Prehistoric human impacts on California birds: evidence from the Emeryville Shellmound Avifauna

The abundance of artiodactyls, marine mammals, waterfowl, seabirds, and other animals in 18th- and 19th-century California astonished early explorers, and the incredible wildlife densities reported in their accounts are routinely taken as analogues for the original or pristine zoological condition. However, recent analyses of archaeological fish and mammal materials from California and elsewhere in western North America document that those early historic-period faunal landcsapes represent poor analogues for prehistoric environments, because they postdate a dramatic 16th- or 17th-century population-crash of native hunters. The superabundance of tame wildlife witnessed during the early historic period may only reflect population irruptions that followed the demise of their main predators. While analyses of archaeological faunas from California have documented that prehistoric peoples had substantial impacts on populations of fish and mammals, harvest pressure on bird populations has yet to be documented. The hypothesis that prehistoric hunters caused depressions of avian taxa is tested here through a description and analysis of the Emeryville Shellmound avifauna: the first substantial, well-documented archaeological bird sequence for the late Holocene of California. A total of 64 species is represented by the 5,736 identified bird specimens derived from the stratified Emeryville deposits that date from between 2,600 and 700 years ago; waterfowl, cormorants, and shorebirds dominate the collection. Chrono-stratigraphic trends in relative taxonomic abundances and age structure within those groups are consistent with long-term anthropogenic depressions resulting from expansion of regional human populations over the occupational history of the mound. In general, large-sized bird species, those that occupied habitats closer to bayshore human residences, and those that were otherwise sensitive to human hunting pressure decreased in numbers over time. In the waterfowl assemblage, geese (Branta canadensis, B. hutchinsii, Anser albifrons, Chen caerulescens, C. rossii) declined significantly over time as compared with ducks, and the remains of the largest-sized geese (B. canadensis moffitti, A. albifrons, C. caerulescens) declined as compared with the smaller ones (e.g. B. hutchinsii, C. rossii). As hunting returns from local patches decreased over time, ever-increasing use was made of more distant, marine-oriented duck taxa - namely scoters (Melanitta fusca and M. perspicillata). Double-crested Cormorants (Phalacrocorax auritis) were especially hard-hit by human harvesting activities, which caused the extirpation of local island-based colonies; changes in the relative age and species composition of the regional Phalacrocorax fauna; and, ultimately, a nearly complete abandonment of cormorant hunting. Finally, the largest species of shorebirds-Marbled God wits (Limosa fedoa), Long-billed Curlews (Numenius americanus), and Whimbrels (N. phaeopus) - declined significantly over time, in comparison with smaller shorebird species. None of those patterns are correlated with changes in pertinent paleoenvironmental records that might indicate that they were caused by climate-based environmental change. They suggest, however, that activities of human foragers had a fundamental influence on the late Holocene avian fauna of the region, and that records of bird abundances, distributions, and behavior from the early historic period are anomalous in the context of the past several thousand years of intensive human harvesting. The conclusions presented here challenge the conventional wisdom regarding prehistoric landscape ecology in North America and have important implications for analyses that require information on long-term population histories, including those involving modern patterns in genetic diversity directed toward conservation related problems.

PREHISTORIC HUMAN IMPACTS ON CALIFORNIA BIRDS: EVIDENCE FROM THE EMERYVILLE SHELLMOUND AVIFAUNAEditor: John Faaborg 224 Tucker Hall Division of Biological Sciences University of Missouri Columbia, Missouri 65211 Project Manager: Mark C. Penrose Managing Editor: Richard D. Earles AOU Publications Office 622 Science Engineering Department of Biological Sciences University of Arkansas Fayetteville, Arkansas 72701 ORNITHOLOGICAL MONOGRAPHS The Ornithological Monographs series, published by the American Ornithologists' Union, has been established for major papers too long for inclusion in the Union's journal, The Auk. Copies of Ornithological Monographs are available from Buteo Books, 3130 Laurel Road, Shipman, VA 22971. Price of Ornithological Monographs no. 56: $10.00 ($9.00 for AOU members). Add $4.00 for handling and shipping charges in U.S., and $5.00 for shipping to other countries. Make checks payable to Buteo Books. Author of this issue, Jack M. Broughton. Library of Congress Control Number 2004113911 Printed by Cadmus Communications, Ephrata, PA 17522 Issued 27 October 2004 Ornithological Monographs, No. 56 xii + 90 pp. Copyright © by the American Ornithologists' Union, 2004 ISBN: 0-943610-62-1 Cover: Design by Jack M. Broughton. Canada geese drawing: Line art by Timothy Knepp, courtesy of U.S. Fish and Wildlife Service.TABLE OF CONTENTS Lists of Tables and Figures .................................................................................................................... vii From the Editor............................................................................................................................................... xi ABSTRACT .................................................................................................................................................... 1 INTRODUCTION....................................................................................................................................... 3 SAN FRANCISCO BAY SHELLMOUNDS AND THE EMERYVILLE SITE AND FAUNA......................................................................................................................................................... 4 Chronology............................................................................................................................................... 6 Avifaunal Materials........................................................................................................................... 7 SYSTEMATICS AND OSTEOLOGY................................................................................................... 8 Order Gaviiformes................................................................................................................................ 8 Family Gaviidae ................................................................................................................................ 8 Order Podicipediformes.................................................................................................................... 13 Family Podicipedidae...................................................................................................................... 13 Order Procellariiformes................................................................................................................. 14 Family Diomedeidae ....................................................................................................................... 14 Family Procellariidae ....................................................................................................................... 14 Order Pelecaniformes ....................................................................................................................... 14 Family Pelecanidae........................................................................................................................... 14 Family Phalacrocoracidae ............................................................................................................. 15 Order Ciconiiformes........................................................................................................................... 17 Family Ardeidae................................................................................................................................ 17 Family Cathartidae........................................................................................................................... 17 Order Anseriformes ............................................................................................................................ 17 Family Anatidae ................................................................................................................................ 17 Order Falconiformes .......................................................................................................................... 23 Family Accipitridae .......................................................................................................................... 23 Family Falconidae.............................................................................................................................. 24 Order Galliformes ............................................................................................................................... 24 Family Phasianidae........................................................................................................................... 24 Family Odontophoridae................................................................................................................. 25 Order Gruiformes................................................................................................................................ 25 Family Rallidae................................................................................................................................... 25 Family Gruidae................................................................................................................................... 25 Order Charadriiformes ..................................................................................................................... 26 Family Charadriidae......................................................................................................................... 26 Family Recurvirostridae................................................................................................................. 26 Family Scolopacidae......................................................................................................................... 26 Family Laridae ................................................................................................................................... 27 Family Alcidae..................................................................................................................................... 28 Order Strigiformes.............................................................................................................................. 28 Family Tytonidae .............................................................................................................................. 28 Family Strigidae ................................................................................................................................ 28 Order Passeriformes........................................................................................................................... 29 Family Corvidae................................................................................................................................ 29 TAXONOMIC SUMMARY AND DEPOSITIONAL ORIGIN ............................................... 29 ARCHAEOLOGICAL MEASURES OF AVIAN RESOURCE DEPRESSION................. 30 Relative-abundance Indices............................................................................................................. 30 v Prey Age Composition.............................................................................................31 Paleoclimatic Variables..........................................................................................32 Summary...................................................................................................................32 ANTHROPOGENIC DEPRESSIONS AND THE EMERYVILLE AVIFAUNAL SEQUENCE....................................................................................................................................32 Waterfowl................................................................................................................32 Cormorants ............................................................................................................39 Shorebirds.......................................................................................................................................43 GENERAL CONCLUSIONS ...........................................................................................................44 ACKNOWLEDGMENTS................................................................................................................46 LITERATURE CITED........................................................................................................................46 APPENDIX .........................................................................................................................................53 vi LIST OF TABLES 1. Emeryville provenience units with associated radiocarbon determinations ... 7 2. Numbers of identified bird specimens per taxon by stratum at the Emeryville Shellmound ..................................................................................................................... 9 3. Numbers of identified subadult bird specimens by stratum at the Emeryville Shellmound ..................................................................................................................... 12 4. Minimum interorbital frontal breadths for recent Melanitta nigra, M. perspicillata, and M. fusca specimens ......................................................................... 21 5. Premaxilla lengths for recent Melanitta nigra, M. perspicillata, and M. fusca specimens......................................................................................................................... 22 6. Average weights of anatid species identified from the Emeryville Shellmound .. 33 7. Numbers of identified specimens (NISP) of anserines and anatinines by major stratum from the Emeryville Shellmound .............................................................. 34 8. Numbers of identified specimens (NISP) of large geese and small geese by major stratum from the Emeryville Shellmound................................................... 35 9. Numbers of identified specimens (NISP) of merginines and other ducks by major stratum from the Emeryville Shellmound................................................... 36 10. Numbers of identified cormorants and all other birds (NISP) by major stratum from the Emeryville Shellmound .............................................................. 40 11. Numbers of identified subadult and adult cormorants (NISP) by major stratum from the Emeryville Shellmound .............................................................. 41 12. Numbers of identified Pelagic, Brandt's, and Double-crested Cormorants (NISP) by major stratum from the Emeryville Shellmound............................... 42 13. Numbers of identified large and small shorebirds (NISP) by major stratum from the Emeryville Shellmound ............................................................................. 