proposition for any artist. Bob Hines is to be congratulated for pulling it off as well as he has, for his was an especially tough assignment with so many birds with similar shapes and colors. Perhaps the most pertinent question that can be asked of a work of this kind is, “who will be able to use it effectively?” On this point there is little doubt that almost any experienced birder, professional or non-professional, will be able to do so. For the inexperienced person it may be a different matter. From the start it was hoped that the handbook could be used effectively by non-birders, even highly trained scientists, in making records during voyages when an ornithologist or bird watcher was not present. But a good friend, an expert on invertebrates, confided that it took more than the handbook to enable her to identify sea birds accurately. She discussed the need for large, highly demonstrative illustrations (preferably color plus black/white prints plus line drawings and written descriptions) of each species in different conformations, which could be mounted (for instant comparison) on the bridge and the pilot house. If such identification sheets were available to relevent ships’ personnel, perhaps more realistic bird censuses could be taken. This is a big order to be sure. Maybe for these special cases the only solution is a short course in bird watching along with the hand- book. But whether best for the experienced or inexperienced, the handbook nevertheless is indispensable for anyone contemplating observations of far southern birds at land or at sea. — David F. Parmelee. Penguins, Past and Present, Here and There. By George Gaylord Simpson. Yale Univ. Press, New Haven and London, 1976: xii + 150 pp., 10 color and 24 black-and- white photographs, 9 maps. $10.00. — In 1933, George Simpson and his party made a collection of fossil mammals in Patagonia and incidentally accumulated the best collec- tion of fossil penguin bones of its time. Unable to find an ornithologist willing to study tlie collection, Simpson took up the cudgel himself and published his well-known mono- graph on fossil penguins in 1946. Since then he has continued his studies of fossils and has seen most of the living species in the wild as well. His infatuation with penguins inspired him to write this book “for adults who do not necessarily know much about penguins but for whom there is nothing that they do not really want to know.” With such readership an author is presumed relieved of the necessity of commanding attention by irresistible prose. The first chapter on the earliest accounts of penguins is nevertheless sprightly and almost irresistible. The next chapter on naming penguins begins in the same vein but it becomes more pedantic and ends with a discussion of each species’ scientific name. Subsecjuent chapters continue as competent accounts of general features of penguin biology, fossil penguins, distribution and speciation of extant penguins, breeding behavior and ecology, and exploitation of penguins by man. Except for occasional light touches, however, this major portion of the book is likely to lose readers who want more than just facts. Unfortunately, neither the black-and-white, nor the color photographs are exceptional, and the distribution maps are sometimes difficult to interpret. Dr. Simpson’s book provides broad coverage of penguins for the layman. It falls short of Pettingill’s “Another Penguin Summer,” however, in presenting “the singular charm of penguins.” — Richard L. Zusi. December 1977 • OKNITHOL()(;iCAL LITER ATU HE (AD Parent Birds and Their Young. By Alexander E. Skuti h. Univ. of Texas Press, Austin, 1976: xviii + 503 pp., 116 plates, 18 tables, 19 figs. $27.50. — In tins large volume, Skutch presents his synthesis of the reproductive activities of birds and the characteristics of young birds, bringing together a wealth of useful facts and his evolutionary interpreta- tions of the major patterns. The facts are drawn from over 40 years of painstaking observations in the New World Tropics and from an extensive literature survey. The text is divided into 34 chapters, each covering one topic. These are arranged as follows: pair formation and mating systems (3 chapters), territoriality (1), timing of nesting (2), nest form, materials, construction, and maintenance (4), egg size and color (1), incubation patterns (5), hatching process (1), developmental state at hatching (1), parental care of nestlings (4), nestling interactions (1), fledging, care and education of fledglings (4), inter- and intraspecific helpers (2), nests as dormitories (1), concealment and direct defense of the nest (2), reproductive rate and its regulation (2). Skutch set himself the formidable task of preparing a comprehensive, yet detailed, treatment of reproduction to satisfy the amateur naturalist as well as the professional investigator. He has succeeded rather well, largely because his mastery of the written language makes for an eminently lucid style, readable by most laymen, and because the examples discussed are interesting. Some of my favorite sections are the accounts of manakin behavior on the lek, megapode habits, procedures whereby non-incubating males learn that their offspring have hatched and require feeding, and birds feeding nestlings of other species. For the investigator, the hook will he valuable chiefly for the sheer mass of facts arranged for easy comparison among birds, and for the 18 tables. These include data on age at first breeding, nesting periodicity of seabirds, incubation patterns and duration, feeding rates, nesting success, and clutch size. The figures are well chosen to illustrate points in the text (except that I could find no text reference to fig. 19). The 116 black-and-white photographs are of generally good quality and appropriate to the discussion at hand, but many could have been omitted. In the text, species are identified by common name and alternate common names and scientific names are given in the index at the end of the book. Common names do not always follow American use, e.g. Common Bluebird for Sialia sialis. The text is nearly error-free; I found only 2 typographical errors and 1 factual mistake ( using a linear instead of a logarithmic model to calculate daily rates of nest loss from overall nest mortality). Nevertheless, the objective scientist may find fault with some of the concepts expressed, as that “birds often sing from pure ebullience or joyousness . . . and females sitting in the nest sometimes hum little ditties expressive of contentment” (p. 336), and that “those (birds) that maintain their population with the smallest reproductive effort, are the truly efficient species. They enjoy the longest, and doubtless most satisfying, lives” (p. 376). It must be pointed out that many of Skutch’s evolu- tionary interpretations do not jibe with currently accepted views on the nature of natural selection. In particular, Skutch fails to provide a plausible selective mechanism when he argues that because species in stable environments produce only enough young to replace annual losses, despite being able to feed additional offspring, they are limiting their reproductive rate to avoid depleting their resources. In summary, in spite of some questionable interpretations. Parent Birds and Their Young is easy to read, well-organized and factual, and, especially for the amateur, may serve to alert observers to phenomena worth recording. — Susan C. Wiiite-Sciiuler. 650 THE WILSON BULLETIN • VoL 89, No. 4 Bird Sounds. By Gerhard A. Thielcke. University of Michigan Press, Ann Arbor, 1976; 190 pp. (First Published as Vogelstimmen, Springer-Verlag, 1970.) $2.95 (paper), $6.95 (cloth). — This book, written in semi-popular style, begins with a chapter on methods of analysis of avian acoustic signals, comparing the oscilloscope and the sound spectrograph as research tools. Thereafter are chapters on vocal versus mechanical sounds, neuro- anatomical studies, functions of calls vs. songs, ontogeny of vocalizations, vocalizations and speciation, evolution of sounds, annual and diurnal cycles, and finally bird sounds and music. The translator has retained the German word “strophe” throughout the text, but the term appears to have two different meanings. On page 3 it is stated that: “Several strophes are called a song.” Presumably then, “strophe” is equivalent to the term “phrase” of other investigators. Thereafter, however, strophe appears to signify “song types” or “themes,” e.g. when discussing songs of Marsh Tits and treecreepers (pp. 35 and 36). Breeding season (Brutperiode) is referred to as “brooding period” throughout. “Clutch” (Gelege) was translated as “egg-lay” (p. 69), and “altricial” appears as “in- sessorial” (p. 65). “Sibling species” appears as “twin species” (p. 137) and “domestic cockerels” (pp. 164 and 165) has become “turkey cockerels.” But these are not serious; all-in-all the translator has done an excellent job. The author states (p. 96) that: “The ability to learn song begins around the thirtieth day of life in song birds.” This is contradicted on p. 115 in which he informs us that White-crowned Sparrows, captured at 30 days of age and then isolated, sang their home dialect the following spring. Marler has refined these data further (J. Comp. Physiol. Psychol., Mongr. 71(2), 1970). In the chapter on sound-production (p. 26) the author informs us that the interplay of sound-production and breathing is an unsolved problem, and that surely birds must inhale and exhale as they sing. Recent attempts to tackle this question have been sum- marized by Gaunt et al. (Condor 78:208-223, 1976). In the section on antiphonal singing, the author asks (p. 45) whether or not duetting is correlated with pair-contact in dense vegetation or with nocturnality when pairs are visually separated. Recent surveys by Payne (Ostrich Suppl. 9:125-146, 1971) and Kunkel (Z. Tierpsychol. 34:265-307, 1974) indicate that duetting is most often found in birds with long-term pair bonds, and not necessarily in areas of poor visibility. Indeed, Kunkel points out that duetting is often associated with elaliorate visual displays during which singing pairs are very close to each other. The author also asks (p. 117) how a new dialect can come to exist. Several authors have tried to provide answers to this. Field studies (Payne, Ornithol. Monogr. 11, 1973; Lemon, Condor 77:385-406, 1975; Baptista, Univ. Calif. Puhl. Zool. 105:1-52, 1975) on song variation in natural populations indicate that song learning is an imperfect process. Due either to copy errors, or in Lemon’s view, and improvisational process that he calls “drift,” new themes or variants of existing themes continue to emerge. Given geographical isolation, oral tradition, the accumulation of copy errors (or Lemon’s “drift”), new dialects may eventually develop. The chapter on “sound parasitism” (pp. 155-159) summarizes Nicolai’s studies on viduine brood parasitism on estrildid finches, in which he suggested that each viduine parasitizes and mimics songs of only one host. Payne (Auk 93:25-38, 1976) has cau- tioned that the one-host-one-parasite theory is good in only some parts of Africa. In other areas (e.g. in West Africa) one viduine form may mimic as many as four different fire- finch hosts. December 1977 • ORNITHOLOGICAL LITERATURE 651 My comments are not meant as criticisms of the author’s work, but merely indicate the popularity of the field and the speed at which new data are being accumulated. Tbe author has brought together a fine review of the field up to 1970. Especially notable are his chapters on functions of calls, functions of songs, evolution of species, and evolution of sounds, in which he summarizes many important studies from the German literature. The latter include his own extensive work on geographical variation and ontogeny of vocalizations, and playback experiments with treecreepers iCerthia spp.), tits {Pams spp.) and other species that will, I am sure, stimulate others to exciting research. “Bird Sounds” is definitely a must for students of ethology, systematics, and avian bioacoustics. The price is right to boot! — Luis F. Baptista. Song Dialects and Demes in Sedentary Populations of the White-Crowned Sparrow (Zonotrichia leucophrys nuttalli) . By Luis F. Baptista. University of California Publications in Zoology, Vol. 105. University of California Press, Berkeley, Los Angeles and London, 1975: 52 pp., 25 plates. $4.00 — In his introduction Baptista speaks of the White-crowned Sparrow as the white rat of the ornithological world. Certainly insofar as vocalizations are concerned it is among the most thoroughly studied of all birds, providing some of the most complete documentation now available on song variation within and among wild populations. The work described here extends and amplifies previous studies by Marler and Tamura (Condor 64:368-377, 1962; Science 146:1483- 1486, 1964) as well as previously reported studies by Baptista himself (Z. Tierpsychol. 