44 Al. Numbers of identified bird specimens by element and portion for stratum 1, Uhle excavation............................................................................................................... 54 A2. Numbers of identified bird specimens by element and portion for stratum 2, Uhle excavation............................................................................................................... 55 A3. Numbers of identified bird specimens by element and portion for stratum 3, LThle excavation............................................................................................................... 57 A4. Numbers of identified bird specimens by element and portion for stratum 4, Uhle excavation............................................................................................................... 59 A5. Numbers of identified bird specimens by element and portion for stratum 5, Uhle excavation............................................................................................................... 60 A6. Numbers of identified bird specimens by element and portion for stratum 6, Uhle excavation............................................................................................................... 61 A7. Numbers of identified bird specimens by element and portion for stratum 7, Uhle excavation............................................................................................................... 62 A8. Numbers of identified bird specimens by element and portion for stratum 8, Uhle excavation............................................................................................................... 63 A9. Numbers of identified bird specimens by element and portion for stratum 9, Uhle excavation............................................................................................................... 65 A10. Numbers of identified bird specimens by element and portion for stratum 10, Uhle excavation............................................................................................................... 66 All. Numbers of identified bird specimens by element and portion for stratum 2, Nelson excavation.......................................................................................................... 68 vii A12. Numbers of identified bird specimens by element and portion for stratum 3, Nelson excavation.......................................................................................................... 68 A13. Numbers of identified bird specimens by element and portion for stratum 4, Nelson excavation.......................................................................................................... 69 A14. Numbers of identified bird specimens by element and portion for stratum 5, Nelson excavation.......................................................................................................... 70 A15. Numbers of identified bird specimens by element and portion for stratum 6, Nelson excavation.......................................................................................................... 70 A16. Numbers of identified bird specimens by element and portion for stratum 7, Nelson excavation.......................................................................................................... 70 A17. Numbers of identified bird specimens by element and portion for stratum 8, Nelson excavation.......................................................................................................... 70 A18. Numbers of identified bird specimens by element and portion for stratum 9, Nelson excavation.......................................................................................................... 70 A19. Numbers of identified bird specimens by element and portion for stratum 10, Nelson excavation.......................................................................................................... 71 A20. Numbers of identified bird specimens by element and portion for stratum 11, Nelson excavation.......................................................................................................... 71 A21. Numbers of identified bird specimens by element and portion for trench 1 (0-1 ft), Schenck excavation ......................................................................................... 72 A22. Numbers of identified bird specimens by element and portion for trench 1 (1-2 ft), Schenck excavation......................................................................................... 72 A23. Numbers of identified bird specimens by element and portion for trench 1 (2-3 ft), Schenck excavation......................................................................................... 73 A24. Numbers of identified bird specimens by element and portion for trench 1 (3-4 ft), Schenck excavation ......................................................................................... 74 A25. Numbers of identified bird specimens by element and portion for trench 1 (4-5 ft), Schenck excavation......................................................................................... 74 A26. Numbers of identified bird specimens by element and portion for trench 1 (5-6 ft), Schenck excavation......................................................................................... 75 A27. Numbers of identified bird specimens by element and portion for trench 1 (6-7 ft), Schenck excavation......................................................................................... 75 A28. Numbers of identified bird specimens by element and portion for trench 1 (7-8 ft), Schenck excavation ......................................................................................... 76 A29. Numbers of identified bird specimens by element and portion for trench 2 (1-2 ft), Schenck excavation......................................................................................... 77 A30. Numbers of identified bird specimens by element and portion for trench 2 (2-3 ft), Schenck excavation......................................................................................... 78 A31. Numbers of identified bird specimens by element and portion for trench 2 (3^ ft), Schenck excavation......................................................................................... 79 A32. Numbers of identified bird specimens by element and portion for trench 2 (4-5 ft), Schenck excavation......................................................................................... 80 A33. Numbers of identified bird specimens by element and portion for trench 2 (5-6 ft), Schenck excavation ......................................................................................... 81 A34. Numbers of identified bird specimens by element and portion for trench 2 (6-7 ft), Schenck excavation ......................................................................................... 82 A35. Numbers of identified bird specimens by element and portion for trench 3 (0-1 ft), Schenck excavation......................................................................................... 83 viii A36. Numbers of identified bird specimens by element and portion for trench 3 (1-2 ft), Schenck excavation......................................................................................... 84 A37. Numbers of identified bird specimens by element and portion for trench 3 (2-3 ft), Schenck excavation......................................................................................... 85 A38. Numbers of identified bird specimens by element and portion for trench 3 (3-4 ft), Schenck excavation ......................................................................................... 86 A39. Numbers of identified bird specimens by element and portion for trench 3 (4-5 ft), Schenck excavation......................................................................................... 87 A40. Numbers of identified bird specimens by element and portion for trench 3 (5-6 ft), Schenck excavation ......................................................................................... 88 A41. Numbers of identified bird specimens by element and portion for trench 3 (6-7 ft), Schenck excavation......................................................................................... 89 A42. Numbers of identified bird specimens by element and portion for trench 3 (7-8 ft), Schenck excavation......................................................................................... 89 A43. Numbers of identified bird specimens by element and portion for trench 3 (8 ft to 8 ft 6 in), Schenck excavation......................................................................... 90 LIST OF FIGURES 1. Map of the San Francisco Bay area, showing location of the Emeryville Shellmound and other shellmound sites ................................................................ 5 2. Demolition of the Emeryville Shellmound by steam shovel, 1924 ...................... 6 3. Right coracoid of Gavia stellata and left proximal carpometacarpus of Phoebastria albatrus.......................................................................................................... 13 4. Left proximal tibiotarsus of Phalacrocorax penicillatus and left femur of Phalacrocorax auritus ...................................................................................................... 15 5. Left proximal humerus of Ardea herodias ..................................................................... 17 6. Premaxilla of Chen caerulescens and frontal of Bran ta canadensis.............................. 18 7. Left coracoid and right ulna of Gallus gallus .............................................................. 25 8. Right proximal carpometacarpus of Numenius americanus...................................... 27 9. Right humerus, missing proximal end, of Otus kennicottii...................................... 28 10. Distribution of the goose index (£NISP anserines/£NISP anatids) by stratum at the Emeryville Shellmound.................................................................................... 34 11. Distribution of the large-goose index (VNISP large and medium anserines/ £NISP anserines) by stratum at the Emeryville Shellmound.............................. 35 12. Distribution of the scoter index (VNISP merginines/£NISP anatinines) by stratum at the Emeryville Shellmound..................................................................... 36 13. Distribution of the cormorant index (£NISP cormorants /TNISP birds) by stratum at the Emeryville Shellmound..................................................................... 40 14. Distribution of the subadult cormorant index (VNISP subadult cormorants/ £NISP cormorants) by stratum at the Emeryville Shellmound........................ 41 15. Holocene variation in the frequency of ENSO events as reconstructed from the sedimentation record of Laguna Pallcacocha, southern Ecuador............. 41 16. Distribution of the Brandt's-Pelagic index (VNISP Brandt's + Pelagic cormorants/ VNISP cormorants) by stratum at the Emeryville Shellmound ... 42 ix From the Editor Everyone is aware that human activities affect birds, usually negatively. In recent years, we responded to apparent declines in bird populations by developing a massive international migratory-bird conservation plan that hopes to "keep common birds common." Yet we continue to face an extinction crisis as habitat dwindles and less and less of the world remains wild. Many of us think of these negative effects on bird populations as a modern phenomenon, one that came along with burgeoning populations virtually throughout the globe. Those of us who study island avifaunas were aware of cases such as Hawaii, where humans caused many extinc­tions through harvest and habitat change over 1,000 years ago, but these we thought of as special cases that revolved around the constraints of naive island faunas. Others were aware of the argu­ments that many of our native North American megafauna, things like mammoths and ground sloths, may have been driven to extinction by the earliest humans on the continent. But many scien­tists familiar with the relevant archaeology and paleontology have argued that climate change is a far more parsimonious explanation for those losses. The general consensus was that pre-European humans living in North America had little or no effect on continental wildlife populations. After you read Ornithological Monograph No. 56,1 think you will agree that we need to reconsider our impressions about human impacts on bird populations in the distant past. Jack Broughton makes an excellent case that native peoples living in the San Francisco Bay area harvested enough birds to deplete populations and even cause some local extinction, perhaps as long as 2,000 years ago. He also notes that proper knowledge of prehistoric bird populations is critical to understand­ing present-day patterns of population change and related factors such as genetic bottlenecks. In this monograph, avian paleontology and archaeology meet modern conservation biology and teach us to be careful about what we assume. As always, reviewing monograph-length manuscripts requires dedicated volunteers. For Ornithological Monograph No. 56, we thank Douglas Causey of the Museum of Comparative Zoology, Harvard University; and R. Lee Lyman, Chair of the Department of Anthropology at the University of Missouri-Columbia. While some of you may want to read around the osteological details neces­sary for our author to make his case, I think that all of you will be impressed by the major impacts indigenous peoples have had on continental bird populations. John FaaborgOrnithological Monographs Volume (2004), No. 56,1-90 PREHISTORIC HUMAN IMPACTS ON CALIFORNIA BIRDS: EVIDENCE FROM THE EMERYVILLE SHELLMOUND AVIFAUNA Jack M. Broughton1 Department of Anthropology, University of Utah, 270 South 1400 East, Room 102, Salt Lake City, Utah 84112, USA Abstract.-The abundance of artiodactyls, marine mammals, waterfowl, seabirds, and other animals in 18th- and 19th-century California astonished early explorers, and the incredible wildlife densities reported in their accounts are routinely taken as analogues for the original or pristine zoological condition. However, recent analyses of archaeological fish and mammal materials from California and elsewhere in western North America document that those early historic-period faunal landcsapes represent poor analogues for prehistoric environments, be­cause they postdate a dramatic 16th- or 17th-century population-crash of native hunters. The superabundance of tame wildlife witnessed during the early historic period may only reflect population irruptions that followed the demise of their main predators. While analyses of ar­chaeological faunas from California have documented that prehistoric peoples had substantial impacts on populations of fish and mammals, harvest pressure on bird populations has yet to be documented. The hypothesis that prehistoric hunters caused depressions of avian taxa is tested here through a description and analysis of the Emeryville Shellmound avifauna: the first substantial, well-documented archaeological bird sequence for the late Holocene of California. A total of 64 species is represented by the 5,736 identified bird specimens derived from the stratified Emeryville deposits that date from between 2,600 and 700 years ago; waterfowl, cormorants, and shorebirds dominate the collection. Chrono-stratigraphic trends in relative taxonomic abundances and age structure within those groups are consistent with long-term anthropogenic depressions resulting from expansion of regional human populations over the occupational history of the mound. In general, large-sized bird species, those that occupied habitats closer to bayshore human residences, and those that were otherwise sensitive to hu­man hunting pressure decreased in numbers over time. In the waterfowl assemblage, geese (Branta canadensis, B. hutchinsii, Anser albifrons, Chen caerulescens, C. rossii) declined significantly over time as compared with ducks, and the remains of the largest-sized geese (B. canadensis mof- fitti, A. albifrons, C. caerulescens) declined as compared with the smaller ones (e.g. B. hutchinsii, C. rossii). As hunting returns from local patches decreased over time, ever-increasing use was made of more distant, marine-oriented duck taxa -namely scoters (Melanitta fusca and M. perspicillata). Double-crested Cormorants (Phalacrocorax auritis) were especially hard-hit by human harvesting activities, which caused the extirpation of local island-based colonies; changes in the relative age and species composition of the regional Phalacrocorax fauna; and, ultimately, a nearly complete abandonment of cormorant hunting. Finally, the largest species of shorebirds-Marbled Godwits (Limosa fedoa), Long-billed Curlews (Numenius americanus), and Whimbrels (N. phaeopus) - declined significantly over time, in comparison with smaller shore- bird species. None of those patterns are correlated with changes in pertinent paleoenvironmen- tal records that might indicate that they were caused by climate-based environmental change. They suggest, however, that activities of human foragers had a fundamental influence on the late Holocene avian fauna of the region, and that records of bird abundances, distributions, and behavior from the early historic period are anomalous in the context of the past several thousand years of intensive human harvesting. The conclusions presented here challenge the conventional wisdom regarding prehistoric landscape ecology in North America and have im­portant implications for analyses that require information on long-term population histories, including those involving modern patterns in genetic diversity directed toward conservation- related problems. Received 10 April 2004, accepted 6 August 2004. Resumen. -La abundancia de artiodactilos, marmferos marinos, aves acuaticas (Anseriformes), aves marinas y otros animales en California durante los siglos 18 y 19 deslumbro a los primeros exploradores, y las densidades increi'bles de fauna silvestre ‘E-mail: [e-mail] 1 ORNITHOLOGICAL MONOGRAPHS NO. 56 mencionadas en sus informes son tomadas de modo rutinario como analogas a la condicion zoologica original o pristina. Sin embargo, analisis recientes de materiales arqueologicos de peces y mamiferos provenientes de California y de otros sitios del oeste de America del Norte senalan que estos escenarios historicos tempranos de la fauna representan analogias equivocadas de los ambientes prehistoricos, ya que ellos son posteriores a una reduccion dramatica de las poblaciones de cazadores nativos ocurrida durante el siglo 16 o 17. La superabundancia de fauna silvestre docil observada durante el periodo historico temprano puede solo reflejar irrupciones en las poblaciones que siguieron a la caida de sus principales depredadorcs. Mientras que los analisis de la fauna arqueologica de California han doeumentado que las poblaciones humanas prehistoricas tuvieron un impacto substancial en las poblaciones de peces y mamiferos, no se ha doeumentado aun la presion de cosecha en las poblaciones de aves. La hipotesis de que los cazadores prehistoricos causaron reducciones de taxa de aves es evaluada aquf a traves de la descripcion y el analisis de la avifauna de Emeryville Shellmound: la primera secuencia arqueologica de aves substancial y bien documentada del Holoceno tardio de California. Un total de 64 especies de aves esta representado por 5736 ejemplares identificados, derivados de los depositos estratificados de Emeryville, que datan de entre 2,600 y 700 anos atras; dominan la coleccion los Anseriformes, los cormoranes y las aves playeras. Las tendencias crono-estratigraficas en las abundancias taxonomicas relativas y en la estructura de edades dentro de esos grupos son consistentes con las disminuciones a largo plazo de las poblaciones humanas, resultantes de la expansion regional de estas poblaciones durante el periodo de ocupacion del sitio arqueologico. En general, las especies de aves de gran tamano, las que ocuparon ambientes cercanos a las playas habitadas por humanos y aquellas que de otro modo eran sensibles a la presion antropica de caza decrecieron en abundancia con el tiempo. Entre los Anseriformes, los gansos (Brnnta canadensis, Anser albifrons, Chen caerulescens, C. rossii) declinaron significativamente con el tiempo en comparacion con los patos, y los restos de gansos de gran tamano (B. canadensis mojfitti, A. albifrons, C. caerulescens) declinaron en comparacion con los mas pequenos (e.g. B. canadensis minima, C. rossii). A medida que la abundancia de presas cazadas en parches locales declino con el tiempo, se incremento el uso de taxa mas afines a ambientes marinos ubicados a mayor distancia, como Melanitta fusca y M. perspicillata. El cormoran Phalacrocorax auritas fue especialmente afectado por las actividades de cosecha de los humanos, que causaron la extirpacion de colonias locales ubicadas en islas, cambios en la edad relativa y la composicion de especies de la fauna regional de Phalacrocorax y, finalmente, el abandono casi total de la caza de cormoranes. Finalmente, las especies de aves playeras de mayor tamano, Limosa fedoa, Numenius americanus y N. phaeopus, declinaron significativamente con el tiempo, en comparacion con especies playeras de menor tamano. Ninguno de estos patrones estan correlacionados con variaciones en los registros paleo-ambientales que puedan indicar que fueron causados por cambios en el clima. Sin embargo, estos patrones sugieren que las actividades de los humanos recolectores tuvieron una influencia fundamental en la avifauna regional del Holoceno tardi'o, y que los registros de las abundancias, distribuciones y comportamiento de las aves del periodo historico temprano son anomalas en el contexto de los ultimos varios miles de anos de cosecha intensa por parte de humanos. Las conclusiones presentadas aquf desafian la creencia convencional sobre la ecologia de paisajes prehistoricos en America del Norte y tienen implicancias importantes para los analisis que requieren informacion de historias poblacionales de largo plazo, incluyendo aquellas que consideran patrones modernos en diversidad genetica dirigidos a problemas de conservacion.PREHISTORIC CALIFORNIA BIRDS 3 Introduction In the fall of that year [1850], my father, while going from San Francisco to San Jose, met with acres of white and gray geese.. .They were feeding near the roadside, indifferent to the presence of all persons, and in order to see how close he could approach he walked directly towards them. When within five or six yards of the nearest ones they stretched up their necks and walked away like domestic geese.. .They seemed to have no idea that they would be harmed, and feared man no more than they did the cattle in the fields...but it must be understood that in those days they were but little hunted...This seems the most plausible accounting for the stupid tameness of the geese.-Bryant (1890), quoted in Grinnell et al. (1918) Accounts of enormous flocks of tame geese are typical of early historical descriptions of California's avifauna. Both the sheer abundance and docility of the birds astonished many who wrote about the region in the years before the Gold Rush and the era of market hunting that came soon after. In 1833, George Yount noted of the San Francisco Bay area that "the wild geese, and every species of waterfowl darkened the sur­face of every bay.. .in flocks of millions. When dis­turbed, they arose to fly, the sound of their wings was like that of distant thunder" (Camp 1923). A decade earlier, in the same area, the Russian explorer Otto von Kotzebue (1830) had observed "flocks of wild geese, ducks, and snipes, so tame that we might have killed great numbers with our sticks." Indeed, some early explorers did kill great numbers with their sticks. William Thomes (1892), for instance, also encountered "thousands of geese and ducks" around San Francisco Bay in the 1840s and claimed to have never seen "so many wild fowl at one time before or since." Because the birds were so abundant and acted "more stupid [than] if they had been hatched in a barnyard, in Rhode Island, and waiting for their daily supply of corn," Thomes and his company found "no pleasure" in shooting them. So, to supply their ship, they simply "threw clubs at them, and knocked them over," thus saving their powder and shot (Thomes 1892). Other vertebrate taxa, too, were extremely abundant in the early historic period of California. Perhaps most noteworthy were the artiodactyls-elk (Cervus elaphus), mule deer (Odocoileus hemionus), and pronghorn (Antilocapra americana)-reported to have "darkened the plains for miles" (Bosqui 1904). The abundance of marine mammals likewise deeply impressed early chroniclers; sea otters (Enhydra lulris), for example, hauled out in such numbers around San Francisco Bay that the shores "appeared covered with black sheets" (Ogden 1941). Predictably, large predators abounded in such an environ­ment: Yount's report of seeing "fifty or sixty" grizzly bears (Ursus arctos) a day is not atypical (Preston 2002). Many such observations were made by vet­eran travelers, who, like Bryant, reasoned that the unwariness of the game must have resulted from a virtual lack of human hunting pressure. Some went so far as to fault the Hispanic settlers for their lack of interest in hunting, and others even "contemplated how the relative ease of hunting contributed to the perceived 'indolence' of both settler and native alike" (Preston 2002). Although the various 19th-century chroni­clers, explorers, and settlers may have had vari­ous motivations for exaggerating in their diaries, ships' logs, and scientific survey reports, the overall consistency of the accounts suggests that their portrayal of California's early-historic fau- nal abundance is generally accurate. Importantly, the abundance reported in those accounts is rou­tinely taken as an analogue for the state's original or natural zoological condition and, as a result, is used as the baseline by which modern popula­tion trends and distributions are measured and compared (e.g. Johnson and Jehl 1994). Recent archaeological analyses suggest, however, that the superabundance of wildlife observed in California in the early historic period is, in fact, an extremely poor analogue for the zoo­logical setting in pre-Columbian times. Guided by models from foraging theory (e.g. Stephens and Krebs 1986), systematic analyses of fish and mammal remains derived from California archaeological sites have indicated that late Holocene (i.e. the last 4,000 years) human popu­lations in the region had substantial impacts on a variety of fish and mammal populations. One of the most detailed of such records has come from a huge stratified archaeologi­cal site once located on the eastern shore of San Francisco Bay: the Emeryville Shellmound (Broughton 1995, 1997, 1999, 2002a). Analysis of the exceptionally rich faunal collection has shown that such large-bodied taxa as elk and sturgeon (Acipenser sp.) provided an ever- decreasing part of human diets across the4 ORNITHOLOGICAL MONOGRAPHS NO. 56 occupational history of the site, which spanned from about 2,600 to 700 years before present (BP). That conclusion is based primarily on trends in relative frequencies of elk and stur­geon bones: both species are very abundant early on, but are virtually absent by the end of the occupation. Demographic signals of harvest pressure, such as trends in age and size profiles, have also been documented for those and other taxa in the deposit (Broughton 1995, 1997, 1999, 2002a). Similar patterns have been reported in a number of archaeological records across the state, and-whereas none appear to correlate with other potential causes for population declines, such as environmental change (e.g. Hildebrandt and Jones 1992, 2002; Broughton 1994a, b, 1999, 2002a, b; Porcasi et al. 2000; Grayson 2001)-they follow predict­ably from foraging theory, given conditions of ever-increasing human population densities and hunting pressure. The patterns appear to represent cases of long-term resource depres­sion (sensu Charnov et al. 1976), or declines in capture rates of prey that result directly from the activities of predators. Archaeological evidence for severe late Holocene depressions in a wide array of vertebrate taxa stands in stark contrast to the fabulous abundances reported in early historical times. It now seems clear that such accounts only reflect the irruption of animal populations after native Californians had experienced dramatic disease-based population declines, apparently initiated by limited coastal contacts between European explorers and California Indians in the early 16th century (Erlandson and Bartoy 1995, Preston 1996, Erlandson et al. 2001). From those isolated encounters, disease apparently spread rapidly through the aboriginal popula­tion of California, well before the arrival of the settlers and travelers who furnished the accounts of wildlife superabundances (Broughton 1994b, 2002b, 2004; Preston 1996, 2002). Thus, the lat­est prehistoric and early-historic baselines or benchmarks for California ecosystems, though separated by mere decades, appear to be worlds apart. Most importantly, those differences and the processes that produced them have implica­tions for the management and conservation of wildlife resources today (Broughton 2004). Analyses of archaeological faunas from California have documented that late Holocene human populations had substantial impacts on fish and mammal populations, but harvest pressure on bird populations has yet to be documented. Indeed, outside of oceanic island contexts where human-caused avifaunal extinc­tions and extirpations are well described (see Steadman 1995, Martin and Steadman 1999, and references therein), there have been no sys­tematic attempts to evaluate evidence for avian resource depression by prehistoric foragers any­where in the world. Here, I document the entire provenienced sample of bird remains recovered from the Emeryville Shellmound. Part of the assemblage was examined by Hildegarde Howard (1929) in her classic study, but most of it has remained unexamined until now. The materials provide a unique, fine-grained anthropogenic sequence of bird harvesting, dating from about 2,600 to 700 years BP. I analyzed the collection to evaluate the role that ancient hunters played in structuring the prehistoric avifauna of the largest estuary and contiguous tidal marsh system on the Pacific coast. The results have implications for the study and management of modern California bird populations for which information on long-term population trends is required. San Francisco Bay Shellmounds and the Emeryville Site and Fauna Around the beginning of the 20th century, surveys of the San Francisco Bay shoreline documented the presence of 425 shellmounds-archaeological sites whose primary visual constituent is shell (Fig. 1). That figure undoubtedly underestimates the true number of sites, given that many had already been obliterated through development and other causes (Nelson 1909). The mounds were made up not only of shells, but of soil, rocks, animal bones, ash, charcoal, and artifacts - all debris and tools from the day-to-day activities of ancient people. The mounds varied substantially in size; some had basal diameters of only a few meters and stood but a few centimeters above the shore, whereas