34:147-171, 1974). As a basic framework for his investigation, Baptista asks the following questions: (1) does gene flow occur between demes, as indicated by song characteristics?; (2) what occurs in contact zones between neighboring demes? ; (3) what barriers exist between populations?; (4) how large is a dialect group?; and (5) what are the effects of visiting migrants on the songs of local populations? The study included various localities in the San Francisco Bay area and extended from April 1968 through July 1971. Based on an analysis of over 2400 song spectrograms of over 400 birds it was found that there is not one single song theme in each area, but rather a dominant motif or “dialect” with a number of minor variations in song. Song variants were not found randomly dispersed throughout the populations, but were clumped, forming a number of subdialect areas within each dialect region. Although most individuals sang only one theme each, a few sang from two to four. Some birds were apparently “misimprinted” with songs sung by wintering migrants belonging to the subspecies pugetensis. Dialect or “theme” areas varied in size from the rather small Brooks Island and adjacent mainland to the much larger Berkeley area. Some populations were separated by ecological barriers, and their song themes thus developed in isolation. Others seem to be separated only by distance. The mosaic distribution of subdialect groups and dialects in a continuum points to a population structure of genetically semi-isolated neighborhoods. Between contiguous dialect areas Baptista found a narrow mixed zone of song themes. Many individuals within this zone maintain their separate and distinct dialect forms, while some sing a mixture of the 2 in a repertoire of more than one song pattern, or a single song containing elements of the two separate dialects. Most of the publication is taken up by the description of song characteristics of the various San Francisco Bay area populations sampled, and therein lies its main strength. The remarkably thorough documentation of song characteristics of the birds throughout 652 THE WILSON BULLETIN • Vol. 89, No. 4 this topographically and ecologically complex area provides the basic data upon which to begin to understand the distribution and dynamics of song dialects. The amount of work involved in the analysis of the recording is enormous, although Baptista does not mention it. Perhaps the greatest weakness in the paper is the author’s failure to nail down solidly the answers to the questions he asked at the beginning. Most of them are answered, more or less, but some, such as question number 4, are left pretty much up to the reader to figure out for himself. There are some grammatical and spelling lapses that indicate a lack of thorough editing. On the whole, however, this publication provides a remarkable record of song variation within and between adjacent populations of the White-crowned Sparrow (Zonotricha leucophrys nuttalli) , and is a very important contribution to the literature on song dialects. — William L. Thompson. Social Organization anu Behavior of the Acorn Woodpecker in Central Coastal California. By Michael H. MacRoberts and Barbara R. MacRoberts. Ornithological Monographs No. 21, American Ornithologists Union, 1976: viii + PP-» 39 figs., 13 tables. $7.50. — The Acorn Woodpecker (Melanerpes formicivorus) is a fascinating spe- cies because of its unusual social system. At least in California, the birds live in per- manent groups of up to 12 or more individuals of both sexes and all ages. In the fall the groups communally harvest, store and defend caches of acorns, which are their major food. In spring the colonies breed as units, with most or all members helping to incubate and feed 1 or (rarely) 2 broods of young. These facts about Acorn Woodpecker natural history were discovered long ago, but no comprehensive studies were undertaken using marked birds to determine such important aspects as the genealogical relationships among group members, the amount of movement between groups, or even who lays eggs fertilized by whom. With the current flurry of interest in group and kin selection, and “sociobiology” in general, it is not surprising that several such studies are now in various stages of comj)letion. This is the first to reach publication. I'he study was conducted at The Hastings Natural History Reservation above tbe Carmel Valley in (California. This is jirime Acorn Woodpecker habitat because of the abundance and diversity of oaks present (6 species and 5 hybrids grow on the reserva- tion). Most of the fieldwork was carried out between October, 1971, and August, 1974; 149 birds were individually color-banded and 60 groups were studied, 32 intensively enough to appear in an appendix describing their individual histories. Extensive ob- servations were made from blinds. The authors did not include a methods section and we do not know exactly how many hours of observation were involved. Also, no mention is made of technicpies for cajituring birds. This is unfortunate, since Acorn Woodpeckers are notoriously elusive and anyone who captures 149 of them has learned something useful. The monograph is w’ell-written and a pleasure to read. Following introductory remarks and a description of habitats occupied, the MacRoberts’ discuss the all-important topic of group composition. They found that group size varied from 2 to 15, with a mean between 5 and 6; all groups had both male and female members, and sex ratios were about equal. Most recruitment to groups was by reproduction, although 9 individuals, mostly adults, were observed to move from one known group to another; some 87 birds disappeared during the course of the study, unknown proportions dying or moving off the reservation. The MacRoberts’ write in their summary (p. 82) that “the Acorn W'oodpecker is a sedentary species,” and that “each group maintains a year-round, all- purpose territory.” This picture of extreme stability appears true only to a degree. For December 1977 • ORNITHOLOGICAL LITERATURE 653 example, of the 32 groups discussed in detail in an appendix, 19 either appeared on the reservation from unknown areas, left the reservation for unknown areas, or made documented moves on the reservation during the study. In some cases these moves were attributable to acorn crop failures. Acorn Woodpeckers, acting as groups, vigorously defend their territories against intra- and interspecific intruders. In a species that caches immense quantities of food, the defense of that food supply becomes a major operation. The MacRoberts’ were fortunate to be able to document several inter-group conflicts that resulted in complete or partial displacement of one colony by another. In no cases did smaller groups displace larger ones. There also is a highly significant figure showing a strong positive correlation between group size and territory size. These observations suggest that one reason for the evolution of group living in this species may be the advantages gained in defense of food resources. The reproductive effort of Acorn Woodpeckers at the Hastings Reservation was very low — between 0.2 and 0.3 young per adult per year. Many groups failed to breed at all in any given year. The MacRoberts’ were unable to explain this phenomenon, although it seems perhaps consistent with a strongly K-selected life history strategy. Indeed, the authors note that Acorn Woodpeckers on their study area live at high, relatively stable population densities. In most years food supplies are good and populations remain at or near carrying capacity. Young stay with their groups, presumably because there is no place for them to go and because there is little or no chance for them to breed on their own. Even though groups may move, they appear to do so as intact units. In this environment, and if the colonies are made up of related individuals, K and kin selection will be powerful forces shaping the social system of the species. It may not be that simple. One would predict that small groups might breed more than large ones, since recruitment would be both possible (more “room” for additional birds) and advantageous (more efficient defense of territories) . However, the MacRoberts’ data show, if anything, that it was the small groups that more often failed to breed. As the authors observe (p. 71), “the exact genetic relationship among group members was not known in most cases.” Their data suggest that most recruitment to groups occurs via reproduction, but in order to say anything conclusive about kin selection or group selection it would be necessary to learn what happened to those 87 individuals that disappeared during the course of the study. The social behavior of Acorn Woodpeckers appears very complex, and is not fully de- scribed in this study. There is a long appendix on visual and vocal displays, with high- quality audiospectrograms. The vocal repertoire is large and the function of many calls remains unclear because of the variety of situations in which they occur. The Acorn Wood- pecker almost certainly evolved from a non-social melanerpine relative, but the MacRoberts’ make no attempt to compare the displays and vocalizations of their species with those of close relatives. This should be done, because it could provide an excellent example of the ways in which display behaviors evolve to fit a new social system. The authors found that dominance hierarchies exist within groups, males and older birds being dominant; but there is no discussion of the role of calls or displays in the maintenance of these hierarchies. In several instances, adults excluded young of the year from acorn stores, but did not expel them from the territory. The young either fed on secondary stores, or were rationed acorns by the adults. The significance of this behavior remains obscure. Much remains to be learned about the Acorn Woodpecker. We still do not know which group members are parents and which are helpers during nesting. Long-term studies 654 THE WILSON BULLETIN • VoL 89, No. 4 will be necessary to determine genealogical relationships among group members. The question of population regulation remains unanswered, and we still have no clear idea why certain groups breed while others do not. These comments are intended not as criticisms but as encouragement for further work. The MacRoberts’ have written an excellent account based on an intensive and highly successful three-year field effort. In so doing they have made by far the most significant contribution to date toward an under- standing of this remarkable species. — Carl E. Bock. The Habitat Guide to Birding. By Thomas P. McElroy, Jr., Albert A. Knopf, New York, 1974: xvi + 257 pp. $8.95. — This book is far more than its title implies. It is really a text which, in addition to habitats, touches on such wide-ranging subjects as moonwatching (bird migration at night), choice of binoculars and spotting scopes, their use and care, precautions and clothing for different habitats, and tips on leading bird trips. It briefly discusses feeding stations and choice of bird food. It takes up tbe subject of bird communities and the niches filled by various species in the same habitat. There are numerous drawings by Matthew Kalmenoff, almost all of which I found very' pleasing. Some readers will not like the physical make-up of the book. The text covers a little over half of each 8 inch X 9 inch page, leaving a 3% inch margin for the artist's illustrations. However, on some pages this margin is blank, and it seems a waste of s])ace, although the author suggests using this space for notes or sketches, thus personalizing the reader's copy. There are no labels under the species drawings, and one must turn to a “Guide to the Illustrations” in the front if one cannot identify the l)ird or scene depicted. Therefore a beginner might well be confused when reading about the meadowlark and Killdeer, with an unlabelled drawing of a Horned Lark beside the text. In one place there are drawings of foot adaptations, but with no accompanying discussion. However, in most cases, text and illustrations are adjacent. In a book wbich deals with a great variety of topics, the author has done a very good job of leading from one subject into the next. His bird categories are interesting and often uniijue. For example, in his chapter on common lurds of country roadsides, he divides them into wire-sitters, fence-post-sitters, road-kill scavengers and “others.” His descriptions and explanations of the stages of development of marshes and swamps, the eventual domination of old fields by forests, and the effect of such plant successions on bird life, I found excellent. As the author himself points out, this book will be a source of help to the amateur and the experienced birder alike. I strongly recommend that the reader note carefully his early chapter called “Some suggestions for using this book.” It should be read more than once. There is a species index and a good bibliography. The book is limited to the eastern half of the continent, and I hope that someday there will be a similar one for the west. — Sally Hoyt Spofford. December 1977 • ORNITHOLOGICAL NEWS 6S5 ORNITHOLOGICAL NEWS LOUIS AGASSIZ FUERTES AND MARGARET MORSE NICE AWARDS Fuertes Awards are devoted to the encouragement and stimulation of young ornithol- ogists. One particular desire is the development of research interests among amateur ornithologists and students. Any kind of ornithological research may be aided. Recipients of grants need not be associated with academic institutions. Each proposal is considered primarily on the basis of possible contribution to ornithological knowledge. Although grantees are not required to publish their studies in The Wilson Bulletin, it is hoped that they will submit their manuscripts to the editor of The Wilson Bulletin for consideration. Most of the statements applicable to the Fuertes Awards are also applicable to the Nice Award. However, the Nice Award is limited to applicants not associated with a college or university. It is intended to encourage the independent researcher without access to funds and facilities generally available at the college. High school students are eligible. In some years two Fuertes Awards have been made, in some years, one. Amounts have been between $200 and $100. One Nice Award is made annually, in the amount of $100. Interested persons may write to Eugene S. Morton, National Zoological Park, Office of Zoological Research, Washington, D.C. 20008. Completed applications must he received by 15 March 1978. Final decisions will be made by the Council at the annual meeting of the Society, 4-7 May 1978. FRANK M. CHAPMAN FUND The Frank M. Chapman Memorial Fund gives grants in aid of ornithological research and also post-doctoral fellowships. While there is no restriction on who may apply, the Committee particularly welcomes and favors applications from graduate students; projects in game management and the medical sciences are seldom funded. Applications are due on 15 September and 15 February. Information on form and content of applications may be obtained from the Frank M. Chapman Memorial Fund Committee, The American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024. 1978 ANNUAL MEETING The 59th annual meeting of The Wilson Ornithological Society will be held at Jackson’s Mill, West Virginia, on 4-7 May 1978. The meeting will be hosted by the Brooks Bird Club, the Department of Wildlife Biology of West Virginia University, and West Virginia University. Information concerning lodging and field trips and abstract forms for sub- mitted papers will be mailed to the membership early in 1978. The deadline for submission of abstracts will be 1 March 1978. A special feature of the meeting will be a symposium titled, “Resource Use Strategies in Birds,” to be held on the afternoon of Friday, 5 May. The symposium is organized by Dr. Elliot J. Tramer. The chairman of the local Committee is Dr. Robert Whitmore, Division of Forestry, West Virginia University, Morgantown, WV 26506. 656 THE WILSOxN BULLETIN • VoL 89, No. 4 INTERNATIONAL BIRDS IN CAPTIVITY SYMPOSIUM The 1st International Birds in Captivity Symposium wdll be held in Seattle, Washington from 8-12 March 1978. Sessions ’will deal with husbandry, nutrition, medicine, sexing techniques, behavior, reproduction, and ornithological studies done in the ’wild. A panel discussion will be held after each topic session and there will be a special panel on conservation. All presented papers will be published after completion of the Symposium. For more information please write: Jan R. van Oosten, Chairman, lECF, 1008 James St., Seattle, WA 98104. A SYMPOSIUM: ROLE OF INSECTIVOROUS BIRDS IN FOREST ECOSYSTEMS The role of insectivorous birds in forest ecosystems will be the theme of a symposium co-sponsored by Stephen F. Austin State University (SFASU) School of Forestry and USDA Forest Service (USFS), Southern Forest Experiment Station, Nacogdoches, Texas. Conference chairmen are Dr. James C. Kroll (SFASU) and Dr. James G. Dickson (USFS). The symposium should provide a forum for exchange of current ideas between researchers and land managers. Tentative subject areas are: sampling bird populations, sampling prey populations, population dynamics of insectivorous birds, birds as biological control agents, and methods to increase bird populations. Technical papers will be presented 13 and 14 July, and a field trip will be scheduled for the morning of 15 July. Invited and volunteer papers selected for presentation by the editorial review board will be presented orally in abbreviated form and published in the symposium proceedings. Persons wishing to present a paper should submit manuscripts by 1 March 1978 to: Dr. James C. Kroll, P.O. Box 6109, Stephen F. Austin State University, Nacogdoches, Texas 75962. OPINIONS OF THE INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE The following Opinions have been published recently by the International Commission on Zoological Nomenclature, c/o British Museum (Natural History), Cromwell Road, London SW7 5BD, United Kingdom ( see Bulletin Zoological Nomenclature Volume 34, part 1). 1078 (p. 14) Anas punctata Burchell, 1822 (Aves suppressed under the plenary powers. 1081 (p. 25) Addition of Family-Group Names based on Alca (Aves) and Alces (Mam- malia) to the Official List of Family-Group Names in Zoology. The Commission cannot supply separates of Opinions. INDEX TO VOLUME 89, 1977 By Bonnie J. Turner and Wendy J. Weber Accipiter cooperii, 63, 360-369, 593 fasciatus vigilax, 619 gentilis, 360-369 nisus, 361, 599, 620 striatus, 63, 362, 620 Aceros leucocephalus, 8 Acrocephalus schoenobaenus, 462 scirpaceus, 462 Actitis macularia, 167-171, 543-561 Adkisson, Curtis S., Morphological vari- ation in North American Pine Gros- beaks, 380-395; see Conner, Rich- ard N. and Aethia cristatella, 618 pusilla, 618 age determination Agelaius phoeniceus, 73 Chlorostilbon canivetii, 160 Ortalis vetula, 48-51 Xanthocephalus xanthocephalus, 73 age determination of embryos Larus delawarensis, 250 Agelaius phoeniceus, 28, 43-44, 73-80, 221, 231, 253-265, 396-403, 486, 543-561, 583-592 Aix sponsa, 231 Ajaia ajaja, 97, 157-158, 344 Alden, Stephen, see Silliman, James G., Scott Mills, and Alexornis, 357 Amadon, D., Further comments on sexual dimorphism in birds, 619-621 Amakihi, see Loxops virens Amazilia saucerottei, 161 tzacatl, 162 yucatanensis, 345 Amazona leucocephala bahamensis, 637 Ammodramus henslowii, 222, 231, 432 savannarum, 222, 231, 427, 486 Ammospermophilus harrisii harrisii, 470 Ammospiza caudacuta, 427, 432 Anas acuta, 154-157, 332 platyrhynchos, 231, 332, 540, 631 rubripes, 540 strepera, 331 Ani, Smooth-hilled, see Crotophaga ani annual meeting. Proceedings of 58th, 506- 519 Aphelocoma coerulescens, 172 ultramarina, 172, 303 Ara spp., 632 Aramus guarauna, 343 Archaeopteryx, 356, 488-492 Archilochus colubris, 163 Ardea herodias, 266 Ardeola ibis, see Buhulcus ibis Arenaria interpres, 167-171, 470 Arnold, Keith A. and Leon J. Folse, Jr., Movements of the Great-tailed Grackle in Texas, 602-608 Asio flammeus, 487 Asyndesmus lewis, 622 Auklet, Crested, see Aethia cristatella Least, see Aethia pusilla Auriparus flaviceps, 572-582 Austin, George T., Production and survival of the Verdin, 572-582 Avery, Michael, Paul F. Springer, and J. Frank Cassel, Weather influences on nocturnal bird mortality at a North Dakota tower, 291-299 awards and grants Aaron M. Bagg Student Membership Award, 503-504 Alexander Wilson Prize, 514 Eastern Bird-Banding Association, 504 Edwards Prize, 514 Frank M. Chapman Memorial Fund, 655 Hawk Mountain Research, 370 Louis Agassiz Fuertes Award, 514 Margaret Morse Nice Award, 514 Balph, Martha Hatch, Sex differences in alarm responses of wintering Eve- ning Grosbeaks, 325-327 Bananaquit, see Coereba flaveola banding Agelaius phoeniceus, 73-80 Canachites canadensis, 67 Carpodacus cassinii, 57-66 Chasiempis sandwichensis, 193-210 657 658 THE WILSON BULLETIN • Vol. 89, No. 4 Falco sparverius, 618-619 Lagopus leucurus, 107-115 Parabuteo unicinctus, 469-470 Passer domesticus, 300-309 Quiscalus mexicanus, 602-608 warblers, migrant, 456-465 Xanthocephalus xanthocephalus, 73-80 Baptista, Luis F., Song Dialects and Demes in Sedentary Populations of the White-crowned Sparrow, reviewed, 651-652; review by, 650-651 Baptornis, 356 Barlow, J. C., see James, R. D., P. L. Mc- Laren, and Bart, Johnathan, Winter distribution of Red-tailed Hawks in central New York State, 624-625 Bartramia longicauda, 543-561 Bateman, Hugh A., see Flickinger, Edward L., David S. Lobpries, and behavior agonistic Calidris alba, 434-438 C. pusilla, 336-338 Hydranassa tricolor, 266-270 Nycticorax nycticorax, 351 Passer domesticus, 477 Pluvialis s(iuatarola, 471 Podiceps grisegena, 44 brooding Bonasa umbellus, 449-450 Ca{)rimulgus vociferus, 476-477 Chasiempis sandwichensis, 204-206 Falco sparverius, 618-619 Hydranasa tricolor, 282-283 Passer domesticus, 303-305 Podiceps grisegena, 41-42 Sphyrapicus varius, 313-315 courtship Chasiempis sandwichensis, 199 Podiceps grisegena, 33-35 Streptopelia risoria, 593-601 defense Otus choliba, 609-612 display Carpodacus cassinii, 61-63 Chasiempis sandwichensis, 199-201 Hydranassa tricolor, 266-285 distress Caprimulgus vociferus, 476-477 Hesperiphona vespertina, 325-327 Nycticorax nycticorax, 351 Vireo bellii, 349 dominance Carpodacus cassinii, 57-66 egg rejection Icterus galbula, 24-27 flocking Calidris alba, 434-438 Lagopus leucurus, 110-112 Pluvialis squatarola, 470-472 Psilorhinus morio, 171-173 Ramphocelus passerinii, 151-153 foraging Amazilia yucatanensis, 345 Buteo jamaicensis, 624-625 Calidris alba, 434-438 C. pusilla, 336-338 Chasiempis sandwichensis, 194 Coragyps atratus, 328-329 Cyanocompsa parellina, 345 Dendroica coronata, 346, 485 D. discolor, 158-159 Euducimus albus, 342-345 Hummingbirds, Costa Rican, 159-164 Lampornis calolaema, 613-616 Larus argentatus, 338, 415-421 Loxigilla noctis, 339 Melanitta deglandi, 331 Pandion haliaetus, 149-150 Phaethornis guy, 613-616 Picoides