others were much larger, covering >3 ha and rising >10 m in height. Radiocarbon dating has indicated that the earliest shellmounds began to form -4,000 years BP and that the San Francisco shoreline was occupied continuously from that time to the his­toric period (Broughton 1994b, 1995, 1999; Lightfoot and Luby 2002). The record clearly reflects a substan­tial prehistoric human presence in the region, popula­tions supported entirely by the hunting and gathering of wild animals and plants. Those populations appear to have increased significantly across much of the late Holocene, judging from the increasing numberPREHISTORIC CALIFORNIA BIRDS 5 Fig. 1. Map of the San Francisco Bay area, showing location of the Emeryville Shellmound and other shell- mound sites (site locations from Nelson 1909).6 ORNITHOLOGICAL MONOGRAPHS NO. 56 of dated sites and human burials over that period (Broughton 1999). Absolute human population sizes are, however, exceedingly difficult to derive archaeo­logically; I suspect that Kay's (2002) estimate of 2-3 million or more for the state of California is within reason. Given the richness of the San Francicso Bay environment, the human population of the region prior to European contact was likely somewhere in the range of 50,000 to 150,000 people. The Emeryville Shellmound, located on the east­ern shore of San Francisco Bay between the cities of Oakland and Berkeley, measured 100 x 300 m in area and extended to a- depth of >10 m (Figs. 1 and 2). It was the largest of what was originally a complex of about six mounds located on the alluvial flat of Emeryville (Broughton 1996). Max Uhle and John Merriam conducted the first excavation of the site in 1902. At the time, the enormous site was the central feature of "Shellmound Park." As part of a recreation ground, the Emeryville mound was crowned with a dance pavilion and cypress hedge. With the pavilion atop the center of the mound, Uhle and Merriam excavated a lateral section of the mound's western slope and a tunnel that extended from there to its center. They dug >200 m3 of midden and removed the sediments "stratum by stratum." They encountered 10 distinct strata and collected and provenienced all artifacts, including a large sample of vertebrate remains, by those strata (Uhle 1907). They collected the vertebrate materials and other artifacts with sieves of an unspecified, but apparently coarse-grain, mesh size (Uhle 1907, Schenck 1926). Four years later, in spring 1906, Nels C. Nelson led the second excavation at Emeryville, in which a 6 x 6 ft unit was stratigraphically dug in the eastern side of the mound. He identified 11 natural strata and collected and provenienced all artifacts, including vertebrate remains, by those strata. Given the smaller volume of sediment excavated, a much smaller sample of verte­brate remains was recovered (Broughton 1996). The Emeryville Shellmound was leveled by a steam shovel in 1924 (Fig. 2). W. E. Schenck salvaged a large series of human burials and associated artifacts, along with a large collection of vertebrate materials, as the mound was being demolished. Unfortunately, Schenck was unable to collect within-site provenience data for the vertebrate bones and teeth collected at that time, because "scientific ends were secondary" (Schenck 1926). However, after the mound had been reduced to the level of the surrounding plain, Schenck excavated three 50 x 6 ft trenches in the base of the deposit, near the center of the mound. Those trenches, excavated in 1-ft arbitrary levels to a depth of >10 ft, produced a sizable faunal collection (Schenck 1926). Chronology Thirteen radiocarbon assays have now been derived from bone and charcoal specimens recovered from various strata throughout the Emeryville deposit; Fig. 2. Demolition of the Emeryville Shellmound by steam shovel, 1924. (Photo courtesy of the Phoebe A. Hearst Museum of Anthropology.)PREHISTORIC CALIFORNIA BIRDS 7 exact proveniences of the dated materials are known for 11 of them (Broughton 1999). On the western side of the mound, dates range from 2,620 ± 70 years BP at the basal contact between the cultural midden and the alluvial clay on which the mound sits to 950 ± 50 years BP for stratum 2. For the Nelson strata on the mound's eastern side, three dates are available: 2,370 ± 70 years BP for basal stratum 11; 1,100 ±50 years BP for stratum 5; and 720 ± 60 years BP for stratum 3 (Table 1). There are no chrono-stratigraphic inconsistencies in the dates from either the Uhle-Merriam or Nelson excavations; in other words, within each excavation, the oldest dates are from the lowest strata, whereas the youngest dates are from the highest ones. A single radiocarbon date was obtained near the top (1-2 ft below the surface) of one of Schenck's trenches, and six dates were obtained for the base of the mound. Together, those dates serve to bracket the deposition of the Schenck trench sediments between 2,600 and 1,970 years BP. That interval incorporates the period of depo­sition for the four basal strata (i.e. strata 10 through 7) from the Uhle-Merriam excavation. Accordingly, I aggregated the 1-ft samples from Schenck's three trenches into a total of four provenience units. The three early-20th-century excavations provided >20 independent sample units that could be assigned to the 10 primary strata of the mound (Table 1). Reporting the identified bird remains from those units will allow for a fine-grained ordinal-scale analysis of changing bird-use patterns over the period from about 2,600 to 700 years BP (see Broughton [1999] for more details on stratigraphic relationships and dating). Avifaunal Materials Reported here are 5,736 identified bird specimens that were collected from the Emeryville provenience units described above. The bird and other verte­brate remains from Emeryville are housed at the Phoebe Apperson Hearst Museum of Anthropology (PAHMA) at the University of California, Berkeley. Given the large size of the collection, I provide cata­logue numbers only for specimens identified here to the species level. Numbers preceded by "EMF" refer to the field catalog of Edna M. Fisher, a curatorial assistant at the Museum of Vertebrate Zoology (MVZ, Berkeley, California) during the early 20th century (see Broughton [1999] for further details on the cura- tion of the Emeryville fauna). As noted above, Howard (1929) described part of the Emeryville avifauna. Although her analysis was exemplary in many ways, she did not provide associated stratigraphic information for any of the specimens she described and, for reasons unknown, Table 1. Emeryville provenience units with associated radiocarbon determinations. Provenience unit Abbreviation Uhle, stratum 1 U1 Uhle, stratum 2 U2 Uhle, stratum 3 U3 Uhle, stratum 4 U4 Uhle, stratum 5 U5 Uhle, stratum 6 U6 Uhle, stratum 7 U7 Uhle, stratum 8 U8 Uhle, stratum 9 U9 Uhle, stratum 10 U10 Nelson, stratum 2 N2 Nelson, stratum 3 N3 Nelson, stratum 4 N4 Nelson, stratum 5 N5 Nelson, stratum 6 N6 Nelson, stratum 7 N7 Nelson, stratum 8 N8 Nelson, stratum 9 N9 Nelson, stratum 10 N10 Nelson, stratum 11 Nil Schenck trench level 1: 0-2' SI Schenck trench level 2: 2-4' S2 Schenck trench level 3: 4-6' S3 Schenck trer.ch level 4: 6-9' S4 Base of southeast corner of mound Base of mound Radiocarbon Stratum determinations (years BP) 1 2 950 ± 50 3 4 5 1,400 ± 50 6 7 1,980 ± 50 8 2,070 ± 60 9 10 2,620 ± 70; 2,400 ± 70; 1,030 ± 60 1 1 720 ± 60 1-3 3 1,110 ±50 4-9 4-9 4-9 4-9 4-9 10 2,370 ± 70 7 1,970 ± 50 8 9 10 -10 2,530 ± 30 -10 2,310 ± 220reported only a sample of the Emeryville avifaunal collection. As a result, the specimens reported here and those that Howard (1929) described represent different subsets of the Emeryville avifauna, though there is substantial overlap. Specifically, almost half of the sample (1,853 of 4,155 bones) identified and reported by Howard was collected by Schenck during the steam-shovel demolition of the mound; hence, within-site provenience information was never obtained for them (see Broughton 1999). Given my interest in examining change through time across the depositional sequence, I do not report on those materials here. In addition, Howard apparently did not have access to and thus did not examine 3,710 specimens that had associated stratigraphic informa­tion. Identifications of those specimens are reported here for the first time. I took the following approach in treating the sample (n = 2,026) of provenienced specimens previ­ously identified to some level by Howard. I refined identifications for specimens that she left at the genus level or higher taxonomic categories (n - 1,515), but report again the relatively small sample of prove­nienced specimens that Howard identified to the species level (n = 511). 1 did not systematically verify the latter identifications, given the widely renowned accuracy of Howard's work (see Campbell 1980); they are simply presented again here with updated taxonomic nomenclature (Banks et al. 2004) and, most importantly, by their associated provenience units. My identifications, listed as "additional elements" below, were based on comparisons with recent bird specimens from the following collections: MVZ; Burke Museum of Natural History and Culture, University of Washington (UWBM); and Utah Museum of Natural History (UMNH). Diagnostic osteological characters were derived from the examination of mul­tiple individuals per species (typically six or more). Anatomical terminology follows Howard (1929) and Baumel (1993). Given the large size of the collec­tion, identifications were not attempted for isolated cervical or thoracic vertebrae, ribs, cuneiforms, and phalanges of the foot. To minimize multiple identi­fications of the same fragmented element portion, only the more complete specimens were examined. Specifically, I attempted identifications for substantia] cranium fragments that included either most of the frontal region, premaxillae, or posterior braincase. For mandibles, fairly complete anterior or posterior portions were identified. Identifications of furculae were based on fragments possessing the furcular process at the symphysis. For the pelvis and the syn- sacrum, specimens that represented >50% of the syn- sacral vertebral column were identified. Only sternum specimens that contained the manubrial process were studied. Finally, identifications of the long bones were attempted only for specimens possessing >60% of a proximal or distal articular surface. 8 All elements were also assigned to one of three broad ontogenetic age categories: chicks, juveniles, and adults. Those assignments were based on the size of the element and its state of development. Specimens were identified as chicks if they were very small in size, porous, and lacking adult cortical bone and muscle attachments. Juveniles were identified as those specimens that approached, or had attained adult size but lacked complete development of corti­cal bone. Remains of chicks clearly represent birds derived from local nesting sites, but juveniles could represent first-year migrants from distant breeding localities. I use the numbers of identified specimens (NISP) as a measure of taxonomic abundances in the analyses that follow. Although clearly imperfect, this is the least contrived and arguably least problematic available measure of relative abundance for archaeological and paleontological faunas (see Grayson 1984). Numbers of identified specimens per taxon by stratum is pro­vided in Table 2; numbers per taxon, element portion, and provenience unit are provided in the Appendix. Chick and juvenile specimens are presented by taxon and stratum in Table 3. Provided below is a systematic list of the taxa and elements represented in the prove­nienced sample from Emeryville, and the osteological criteria that I used to identify them. ORNITHOLOGICAL MONOGRAPHS NO. 56 Systematics and Osteology Order Gaviiformes Family Gaviidae Gavia stellata (Pontoppidan 1763) or G. pacifica (Lawrence 1858) Referred material.-Reported in Howard (1929): tarsometatarsi (EMF A1135, 7861, 8595, A3062, 6385, 8592, 7076, 10299). Additional elements: carpometacarpus (EMF A4012), syn- sacrum (EMF 8745), and tarsometatarsus (EMF A3482). Gavia stellata (Pontoppidan 1763) Referred material.-Reported in Howard (1929): cranium (EMF 7996; listed as "no. 2996" in Howard 1929), coracoids (EMF 7043, 8577), scapula (EMF 8706), humerus (EMF 8706), ulna (EMF 10522), carpometacarpi (EMF 8324, 8019, 8054), tibiotarsi (EMF 10394, A4847, 8014, 8348). Additional elements: coracoids (EMF 17293, A5219, A5311, A11526 [Fig. 3A]), humeri (EMFTable 2. Numbers of identified bird specimens per taxon by stratum at the Emeryville Shellmound (abbreviations follow text). Stratum Taxon U1 U2 U3 U4 U5 U6 U7 U8 U9 U10 N2 N3 N4 N5 N6 N7 N8 N9 N10 Nil SI S2 S3 S4 Z Gavia stellata 2 8 2 2 1 1 5 3 4 28 G. stellata/pacifica 1 1 1 1 2 2 3 11 G. imrner 1 5 1 1 2 4 2 3 19 Podiceps auritus 1 1 1 3 P. nigricollis 2 1 1 1 1 1 1 8 P. auritus/nigricollis 1 4 3 1 1 1 11 Aechmophorus occidentalis/clarkii 5 10 4 1 2 2 1 3 2 1 31 Phoebastria albatrus 1 1 Fulmarus glacialis 3 3 Pelcanus sp. 1 1 1 3 3 2 1 12 P. erythrorhynchos 1 1 P. occidentalis 1 3 1 4 4 2 15 Phalacrocorax sp. 3 7 1 4 15 22 42 98 26 71 14 106 100 38 62 609 P. penicillatus 4 3 4 2 2 3 4 2 1 1 1 9 13 10 3 62 P. auritus 2 3 8 11 8 10 17 29 2 23 1 6 44 46 24 30 264 P. penicillatuslauritus 2 1 1 1 1 6 P. pelagicus 1 1 1 1 2 1 1 1 9 Ardeidae (Bittern sized) 1 1 Botaurus lentiginosus 1 1 Ardea herodias 1 2 1 3 1 3 2 2 1 16 Cathartes aura 1 1 1 1 4 Anserinae (small) 14 55 39 41 28 15 34 58 18 49 1 4 5 2 32 59 49 34 537 Anserinae (medium) 11 112 86 64 36 15 42 160 18 115 5 11 13 1 6 2 7 94 195 146 83 1213 Chen caerulescens 1 2 1 5 5 3 6 5 1 29 C. rossii 3 1 1 1 1 1 8 Branta hutchinsii cf. minima 1 1 1 3 B. canadensis cf. parvipes 2 1 1 2 1 1 1 3 1 2 15 B. c. cf. moffitti 2 1 3 5 1 4 3 1 20 Anatinae (small) 3 37 10 6 2 5 4 14 2 18 9 1 1 1 113 Anatinae (large) 86 532 108 38 23 15 17 83 22 116 4 8 50 4 3 15 3 3 23 24 66 30 1264 Anas sp. (teal) 2 5 4 3 1 2 1 2 1 21 Anas sp. 1 10 3 7 1 1 1 9 4 1 1 1 4 2 5 51 A. platyrhynchos 1 4 2 1 1 1 2 1 2 2 2 19 A. clypeata 2 1 1 1 5 PREHISTORIC CALIFORNIA BIRDSTable 2. Continued. o Stratum Taxon U1 U2 U3 U4 U5 U6 U7 U8 U9 UlO N2 N3 N4 N5 N6 N7 N8 N9 N10 Nil SI S2 S3 S4 I Aythya sp. 2 6 2 1 1 1 1 2 4 1 1 1 23 Aythya sp. (large) 1 1 1 1 4 A. cf. valisneria 1 1 2 A. valisneria 1 1 2 A. cf. marila 1 1 1 1 4 A. marila 1 1 2 A. ajfinis 1 1 2 Melanitta sp. 38 209 106 9 3 2 10 12 14 1 5 20 12 4 1 4 5 9 11 466 M. perspicillata 3 1 1 5 M. fusca 2 6 8 M. perspicillata/fusca 5 1 6 M. perspicillata/nigra 1 1 Bucephala sp. 1 1 B. albeola 1 1 1 2 1 6 B. clangula/islandica 1 3 4 2 1 1 12 Mergus sp. 1 1 2 M. cf. serrator 2 2 M. serrator 1 1 1 1 4 Oxyura jamaicensis 1 2 3 Accipitridae 1 1 Accipitridae (small) 1 1 2 Elanus leucurus 1 1 1 3 Haliaeetus leucocephalus 1 4 1 1 1 8 Circus cyaneus 1 1 2 Accipiter cooperii 1 1 Buieo sp. 1 3 2 1 2 5 1 3 1 3 1 23 B. lineatus 1 1 2 4 B. jamaicensis 1 1 1 7 1 11 B. regalis 1 1 1 3 B. jamaicensishegalis 1 1 2 1 1 6 B. jamaicensis/lagopus 1 1 2 Falco sparverius 1 1 F. columbarius 1 1 2 F. peregrinus 1 1 1 2 5 F. mexicanus 1 1 ORNITHOLOGICAL MONOGRAPHS NO. 56Table 2. Continued. Stratum Taxon U1 U2 U3 U4 U5 U6 U7 U8 U9 U10 N2 N3 N4 N5 N6 N7 N8 N9 N10 Nil SI S2 S3 S4 7. F. peregrinus/mexicanus 1 5 2 Dendragapus obscurus 2 Callipepla californica/ Oreortyx pictus 1 C. californica 1 2 1 1 Rallus longirostris 2 Fulica americana 2 2 Grus canadensis 2 1 4 1 6 G. canadensis tabida 1 Pluvialis squatarola 2 1 1 Charadrius vocifents 2 Recurvirostra americana 1 Catoptrophorus semipalmalus 1 9 5 3 2 Numenius americanus 6 19 7 12 7 3 2 11 7 N. phaeopus Limosa fedoa 1 3 1 4 3 1 L.fedoa/N. phaeopus 3 1 1 2 2 Calidris alba 1 C. alba/alpina 1 Limnodromus sp. 2 25 5 1 Larus glaucescens/hyperboreus Larus sp. (large) 3 5 1 1 2 1 5 Larus sp. (small) 1 2 1 2 Uria aalge 2 1 1 Uria sp. 8 23 26 16 5 6 7 15 2 Cepphus columba 1 Strigiformes Strigidac (medium) 1 Otus kennicottii 1 Asio flammeus Asio sp. 1 1 Bubo virginianus Tyto alba 1 1 1 10 2 Corvus brachyrhynchos 2 2 1 5 3 6 5 29 1 C. corax 1 1 3 2 1 I : 202 1127 482 259 156 121 191 607 13C 1 9 2 1 2 1 1 1 8 1 2 5 5 9 3 5 2 1 4 29 1 1 1 6 2 1 2 3 3 1 27 6 1 2 5 8 11 5 112 1 1 2 1 1 1 1 4 23 1 10 1 1 6 1 40 1 1 2 4 3 3 6 1 35 1 1 8 1 1 6 10 2 1 14 17 9 6 167 1 1 1 1 1 2 2 1 3 1 4 4 9 6 3 2 10 36 13 1 5 15 14 7 109 1 1 6 5 1 22 523 7 28 111 7 9 1 1 18 10 41 374 569 449 313 5736 PREHISTORIC CALIFORNIA BIRDSPREHISTORIC CALIFORNIA BIRDS 13 Fig. 3. (A) Right coracoid of Gavia stellata (EMF A11526) and (B) left proximal carpometacarpus of Phoebastria albatrus (EMF 8635). A9151, PAHMA 12-1437), radius (EMF A8008), carpometacarpi (EMF A2688, A5216, A5561), synsacra (EMF A5899, PAHMA 12-1430), tibio- tarsi (EMF A8688, A5354, A12266). Remarks.