arizonae, 485 P. scalaris, 485 Podiceps grisegena, 42-43 Psilorhinus morio, 171-173 Rallus longirostris yumanensis, 332- 336 Ramphocelus passerinii, 151-153 Regulus calendula, 485 Sialia sialis, 404-414 Sphyrapicus varius, 315-316 Spizella pusilla, 625-627 Uropsila leucogastra, 345 Vireo flavoviridis, 345 foraging, ground Caprimulgus carolinensis, 165-166 hunting Pithecophaga jefferyi, 2-7 December 1977 • INDEX TO VOLUME 89 659 inter-brood movements Canachites canadensis, 61-12 nesting Agelaius phoeniceus, 73-75 Anas acuta, 154^157 Bonasa umbellus, 439-455 Capella gallinago, 118 Caprimulgus vociferus, 476-477 Cathartes aura, 328-329 Chasiempis sandwichensis, 199-206 Coragyps atratus, 328-329 Larus argentatus, 472-473 Loxigilla noctis, 339-340 Podiceps grisegena, 35-41 Sphyrapicus varius, 310-324 Xanthocephalus xanthocephalus, 73-75 nestling Otus choliba, 609-612 pairing Agelaius phoeniceus, 583-592 Eudocimus albus, 97 social Carpodacus cassinii, 57-66 Berger, A. J., see Whittow, G. C. and Best, Louis B., Patterns of feeding Field Sparrow young, 625-627 Bison, see Bison bison antiquus Bison bison antiquus, 622 Bittern, American, see Botaurus lentigino- sus Least, see Ixobrychus exilis Blackbird, Yellow-headed, see Xanthoceph- alus xanthocephalus Red-winged, see Agelaius phoeniceus Blankenship, David R., see King, Kirke A., , Richard T. Paul and Rohin C. A. Rice Bluebird, Eastern, see Sialia sialis Mountain, see Sialia currucoides Bobolink, see Dolichonyx oryzivorus Bobwhite, see Colinus virginianus Bock, Carl E,, review by, 652-654 Bohlen, H. David, see Seets, James W. and Bolen, Eric G., see McCamant, Richard E. and Bombycilla cedrorum, 28, 102, 158, 345, 543-561 Bonasa umbellus, 439-455 Borecky, Stephen R., review by, 354-355 Borror, Donald J., Rufous-sided Towhees mimicking Carolina Wren and Field Sparrow, 477-480 Botaurus lentiginosus, 231 Brachyramphus brevirostris, 467-469 marmoratus, 467 Brant, see Branta bernicla Branta bernicla, 528 canadensis, 469, 523-531 Braun, Clait E., James H. Enderson, Charles J. Henny, Heinz Meng, and Alva G. Nye, Jr. Conservation Com- mittee Report, 360-369; , Keith W. Harmon, Jerome A. Jack- son, and Carroll D. Littlefield. Re- port of the Conservation Committee, 511; see Hoffman, Richard W. and breeding biology Agelaius phoeniceus, 73-80 Auriparus flaviceps, 572-582 Bonasa umbellus, 439-455 Capella gallinago, 116-119 Chasiempis sandwichensis, 193-210 Dolichonyx oryzivorus, 78 Hydranassa tricolor, 266-285 Passer domesticus, 300-309 Petrochelidon pyrrhonota, 286-290 Xanthocephalus xanthocephalus, 73-78 breeding bird survey, 543-561 Brewer, Richard and Lynda Swander, Life history factors affecting the intrinsic rate of natural increase of birds of the deciduous forest biome, 211-232 Brilliant, Green-crowned, see Heliodoxa jacula Brolga, see Crus rubicundus Brudenell-Bruce, P. G. C., The Birds of the Bahamas (reviewed) , 636-637 Bubo virginianus, 231, 422 Bubulcus ibis, 158, 266, 350 Buceros hydrocorax, 7 Bullfinch, see Pyrrhula pyrrhula Greater Antillean, see Loxigilla violacea Lesser Antillean, see Loxigilla noctis Bunting, Blue, see Passerina parellina Indigo, see Passerina cyanea 660 THE WILSON BULLETIN • VoL 89, No. 4 Lazuli, see Passerina ainoena Orange-breasted, see Passerina leclan- cherii Painted, see Passerina ciris Rose-bellied, see Passerina rositae Varied, see Passerina versicolor Burger, Joanna, review by, 645-646; and D. Caldwell Hahn, Crow pre- dation on Black-crowned Night Heron eggs, 350-351 Burns, Robert D., new life member, 561 Burton, Jean, see McNeil, Raymond and Buteo jamaicensis, 237, 361, 619, 624-625, 629 lagopus, 629 swainsoni, 623 Butorides striatus, 231, 266 virescens, see Butorides striatus Caccamise, Donald F., Breeding success and nest site characteristics of the Red-winged Blackbird, 396-403; and Wesley W. Weathers, Winter nest microclimate of Monk I’arakeets, 346-349 Calaby, J. H., see Frith, H. J. and Calidris alha, 167-171, 434-438, 470 alpina, 167-171, 613 a. pacifica, 613 bairdii, 612 canutus, 167-171, 470 fuscicollis, 167-171 melanotos, 167-171 minutilla, 167-171, 178 pusillus, 167-171, 178, 336, 437, 469 (Jalyptorhynchus baudini, 495 Camelops, 622 Campylopterus hemileucurus, 160-165 Campylorhynchus brunneicapillus, 576 Canachites canadensis, 67-72, 452 c. franklinii, 452 Canis latrans, 470, 622 Capella gallinago, 116-121, 169 Capen, David E., Eggshell thickness vari- ability in the White-faced Ibis, 99- 106 Capercaillie, see Tetrao urogallus Caprimulgus carolinensis, 165-166, 231 vociferus, 231, 476 Caracara cheriway, 618 Cardinal, see Cardinalis cardinalis Vermilion, see Cardinalis phoeniceus Cardinalinae, 130-148 Cardinalis cardinalis, 130-148, 232, 543- 561 phoeniceus, 130-148 sinuatus, see Pyrrhuloxia sinuatus Carduelis tristis, 232, 543-561 Carpenter, F. Lynn, Ecology and evolution of an Andean hummingbird ( Oreo- trochilus estella), reviewed, 352-354 Carpodacus cassinii, 57-66 mexicanus, 57, 253-265, 480 purpureus purpureus, 57 Carter, Brenda, see Manning, T. H. and Caryothraustes canadensis, 130-148 humeralis, 130-148 Cassel, J. Frank, see Avery, Michael, Paul F. Springer, and Casmerodius albus, 350 Catbird, Gray, see Dumetella carolinensis Cathartes aura, 328-329 burrovianus, 329 Catharus fuscescens, 231, 425, 543-561 guttatus, 425 minimus, 425 mustelina, see Hylocichla mustelina ustulatus, 425 Catoptrophorus semipalmatus, 167-171 census Capella gallinago, 116-121 Dendrocygna bicolor, 329-331 Centrocercus urophasianus, 523 Certhia familiaris, 424 Chachalaca, Plain, see Ortalis vetula Chaetura pelagica, 543-561 Chamberlin, Michael L., Observations on the Red-necked Grebe nesting in Michigan, 33-46 Charadrius semipalmatus, 167-171, 470 vociferus, 167-171, 543-561 Chasiempis sandwichensis, 193-209; color- plate facing 193 s. gayi, 193-209; colorplate facing 193 s. sandwichensis, 193 s. sclateri, 193, 202 Chat, Yellow-breasted, see Icteria virens chemical residues Lophodytes cucullatus, 532-542 December 1977 • INDEX TO VOLUME 89 661 Mergus merganser, 532-542 M. serrator, 532-542 Streptopelia risoria, 593-601 Chen hyperborea, 501-502 rossii, 501-502, 528 Chendytes lawi, 358 Chickadee, Black-capped, see Parus atri- capillus chicken, domestic, see Callus gallus Chlidonias niger, 231 Chlorestes notatus, 615 Chlorophanes spiza, 152 Chlorostilbon canivetii, 160 Chondestes grammacus, 222 Chuck- Will’s- Widow, see Caprimulgus caro- linensis Cinclorhamphus cruralis, 619-620 Cink, Calvin L., Snake predation on Bell’s Vireo nestlings, 349-350 Cistothorus platensis, 222, 231, 424 Clangula hyemalis, 81-91 Clench, Mary H., review by, 495-496, 636- 638 clutch size Agelaius phoeniceus, 74-77 Auriparus flaviceps, 572-574 Chasiempis sandwichensis, 202 deciduous forest biome birds, 212-215 Dendroica petechia, 153 Empidonax minimus, 153 Falco sparverius, 618 Icterus galbula, 23 Larus delawarensis, 246 Petrochelidon pyrrhonota, 286-290 Plegadis chihi, 104 Podiceps grisegena, 37 Xanthocephalus xanthocephalus, 74-77 Coccyzus americanus, 231 erythrophthalmus, 232 Cockatoo, White-tailed Black, see Calypto- rhynchus baudini Coereba flaveola, 637 Colaptes auratus, 122-129, 165, 232, 312, 424, 543-561 Cole, William W., Jr,, new life member, 232 Colibri delphinae, 160 thalassinus, 160 Colinus virginianus, 231 color-marking Progne subis, 177 Calidris pusillus, 178 C. minutilla, 178 color measurement Cardinalinae, 130-148 Pinicola enucleator, 387-388 Coluber constrictor, 349 Columba livia, 543-561 nigrirostris, 152 palumbus, 64, 434 Conant, Sheila, The breeding biology of the Oahu ‘Elepaio, 193-210 Connecticut Warbler, see Oporornis agilis Conner, Richard N. and Curtis S, Adkisscn, Principal component analysis of woodpecker nesting habitat, 122-129 Conservation Committee Report, Falconry: Effects of Raptor populations and management in North America, 360- 369 Contopus sordidulus, 253-265 Virens, 212, 226, 231, 543-561 copulation Chasiempis sandwichensis, 199 Hydranassa tricolor, 280 Podiceps grisegena, 37 Coragyps atratus, 328-329 Corvus brachyrhynchos, 350, 543-561 corax, 233-242 corone, 239 coronoides, 495 cryptoleucus, 239 frugilegus, 239 macrorhynchos, 8 ossifragus, 350 Cosgrove, Ed, see Cosgrove, Irene and Cosgrove, Irene and Ed Cosgrove, My reci- pes are for the Birds, reviewed, 359 Costa Rica, 151-153, 159-164, 613-616 cottontail, desert, see Sylvilagus auduhoni Coturnicops noveboracensis, 424 Coturnix japonica, 102 Coughley, Graeme, Analysis of Vertebrate Populations, reviewed, 359 Cowbird, Brown-headed, see Molothrus ater Cowgill, Ursula M., review by, 501-502 coyote, see Canis latrans 662 THE WILSON BULLETIN • Vol 89, No. 4 Cracraft, Joel, Special Review — John Os- trom’s Studies on Archaeopteryx, The Origin of Birds, and the Evolu- tion of Avian Flight, 488-492 Cram, Roger F., Bird Flight Photography, reviewed, 359 Crawford, Richard D., Breeding biology of year-old and older female Red- winged and Yellow-headed Black- birds, 73-80 Creeper, Brown, see Certhia familiaris Crocethia alba, see Calidris alba Cromartie, Eugene, see White, Donald H. and Crotophaga ani, 303 Crow, Common, see Corvus brachyrhynchos Fish, see Corvus ossifragus Large-billed, see Corvus macrorhynchos Cuckoo, Black-billed, see Coccyzus ery- throphtbalmus Yellow-billed, see Coccyzus americanus Cyanocitta cristata, 171, 231, 351, 543-561 stelleri, 171 Cyanocompsa cyanea, 130-148 parellina, 130-148, 345 Cyanoloxia glaucocaerulea, 130-148 Dacelo gigas, 495 Dawson, W illiam R., review by, 496-497 Dendragapus obscurus, 108 Dendrocygna autumnalis, 621-622 bicolor, 329, 621 Dendroica caerulescens, 167, 426, 458, 466 castanea, 166-167, 426 cerulea, 231 coronata, 167, 346, 426, 485 c. auduboni, 253-265 discolor, 158, 458, 466 dominica, 466, 637 fusca, 426, 466 magnolia, 166-167, 425, 466 palmarum, 426, 466 pensylvanica, 166-167, 232, 426 petechia, 153-154, 231, 253-265, 425, 543-561 pinus, 426, 466 striata, 166-167, 426, 466 tigrina, 425, 466 virens, 426, 466 density, breeding Capella gallinago, 116-119 Diamond, A. W., P. Lack, and R. W, Smith, Weights and fat condition of some migrant warblers in Jamaica, 456- 466 Dichromanassa rufescens, 158, 266 Dickcissel, see Spiza americana Dickerman, Robert W., Three more new specimen records for Guatemala, 612-613 Diglossa plumbea, 615 dimorphism, sexual Eudocimus albus, 92-98 display Hydranassa tricolor, 270-285 distribution Buteo jamaicensis, 624-625 Chasiempis sandwichensis, 194-195 Pithecophaga jefferyi, 11-17 Dixon, Keith L., review by, 642-644 Dolicbonyx oryzivorus, 78, 212, 224, 231, 426, 487, 543-561 Doryfera ludovicae, 160 Douglass, John F., Prairie Warbler feeds from spider web, 158-159 Dove, Ringed Turtle, see Streptopelia ri- soria Rock, see Columba livia White-bellied, see Leptotila jamaicensis Dowitcher, Short-billed, see Limnodromus griseus Dryocopus pileatus, 122-129 DuBois, Alexander Dawes with Charlotte A. DuBois, Birds and Their Ways, reviewed, 502 Duck, Black, see Anas rubripes Wood, see Aix sponsa Dumetella carolinensis, 231, 424, 543-561 Dunlin, see Calidris alpina Dyer, Eric L., Rapid chick separation in W hip-poor-wills, 476-477 Eagle, Bald, see Haliaeetus leucocephalus Monkey-eating, see Pithecophaga jefferyi Ectopistes migratorius, 219 Editor’s Report, 509-510 egg size Agelaius phoeniceus, 74—75 Branta canadensis, 469 December 1977 • INDEX TO VOLUME 89 663 Calidris pusilla, 469 Chasiempis sandwichensis, 202 Icterus galbula, 31 Lanius ludovicianus, 31 Larus delawarensis, 473-475 Xanthocephalus xanthocephalus, 74-75 eggs, recognition Icterus galbula, 30-31 runt Branta canadensis, 469 Calidris pusilla, 469 Larus spp., 469 Sturnus vulgaris, 469 eggshell thickness Lophodytes cucullatus, 532-542 Mergus merganser, 532-542 M, serrator, 532-542 Plegadis chihi, 99-106 Egret, Cattle, see Bubulcus ibis Common, see Casmerodius albus Reddish, see Dichromanassa rufescens Snowy, see Egretta thula Egretta thula, 266, 350 Elaenia, Caribbean, see Elaenia martinica Elaenia martinica, 341 ‘Elepaio, see Chasiempis sandwichensis Elaphe ohsoleta, 349 elephant seals, see Mirounga Ellarson, Robert S., see Peterson, Steven R. and Elvira cupreiceps, 162 embryo development Larus delawarensis, 243-252 Emerald, Coppery-headed, see Elvira cuprei- ceps Empidonax alnorum, 424, 543-561 difficilis, 253-265 flaviventris, 424 minimus, 153-154, 166-167, 424 traillii, 231, 253-265, 543-561 virescens, 212, 226, 231 endangered species and populations Dendrocygna bicolor, 329-331 Picoides borealis, 164-165 Pithecophaga jefferyi, 1-20 Rallus longirostris yumanensis, 332-336 Enderson, James H., see Braun, Clait E., , Charles J. Henny, Heinz Meng, and Alva G. Nye, Jr. Eremophila alpestris, 231, 543-561, 618 Erythrotriorchis radiatus, 620 Eudocimus albus, 92, 342 ruber, 97, 343 Eulampis, 342 Euplectes progne delamerei, 620 Eupherusa examia, 162, 615 Falco mexicanus, 360-369, 598 peregrinus, 239, 598 p. anatum, 360-369 severus, 8 sparverius, 360-369, 486, 593, 618, 623- 624 Falcon, Prairie, see Falco mexicanus falconry, see conservation committee report, 360-369 Farner, Donald S. and James R. King, eds.. Avian Biology, Vol. 5, reviewed, 181- 183 fat, 456-465 Feinsinger, Peter, Notes on the humming- birds of Monteverde, Cordillera de Tilaran, Costa Rica, 159-164 Ficedula hypoleuca, 233, 323 Finch, Blue, see Passerina caerulescens and Porphyrospiza caerulescens Black Rosy, see Leucosticte atrata Cassin’s, see Carpodacus cassinii Gray-crowned Rosy, see Leucosticte te- phrocotis House, see Carpodacus mexicanus Purple, see Carpodacus purpureus fishing behavior, see behavior, foraging Flack, J. A. Douglas, Bird Populations of Aspen Forests in Western North America, reviewed, 642-644 Flamingo, American, see Phoenicopterus ruber fledging success Agelaius phoeniceus, 74-76 Parabuteo unicinctus, 469-470 Xanthocephalus xanthocephalus, 74—76 Flicker, Common, see Coleptes auratus Flickinger, Edward L., David S. Lobpries and Hugh A. Bateman, Fulvous Whistling-duck populations in Texas and Louisiana, 329-331 flock size Calidris alba, 434-438 664 THE WILSON BULLETIN • VoL 89, No. 4 Dendrocygna bicolor, 329-331 Lagopus leucurus, 110-112 Psilorhinus morio, 171 Ramphocelus passerinii, 151-153 flocks, mixed species, 152, 470-472 Flower-piercer, Slaty, see Diglossa plumbea Flycatcher, Acadian, see Empidonax vires- cens Alder, see Empidonax alnorum Ash-throated, see Myiarchus cinerascens Great Crested, see Myiarchus crinitus Least, see Empidonax minimus Pied, see Ficedula hypoleuca Western, see Empidonax difficilis Willow, see Empidonax traillii Yellow-bellied, see Empidonax flaviven- tris Folse, Leon, J. Jr., see Arnold, Keith A. and food habits Caprimulgus carolinensis, 166 Clangula hyemalis, 81-91 Dendroica coronata, 485 Eudocimus albus, 94—97 Falco sparverius, 623-624 Larus argentatus, 338, 415-421 Melanerpes formicivorus, 150-151 Pandion baliaetus, 625 Picoides arizonae, 485 P. scalaris, 485 Rallus longirostris yumanensis, 332-336 Regulus calendula, 485 Sialia sialis, 407 food, seasonal change Clangula hyemalis, 81-91 food storage Falco sparverius, 623-624 Melanerpes formicivorus, 150-151 foraging, see behavior, foraging Fowl, Mallee, see Leipoa ocellata fox, red, see Vulpes vulpes Frith, H. J. and J. H. Calaby (eds,). Pro- ceedings of the 16th International Ornithological Congress, reviewed, 493-494 frog, squirrel tree, see Hyla squirella Gadwall, see Anas strepera Callus gallus, 99, 243, 452 Gaunt, Abbot S., Membership Committee Report, 510 Gauthreaux, Sidney A. Jr., review by, 181- 183; painting by, frontispiece facing 373 Genoways, Hugh H., see McCaugh, M. Houston and Geothlypis rostrata, 637 trichas, 214, 231, 253-265, 349, 426, 461, 466 Giezentanner, Keith, photograph by, frontis- piece facing 521 Glaucomys volans, 165 Gnatcatcher, Blue-gray, see Pclioptila caerulea Goldentail, Blue-throated, see Hylocharis eliciae Goldfinch, American, see Carduelis tristis Goodwin, Derek, Crows of the World, re- viewed, 354^-355 Goose, Blue, see Chen hyperborea Canada, see Branta canadensis Ross’, see Chen rossii Snow, see Chen hyperborea Goossen, J. Paul, Yellow Warbler nest used by a Least Flycatcher, 153-154 Goshawk, see Accipiter gentilis Crackle, Boat-tailed, see Quiscalus major Common, see Quiscalus quiscula (ireat-tailed, see Quiscalus mexicanus Grant, Gilbert, S. and Thomas L. Quay, Breeding biology of Cliff Swallows in Virginia, 286-290 Grebe, Eared, see Podiceps nigricollis Horned, see Podiceps auritus Pied-billed, see Podilymbus podiceps Red-necked, see Podiceps grisegena Green, Richard F., Do more birds produce fewer young? A comment on May- field’s measure of nest success, 173- 175 Griese, Herman J., see Stahlecker, Dale W. and grit, 82 Grosbeak, Black-backed, see Pheucticus aureoventris Black-headed, see Pheucticus melano- cephalus December 1977 • INDEX TO VOLUME 89 665 Blue, see Guiraca caerulea Blue-black, see Passerina cyanoides and Cyanocompsa cyanea Crimson-collared, see Rhodothraupis celaeno Evening, see Hesperiphona vespertina Indigo, see Passerina glaucocaerulea and Cyanoloxia glaucocaerulea Pine, see Pinicola enucleator Red-and-black, see Periporphyrus eryth- romelas Rose-breasted, see Pheucticus ludovici- anus Slate-colored, see Pitylus grossus Ultramarine, see Passerina brisscnii and Cyanocompsa cyanea Yellow, see Pheucticus chrysopeplus Yellow-green, see Caryothraustes cana- densis Yellow-shouldered, see Caryothraustes humeralis ground squirrel, Harris’ antelope, see Am- mospermophilus harrissii harrissii Grouse, Black, see Lyrurus tetrix Blue, see Dendragapus obscurus Red, see Lagopus 1. scoticus Ruffed, see Bonasa umbellus Sage, see Centrocercus urophasianus Sharp-tailed, see Pedioecetes phasianellus Spruce, see Canachites canadensis Groves, Sarah, see Harrington, Brian A. and growth, nestling Agelaius phoeniceus, 75 Otus choliba, 609-612 Xanthocephalus xanthocephalus, 75 Grubb, Thomas C., Jr., Why Ospreys hover, 149-150 Grubbs, Vera Lee, An intraspecific mortal attack, 477 Grus canadensis, 635 rubicundus, 495 Guatemala, 612-613 Guillory, Harland, D., see Pratt, H. Doug- las, Brent Ortego, and Guiraca caerulea, 130-148, 253-265 Gull, Black-headed, see Larus ridibundus California, see Larus californicus Great Black-backed, see Larus marinus Herring, see Larus argentatus Laughing, see Larus atricilla Lesser Black-backed, see Larus fuscus request for assistance, 421 Ring-billed, see Larus delawarensis Western, see Larus occidentalis Gymnorhina spp., 495 Gymnorhinus cyanocephalus, 171 habitat choice Pithecophaga jefferyi, 8-9 habitat destruction Anus acuta, 154-157 habitat partitioning Branta canadensis, 523-531 Chasiempis sandwichensis, 219-226 Eudocimus albus, 94 grassland birds, 486-487 Lagopus leucurus, 109-110 Lampornis calolaema, 614-616 passerine birds, 253-265 Phaethornis guy, 614-616 Sialia sialis, 410 habitat, nesting Colaptes auratus, 122-129 Dryocopus pileatus, 122-129 Melanerpes erythrocephalus, 122-129 Picoides pubescens, 122-129 P. villosus, 122-129 habitat resource use, 253-265 Haegele, M. A. and Rick H. Hudson, Re- duction of courtship behavior in- duced by DDE in male Ringed Turtle Doves, 593-601 Hahn, D, Caldwell, see Burger, Joanna and Hailman, Jack P., see Waide, Robert B, and Haliaetus leucocephalus, 178 Hall, George A., Proceedings of the Lifty- eighth Annual Meeting, 506-519 Hamel, Paul B., see Hooper, Robert G. and Hanson, Harold C. and Robert L. Jones, The biogeochemistry of Blue, Snow and Ross’ geese, reviewed, 501-502 Harmon, Keith W., see Braun, Clait E. et ah, Conservation Committee Report, 1976 Harrington, Brian A. and Sarah Groves, 666 THE WILSON BULLETIN • VoL 89, No. 4 Aggression in foraging migrant Semi- palmated Sandpipers, 336-338 Harrison, Keith G., Perch height selection of grassland birds, 486-487 hatching Podiceps grisegena, 41 Hawk, Cooper’s, see Accipiter cooperii Harris’, see Parabuteo unicinctus Red-tailed, see Buteo jamaicensis Rough-legged, see Buteo lagopus Sharp-shinned, see Accipiter striatus Swainson’s see Buteo swainsoni Heintzelman, Donald S., A Guide to East- ern Hawk Watching, reviewed, 358 Heliodoxa jacula, 162 Heliomaster constantii. 162 Hellack, Jenna J. and Gary D. Schnell, Phenetic analysis of tlie subfamily Cardinalinae using external and skeletal characters, 130-148 Helminthophis, 610 Helmitheros vermivora, 466 Henny, Charles J., see Braun, Clait E., James H. Enderson, , Heinz Meng, and Alva G. Nye, Jr. Hermit, Guy’s, see Pliaethornis guy Green, see Pliaethornis guy Little, see Pliaethornis longueniareus Heron, Black-crowned Night, see Nyctico- rax nycticorax Great Blue, see Ardea herodias Green, see Butorides striatus Louisiana, see Hydranassa tricolor Hesperiphona vespertina, 325-327 Hesperornis, 356 Hieraaetus kienerii, 620 Hirundo rustica, 176, 487, 543-561, 562-571, 632-635 Hobby, Oriental, see Falco severus Hoff, Janies G., Slipper shells, a major food item for White-winged Scoters, 331 Hoffman, Richard W. and Clait E. Braun, Characteristics of a wintering popu- lation of White-tailed Ptarmigan in Colorado, 107-115 home range Pithecophaga jefferyi, 9-11 Quiscalus mexicanus, 483-485 Honeycreeper, Green, see Chlorophanes spiza Hooper, Robert G., Nesting habitat of Com- mon Ravens in Virginia, 233-242; and Paul B. Hamel, Nesting habitat of Bachman’s Warbler — a Review, 373-379 Hornbill, Rufous, see Bucercs hydrccorax Writhed-billed, see Aceros leucocephalus Hudson, Rick H., see Haegele, M. A. and Hummingbird, Blue-vented, see Amazilia saucerottei Fiery-throated, see Panterpe insignis Fawn-breasted, see Amazilia yucatanensis Rufous-tailed, see Amazilia tzacatl Ruby-throated, see Archilochus colubris Scintillant, see Selasphorus scintilla Stripe-tailed, see Eupherusa eximia Hunt, George L., and Max C. Thompson, Black-legged Kittiwakes nesting on snowbank, 616-618 Hydranassa tricolor, 266-285 Hyla scjuirella, 166 Hylocharis eliciae, 161 Hylocichla mustelina, 231, 425 Ibis, Australian White, see Threskiornis molucca Glossy, see Plegadis falcinellus Sacred, see Threskiornis aethiopiea Scarlet, see Eudocimus ruber White, see Eudocimus albus White-faced, see Plegadis chihi Icteria virens, 253-265 Icterus cucullatus, 253-265 dominicensis northropi, 637 gall)ula, 21-32, 212, 232, 426, 543-561, 637 g. hullocki, 253-265 Ictinia mississippiensis, 176 incubation Bonasa umbellus, 443-449 Chasiempis sandwichensis, 203-204 Dendrocygna autumnalis, 621 -622 Passer domesticus, 303-304 Petrochelidon pyrrhonota, 287 Podiceps grisegena, 37-41 Sphyrapicus varius, 312-313 December 1977 • INDEX TO VOLUME 89 667 Iridoprocne bicolor, 289 Ixobrychus exilis, 231 Jackson, Jerome A., Editor’s Report, 509- 510; see Braun, Clait E. et al.. Con- servation Committee Report Jamaica, 456-465 James, Douglas, The President’s page, 505; student membership committee, 511 James, R. D., P. L. McLaren and J. C. Bar- lov\r. Annotated Checklist of the Birds of Ontario, reviewed, 359 Jansma, Roxana, see Stacey, Peter B. and Jay, Blue, see Cyanocitta cristata Brown, see Psilorhinus morio Florida Scrub, see Aphelocoma c. coeru- lescens Mexican, see Aphelocoma ultramarina Pihon, see Gymnorhinus cyanocephalus Steller’s, see Cyanocitta stelleri Jehl, Joseph R. Jr., Pluvianellus socialis: Biology, Ecology and Relationships of an enigmatic Patagonian shore- bird, reviewed, 645-646 Johnson, Bruce R, and Ronald A. Ryder, Breeding densities and migration pe- riods of Common Snipe in Colorado, 116-121 Johnson, Robert F., Jr. and Leo M. Kirsch, Egg movement by a female Gadwall between nest bowls, 331-332 Jones, Robert L., see Hanson, Harold C. and Junco hyemalis, 427 Kaminski, Richard M. and Harold H. Prince, Nesting habitat of Canada Geese in southeastern Michigan, 523-531 Kaufman, John and Heinz Meng, Falcons Return, reviewed, 498-499 Kear, Janet, review by, 179-180 Kennedy, Robert S., Notes on the biology and population status of the Monkey- eating Eagle of the Philippines, 1- 20; new life member, 20 Keppie, Daniel M., Inter-brood movements of juvenile Spruce Grouse, 67-72 Kestrel, American, see Falco sparverius Kilgore, Delbert L., Jr. and Kathy L. Knud- son. Analysis of materials in Cliff and Barn swallow nests: relationship between mud selection and nest architecture, 562-571 Kilham, Lawrence, Nest-site differences be- tween Red-headed and Red-bellied woodpeckers in South Carolina, 164- 165; Nesting behavior of Yellow-bel- lied Sapsuckers, 310-324 Killdeer, see Charadrius vociferus King, James R., see Earner, Donald S. and King, Kirke A., David R. Blankenship, Richard T. Paul, and Robin C. A. Rice, Ticks as a factor in the 1975 nesting failure of Texas Brown Peli- cans, 157-158 Kingbird, Eastern, see Tyrannus tyrannus Western, see Tyrannus verticalis Kinglet, Golden-crowned, see Regulus sa- trapa Ruby-crowned, see Regulus calendula Kirsch, Leo M., see Johnson, Robert F., Jr. and — ■ Kite, Mississippi, see Ictinia mississippien- sis Kittiwake, Black-legged, see Rissa tridac- tyla Klimosewski, John, see Radke, Eleanor L. and Knot, see Calidris canutus Knudson, Kathy L., see Kilgore, Delbert L., Jr. and Kookaburra, see Dacelo gigas Krapu, Gary L., Pintail reproduction ham- pered by snowfall and agriculture, 154-157 Kushlan, James A., Sexual dimorphism in the White Ibis, 92-98; Foraging be- havior of the White Ibis, 342-345 Lack, P., see Diamond, A. W., — , and R. W. Smith Lagopus lagopus, 109, 451 1. scoticus, 451 leucurus, 107-115 mutus, 109 Lampornis calolaema, 162, 613 hemileucus, 162 668 THE \^'ILSON BULLETIN • VoL 89, No. 4 Lanius excubitor, 63 ludovicianus, 30 Lark. Horned, see Eremophila alpestris Larus argentatus, 243, 338, 415-421. 