-Identifications were based on criteria presented in Howard (1929). Proximal tibiotarsus of G. stellata is further distinguished from G. pacifica by a much deeper lateral under-' cut of the external articular surface. Gavia immer (Briinnich 1764) Referred material.-Reported in Howard (1929): coracoids (EMF 8168, A3117, 16882, A3231), humeri (EMF 16870, A4052, 6896), tarsometatarsi (EMF 7151, 8619, A4350). Additional elements: mandible (EMF A1204), scapula (EMF A9541), humerus (EMF 8722), radius (EMF A8678), carpometacarpus (EMF A11513), femur (EMF A5378), tarsometatarsi (EMF A6092, PAHMA 12-1427). Remarks. -Gavia immer is easily distinguished from G. pacifica and G. stellata by its large size. Although the average sizes of G. immer ele­ments are smaller than those of G. adamsii, their ranges overlap (see Fitzgerald 1980). However, postcranial materials more closely match G. immer than G. adamsii in size, and the mandible is referable to G. immer on the basis of the distinctive hook-shaped external articular process. Order Podicipediformes Family Podicipedidae Podiceps auritus (Linnaeus 1758) or P. nigricollis Brehm 1831 Referred material.-Reported in Howard (1929): synsacrum (EMF 8219; listed as "no. 2219" in Howard 1929). Additional elements: sternum (PAHMA 12-1434), coracoids (EMF A3055, A9220), humeri (EMF16883, A8696, A9213; PAHMA 12-1476), ulna (PAHMA 12­1437), tibiotarsi (PAHMA 12-1437,12-1437). Podiceps auritus (Linnaeus 1758) Referred material.-Cranium (EMF A206), humerus (EMF A5490), ulna (EMF A2825). Podiceps nigricollis Brehm 1831 Referred material. - Coracoid (EMF A6097), ulna (EMF A1879), carpometacarpus (PAHMA 12-1437), tibiotarsi (EMF 5359, A4253, 8022, 8357, A12624). Remarks. - Howard (1929) lacked sufficient reference specimens to confidently distinguish the small grebes represented in the Emeryville collection, though she thought that P. nigricollis and possibly P. auritus were represented. Both species are clearly present in the Emeryville fauna. The anterior cranium of P. auritus is distin­guished from that of P. nigricollis by having a deeper tip of the premaxilla and longer external nares. Compared with that of P. auritus, the car­pometacarpus of P. nigricollis has a more attenu­ated metacarpal I and a less extensive external projection of the pollical facet. The morphology of the cnemial crest of the tibiotarsus differs: in P. auritus, it tapers into a well-defined ridge along the proximal shaft; in P. nigricollis, it ter­minates abruptly, just proximal to condyles. The tibiotarsus of P. auritus is also distin­guished by a pronounced anterior depression just distal to the external articular surface and a much deeper tendinal groove of the distal end. The remaining elements were assignable to P. nigricollis or P. auritus on the basis of their very small or very large size, respectively.Aechmophorus occidentalis (Lawrence 1858) or A. clarkii (Lawrence 1858) Referred material. - Reported in Howard (1929) as A. occidentalis: ulna (EMF 10007), synsacra (EMF 8184, 8605), femora (EMF A4325, 6340), tarsometatarsus (EMF 10314), tibiotarsus (EMF A3275). Additional elements: cranium (EMF A6771), sterna (EMF A12672, A6515; PAHMA 12-1157), coracoid (EMF A10335), humerus (PAHMA 12-1437), radius (EMF A4701), ulna (PAHMA 12-1437), carpometacarpus (EMF A8538), synsacra (EMF A8583 A6258, A4110; PAHMA 12-1430, 12-1434, 12-1434, 12-1434), femora (EMF A185, PAHMA 12-1313), tibiotarsi (EMF A9209, A9221, A9238), fibulae (EMF A4017, 6279), tarsometatarsus (PAHMA 1-9777). Remarks.-In addition to size differences and the criteria presented in Howard (1929), the following features characterize A. occidentalis- clarkii and distinguish them from P. grisegena. (1) Cranium, posterior: The sagittal nuchal crest is well defined and extends posteriorly to form a sharp process. (2) Sternum: The coracoidal facets are farther apart at midline. (3) Coracoid: The coracoid is longer and thinner, with more pro­nounced ventral projections of the distal (furcu- lar) end. (4) Scapula: The coracoidal articulations are less developed and less anteriorly projecting, and the ventrolateral portion of the neck bears a more prominent depression. (5) Radius, proxi­mal: The region of the ulnar facet is more steeply sloping. (6) Femur: The femur is much stouter for its length, with more bulbous proximal and distal ends. Distally, the pronounced tubercle on the posterodistal surface in the popliteal area is not connected to the external condyle by a well- developed ridge. (7) Tibiotarsus: The proximal tibiotarsus differs by having a less laterally (fibular) extended external articular surface, an outer cnemial crest that extends distally past the external articular surface, and a profile that is rounded rather than oblong. (8) Fibula: The heads are larger, and the proximal shafts are thinner and flatter. Order Procellariiformes Family Diomedeidae Phoebastria albatrus (Pallas 1769) Referred material.-Carpometacarpus (EMF A8635 [Fig. 3B]). 14 Remarks.-The specimen is far too large to represent P. immutabilis or P. nigripes. Howard (1929) made one probable identification of this species, a radius, from the unprovenienced Emeryville sample. Family Procellariidae Fuimarus glacialis (Linnaeus 1761) Referred material.-Humerus (EMF 17286), radius (EMF 17277), ulna (EMF 17305). Remarks.- These elements are distinguished from Puffinus as follows. The distal humerus exhibits (1) a much deeper depression for the brachialis anticus, (2) a more rounded entepi- condyle, and (3) a less laterally projecting ect- epicondylar prominence. In addition to being straighter and more robust for its length, the radius differs by having distal ends that are more expanded and show an obvious "neck." The scapholunar facets of this element are also oriented at more of an angle (45°) to the long axis of the bone. The proximal ulna is much broader, the dorsal (palmar) projections of the internal and external cotylae are less pro­nounced, and the olecranon is less pointed and prominent. Order Pelecaniformes Family Pelecanidae Pelecanus sp. Referred material.-Humeri (EMF A3598, A3622, A4003, A195, A3626, A206), ulna (PAHMA 12-1476), carpometacarpus (EMF A3099), femur (EMF A3113), synsacra (PAHMA 1-9831, 12-1363). Pelecanus erythrorhynchos Gmelin 1789 Referred material.-Reported in Howard (1929): femur (EMF 8723). Pelecanus occidentalis Linnaeus 1766 Referred material.-Reported in Howard (1929): mandibles (EMF 5769, 7424), humeri (EMF 6760, 7397, 7398, 10420), ulnae (EMF 6303, 7775), femur (EMF 10582), tarsometatar­sus (EMF 7848). Additional elements: humerus ORNITHOLOGICAL MONOGRAPHS NO. 56 PREHISTORIC CALIFORNIA BIRDS 15 (EMF A11185), radius (EMF A2457), ulnae (PAHMA 1-9795, 12-1329), carpal digit 2 pha­lanx 1 (EMF A3105). Remarks.-The additional elements of P. occi­dentalis are distinguished from those of P. erythro­rhynchos in the following ways. The radius differs by having (1) a convexity or bossing just medial to the prominent fossa on the palmar aspect of the distal end, (2) a sharp line or ridge extend­ing along the distal shaft, and (3) a prominent rounded eminence just proximal to the scaph- olunar facet on the palmar surface. The ulna has a flatter palmar aspect of the proximal shaft, just distal to the external cotyla, and lacks a marked depression in that region. The small size, alone, of carpal digit 2 phalanx 1 distinguishes P. occi­dentalis from P. erythrorhynchos. The humerus was identified by criteria in Howard (1929). Family Phalacrocoracidae Phalacrocorax sp. Referred material.-A total of 608 specimens, including all major elements of the skeleton, most representing chicks and juveniles (Table 3). Phalacrocorax penicillatus (Brandt 1837) or P. auritus (Lesson 1831) Referred material. - Cranium (EMF 17487), ulna (EMF 7971), synsacra (EMF A157, 7370, 8364), tarsometatarsus (EMF 9262). Phalacrocorax penicillatus (Brandt 1837) Referred material.-Reported in Howard (1929): mandibles (EMF 8313, 8329, 8337, 8639), coracoid (EMF 8265), scapula (EMF 8302), humeri (EMF 7379, 7432, 7969, 8157, 8267), radii (EMF 8025, 8028, 5326, 7391, 7980, 8004), synsacra (EMF 5365, 7774, 7819), femora (EMF 6888, 7065, 6749, 7443, 8196), tibiotarsus (EMF 5322), tarsometatarsi (EMF 6363, 10122, A3458, 6360, 8583, 9157). Additional elements: crania (EMF A10822, A10824, A10797; PAHMA 12­1356), sternum (EMF A10125), coracoid (EMF A1143), scapula (EMF 8594), humeri (EMF 8293, 9891; PAHMA 1-9736), radii (EMF A9229, A9210, A10484; PAHMA 12-1437), ulnae (EMF 6350, PAHMA 12-1437), carpometacarpus (EMF A5274), femora (EMF 8910, A2695, A5479, A11610), tibiotarsi (EMF A5244 [Fig. 4A], A11311, A11502, A8194, A11504; PAHMA 12­1453), tarsometatarsus (EMF A11534). Phalacrocorax auritus (Lesson 1831) Referred material.- Reported in Howard (1929): crania (EMF 7255, 8335, 8553, 9918, 10141, 10545), coracoids (EMF 8735, 9227, 7030, 6344, 7766, 7790, 8282, 8327, 8727, 8733, 9232, 10110, 10287, 10527), humerus (EMF 8718), radii (EMF 6316, 8018, 8041, 8244, 6290, 7832, 8356, 9208, 10587, 8211), carpometacarpi (EMF 8007, 8040, 8347, 7983, 8000, 8001, 8836, 9827), synsacra (EMF 7256, 5793, 5364, 5366, 6254, 6319, 6781, 8856, 9917), femora (EMF 5332, 5206, 8740, 6335, 8608, 8710), tarsometatarsi (EMF 6747, 7157, 5344, 6325, 6343, 8012, 8580, 9174, 10520, A3264), tibiotarsi (EMF 6305, 7083, 5325). Additional elements: crania (EMF A2486, A2852, 5369; PAHMA 12-1345, 12-1363), mandibles (EMF A9104, A88, 17103, A11242, 9980, A2102; PAHMA 12-1441), carpometa­carpi (EMF A11368, A10482, A10585, A9523, A10573; PAHMA 12-1437), coracoids (EMF 17303, A100, A8725, A10897, A11284, A11321, A8726, A12639, A10889, A10835, A11296, A11312, A11306, 7811; PAHMA 12-1437, 12­1309, 1-9840), scapulae (EMF 17298, A91, 8271, 8734, 9247, 10171, 10436, 10422, A8628, A12753, A12767, A7686, A53, A12602; PAHMA 12-1449, 12-1157), humeri (EMF A824, A867, A4034, 7059, 5762, 6262, 6265, 6775, 7437, 7781, 7807, 7831, 7988, 8204, 8285, 8291, 8582, 8709, 8725, 8736, 9874, 9905, A3230, A3261, A534, A854, A lcm B Fig. 4. (A) Left proximal tibiotarsus of Phalacrocorax penicillatus (EMF A5244) and (B) left femur of P. auri­tus (PAHMA 12-1462).16 ORNITHOLOGICAL MONOGRAPHS NO. 56 5776, 8254, 8600, 8720, 9223, 9233, A1730, A2389, 8721, A12622, A10315, A11289, A7153; PAHMA 1-9823), radii (EMF A83, A4073, A2409, A2402), ulnae (EMF 5783, 7409, 8610, 7760, 7963, 8595, A237, A8640; PAHMA 1-9721), synsacra (EMF A11238, A11439, A11443; PAHMA 12-1430), femora (EMF 6302, 8637, 10166, A62, A70, 17429, 16968, 16971, A74, A85, A97, 6910, 5777, 5784, 5800, 5337, 5350, 5336, 5338, 5339, 6271, 6286, 6365, 6754, 7380, 7765, 7808, 7809, 8823, 8832, 8843, 8831, 9445, 9164, 9204, 9158, 9187, 9197, 9886, 10536, A1929, A2095, A2459, A1546, A1715, A2097, A2106, A2541, A2951, A3315, A543, A8750, A10380, A9546, A10675; PAHMA 12-1437, 12-1309, 12-1462, 12-1462 [Fig. 4B], 1-9831, 1-9832, 1-9838), tibiotarsi (EMF 5343, A8622, 16957, A694, A3101, A2820, A9217, A10579, A10600, A10609; PAHMA 12-1437, 12-1456, 12-1157), tarsometatarsi (EMF A9581, A9617, A8634, A8650, A9554, A11308, A11317, A10451, A10464, A11508, 5360, 7460, 9199; PAHMA 1-9840). Phalacrocorax pelagicus Pallas 1811 Referred material.-Reported in Howard (1929): femur (EMF 8262), tibiotarsus (EMF 7976). Additional elements: coracoids (EMF A1719, A10591), humeri (EMF A82, PAHMA 1­9760), carpometacarpus (EMF A10850), femora (EMF A838, A10368). Remarks. - (1) Cranium: P. auritus has a sharply defined external midsagittal occipital crest not present in P. pelagicus or P. penicillatus. Phalacrocorax penicillatus is distinguished from P. auritus by smaller, less anteriorly projecting postorbital processes. The rostrum of P. peni­cillatus is longer, less deep dorsoventrally at the base, and narrower than that of P. auritus. Paired grooves extending distally from external nares are deeper in P. penicillatus than in P. auri­tus. The cranium of P. pelagicus is easily distin­guished by its small size. (2) Mandible: Criteria used to distinguish mandibles of P. penicillatus from those of P. auritus are in Howard (1929); P. pelagicus is easily distinguished from the for­mer species by its small size. (3) Sternum: The ventral manubrial spine is larger and thicker (mediolaterally enlarged) in P. penicillatus as compared with both P. auritus and P. pelagicus. Pneumatic foramina on the dorsal surface of the anterior sternum are present in P. auritus and P. penicillatus but lacking in P. pelagicus. (4) Furculum: An intramuscular line on the internal shaft approaches the posterior border more abruptly in P. auritus than in P. penicillatus (Howard 1929). Phalacrocorax pelagicus is like P. penicillatus in this feature, but the line is located more anteriorly on the shaft. (5) Coracoid: The head is thicker mediolaterally in P. auritus than in P. penicillatus (cf. Howard 1929) and P. pelagicus. Also, the neck between the glenoid facet and the bicepital attachment is markedly depressed in P. auritus and P. penicillatus but relatively flat in P. pelagicus. (6) Scapula: The internal margin of the acromion is smoothly rounded from blade to dorsal tip in P. auritus, but angular in P. penicillatus (Howard 1929) and P. pelagicus. In addition, the medial portion of the neck just distal to the acromion bears an elongated furrow in P. pelagicus that is absent in P. penicillatus and P. auritus. (7) Humerus: Proximal end: The capital groove is deeper in P. auritus than in P. penicillatus (Howard 1929); the bicepital crest (anconal view) forms a more well-defined ridge in P. penicillatus than in P. auritus and P. pelagicus; the internal tuberos­ity is larger, with a steeper internal face in P. auritus and P. pelagicus than in P. penicillatus; and the distal face projects farther distally in P. pelagicus and P. penicillatus than in P. auri­tus. Distal end: In addition to the features described in Howard (1929), P. penicillatus and P. auritus differ from P. pelagicus by having an entepicondyle with two prominent tubercles separated by a well-defined depression. (8) Radius: Criteria in Howard (1929) were used to distinguish P. penicillatus from P. auritus; P. pelagicus was distinguished by its smaller size. (9) Ulna: The external cotylae of the proximal end are longer, more sharply hooked, and undercut in P. auritus as compared with P. penicillatus. The palmar margin of the internal condyle is smoothly rounded in P. pelagicus but flatter with a sharp bend medially in P. auritus and P. penicillatus. I failed to observe consistent criteria to distinguish the distal ulna among these species. (10) Carpometacarpus: P. auritus differs from P. penicillatus and P. pelagicus by having a more sharply defined posterodistal limit of the internal articular ridge of the car­pal trochlea, and a tubercle at the same limit of the parallel articular ridge. In addition, the intermetacarpal space is narrower distally in P. auritus and P. pelagicus than in P. penicillatus. (11) Synsacrum: Criteria in Howard (1929) werePREHISTORIC CALIFORNIA BIRDS 17 used to distinguish P. penicillatus from P. auri- tus; the iliac process in P. pelagicus is like that of P. penicillatus. (12) Femur: The posterior aspect of the trochanter is more rugose in P. penicil­latus than in P. auritus. Moreover, the anterior aspect of the proximal end is less depressed in P. auritus than in P. penicillatus (Howard 1929); P. pelagicus shows an intermediate expression of this feature. The anteromedial section of the proximal shaft, just distal to the head, is marked by a distinctive groove or depression in P. penicillatus but is smooth to slightly rough­ened in P. auritus; this region exhibits a small tubercle in P. pelagicus. Relative depths of the external and fibular condyles of the distal end are distinctive in P. auritus and P. penicillatus, as described by Howard (1929); P. pelagicus is like P. penicillatus in this feature. (13) Tibiotarsus: Criteria in Howard (1929) were used to distin­guish P. penicillatus from P. auritus; the tibiotar­sus of P. pelagicus is substantially smaller than that of those two species. (14) Tarsometatarsus: Criteria in Howard (1929) were used to distin­guish P. penicillatus from P. auritus. Also, troch­lea for digit III projects less anteriorally in P. auritus than in P. penicillatus or P. pelagicus. The tarsometatarsus of P. pelagicus is distinctively short and stout, compared with those of P. auri­tus and P. penicillatus. Order Ciconiiformes Family Ardeidae Botaurus lentiginosus (Rackett 1813) Referred material.