472, 618 atricilla. 158, 474 californicus, 243 delawarensis, 243-252, 473 fuscus, 415 glaucescens, 419, 472 marinus, 338, 618 occidentalis, 243 ridibundus, 472-473 Leberman, Robert C., The Birds of the Ligonier Valley, reviewed. 358-359 Leiothrix lutea, 208 Leipoa ocellata, 495 Lek Phaethornis guy, 160 Leptotila jamaicensis, 637 Leptoptilos crumeniferus, 97 Leucosticte atrata, 64 tepbrocotis. 64 Libran- Committee Report — 1976, 511-512 Limnodromus griseus, 167-171 Limnothlypis swainsonii, 374. 466 Limpkin, see Aramus guarauna Littlefield. Carroll D., see Braun, Clait E, et ab. Conservation Committee Re- port Lloyd-Evans. Trevor L., review by, 191-192 Lobpries, David S., see Flickinger, Edward L.. , and Hugh A. Bateman Lophcdytes cucullatus. 532-542 Lophortyx gambelii, 470 Loxigilla noctis, 338-342 n. con ii, 340 n. ridgwayi. 340 portoricensis. 342 violacea. 637 Loxops virens virens. 480 Lunk, William A., Librar}- Committee Re- port. 511-512 Lyon, David L., review by. 352-354 Lyrurus tetrix, 112 Lysaght. A. M., The Book of Birds: Five Centuries of Bird Illustration, re- viewed, 189-191 McCamant, Richard E. and Eric G. Bolen, Response of incubating Black-bellied Whistling-ducks to loss of mates, 621-622 McClure, H. Elliott, review by, 497-498 McElroy, Thomas P., Jr., The Habitat Guide to Birding, reviewed, 654 McGaugh. M. Houston and Hugh H. Geno- ways. State Laws as they pertain to Scientific Collecting Permits, re- viewed, 192 McHargue, Laurie A., Nesting of Turkey and Black vultures in Panama. 328- 329 McLaren. P. L., see James, R. D., , and J. C. Barlow McNeil. Raymond and Jean Burton, South- bound migration of shorebirds from the Gulf of St. Lawrence, 167-171 MacFarlane, Anne E., Roof-nesting by Common Terns, 475-476 MacRoberts, Barbara R., see MacRoberts, Michael H. and MacRoberts. Michael H. and Barbara R. MacRoberts, Social Organization and Behavior of the Acorn Woodpecker in Central Coastal California, re- viewed, 652-654 Macaw, see Ara spp. Magpies, Australian, see Gymnorhina spp. Maher. \^’illiam J,, review by, 641-642 Malaclemys terrapin, 350 Mallard, see Anas platyrhynchos Malurus cyaneus, 495 management recommendations Pithecophaga jeffeiyi, 18 Manakin, Red-capped, see Pipra mentalis Manning. T. H, and Brenda Carter. Inci- dence of runt eggs in the Canada Goose and Semipalmated Sandpiper, 469 Manorina spp., 495 Marabou, see Leptoptilos crumeniferus Mares, Michael A., reviews by, 359, 499- 501, 644 Margarops. 341 Marion, ayne R., Growth and develop- ment of the Plain Chachalaca in south Texas, 47-56 December 1977 • INDEX TO VOLUME 89 669 Martin, Purple, see Progne subis Maurer, David R., reviews by, 498-499, 502 Maxson, Stephen J., Activity patterns of female Ruffed Grouse during the breeding season, 439-455 Mayr, Ernst, Evolution and the Diversity of Life: Selected Essays, reviewed, 352 Meadowlark, Eastern, see Sturnella magna Medway, Lord and David R. Wells, The Birds of the Malay Peninsula, re- viewed, 497-498 Melanerpes, 622 aurifrons, 322 carolinus, 122-129, 164, 231, 424, 432 chrysauchen, 318, 322 erythrocephalus, 122-129, 164, 424 formicivorus, 150-151, 322, 652-654 rubricapillus, 322 superciliaris, 637 Melanitta deglandi, 331 Melospiza georgiana, 231, 427 lincolnii, 427 melodia, 173, 231, 253-265, 427, 543-561 Membership Committee Report, 510 Meng, Heinz, see Braun, Clait E., James H. Enderson, Charles J. Henny, — , and Alva G. Nye, Jr.; Kaufmann, John and Merganser, Common, see Mergus merganser Hooded, see Lophodytes cucullatus Red-breasted, see Mergus serrator Mergus merganser, 532-542 serrator, 532-542 Mexico, 171-173, 571 Micropalama himantopus, 167-171 Microtus pennsylvanicus, 624^625 migration, 166-167 Capella gallinago, 119-120 shorebirds, 167-171 tower kills, 291-299 warblers, 456-465 Mills, G. Scott, American Kestrel rejects captured spadefoot toad, 623-624; see Silliman, James, , and Stephen Alden Mimus polyglottos, 480, 637 Miners, see Manorina spp. Mirounga, 621 Mniotilta varia, 231, 425, 457-458, 466 Mock, Douglas W., review by, 639 641 Mockingbird, see Mimus polyglottos Molothrus ater, 21-32, 253-265, 487, 543- 561, 584 molt Brachyramphus brevirostris, 467-469 Caprimulgus carolinensis, 165-166 Ortalis vetula, 54-55 Monareba spp., 193 Moriarty, David J., Flocking and foraging in the Scarlet-rumped Tanager, 151- 153 morphological variation, see Pinicola enu- cleator, 380-395 Morrison, Michael L. and R. Douglas Slack, Flocking and foraging behavior of Brown Jays in northeastern Mexico, 171-173 mortality, tower kills, 291-299, 422-433 Mountain-gem, Purple-throated, see Lam- pornis calolaema White-bellied, see Lampornis hemileucus Murrelet, Kittlitz’s, see Brachyramphus brevirostris Marbled, see Brachyramphus marmoratus Muscicapa hypoleuca, see Ficedula hypo- leuca muskrat, see Ondatra zibethica Mutton-birds, Tasmanian, see Puffinus tenuirostris Mycteria americana, 97, 343 Myiarchus cinerascens, 253-265 crinitus, 226, 231, 424, 543-561 Myiopsitta monachus, 346-349 Myres, M. Timothy, see Sadler, Thomas S. and Neanis, 357 nest, abandonment Dendrocygna autumnalis, 621-622 building Chasiempis sandwichensis, 199-202 Hirundo rustica, 562-571 Loxops virens, 480-483 Passer domesticus, 302-303 Petrochelidon pyrrhonota, 562-571 Podiceps grisegena, .35-37 Sphyrapicus varius, 310-312 sanitation Chasiempis sandwichensis, 204-206 Sphyrapicus varius, 318 670 THE WILSON BULLETIN • VoL 89, No. 4 sites Agelaius phoeniceus, 396-403 Branta canadensis, 525-530 Capella gallinago, 118 Chasiempis sandwichensis, 199-202 Colaptes auratus, 122-129 Corvus corax, 233-242 Dendroica petechia, 153-154 Dryocopus pileatus, 122-129 Empidonax minimus, 153-154 Hirundo rustica, 632-635 Ictinia mississippiensis, 176 Larus delawarensis, 473-475 Melanerpes carolinus, 122-129, 164- 165 M. erythrocephalus, 122-129, 164—165 Myiopsitta monachus, 346-349 Parabuteo unicinctus, 469-470 Petrochelidon pyrrhonota, 286 Picoides pubescens, 122-129 P. villosus, 122-129 Podiceps grisegena, 35-37 Rissa tridactyla, 616-618 Sayornis phoebe, 632-635 Sterna hirundo, 475-476 nesting failure Anas acuta, 154-157 Corvus corax, 235 Pelecanus occidentalis, 157-158 Sterna fuscata, 157-158 S. hirundo, 157-158 habitat Agelaius phoeniceus, 397 Branta canadensis, 523-531 breeding bird surveys, and, 543-561 Cathartes aura, 328-329 Chasiempis sandwichensis, 197-199 Coragyps atratus, 328-329 Corvus corax, 233-242 Hirundo rustica, 562-571, 632-635 Parabuteo unicinctus, 469-470 Petrochelidon pyrrhonota, 562-571 Sayornis phoebe, 632-635 \ ermivora bachmanii, 373-379 materials Hirundo rustica, 562-571 Petrochelidon pyrrhonota, 562-571 phenology Agelaius phoeniceus, 74 Xanthocephalus xanthocephalus, 74 success, 173-175 x-Vgelaius phoeniceus, 397-398 Auriparus flaviceps, 576-580 Chasiempis sandwichensis, 206-207 Corvus corax, 234-235 Larus delawarensis, 473-475 Nycticorax nycticorax, 351 Passer domesticus, 307 Petrochelidon pyrrhonota, 287-289 nestling mortality Agelaius phoeniceus, 399-400 Auriparus flaviceps, 579-580 Ictinia mississippiensis, 176 Petrochelidon pyrrhonota, 288-289 Numenius phaeopus, 167-171 Nuthatch, Red-breasted, see Sitta canaden- sis White-breasted, see Sitta carolinensis Nycticorax nycticorax, 104, 158, 350-351, 538 Nye, Alva G. Jr., see Braun, Clait E., James H. Enderson, Charles J. Henny, Heinz Meng, and officers, new, 512 Ohmart, R. D, and R. E. Tomlinson, Foods of western Clapper Rails, 332-336 Oldsquaw, see Clangula hyemalis Olson, Storrs L., ed.. Collected Papers in Avian Paleontology Honoring the 90th Birthday of j\lexander Wet- more, reviewed, 355-358 Ondatra zibethica, 45, 524 O’Neill, John P., painting by, frontispiece facing 1 Oporornis agilis, 426 Philadelphia, 426 tolmiei, 253-265 Oretrochilus estella, 352 Oriole, Black-cowled, see Icterus domini- censis northropi Hooded, see I. cucullatus Northern, see I. galbula Orr, Donald E., see Ryder, John P., , and Ghomi H. Saedi Ortalis vetula, 47-56 Ortego, Brent, see Pratt, J. Douglas. , and Harland D. Guillory December 1977 • INDEX TO VOLUME 89 671 Orthorhyncus, 341 Oryzomys, 610 Osprey, see Pandion haliaetus Ostrom, John, John Ostrom’s Studies on Archaeopteryx, The Origin of Birds, and the Evolution of Avian Flight, reviewed, 488-492 Otus asio, 609 choliba, 609-612 flammeolus, 609 trichopsis, 609 Ovenbird, see Seiurus aurocapillus Owl, Barn, see Tyto alba Barred, see Strix varia Great Horned, see Bubo virginianus Short-eared, see Asio flammeus Tropical Screech, see Otus choliba Palmer, Ralph S., ed.. Handbook of North American Birds, vols. 2 & 3 (Water- fowl), reviewed, 179-180 Panama, 328-329 Pandion haliaetus, 149-150, 364, 625 Panterpe insignis, 161, 615 Parabuteo unicinctus, 469, 618-619 Parakeet, Monk, see Myiopsitta monachus parasitism, brood, 21-32 Parker, James W., Mortality of nestling Mississippi Kites by ants, 176 Parkes, Kenneth C., reviews by, 189-191, 638-639 Parmalee, W., Avian bone pathologies from Ankara sites in South Dakota, 628- 632 Parmelee, David F., review by, 646-648 Parrot, Cuban, see Amazona leucocephala bahamensis Parula americana, 231, 425, 466 Parula, Northern, see Parula americana Parus atricapillus, 231, 543-561 bicolor, 231 caeruleus, 412, 580 major, 173, 223, 315 Pasquier, Roger F,, Herring Gull eating bayberry, 338; Watching Birds, re- viewed, 644 Passer domesticus, 22, 286-290, 300-309, 340, 477, 543-561 Passerculus sandwichensis, 231, 427, 486, 543-561 Passerherbulus henslowii, see Ammodramus henslowii Passerina amoena, 130-148, 253-265 brissonii, 130-148 caerulescens, 130-148 ciris, 130-148 cyanea, 130-148, 232, 253-265, 427, 543- 561 cyanoides, 130-148 glaucocaerulea, 130-148 leclancherii, 130-148 parellina, 130-148, 345 rositae, 130-148 versicolor, 130-148 Paul, Richard T., see King, Kirke A„ David R. Blankenship, , and Robin C. A. Rice Pavo, 621 Paynter, Raymond A., Jr. and Melvin A. Traylor, Jr., Ornithological Gazeteer of Ecuador, reviewed, 638-639 Peaker, Malcolm, ed.. Avian Physiology: Symposia of the Zoological Society of London, No. 35, reviewed, 183- 189 Pedioecetes phasianellus, 336, 522 Pelecanus occidentalis, 157-158, 416, 474, 593, 639-641 Pelican, Brown, see Pelecanus occidentalis Penguin, Adelie, see Pygoscelis adeliae Peregrine, see Falco peregrinus Periporphyrus erythromelas, 130-148 Peterson, Steven R. and Robert S. Ellarson, Food habits of Oldsquaws wintering on Lake Michigan, 81-91 Petrochelidon pyrrhonota, 286-290, 543- 561, 562-571 Pewee, Eastern Wood, see Contopus virens Western Wood, see Contopus sordidulus Pettingill, Olin Sewall, Jr., A Guide to Bird Finding East of the Mississippi, 2nd ed., reviewed, 644-645 Phaethornis guy, 160, 613 longuemareus, 160 Phainopepla, see Phainopepla nitens