-Scapula (EMF A10870). Remarks.-The specimen compares in size with Nycticorax nycticorax and B. lentiginosus. However, the outline of the proximal mar­gin between the furcular articulation and the coracoidal articulation is gently curved, as in B. lentiginosus, not sharply bent, as in N. nyc­ticorax. This species was not identified in the sample of Emeryville material that Howard (1929) examined. Ardea herodias Linnaeus 1758 Referred material.- Reported in Howard (1929): scapulae (EMF 7252, 8849), femur (EMF 6778), tarsometatarsi (EMF 7475, 7975, 8299). Additional elements: cranium (EMF A1852), mandibles (EMF A10436, 8199; PAHMA 12-1449), humeri (EMF A1732, A9520 [Fig. 5]; PAHMA 12-1476), radii (EMF 5371, A3274), femur (EMF A3648). Remarks.-The large size of the elements rules out all other ardeids, including A. alba. Family Cathartidae Cathartes aura (Linnaeus 1758) Referred material.-Reported in Howard (1929): radius (EMF 10544), synsacrum (EMF 8287). Additional elements: humerus (EMF A9165), synsacrum (EMF A10116). Remarks.-The synsacrum and humerus are too large for Coragyps atratus and too small for Gymnogyps californicus. The G. californicus mate­rial that Howard (1929) identified lacks within- site provenience information. Order Anseriformes Family Anatidae Subfamily Anserinae (small) Referred material.-A total of 536 miscella­neous elements are represented. Subfamily Anserinae (medium) Referred material.-A total of 1,212 specimens representing all major elements of the skeleton were identified. 1 cm Fig. 5. Left proximal humerus of Ardea herodias (EMF A9520).18 ORNITHOLOGICAL MONOGRAPHS NO. 56 Remarks.-Given the extensive intraspecific variation in the osteological characters of geese, species-level identifications were made only for anserine cranial elements and postcranial elements so large as to rule out all taxa but the largest subspecies of Branta canadensis (e.g. B. c. moffitti). Specimens identified as small anserines are similar in size to B. bernicla, B. hutchinsii min­ima, and Chen rossii. Medium anserines match the size of C. caerulescens caerulescens, Anser albi­frons, and small subspecies of B. canadensis (i.e. B. c. parvipes). Chen caerulescens (Linnaeus 1758) Referred material. - Crania (EMF 7852, 7876 [Fig. 6A], 10415, A3224), mandibles (EMF A109, 17101, 17388, 6940, 5367, 7460, 9971, 10010, 10337, 10340, A1495, A2464, 17499, 8036, A516, A923, 6824, 7407, 8223, 10291, A11866, A8173; PAHMA 12-1348, 12-1454, 12-1454). Remarks. -(I) Cranium: C. caerulescens is dis­tinguished from A. albifrons and B. canadensis by a greater dorsoventral depth of the premaxilla, a more anterodorsally depressed frontal, smaller anterior supraorbital processes, and a steeper slope of the anterior margin of the interorbital septum. (2) Mandible: The dentary is distin­guished from that of other geese by a greater depth and thickness of the body and a deeper ventrolateral groove. The posterior mandible is larger and more robust than that of A. albifrons, and distinguished from that of larger subspe­cies of B. canadensis by a shorter length between the articular facet and the coronal process. The anterior extension of the external articular pro­cess is more prominent than in Branta or Anser. The large size of the elements rules out C. rossii. 1 cm Fic. 6. (A) Premaxilla of Chen caerulescens (EMF 7876) and (B) frontal of Branta canadensis (EMF A5806). Chen rossii (Cassin 1861) Referred material.-Mandibles (EMF A9798, A4700, A178, A3621, A1861, A6984, A7127; PAHMA 12-1437). Remarks.-The mandibles of C. rossii are eas­ily distinguished from those of all other geese by having great depth and thickness for a very short length. Branta hutchinsii (Richardson 1832) cf. minima Referred material.-Mandibles (EMF A8177; PAHMA 12-1455,12-1441). Remarks.- The mandibles of B. hutchinsii minima are smaller and more gracile than those of all other geese. Branta canadensis (Linnaeus 1758) cf. parvipes Referred material. - Crania (EMF A5806 [Fig. 6B], 7849, 8015, 8016, A2485, A2853, 10412; PAHMA 12-1342, 1-9802), mandibles (EMF A6792, A638, 10138, 10413, 17232; PAHMA 12­1449). Remarks.-(1) Cranium: The crania of B. canadensis are distinguished from those of A. albi­frons and C. caerulescens by a narrow and mark­edly roughened or sculptured frontal between the orbits, typically with a slight ridge or line of bone encircling the dorsal margin of the orbits (Fig. 6B). These specimens are too large for B. bernicla and B. hutchinsii and too small for B. c. moffitti. (2) Mandible: The mandible specimens are too small and gracile to represent A. albifrons, C. caerulescens, or B. c. moffitti, and too large for B. hutchinsii. They are distinguished from those of C. rossii by a thinner dentary, a more attenuate coronal process, and a more elongate and anteri­orly tapered external articular process. They are distinguished from B. bernicla by a more vertically oriented internal articular process, a higher coro- noid process, and a more robust coronal process. Branta canadensis (Linnaeus 1758) cf. moffitti Referred material.-Mandibles (EMF 8047, A514, A6328, A11813), coracoids (EMF A1031, A855, A4028, 6311; PAHMA 1-9833, 1-9848), furculum (PAHMA 1-9848), scapulae (EMF A71, 6263), humeri (EMF A2536, A484, 7031; PAHMA 1-9783, 1-9833), ulnae (EMF A96, PAHMA 12-1462). Remarks.-The mandibles of B. canadensis cf.moffitti are distinguished from those of C. cae­rulescens as noted above, and from A. albifrons by a greater distance between the articular facet and the coronal process. Identifications of postcranial elements listed above were based on their large size. Subfamily Anatinae (small) Referred material.-Atotalof 113 specimens rep­resenting all the main elements of the skeleton. Subfamily Anatinae (large) Referred material.-A total of 1,263 elements were identified. Remarks.-Genus- or species-level identifica­tions of ducks were attempted for all cranial specimens but for only a subset of postcranial elements. That subset includes the humerus, sternum, tarsometatarsus, and synsacrum. These are among the most diagnostic anati- nine postcranial elements (Woolfenden 1961). Osteological criteria are presented that distin­guish these elements among the different duck genera represented at Emeryville: Anas, Aythya, Mergus, Melanitta, Bucephala, and Oxyura. Distinctive features of these elements were also observed for genera not recovered from the site (Aix, Somateria, Polysticta, Clangula, Lophodytes, Histrionicus), mostly following Woolfenden (1961), but are not described here. All other postcranial duck elements were assigned to one of two broad size categories: Anatinae (small) or Anatinae (large). Small anatinines are similar in size to Anas sp. (teals), Bucephala albeola, and Oxyura jamaicensis; the Anatinae (large) category may include any number of the larger duck species. Anas sp. Referred material.-Humeri, sterna, synsacra, and tarsometatarsi: 51 elements. Remarks. - Specimens identified as Anas sp. could represent any number of the Anas species larger than the teals (Anas cyanoptera, Anas dis- cors, Anas crecca). Features used to distinguish Anas from the other duck genera are as follows. (1) Mandible: The postarticular process is long and thin, with a long, tapering proximodorsal spine; the lateral cotylae are poorly developed. (2) Sternum: The ventral manubrial spine is long and peg-like, and projects anterodorsally; PREHISTORIC CALIFORNIA BIRDS pneumatic foramena are round or ellipti­cal (Woolfenden 1961); the dorsal manubrial spine is small and typically chevron-shaped. (3) Humerus: The bone is thick for its length; the internal tuberosity is robust; the pneumatic foramen is typically deeply open, with bony struts visible inside; the facet for the anterior articular ligament is elevated (Woolfenden 1959, 1961). (4) Synsacrum: The ventral surface of the anterior ilium grades smoothly into the ischium; the median dorsal ridge is broad and flat; a well-defined ridge is absent along the ventromedial surfaces of the anterior synsacral vertebrae. (5) Tarsometatarsus: Distinguished from that of Aythya by a much thinner bone for a given length and a less acute angle formed between the proximal margin of the trochlea for digit II and the adjacent distal shaft. The proximal ligamental attachment is small as compared with Mergus. The element is not as stout for a given length as that of Bucephala. As distinguished from Melanitta, the external cotyla does not extend into the anteroproximal portion of the shaft, and the ridge on the correspond­ing lateral portion of the shaft is not as sharp. 1 could not observe criteria to distinguish the distal tarsometatarsi of Anas and Melanitta. Anas sp. cf. Anas discors Linnaeus 1766 or Anas cyanoptera Vieillot 1816 or Anas crecca Linnaeus 1758 Referred material.-Cranium, mandibles, sterna, synsacra, and tarsometatarsi: 21 elements. Remarks.-Their small size makes the teals (Anas cyanoptera, A. discors, A. crecca) distinctive as a group from the other species of the genus; species-level identifications were not, however, attempted among them. The cranial specimen was distinguished from B. albeola by (1) a less ventrally angled foramen magnum and (2) a more vertically oriented transverse nuchal crest. The posterior cranium of Oxyura is substantially larger. Anas platyrhynchos Linnaeus 1758 Referred material.-Sternum (EMF A6278), humeri (EMF 8575, 7032, 9948, A10318, A10582, A7034, A8624, A3464, A5353, 6767, 8729, 10099, A8529; PAHMA 12-1314), synsacra (EMF A5886, A2904; PAHMA 12-1430), tarsometatarsus (EMF A11556). 19 Remarks.-These elements are substantially larger than those from all other Anas species. Anas clypeata Linnaeus 1758 Referred material. - Mandibles (EMF A4444, A 5607), sternum (EMF 7444), humerus (PAHMA 1-9722), synsacrum (EMF 9900). Remarks.- The lateromedially enlarged dis­tal end of the dentary distinguishes A. clypeata from all other species of the genus. The other elements were identified on the basis of size: too small to represent A. americana or A. strepera but too large to represent Anas sp. (teal). Aythya sp. Referred material.-Mandibles, humeri, sterna, synsacra and tarsometatarsi: 23 specimens. Remarks.-Features used to distinguish Aythya from Anas, Melanitta, Bucephala, and Mergus are as follows. (1) Mandible: Compared with those of Anas, the postarticular process has greater depth and the lateral cotylae are larger. The flattened ventral surface of the dentary extends farther proximally than in Melanitta and Anas, and the element is larger and more dis­tally flaring than those of Bucephala and Oxyura. The dentaries of Mergus are distinctively long and narrow. (2) Sternum: The ventral manubrial spines, if present, are paired, short, thin, and pointed. (3) Humerus: The humerus is thinner and narrower for a given length than in Anas. On the proximal end, the internal tuberosity is near the height of the head when the bone is laid flat, palmar side down; the head is minimally undercut; the anconal aspect of the deltoid crest is relatively smooth and straight laterally; the bicepital crest is straight, not flaring; the pneu­matic foramen is deep, but lacks bony spicules, as in Anas (Woolfenden 1961). On the distal end, the olecranon fossa is not as deep and the margins are not clearly defined as in Melanitta; the depression for the brachialis anticus is well developed, with a sharply defined distomedial rim; the internal condyle is more deeply under­cut proximally than in Anas; the intercondylar furrow is confluent with the olecranal fossa, not separated by a transverse ridge as in Melanitta (Woolfenden 1961); the attachment of the ante­rior articular ligament is less elevated than in Anas (Woolfenden 1961). (4) Synsacrum: The pre-acetabular ala of the ilium is steep and 20 narrow, and a flat plateau of the ventral surface of the synsacral vertebrae begins near the first lumbar vertebrae and narrows anteriorly to form a sharp ridge along the synsacral thoracic vertebrae. (5) Tarsometatarsus: Distinguished from that of Anas by the criteria above, and from that of Mergus by a substantially shorter length, thicker bone, and smaller most-proximal ligamental attachments. Distinguished from Melanitta by a greater width for a given length (Woolfenden 1961). The smaller species of Aythya (A. marila, A. affinis) are most similar to Bucephala; the latter is distinguished from Aythya by a more deeply incised anteroproximal shaft near the proximal foramina, smaller and less laterally rotated trochlea for digit III, and a much smaller protuberance on the anconal sur­face just distal to the internal cotyla. Aythya sp. (large) Referred material.-Mandible, sternum, and humeri: 4 specimens. Aythya cf. A. valisneria (Wilson 1814) Referred material.-Humerus (EMF 8738), and synsacrum (EMF A6496). Aythya valisneria (Wilson 1814) Referred material.-Cranium (EMF 16927) and humerus (EMF 8200). Aythya cf. A. marila (Linnaeus 1761) Referred material.-Mandibles (EMF A3540, A511), humerus (EMF 17241), and synsacrum (EMF 12-1434). Aythya marila (Linnaeus 1761) Referred material.-Mandibles (EMF A7406, PAHMA 12-1449). Aythya affinis (Eyton 1838) Referred material.-Humerus (A12768), synsa­crum (A6533). Remarks. - Mandibles of A. marila are distin­guished from those of A. americana and A. col- laris by a broader, deeper, and more flared distal end; and from those of A. valisneria by a greater ORNITHOLOGICAL MONOGRAPHS NO. 56 PREHISTORIC CALIFORNIA BIRDS 21 depth and shorter length. In comparison with those of A. affinis, they are deeper and thicker for specimens of comparable length. Other Aythya species-level identifications are based on size. Melanitta sp. Referred material.-Crania, humeri, synsacra, sterna, and tarsometatarsi: 466 elements. Remarks. - (1) Mandible: Melanitta is distin­guished from all other duck genera by a dorsoven- trally flattened distal end of the dentary, with an elongated cavity on the ventral surface proximal to the symphysis. Melanitta is also distinguished by a short, proximally squared-off postarticular process with a short proximodorsal spine and a maximum depth at the surangular (angle of man­dible) that is greater than in the largest Anas and Aythya. Melanitta also exhibits well-developed lateral cotylae. (2) Sternum: Distinguished by a lack of ventral manubrial spines and by small, paired, and widely spaced dorsal manubrial spines. (3) Humerus: On the proximal end, the head is undercut anconally, but not as deeply as in Bucephala; the internal tuberosity does not extend above the head, as in Bucephala; the deltoid crest extends farther distally, with the distal end prominently flared (Woolfenden 1961). On the distal end, the attachment of the anterior articu­lar ligament is much less elevated than in Anas; the olecranal fossa is large, deep, and rectangu­lar in shape, with steep sides and well-defined margins; the intercondylar furrow is separated from the olecranal fossa by a transverse ridge; the olecranal fossa is shallower and less well defined in Anas and Aythya of similar size; Mergus is most similar to Melanitta, but the walls of the olecranal fossa are not as steep. (4) Synsacrum: Distinguished from all other genera by large size; by a well-defined midsagittal ridge run­ning along the ventral surface of the thoracic and lumbar synsacral vertebrae; by a greater length of the synsacral sacral column; and by a greater width of the ilium at the parapophysis of the last sacral synsacral vertebrae. (5) Tarsometatarsus: This element is distinctively long and thin as compared with those of other duck genera, with the exception of Mergus. On the distal end, it is distinguished from that of Mergus by a greater proximal extension of the posterior surface of the trochlea for digit II; by a greater relative depth of the lateral-posterior ridge of the trochlea for digit IV, as compared with the medial-posterior ridge of that trochlea; and by a broad, not taper­ing, posteroproximal extension of the trochlea for digit III. On the proximal end, the internal cotylae are deeper, with more pronounced margins than in Mergus. Melanitta perspicillata (Linnaeus 1758) or M. fusca (Linnaeus 1758) Referred material.- Crania (EMF A6779, A6782, A6785, A6786, A6794, A6776; PAHMA 1-9780). Melanitta perspicillata (Linnaeus 1758) Referred material.-Crania (EMF A1102, A5992; PAHMA 12-1157,12-1354,12-1430). Melanitta fusca (Linnaeus 1758) Referred material.- Crania (EMF A5206, A6773, A6775, A6778, A6780, A6787; PAHMA 12-1354, 12-1157). Remarks.-Species-level identifications were made for Melanitta cranial material using the following criteria. The minimum interorbital breadth of the frontal bone is greater in M. fusca than in M. nigra and M. perspicillata (Table 4). Two specimens with frontal breadths of 10.52 mm (PAHMA 12-1354) and 10.92 mm (EMF A5206) are beyond the range of M. nigra and M. perspi­cillata and were assigned to M. fusca accordingly. The length of the premaxilla, measured from bill tip at the midline to anterior margin of external nares, is also greater in M. fusca than in M. nigra and M. perspicillata (Table 5). With a premaxilla width of 27.08 mm, specimen PAHMA 12-1157 Table 4. Minimum interorbital frontal breadths (mm) for recent Melanitta nigra, M. perspicillata, and M. fusca specimens*. Species n Mean Range SD SE M. nigra 5 7.21 6.73-7.50 0.300 0.134 M. perspicillata 15 7.62 5.95-9.60 0.948 0.245 M. fusca _______13 9.60 7.86-10.72 0.899 0.249 ‘Specimens from MVZ, UWBM, and UMNH (sec text).Table 5. Premaxilla lengths (mm) for recent Melanitta nigra, M. perspicillata, and M. fusca specimens3. 22 ORNITHOLOGICAL MONOGRAPHS NO. 56 Species n Mean Range SD SE M. nigra 6 21.423 20.500-22.230 0.635 0.259 M. perspicillata 17 21.721 19.110-23.780 1.769 0.429 M.fusca _ 22 26.223 23.140-30.350 1.723 0.367 •'Specimens from MVZ, UWBM, and UMNH (see text). is beyond the range of M. nigra and M. perspicil­lata and was identified as M. fusca. In addition, lachrymal bones in M. fusca are large and contain prominent sinuses; these bones are much smaller and lack sinuses in M. perspicillata and M. nigra. Posteriorly oriented supraorbital processes are well developed in all three species, but are longest and thinnest in M. nigra. The frontal is flat along its entire length in M. perspicillata and M. nigra; in M. fusca, the anterior portion slopes ventrally at the supraorbital processes. The anterior frontal exhibits a deep, well-defined midsagittal groove in M. perspicillata and M. fusca, but is smoothly concave inM. nigra. Melanitta fusca exhibits a deep depression just medial to the postorbital process in the posterior wall of the orbit, a feature lacking in M. perspicillata and M. nigra. Muscle attach­ments of the posterior cranium (e.g. crista tempo­ralis and nuchal crests) are more pronounced in M. fusca than in M. nigra and M. perspicillata. The temporal fossae are more triangular-shaped and converge to a point, dorsally, in M. fusca; they are more constricted in M. perspicillata, forming an elongate and rounded dorsal end. Bucephala sp. Referred material.- Humerus (EMF 10131). Remarks.-The specimen may be a very large B. albeola or a very small B. clangula-islandica, or may represent a hybrid. Bucephala albeola (Linnaeus 1758) Referred material.-Sternum (A5918) and humeri (EMF A1040, A90, A3130, 7800; PAHMA 12-1437). Bucephala clangula (Linnaeus 1758) or B. islandica (Gmelin 1789) Referred mat erial.-Sterna (EMF 17263, A5919, A5808, A12276; PAHMA 12-1157, 12-1157), humeri (EMF 17309, A4206, 10519; PAHMA 12­1437), synsacrum (EMF 8300), tarsometatarsus (PAHMA 12-1437). Remarks.-Bucephala albeola is distinguished from B. clangula-islandica on the basis of small size. Features that distinguish these Bucephala ele­ments from those of Anas, Aythya, and Melanitta are described above; features that distinguish them from those of Mergus and Oxyura are as follows. (1) Sternum: The ventral manubrial spine is absent, and the paired dorsal manubrial spines are very small and widely spaced. The coracoidal sulcus has a strong ventral projection of the ventral lip and a sharp posterior curve of the lateroventral lip. The carina is strongly projected anteriorally, not to the extreme seen in Mergus (see below), but more pronounced than in any other duck genus. In O. jamaicensis, the dor­sally projected ventral manubrial spine is short, squared off, and bifurcated distally. (2) Humerus: The head is deeply undercut anconally, and the internal tuberosity does not rise above it when the bone is placed flat, palmar side down. The lateral margin of the deltoid crest is concave; the bicepital crest is widely flaring, with a proximal depression; the pneumatic foramen is closed and internally smooth. The distal end is very similar to that of the smaller Melanitta, but the internal condyle is more deeply undercut proximally, and the entepicondylar prominence is less strongly margined laterally. (3) Synsacrum: Like Melanitta, Bucephala and Oxyura both have sharp ventral ridges along the lumbar and thoracic synsacral vertebrae. Bucephala lacks the strong ventral projection of the ilium at the pectineal process as found in Oxyura. (4) Tarsometatarsus: Bone length is consistently shorter than in Mergus, but longer than in Oxyura; it also lacks the heavily sculptured anterior surface of the trochlea for digit 3 found in the latter. Mergus sp. Referred material.-Two humeri. Mergus cf. M. serrator Linnaeus 1758 Referred material.-Humerus (EMF A5267), tarsometatarsus (EMF A4260).PREHISTORIC CALIFORNIA BIRDS 23 Mergus serrator Linnaeus 1758 Referred material.-Sterna (EMF A6276, A10787), synsacra (PAHMA 12-1441, 12-1336). Remarks. -(1) Sternum: In Mergus, this ele­ment displays the greatest anterior projection of the carina of any anatinine genus and has no ventral manubrial spine. The anterior carinal margin of M. serrator differs from that of M. merganser in having a triangular depression just ventral to the coracoidal sulcus. (2) Humerus: The deltoid crest is "sharply angular" in Mergus (Woolfenden (1961). The pneumatic foramen is open, with bony struts, more deeply excavated than in Anas. In M. merganser, the proximal head is undercut anconally by a distinctive crescent­shaped depression (for external head of triceps), and a clear bony ridge separates the depression from the capital groove; M. serrator lacks this feature. (3) Synsacrum: In Mergus, this element is distinguished by a strongly waisted ilium just anterior to the acetabulum, a strong anterior projection of the pectineal processes, and a prominent ventral projection of the ilium just anterior to the acetabulum. Mergus serrator is distinguished from M. merganser by its smaller size. (4) Tarsometatarsus: This element is most similar to that of Melanitta, and differences are described above. Identification of M. cf. serrator is based on the small size of the element. Oxyura jamaicensis (Gmelin 1789) Referred material.-Sterna (EMF A9115, PAHMA 12-1157), synsacrum (EMF A9098). Remarks.-Sterna are distinguished by the criteria described under Bucephala above. The synsacrum in O. jamaicensis is easily distin­guished from that of all other smaller ducks (B. albeola, A. affinis, Anas sp. [teals]) by a strong ventral projection of the ilium just anterior to the acetabulum. Order Falconiformes Family Accipitridae Elanus leucurus (Vieillot 1818) Referred material.-Reported in Howard (1929): coracoid (EMF 6387), ulna (EMF 8366). Additional element: tibiotarsus (PAHMA 12-1476). Remarks.-The tibiotarsus in Elanus differs from that of all other genera of accipitrids and falconids by having (1) a prominent depression on the posterior shaft just proximal to the poste­rior intercondylar sulcus and (2) a more limited proximal extension of the intercondylar sulcus. Haliaeetus leucocephalus (Linnaeus 1766) Referred material.-Reported in Howard (1929): coracoid (EMF A2999), humerus (EMF A3100), carpometacarpi (EMF A3059, 8708). Additional elements: sternum (PAHMA 12­1427), furculum (EMF A317), coracoid (EMF A194), femur (EMF 7757). Remarks.-Distinguished from Aquila chrys- aetos as follows: (1) the coracoid is longer and lacks a deeply undercut furcular facet; (2) the furculum has a longer groove ventral to the furcular process; (3) the right coracoidal sulcus of the sternum extends medially onto the dorsal aspect of the ventral manubrial spine; and (4) the external condyle of the femur has a greater anteroposterior depth and posteroproximal extension. Circus cyaneus (Linnaeus 1766) Referred material. - Coracoids (EMF 16964, A11322). Remarks.-The coracoid is shorter and stouter than in Accipiter cooperi and A. striatus. Accipiter cooperi (Bonaparte 1828) Referred material.-Mandible (EMF A9237). Remarks.-The dentary is shorter, stouter, and less decurved than in Elanus and Circus and is too large to represent A. striatus. Buteo lineatus (Gmelin 1788) Referred material.-Radius (EMF 5358), ulnae (EMF A471, 5311), femur (PAHMA 12-1437). Remarks.-Six Buteo species occur in California, and they range in size from smallest to largest as follows: B. lineatus, B. swainsoni, B. lagopus, B. jamaicensis, and B. regalis. The radius of B. linea­tus is distinguished by its very small size and by having a smooth and rounded ligamental protu­berance of the distal end. The ulnae identified as B. lineatus are substantially smaller than those of B. swainsoni. The femur was distinguished by24 ORNITHOLOGICAL MONOGRAPHS NO. 56 its very small size and a squared-off distolateral process of the fibular condyle. Buteo jamaicensis (Gmelin 1788) Referred material.-Reported in Howard (1929): sternum (EMF 9884), humeri (EMF 7838, 7788, 8851, 9940, 10270), synsacrum (EMF 9998), tarsometatarsi (EMF 9986, 9999). Additional ele­ments: tarsometatarsi (EMF A10328, A1021). Remarks.-The tarsometatarsi were distin­guished from those of B. regalis by greater length, thinner width, and characters described in Howard (1929); the specimens are too large to represent B. lagopus. Buteo jamaicensis (Gmelin 1788) or B. lagopus (Pontoppidan 1763) Referred material.-Carpometacarpus (EMF 5789), femur (EMF 17098). Remarks.-The elements are too large to rep­resent B. swainsoni, too small to represent B. regalis, and within range of both B. jamaicensis and B. lagopus. Buteo jamaicensis (Gmelin 1788) or B. regalis (Gray 1844) Referred material.-Coracoids (EMF A2125, A5251), humerus (PAHMA 12-1476), femur (EMF A3119), tibiotarsi (EMF 7338, 6248). Remarks.-The specimens are too large to rep­resent B. swainsoni, B. lagopus, or B. lineatus. Buteo regalis (Gray 1844) Referred material.-Radius (EMF 8635), tib­iotarsus (EMF A86), tarsometatarsus (PAHMA 12-1449). Remarks.-The width of the radius head (7.60 mm) exceeds that of 32 measured B. jamai­censis reference specimens and falls within range of B. regalis. The tibiotarsus and tarsometatarsus were identified by characters described in Howard (1929) and by their large size. Family Falconidae Falco sparverius Linnaeus 1758 Referred material.-Femur (EMF A3684). Remarks.-The distal femur specimen is too small to represent F. columbarius. Falco columbarius Linnaeus 1758 Referred material.-Reported in Howard (1929): humerus (EMF 9913). Additional ele­ment: coracoid (EMF A2137). Remarks.-Identification of the F. columbarius coracoid was based on size: too large to repre­sent F. sparverius and too small to represent F. mexicanus. Falco peregrinus Tunstall 1771 or F. mexicanus Schlegal 1850 Referred material.-Reported in Howard (1929): coracoid (EMF 7081). Additional ele­ments: humerus (EMF A4067), carpometacarpi (EMF A9357, A9208), ulna (EMF A9362), femora (EMF A10363, A9365), tibiotarsi (EMF A10347, A9363). Falco peregrinus Tunstall 1771 Referred material.- Reported in Howard (1929): ulna (EMF 9989), tarsometatarsus (EMF 10292). Additional elements: mandible (EMF 10160), ulna (EMF 17099), tibiotarsus (PAHMA 12-1476). Remarks.-These elements differ from those of F. mexicanus as follows. (1) Ulna: The cotyla and condyles are large, with external cotyla more sharply hooked and distopalmarly projecting. (2) Mandible: Greater depth of the dentary. (3) Tibiotarsus: The fossa between the inner and outer cnemial crests on the proximal shaft is shallower, and the distal extremity of the outer cnemial crest lacks a sharp spine. Falco mexicanus Schlegal 1850 Referred material.-Femur (EMF 17220). Remarks.-Distinguished from F. peregrinus by having a posterior rather than a lateral orien­tation of the impression for the ansae iliofibularis muscle. Order Galliformes Family Phasianidae Gallus gallus (Linnaeus 1758) Referred material.-Coracoid (EMF A11518 [Fig. 7A]) and ulna (EMF A11490 [Fig. 7B]).PREHISTORIC CALIFORNIA BIRDS Fig. 7. (A) Left coracoid (EMF A11518) and (B) right ulna (EMF A11490) of Gallus gallus. Remarks.-These elements are very similar in size and morphology to those of Dendragapus obscurus, but differ from that species in the fol­lowing ways. (1) Coracoid: Thinner in general, with a concave internal surface at the sternal end and a more lateroproximal position of the pneumatic foramen. (2) Ulna: Less curvature of the shaft, with a more prominent extension of the external condyle of the distal end, and a smoother distal attachment of the internal cotyla to the palmar shaft. Chickens were not identified from the Emeryville sample that Howard (1929) exam­ined, nor were any domestic species present in the large collection of mammal materials recovered from the site (Broughton 1999). Both these elements originated from the top layer of the mound (Stratum 1) and were clearly derived from historic-period activities that took place on its surface. Family Odontophoridae Oreortyx pictus (Douglas 1829) or Callipepla californica (Shaw 1798) Referred material.