Phainopepla nitens, 253-265 Phalacrocorax aristotelis, 473 Pharomachrus mocinno, 621 Phasianus colchicus, 55, 452 672 THE WILSON BULLETIN • Vol. 89, No. 4 Pheasant, Ring-necked, see Phasianus col- chicus phenetic analysis Cardinalinae, 130-148 phenograms Cardinalinae, 130-148 Pheucticus aureoventris, 130-148 chrysopeplus, 130-148 ludovicianus, 130-148, 214, 231, 426, 543- 561 melanocephalus, 130-148, 253-265 Philippines, 1-20 Philodice bryantae, 162 Philohela minor, 543-561 Phoebe, Black, see Sayornis nigricans Eastern, see Sayornis phoebe Phoenicopterus ruber, 637 Picoides arizonae, 485 borealis, 164 major, 322 pubescens, 122-129, 231, 311 scalaris, 485 tridactylus, 321 villosus, 122-129, 231, 322, 622 Pigeon, Passenger, see Ectopistes migra- torius Short-billed, see Colurnba nigrirostris Wood, see Colurnba palumbus Pinicola enucleator, 380-395 e. alascensis, 393 e. californica, 392 e. carlottae, 392 e. esebatosus, 388 e. flammula, 392 e. leucura, 388, 392 e. montana, 392 Pinkowski, Benedict C., Foraging behavior of the Eastern Bluebird, 404-414 Pintail, see Anas acuta Pipilo aberti, 253-265 erythrophthalmus 232, 253-265 Pipra mentalis, 152 Piranga olivacea, 231, 426 rubra, 253-263 Pithecophaga jefferyi, 1-20, colorplate fac- ing 1 Pitylus grossus, 130-148 Platalea, 343 Plegadis chihi, 97, 99-106 falcinellus, 97, 343 Plover, American Golden, see Pluvialis dominica Black-bellied, see Pluvialis squatarola Semipalmated, see Charadrius semipal- matus Plumage, 620-621 Ortalis vetula, 54-55 Tympanuchus pallidicinctus, 521-522 Pluvialis dominica, 167-171 squatarola, 167-171, 470-472 Pluvianellus socialis, 645-646 Podiceps auritus, 34 grisegena, 33-46 nigricollis, 34 Podilymbus podiceps, 231, 424 Polioptila caerulea, 231, 253-265 Ponshair, James E., Auditor’s Report, 512 Pooecetes gramineus, 231, 486, 543-561 population age composition, 215-216 Carpodacus cassinii, 58-59 Lagopus leucurus, 108-109 estimates breeding bird survey, 549-561 Capella gallinago, 116-121 Carpodacus cassinii, 58-59 Dendrocygna bicolor, 329-331 Pithecophaga jefferyi, 12-17 Porphyrospiza caerulescens, 130-148 Porzana Carolina, 424 Prairie Chicken, Greater, see Tympanuchus cupido Lesser, see Tympanuchus pallidicinctus Pratt, H. Douglas, Brent Ortego, and Har- land D. Guillory, Spread of the Great-tailed Grackle in southwestern Louisiana, 483-485; painting by, frontispiece facing 193 predation on Larus argentatus, 472-473 Nycticorax nycticorax, 350-351 Vireo bellii, 349-350 President’s Page, 505 Prince, Harold H., see Kaminski, Richard M. and Proctor, Noble S., Osprey catches vole, 625 Progne subis, 177, 543-561 December 1977 • INDEX TO VOLUME 89 673 Protonotaria citrea, 231, 425 Protornis, 357 Psilorhinus morio, 171-173 Ptarmigan, Rock, see Lagopus mutus White-tailed, see Lagopus leucurus Willow, see Lagopus lagopus Puffinus tenuirostris, 495 Pygoscelis adeliae, 473 Pyrrhula pyrrhula, 57 Pyrrhuloxia, see Pyrrhuloxia sinuatus Pyrrhuloxia sinuatus, 130-148 Quail, Coturnix, see Coturnix japonica Gambel’s, see Lophortyx gambelii Quay, Thomas L., see Grant, Gilbert S. and Quetzal, see Pharomachrus mocinno Quiscalus major, 483 mexicanus, 483, 602-608 quiscula, 28, 543-561 racer, black, see Coluber constrictor Radke, Eleanor L., and John Klimosewski, Late fledging date for Harris’ Hawk, 469-470 radiotelemetry Bonasa umbellus, 439-455 Raffaele, Herbert A. and Daniel Roby, The Lesser Antillean Bullfinch in the Virgin Islands, 338-342 Raikow, Robert J., reviews by, 192, 352, 355-358, 359, 493-494, 644-645 Rail, California Clapper, see Rallus longi- rostris obsoletus Clapper, see Rallus longirostris Georgia Clapper, see Rallus longirostris crepitans, R. 1. waynei Virginia, see Rallus limicola Yellow, see Coturnicops noveboracensis Rallus limicola, 231, 424 longirostris, 332-336, 350 1. crepitans, 335 1. nayaritensis, 332-336 1. obsoletus, 335 1. rhizophorae, 332-336 1. waynei, 335 1. yumanensis, 332-336 Ramphocelus passerinii, 151-153 range expansion of Quiscalus mexicanus, 483-485, 602-608 rat, rice, see Oryzomys Raven, Australian, see Corvus coronoides Common, see Corvus corax White-necked, see Corvus cryptoleucus Rea, Amadeo M., see Wilson, Michael and Redshank, see Tringa totanus Redstart, American, see Setophaga ruticilla Regulus calendula, 166-167, 425, 485 satrapa, 223, 425 Rheinardia, 621 Rhodothraupis celaeno, 130-148 Rhynchops niger, 158 Rice, Robin C. A., see King, Kirke A., David R. Blankenship, Richard T. Paul, and Ridgely, Robert S., A Guide to the Birds of Panama, reviewed, 192 Riparia riparia, 543-561 Rissa tridactyla, 243, 473, 616-618 Robertson, Raleigh J., see Weatherhead, Patrick J. and Robin, American, see Turdus migratorius Roby, Daniel, see Raffaele, Herbert A. and Rodgers, James A., Jr., Breeding displays of the Louisiana Heron, 266-285 Rohwer, Sievert, Ground foraging and rapid molt in the Chuck-will’s-widow, 165- 166 Rothstein, Stephen L, Cowbird parasitism and egg recognition of the Northern Oriole, 21-32 Rowley, Ian, Bird Life, reviewed, 495-496 Ryder, John P. Donald E. Orr, and Ghomi H. Saedi, Egg quality in relation to nest location in Ring-billed Gulls, 473-475; and Lynn Somppi, Growth and development of known- age Ring-billed Gull embryos, 243- 252; see Johnson, Bruce R. and Sabrewing, Violet, see Campylopterus hemi- leucurus Sadler, Thomas S. and M. Timothy Myres, Alberta Birds, 1961-1970, with Par- ticular Reference to Migration, re- viewed, 641-642 674 THE WILSON BULLETIN • Vol. 89, No. 4 Saedi, Ghomi H., see Ryder, John T., Donald E. Orr, and Salpinctes obsoletus, 253-265 Saltator spp, (albicollis, atriceps, atricollis, atripennis, aurantiirostris, cinctus, coerulescens, maximus, maxillosus, orenocensis, rufiventris, similis), 130-148 Samson, Fred B., Social dominance in winter flocks of Cassin’s Finch, 57- 66 Sanderling, see Calidris alba Sandpiper, Baird’s, see Calidris bairdii Least, see Calidris minutilla Pectoral, see Calidris melanotos Semipalmated, see Calidris pusillus Solitary, see Tringa solitaria Spotted, see Actitis macularia Stilt, see Micropalama bimantopus Upland, see Bartramia longicauda bite-rumped, see Calidris fuscicollis Sapphire, Blue-cbinned, see Cblorestes no- tatus Sappington, James N., Breeding biology of House Sparrows in north Mississippi, 300-309 Sapsucker, Yellow'-bellied, see Sphyrapicus varius illiamson’s, see Sphyrapicus thyroideus Sayornis nigricans, 253-265 pboebe, 226, 632-635 Scapbiopus bammondi, 623-624 Scbarf. \\ illiam C., see Sbugart, Gary ^ . and Schmitt. Carl Gregory, Summer Birds of the San Juan Valley, New Mexico, reviewed, 645 Scbnell, Gary D., see Hellack, Jenna J. and Schreiber, Ralph W'., Maintenance Behavior and Communication in the Brown Pelican, reviewed, 639-641 Scoter, White-winged, see Melanitta de- glandi Sealy, Spencer G., Wing molt of the Kittlitz’s Murrelet, 467-469 Seedeater, Variable, see Sporophila aurita Seets, James W. and H. David Bohlen, Comparative mortality of birds at television towers in central Illinois, 422-433 segregation, sexual Lagopus lagopus, 109 L. leucurus, 108-110 L. mutus, 109 Seiurus aurocapillus, 166, 231, 426, 458, 466 noveboracensis, 426, 466 Selasphorus scintilla, 163 Sericotes, 341 Setophaga ruticilla, 166-167, 231, 426, 461, 466 sex determination Eudocimus albus, 92-98 Ortalis vetula, 48-51 sex ratio Carpodacus cassinii, 58-59 Lagopus leucurus, 108-109 Passer domesticus, 302 sexual dimorphism, 619-621 Chlorostilbon canivettii, 160 Eudocimus albus, 92-98 Pinicola enucleator, 385-387 Shag, see Phalacrocorax aristotelis Shrike, Loggerhead, see Lanius ludovici- anus Northern, see Lanius excubitor Sbugart, Gary W. and William C. Scharf, Predation and dispersion of Herring Gull nests, 472-473 Sialia currucoides, 407 sialis, 232, 404-414 Silliman, James, G. Scott Mills, and Ste- phen Alden, Effect of flock size on foraging activity in wintering San- derlings, 434-438 Simpson, George Gaylord, Penguins, Past and Present, Here and There, re- viewed, 648 Sitta canadensis, 424 carolinensis, 231 Skimmer, Black, see Rhynchops niger Skutch, Alexander F., Parent Birds and Their Young, reviewed, 649 Slack, R. Douglas, see Morrison, Michael L. and Sloan, Norman F., new life member, 210 Slud, Paul, review by, 183-189; Geographic December 1977 • INDEX TO VOLUME 89 675 and Climatic Relationships of Avi- faunas with Special Reference to Comparative Distribution in the Neo- tropics, reviewed, 499-501 Smith, R. W., see Diamond, A. W., P. Lack, and snake black rat, see Elaphe obsoleta red-sided garter, see Thamnophis sirtalis parietalis Snipe, Common, see Capella gallinago Snow, Barbara K., Feeding behavior of two hummingbirds in a Costa Rican mon- tane forest, 613-616 social dominance Carpodacus cassinii, 59-60 Somppi, Lynn, see Ryder, John P. and song patterns Chasiempis sandwichensis, 195-197 Pipilo erythrophthalmus, 477-480 Songlark, Australian Brown, see Cinclo- rhamphus cruralis, 619-620 Sora, see Porzana Carolina Sparrow, Chipping, see Spizella passerina Field, see Spizella pusilla Grasshopper, see Ammodramus savan- narum Henslow’s, see Ammodramus henslowii House, see Passer domesticus Lark, see Chondestes grammacus Lincoln’s, see Melospiza lincolnii Savannah, see Passerculus sandwichensis Sharp-tailed, see Ammospiza caudacuta Song, see Melospiza melodia Swamp, see Melospiza georgiana Vesper, see Pooecetes gramineus White-crowned, see Zonotrichia leuco- phrys White-throated, see Zonotrichia albicollis Sparrowhawk, see Accipiter nisus Sphyrapicus thyroideus, 622-623 varius, 310-324, 424, 622 V. nuchalis, 622 V. varius, 622 Spinus psaltria, see Carduelis psaltria Spiza americana, 130-148, 222, 224, 231, 349, 427 Spizella passerina, 232, 427, 543-561 pusilla, 173, 231, 297, 427, 432, 479, 543- 561, 625-627 Spofford, Sally Hoyt, Poplar leaf-stem gall insects as food for warblers and woodpeckers, 485; review by, 654 Spoonbill, Roseate, see Ajaia ajaja Sporophila aurita, 152 Springer, Paul F., see Avery, Michael, , and J. Frank Cassel squirrel, flying, see Glaucomys volans Staey, Peter B. and Roxana Jansma, Storage of pihon nuts by the Acorn Wood- pecker in New Mexico, 150-151 Stahlecker, Dale W. and Herman J. Griese, Evidenee of double brooding by American Kestrels in the Colorado high plains, 618-619 Starling, see Sturnus vulgaris Starthroat, Plain-capped, see Heliomaster constantii Stelgidopteryx ruficollis, 410, 580 Sterna alhifrons, 475, 613 a. antillarum, 613 a. hrowni, 613 fuscata, 157 hirundo, 157-158, 475, 599 maxima, 416 sandvicensis, 416 Stinson, Christopher H., The spatial distri- bution of wintering Black-bellied Plovers, 470-472 Stork, American Wood, see Mycteria amer- icana Streptopelia risoria, 593-601 Strix varia, 231 Sturkie, Paul B., ed.. Avian Physiology, 3rd ed., reviewed, 496-497 Sturnella magna, 212, 231, 486, 543-561 Sturnus vulgaris, 165, 469, 487, 543-561 Sutton, George Miksch, The Lesser Prairie Chicken’s inflatable neck sacs, 521- 522 Svensson, Lars, Identification Guide to European Passerines, 2nd (revised) ed., reviewed, 191-192 Swallow, Bank, see Riparia riparia Barn, see Hirundo rustica Cliff, see Petroehelidon pyrrhonota 676 THE WILSON BULLETIN • Vo/. 89, No. 4 Rough-winged, see Stelgidopteryx ruficol- lis Tree, see Iridoprocne bicolor Swander, Lynda, see Brewer, Richard and Swierczewski, Edward V., review Ijy, 502- 503 Swift, Chimney, see Chaetura pelagica Sylvilagus auduboni, 470 Syntarus, 489 Tachyphonus luctosus, 152 Tanager, Masked, see Tangara larvata Scarlet, see Piranga olivacea Scarlet-rumped, see Ramphocelus pas- serinii Summer, see Piranga rubra White-shouldered, see Tachyphonus luc- tosus Tangara larvata, 152 nigroviridis consobrina, 639 taxonomy, numerical Cardinalinae, 130-148 Telmatodytes palustris, 231, 424 Tern, Black, see Chilidonias niger Common, see Sterna hirundo Least, see Sterna albifrons Royal, see Sterna maxima Sandwich, see Sterna sandvicensis Sooty, see Sterna fuscata territoriality Agelaius phoeniceus, 585-589 Chasiempis sandwichensis, 194-195 Corvus corax, 236-237 Philodice hryantae, 159-160 Pluvialis squatarola, 470-472 Podiceps grisegena, 44-45 Tetrao urogallus, 112, 452, 620 Thamnophis sirtalis parietalis, 349 Theherge, John B., see Weber, Wayne C. and — - Thielcke, Gerhard A., Bird Sounds, re- viewed, 650-651 Thomas, Betsy Trent, Tropical Screech Owl nest defense behavior and nestling growth rate, 609-612 Thompson, Max C., see Hunt, George L. and Thompson, William L., review by, 651-652 Thrasher, Brown, see Toxostoma rufum Threskiornis aethiopica, 97 molucca, 342-345 Thrush, Gray-cheeked, see Catharus mini- mus Hermit, see C. guttatus Swainson’s see C. ustulatus Wood, see Hylocichla mustelina Thryomanes bewickii, 253-265 Thryothorus ludovicianus, 231, 477 Tit, Blue, see Parus caeruleus Great, see P. major Titmouse, Tufted, see Parus bicolor Tityra, Black-crowned, see Tityra inquisitor Masked, see T. semifasciata Tityra inquisitor, 152 semifasciata, 152 toad, western spadefoot, see Scaphiopus hammondi Tomlinson, R. E., see Ohmart, R. D. and Towhee, Albert’s, see Pipilo aberti Rufous-sided, see P. erythrophthalmus Toxostoma rufum, 231, 424, 543-561 trachea Ortalis vetula, 48-51 Trainer, Elliot J. and Flora E. Tramer, Feeding responses of fall migrants to prolonged inclement weather, 166- 167 Tramer, Flora E., see Tramer, Elliot J. and Traylor, Melvin A. Jr., see Paynter, Ray- mond A. Jr. and Treasurer’s Report, 507-509 Tringa flavipes, 167-171 melanoleuca, 167-171 solitaria, 167-171 totanus, 437 Troglodytes aedon, 232, 253-265, 424, 543- 561 troglodytes, 424 Turdus migratorius, 232, 487, 543-561 Turnstone, Ruddy, see Arenaria interpres turtle, diamondback, see Malaclemys terra- pin Tympanuchus cupido, 522 pallidicinctus, 521-522, colorplate facing 521 Tyrannus spp., 407 December 1977 • INDEX TO VOLUME 89 677 tyrannus, 231, 487, 543-561 verticalis, 29, 253-265 Tyto alba, 303 Uintornis, 357 Uropsila leucogastra, 345 Veery, see Catharus fuscescens Verbeek, Nicolaas A. M., Comparative feed- ing behavior of immature and adult Herring Gulls, 415-421 Verdin, see Auriparus flaviceps Vermivora bachmanii, 373-379, colorplate facing 373 celata, 253-265, 425 chrysoptera, 425 luciae, 253-265 peregrina, 166, 425, 466 pinus, 231 ruficapilla, 425 Violet-ear, Green, see Colibri thalassinus Brown, see Colibri delphinae Vireo, Bell’s, see Vireo bellii Philadelphia, see V. pliiladelphicus Red-eyed, see V. olivaceus Solitary, see V. solitarius Warbling, see V. gilvus White-eyed, see V, griseus Yellow-green, see V. flavoviridis Yellow-throated, see V. flavifrons Vireo bellii, 349 flavifrons, 425 flavoviridis, 345 gilvus, 232, 253-265, 543-561 griseus, 231 olivaceus, 212, 231, 425, 543-561 philadelphicus, 425 solitarius, 425 Virgin Islands, 338-342 vocalization Chasiempis sandwichensis, 195-197 Pipilo erythrophthalmus, 477-480 Sphyrapicus varius, 316-318 vole, meadow, see Microtus pennslyvanicus Vulpes vulpes, 472 Vulture, Black, see Coragyps atratus Yellow-headed, see Cathartes burrovianus Waide, Robert B. and Jack P. Hailman, Birds of five families feeding from spider webs, 345-346 Warbler, Bachman’s, see Vermivora baeli- manii Bay-breasted, see Dendroica castanea Black and White, see Mniotilta varia Blackburnian, see Dendroica fusca Blackpoll, see Dendroica striata Black-throated Blue, see Dendroica ca- erulescens Black-throated Green, see Dendroica vi- rens Blue- winged, see Vermivora pinus Canada, see Wilsonia canadensis Cape May, see Dendroica tigrina Cerulean, see Dendroica cerulea Chestnut-sided, see Dendroica pensyl- vanica Connecticut, see Oporornis agilis Golden-sided, see Vermivora chrysoptera Hooded, see Wilsonia citrina Lucy’s, see Vermivora luciae MacGillivray’s, Oporornis tolmiei Magnolia, see Dendroica magnolia Mourning, see Oporornis Philadelphia Nashville, see Vermivora ruficapilla Orange-crowned, see Vermivora celata Palm, see Dendroica palmarum Pine, see Dendroica pinus Prairie, see Dendroica discolor Prothonotary, see Protonotaria citrea Reed, see Acrocephalus scirpaceus Sedge, see Acrocephalus schcenohaenus Swainson’s, see Limnothlypis swains^nii Tennessee, see Vermivora peregrina Worm-eating, see Helmitheros vermivora Yellow, see Dendroica petechia Yellow-rumped, see Dendroica coronata Yellow-throated, see Dendroica dominica Waxwing, Cedar, see Bombycilla cedrorum weather, effects on breeding bird surveys, 555-558 foraging, 166-167 Sialia sialis, 408-409 mortality, 291-299, 422-433 nesting Anas acuta, 154-157 Bonasa umbellus, 449 Iridoprocne bicolor, 289 678 THE ILSON BULLETIN • Vol. 89, \o. 4 Petrochelidon pyrrhonota, 289 Sterna hirundo, 475 W eatherhead. Patrick J. and Raleigh J. Robertson. Male behavior and female recruitment in the Red-^vinged Blackbird, 583-592 \^’eathers, esley W., see Caccamise, Don- ald F. and eber. ayne C. and John B. Theberge, Breeding bird survey counts as re- lated to habitat and date, 543-561 eeks, Harmon P., Nest reciprocity in Eastern Phoebes and Barn S^vallows, 632-635 weight Eudocimus albus, 92-94 migrant warblers, 456-465 Ortalis vetula. 51-54 \^’ells, David R., see Medway, Lord and \^’himbrel. see Numenius phaeopus \^'hip-poor-will, see Caprimulgus vociferus Whistling-duck. Black-bellied, see Dendro- cygna autumnalis Fulvous, see D. bicolor bite. Donald H. and Eugene Cromartie. Residues of environmental pollutants and shell thinning in Merganser eggs. 532-542 liite-Schuler. .Susan C.. review by. 649 \^’hitmore. Robert C., Habitat partitioning in a community of passerine birds. 253-265 liittow. G. C. and A. J. Berger, Heat loss from the nest of the Hawaiian Honeycreeper, “Amakihi,*’ 480-483 hydah. African Long-tailed, see Euplectes progne delamerei ickstrom, George M., Auditor’s Report. 512 illet, see Catoptrophorus semipalmatus \^'ilson. Michael and Amadeo M. Rea, Late Pleistocene illiamson’s Sapsucker from \^’yoming, 622-623 \^'ilsonia canadensis, 426 citrina. 212, 231 pusilla, 253-265, 426 oodcock, American, see Philohela minor oodpecker. Acorn, see Melanerpes for- mic ivor us Arizona, see Picoides arizonae Downy, see Picoides pubescens Golden-fronted, see Melanerpes aurifrons Golden-naped, see Melanerpes chrysau- chen Great Spotted, see Picoides major Ladder-backed, see Picoides scalaris Pileated, see Dryocopus pileatus Red-bellied, see Melanerpes carolinus Red-cockaded, see Picoides borealis Red-crowned, see Melanerpes rubrieapil- lus Red-headed, see Melanerpes ei^throce- phalus Three-toed, see Picoides tridactylus est Indian Red-bellied, see Melanerpes superciliaris \^'ood-star. Magenta-throated, see Philodice bryantae ren. Bewick's, see Thryomanes bewickii Cactus, see Campylorhynchus brunneica- pillus Carolina, see Thryothorus ludovicianus House, see Troglodytes aedon Long-billed Marsh, see Telmatcdytes pa- lustris Rock, see Salpinctes obsoletus Short-billed Marsh, see Cistothorus pla- tensis Superb Blue, see Malurus cyaneus hite-bellied, see Uropsila leucogastra inter, s:e Troglodytes troglodytes Xanthocephalus xanthccephalus, 73-80, 253-265 Yellowlegs, Greater, see Tringa melano- leucas Lesser, see Tringa flavipes Yellowthroat. Bahama, see Geothlypis ro- strata Common, see Geothlypis trichas Zeleny, Lawrence, The Bluebird: How You Can Help Its Fight for Survival, re- viewed, 502-503 Zenaida macroura, 231, 487, 543-561 Zernickow, Hubert P., new life member, 46 Zonotrichia albicollis, 427 leucophrys, 88, 173, 427 1. nuttalli, 651-652 Zosterops japonica, 208 Zusi, Richard L., review by, 648 This issue of The Wilson Bulletin was published on 27 December 1977. The Wilson Bulletin Editor Jerome A. Jackson Department of Zoology P.O. Drawer Z Mississippi State University Mississippi State, MS 39762 Editorial Assistants Bette J, Schardien Patricia Ramey Bonnie J. Turner Martha Hays Wayne C. Weber Review Editor Robert Raikow Color Plate Editor William A. Lunk Department of Life Sciences 865 North Wagner Road University of Pittsburgh Ann Arbor, MI 48103 Pittsburgh, PA 15213 Suggestions to Authors See Wilson Bulletin, 87:144, 1975 for more detailed “Suggestions to Authors.” Manuscripts intended for publication in The Wilson Bulletin should be submitted in dupli- cate, neatly typewritten, double-spaced, with at least 3 cm margins, and on one side only of good quality white paper. Do not submit xerographic copies that are made on slick, heavy paper. Tables should be typed on separate sheets, and should be narrow and deep rather than wide and shallow. Follow the AOU Check-list (Fifth Edition, 1957) and the 32nd Supplement (Auk, 90:411-419, 1973), insofar as scientific names of U.S. and Canadian birds are concerned. Summaries of major papers should be brief but quotable. Where fewer than 5 papers are cited, the citations may be included in the text. All citations in “General Notes” should be included in the text. Follow carefully the style used in this issue in listing the literature cited ; otherwise, follow the “CBE Style Manual” (1972, AIBS). Photographs for illustrations should have good contrast and be on gloss paper. Submit prints unmounted and attach to each a brief but adequate legend. Do not write heavily on the backs of photographs. Diagrams and line drawings should be in black ink and their lettering large enough to permit reduction. Original figures or photographs submitted must be smaller than 22 X 28 cm. Alterations in copy after the type has been set must be charged to the author. Notice of Change of Address If your address changes, notify the Society immediately. Send your complete new address to the Treasurer, Ernest E. Hoover, 1044 Webster St., N.W., Grand Rapids, Michigan 49504. He will notify the printer. The permanent mailing address of the Wilson Ornithological Society is: c/o The Museum of Zoology, The University of Michigan, Ann Arbor, Michigan 48104. Persons having business with any of the officers may address them at their various addresses given on the back of the front cover, and all matters pertaining to the Bulletin should be sent directly to the Editor. CONTENTS THE LESSER PRAIRIE CHICKEN’s INFLATABLE NECK SACS George Miksch SuttOJl 521 NESTING HABITAT OF CANADA GEESE IN SOUTHEASTERN MICHIGAN Richard M. Kaminski and Harold H. Prince 523 RESIDUES OF ENVIRONMENTAL POLLUTANTS AND SHELL THINNING IN MERGANSER EGGS Donald H. White and Eugene Cromartie 532 BREEDING BIRD SURVEY COUNTS AS RELATED TO HABITAT AND DATE Wayne C. Weber and John B. Theberge 543 ANALYSIS OF MATERIALS IN CLIFF AND BARN SWALLOW NESTS: RELATIONSHIP BETWEEN MUD SELECTION AND NEST ARCHITECTURE „ Delbert L. KUgore, Jr. and Kathy L. Knudsen 562 PRODUCTION AND SURVIVAL OF THE VERDIN George T. Austin 572 MALE BEHAVIOR AND FEMALE RECRUITMENT IN THE RED-WINGED BLACKBIRD Patrick J. W eatherhead and Raleigh J. Robertson 583 REDUCTION OF COURTSHIP BEHAVIOR INDUCED BY DDE IN MALE RINGED TURTLE DOVES M. A. Haegele and Rick H. Hudson 593 MOVEMENTS OF THE GREAT-TAILED GRACKLE IN TEXAS „ Keith A. Arnold and Leon J. Folse, Jr. 602 GENERAL NOTES TROPICAL SCREECH OWL NEST DEFENSE BEHAVIOR AND NESTLING GROWTH RATE Betsy Trent Thomas 609 THREE MORE NEW SPECIMEN RECORDS FOR GUATEMALA Robert W. Dickerman 612 FEEDING BEHAVIOR OF TWO HUMMINGBIRDS IN A COSTA RICAN MONTANE FOREST Barbara K. Snow 613 BLACK-LEGGED KITTIWAKES NESTING ON SNOWBANK George L. Hunt, Jr. and Max C. Thompson 616 EVIDENCE OF DOUBLE BROODING BY AMERICAN KESTRELS IN THE COLORADO HIGH PLAINS Dale W. Stahlecker and Herman J. Griese 618 FURTHER COMMENTS ON SEXUAL SIZE DIMORPHISM IN BIRDS D. Amadon 619 RESPONSE OF INCUBATING BLACK-BELLIED WHISTLING-DUCKS TO LOSS OF MATES Richard E. McCamant and Eric G. Bolen 621 LATE PLEISTOCENE WILLIAMSON’s SAPSUCKER FROM WYOMING Michael Wilson and Amadeo M. Rea 622 AMERICAN KESTREL REJECTS CAPTURED SPADEFOOT TOAD G. ScOtt MUls 623 WINTER DISTRIBUTION OF RED-TAILED HAWKS IN CENTRAL NEW YORK STATE .. Jonathan Bart 623 OSPREY CATCHES VOLE Noble S. Proctor 625 PATTERNS OF FEEDING FIELD SPARROW YOUNG Louis B. Best 625 AVIAN BONE PATHOLOGIES FROM ARIKARA SITES IN SOUTH DAKOTA Paul W. Parmalee 628 NEST RECIPROCITY IN EASTERN PHOEBES AND BARN SWALLOWS Harmon P. Weeks, Jr. 632 ORNITHOLOGICAL LITERATURE 636 ORNITHOLOGICAL NEWS 655 REQUESTS FOR ASSISTANCE 1 571, 635 657 INDEX TO VOLUME 89 JW ' T' . ff |v ■H ivJ