-Sterna (PAHMA 12-1441, EMF A886). Remarks. - Substantial overlap in size of the sternum between C. californica and O. pictus and the fragmentary nature of the anterior sternum specimens precluded species-level identifica­tions. Callipepla californica (Shaw 1798) Referred material.-Humeri (EMF A10868, PAHMA 1-9733), tibiotarsus (EMF A5305), tar­sometatarsus (EMF A9562). Remarks.-These elements are too small to represent Oreortyx. Order Gruiformes Family Rallidae Rallus longirostris Boddaert 1783 Referred material.-Reported in Howard (1929): humeri (EMF 10130, 10309), femur (EMF 8049). Additional elements: femora (EMF A2160, A2698). Remarks.-The prominent, crest-like obtura­tor ridge of the femur distinguishes R. longiro­stris from Fulica and Gallinula. Fulica americana Gmelin 1789 Referred material.- Reported in Howard (1929): scapula (EMF 6268), humerus (EMF 6368), ulna (EMF 6275), tarsometatarsus (EMF 6374). Additional elements: sternum (EMF A2626), radius (EMF A1736), ulna (EMF 8003), carpo­metacarpus (PAHMA 12-1437), synsacrum (EMF A8564). Remarks.-These elements are too large to represent R. longirostris or Gallinula. Family Gruidae Grus canadensis (Linnaeus 1758) Referred material.-Reported in Howard (1929): mandible (EMF 8639), coracoids (EMF 16967, 7417, 8572), scapulae (EMF 5362, 7421, 8180, 9929), humeri (EMF 6309, 10421), car- pometacarpi (EMF 9240, 10521), carpal digit 2 phalanx 1 (EMF 9465), femur (EMF A2410), tibiotarsus (EMF A4892), tarsometatarsi (EMF A2384, A2606). Additional elements: sternum (PAHMA 12-1360), coracoid (PAHMA12-1462), humeri (EMF A8616; PAHMA 12-1447,12-1437), radius (EMF A4336), ulna (EMF A8664), carpal digit 2 phalanx 1 (EMF 17086, A2709), synsa­crum (EMF A9095), tarsometatarsi (EMF A3273, A134, A11645). Remarks.-Both small and large subspecies of G. canadensis are represented in the collection. 25 Order Charadriiformes Family Charadriidae Pluvialis squatarola (Linnaeus 1758) 26 Referred material.-Radius (EMF A3101), ulnae (EMF A6164, 9166), carpometacarpus (PAHMA 12-1472), synsacrum (PAHMA 12-1157), tarso­metatarsus (EMF A5595). Remarks.-The ulna of P. squatarola is most similar to that of Tringa melanoleuca but differs from it by having a longer, more pronounced crest running distally from the internal cotyla along the palmar surface of the shaft; a deeper proximal radial depression; and greater breadth of the distal end. The other elements are smaller than those of Catoptrophorus, and larger than those of T. melanoleuca, Limnodromus, Charadrius, and P. dominica. Referred material.-Ulnae (EMF A7478, A7035). Remarks. - The ulnae are too small to rep­resent Pluvialis and too large to represent any other Charadrius species. Referred material. - Humerus (PAHMA 12­1437), tarsometatarsus (PAHMA 1-9724). Remarks.-The proximal humerus is distin­guished from those of Numenius and Limosa fedoa by a more pronounced excavation of the anconal surface, including undercutting of the head; it is too large to represent Himantopus mexicanus. The trochlea of the distal tarsometa- tarsi are too large to represent H. mexicanus, N. phaeopus, and L. fedoa and too small to represent N. americanus. Referred material.-Reported in Howard (1929): humerus (EMF 9942). Additional ele­ments: sterna (PAHMA 12-1157; EMF 10134, A4396), scapula (PAHMA 12-1437), humerus (EMF A3122), radii (PAHMA 12-1427; EMF 6382, A699, A5248, A2824, A5231, A3164, A4427), ulnae (EMF 6287, All 8, A3452), carpometacarpus (PAHMA 12-1427), synsacra (PAHMA 12-1275, 12-1157; EMF A5910), femur (EMF A6896), tibiotarsi (EMF 16885, A3465, A4177; PAHMA 12-1437), tarsometatarsus (EMF A5508). Remarks.-The sternum of C. semipalmatus differs from that of L. fedoa by having a shorter, more anteriorly flattened manubrial process and a narrower anterodorsal base of the carina; it differs from that of N. phaeopus by having a smaller coracoidal sulcus. The distal humerus is distinguished from that of R. americana by smaller size and a sharper ectepicondylar prominence. The proximal ulna is distinguished from that of Recurvirostra by having a shorter proximodistal length of the external cotyla and by lacking a prominent V-shaped plateau formed by ridges extending distomedially from internal and external cotylae. The distal ulna differs from that of Recurvirostra by having a less prominent carpal tuberosity. Identifications of other elements were based on size. Numenius phaeopus (Linnaeus 1758) Referred material-Reported in Howard (1929): tibiotarsus (EMF 9973). Numenius americanus Bechstein 1812 Referred material. -Reported in Howard (1929): cranium (EMF 10339), mandible (EMF 17029), sterna (EMF 7991, 8612), coracoids (EMF 7047, 6386, 10169, 6349), carpometacarpi (EMF 6792, 9152, 10524). Additional elements: crania (EMF A4700, 5795, 8365, 10171, A2663), sterna (EMF A9113, A9236, A3505, A3506, A3511, A3513, A5518, 9961, 10102, A8059; PAHMA 12-1441, 12-1157, 12-1434, 12-1434), furculae (EMF A5980, PAHMA 12-1430), coracoids (EMF A10330, A4347, A8541, A9552, A9558, A3449, A4156, A4259, A5501), scapulae (EMF A10342, A8663, A8701, A843, A4412, A10848, A11559, 6932, 69335, 8197, 9228, A2830, A1629, A2129, A3249; PAHMA 12-1437), humeri (EMF A8677, A81, A3469, A4285, A6118, A6125, 8038, 9956, A2090, A2461, A3143, A9514, A9728, A9194, 17239, A5281, A5607,7038, 8639, 8719,9219,9927,10127; PAHMA 1-9708, 12-1476, 12-1476), radii (EMF 16887, A10628, A12473, A3361, A5094, A8720), carpometacarpi (EMF A10478, A10624, A9206 ORNITHOLOGICAL MONOGRAPHS NO. 56 Charadrius vociferus Linnaeus 1758 Family Recurvirostridae Recurvirostra americana Gmelin 1789 Family Scolopacidae Catoptrophorus semipalmatus (Gmelin 1789)PREHISTORIC CALIFORNIA BIRDS 27 [Fig. 8], A1641, A9531), carpal digit 2 phalanx 1 (PAHMA 1-9736; EMF A11094, A8407), synsacra (EMF 9266, 16919, A5890, A6301, A12601, A1469, A2365), femur (EMF 7161), tibiotarsi (EMF A9207, A4303, A5475, A5411; PAHMA 1-9739), tarsometatarsi (EMF A9214, A11600). Remarks.-The elements of N. americanus are easily distinguished from those of all other scol- opacids by their large size. Numenius phaeopus (Linnaeus 1758) or Limosa fedoa (Linnaeus 1758) Referred material. - Sternum (EMF A10127), ulna (EMF A10473), carpometacarpus (EMF A10574), synsacrum (EMF 10147), tibiotarsi (EMF A5394, A5471, A5485, A2713). Limosa fedoa (Linnaeus 1758) Referred material. - Reported in Howard (1929): humeri (EMF 5813, 6277, 6772). Additional ele­ments: cranium (EMF 7797), sterna (EMF A6360, A2199), scapulae (EMF 6397, A3047; PAHMA 1-9747), humeri (EMF A9190, 16955, A2711; PAHMA 12-1437, 1-9690), carpometacarpi (EMF A9218, A8685, A791, A3475, A11599; PAHMA 1-9756), synsacra (PAHMA 12-1157, 12-1311), tarsometatarsi (EMF A8723, A4258, A5270). Remarks.-The elements of L. fedoa are con­siderably smaller than those of N. americanus. The cranium was readily identified by the large recurved rostrum. Other elements are very similar to those of N. phaeopus and were distinguished from that species as follows. The humerus exhibits a less pronounced depression distal to the head on the anconal surface; a less undercut head; a steeper slope of the medial rise to the internal tuberosity; and a shallower impression for the brachialis anticus of the distal end. The carpometacarpus has a more 1 cm Fig. 8. Right proximal carpometacarpus of Numenius americanus (EMF A9206). pronounced proximal slope of the process of metacarpal 1 and a less distinctly rounded distal end. The scapula has a more pronounced, medi­ally oriented tubercle at the furcular articula­tion. The distal tarsometatarsus has a deeper groove between the medial and lateral portion of the trochlea for digit 4 and a less tapered (more squared-off) postero-proximal extension of the trochlea for digit III. The sternum has a more posteriorly oriented anterior carinal mar­gin, with a thicker dorsal base. Calidris alba (Pallas 1764) Referred material.- Ulna (EMF A4214). Remarks.-The ulnae of C. alba are most simi­lar to those of C. alpina; they are distinguished from the latter by a shorter total length and a less robust olecranal process. Calidris alba (Pallas 1764) or C. alpina (Linnaeus 1758) Referred material.-Ulna (EMF A8727). Remarks.-This specimen falls within the range of size overlap between C. alba and C. alpina. Limnodromus sp. Referred material.- One ulna (EMF 7866) reported in Howard (1929) as Limnodromus gri- seus. Additional material includes 39 specimens, including all major elements of the skeleton. Remarks.-No criteria were found to distinguish L. griseus from L. scolopaceus; the two were consid­ered a single species until 1950 (Pitelka 1950). Family Laridae Larus sp. (large) Referred material. -A total of 34 specimens, including all major elements of the skeleton. Larus sp. (small) Referred material. - Nine elements. Larus glaucescens Naumann 1840 or L. hyperboreas Gunnerus 1767 Referred material.-Carpal digit 2 phalanx 1 (EMF A856), tibiotarsus (PAHMA 1-9739).28 ORNITHOLOGICAL MONOGRAPHS NO. 56 Remarks.- The two L. glaucescens-hyperboreas specimens were identified on the basis of their very large size. No other species-level identifi­cations were attempted for the gulls, owing to the extensive interspecific overlap of osteologi- cal features. Specimens identified to the Larus sp. (small) category were similar in size to L. canus, L. heermani, L. pipixcan, L. philadelphia, L. delawarensis; the Larus sp. (large) category may include any of the other larger species. Family Alcidae Uria sp. Referred material.-A total of 167 specimens, including all major elements of the skeleton, including 62 reported in Howard (1929) as 11. troille (= aalge). Uria aalge (Pontoppidan 1763) Referred material. - Six crania (EMF 7400, A7041, A5208, A10695; PAHMA 12-1354, 1-9787). Remarks.-Howard (1929) distinguished the elements of U. aalge from the other alcids "on the basis of larger size." She did not, however, consider whether U. lomvia may have been represented in the Emeryville collection, and there is considerable overlap in size of U. aalge and U. lomvia elements. Although U aalge is far more abundant than U. lomvia along the central California coast, the latter occurs casu­ally in the area-most records are from the Monterey area (Small 1994). Clear differences do exist in the crania of these species. Uria aalge is distinguished from U. lomvia by having a lon­ger, thinner, and straighter premaxilla; smaller foramina in the anterior portion of the nasal fossa; deeper depressions anterior to the trans­verse nuchal crests; and more ventrally project­ing and squared-off opistotic processes. Cepphus columba Pallas 1811 Referred material.-Ulna (EMF A11501). Remarks.-The ulnae of C. columba differ from those of Cerorhinca monocerata and Fratercula cor- niculata in the orientation of the carpal tuberos­ity; it is oriented near a right angle (90°) to the main axis of the bone in C. columba, but at a more obtuse angle in C. monocerata and F. corniculata. Order Strigiformes Referred material.- One synsacrum fragment. Family Tytonidae Tyto alba (Scopoli 1769) Referred material.-Reported in Howard (1929): coracoids (EMF 16964, A95, 9924, A2544, A2828), humeri (EMF 6901, 6903, 6922, 6919, 7039), ulna (EMF 8205), femur (EMF A3647), tibiotarsi (EMF 5788, 5363), tarsometatarsi (EMF 6918, 8258, 10113). Additional elements: cora­coid (PAHMA 12-1323), humerus (EMF A3124), radii (EMF A253, 6928, 5807), ulnae (EMF A2406, A2607, A3682), carpometacarpi (EMF A3053, A3116, A10616), femur (EMF A256), tar­sometatarsi (EMF A10895, A2135, A3123, A265; PAHMA 12-1476), tibiotarsi (EMF 7925, A341), ulnae (EMF A2406, A2607, A3682). Remarks.-For most of the represented ele­ments, I used the criteria presented in Howard (1929) to separate T. alba from Strix. Tyto alba is further distinguished from Strix by having shal­lower depressions on the anterior and posterior surfaces of the distal tibiotarsus, and a distinct concavity on the distal radius just proximal to the ligamental prominence. Family Strigidae Referred material.- One humerus. Otus kennicottii (Elliot 1867) Referred material.-Distal humerus (EMF A9573 [Fig. 9]). 1 cm Fig. 9. Right humerus, missing proximal end, of Otus kennicottii (EMF A9573).Remarks.-The distal humerus of O. kennicottii differs from that of Athene cunicularia by having an ectepicondylar prominence that bears a small papilla and a deeper tricipital groove. Otus ken­nicottii was not identified in the Emeryville sample that Howard (1929) examined. Bubo virginianus (Gmelin 1788) Referred material.- Reported in Howard (1929): coracoid (EMF 9898), scapula (EMF 9949), tibiotarsus (EMF 8290), tarsometatarsi (EMF 8039, 8191, 8193). Additional elements: tarsometatarsi (EMF 10098, 7992; PAHMA 1­9823). Remarks. - The tarsometatarsi of Bubo virgin­ianus were distinguished from those of Strix nebulosa and Nyctea scandiaca by criteria in Howard (1929). Asio sp. Referred material.- Reported in Howard (1929): humerus (EMF 7867). Additional ele­ments: carpometacarpus (EMF A10362), femur (EMF A10476). Asio flammeus (Pontoppidan 1763) Referred material. - Carpometacarpi (EMF A836, A924). Remarks.-The total lengths of both carpo­metacarpi exceed the upper limit for A. otus as reported in Emslie (1982). Order Passeriformes Family Corvidae Corvus brachyrhynchos Brehm 1822 Referred material.- Reported in Howard (1929): coracoid (EMF 10333), scapula (EMF 7842), humeri (EMF 8320, A2977, 7813, 9221, 10118, 10293, 10417, 17097), ulnae (EMF 7442, 7840, 10162, 10319, 8020, 8362, 8623, 10010, 10588, 7254, 7441, 7972), carpometacarpi (EMF 5370, 6258, 6292, 6771, 8310, 8368, 9912, 9916), tibiotarsi (EMF 8056, 8213, 8221), tarsometatarsi (EMF A1050, A1138, A263, 7461, A2134, A2960, 6904). Additional elements: mandibles (EMF A1555, A3146), sterna (EMF 7861, A12674), cora­coids (EMF A1642, A2726, A4702, A1636, A2538, A1379, A120, A1034, A10834, A10340, A9361, PREHISTORIC CALIFORNIA BIRDS A8557; PAHMA 12-1454, 1-9703, 12-1462), scapula (EMF A1628), femora (EMF 7164, A246, A3455, 10429, 17279; PAHMA 12-1476), humeri (EMF A240, A1598, A9175, A8680, A10595, A10602, 6929, 7789, 10435), radii (EMF 6926, A5082, A11359, A3144, A3287), ulnae (EMF A9584, A11346, A11352, A11573, A3129, A4213; PAHMA 12-1449, 12-1314, 12-1314, 12-1314, 12-1314), carpometacarpi (EMF 17288, A2832, A10838, A10599, A10468; PAHMA 12-1462, 12­1314, 12-1314), carpal digit 2 phalanx 1 (EMF 6390, PAHMA 12-1462), tarsometatarsi (EMF 17302, A2612), tibiotarsi (EMF A2613, A2963, A2968, A9358, A10630; PAHMA 12-1462). Corvus corax Linnaeus 1758 Referred material.-Reported in Howard (1929): humeri (EMF 8249, 7377), radii (EMF 7091, 8315), ulnae (EMF 10009, 9914, 10116), tibiotarsus (EMF 7845), tarsometatarsi (EMF 7843, 8713). Additional elements: radii (EMF A3629, PAHMA 12-1314), ulnae (EMF A10852, 10135), carpometacarpus (EMF 17434), carpal digit 2 phalanx 1 (EMF A10471, 9991), femora (EMF 7844, A3238), tibiotarsus (EMF A11361), tarsometatarsus (EMF A10862). Remarks.-Corvus brachyrhynchos and C. corax are easily distinguished from each other and from other passerines on the basis of size. Taxonomic Summary and Depositional Origin Sixty-four species are represented by the 5,736 identified bird specimens derived from the provenienced Emeryville sample. All the species are either present in the San Francisco Bay today or, if not, occurred there in his­toric times (Grinnell and Wythe 1927, Small 1994). Forty-five of the 64 species (69.8%) are waterbirds, 15 (23.8%) are raptors, and two each (3.2%) are Galliformes or large corvids. Twenty-five of those species were not reported by Howard (1929) in the Emeryville sample that she examined. Although many of the newly identified species are anatids (n = 15), a group that Howard did not study in any detail, the new species also include Podiceps auritus, Fulmarus glacialis, Botaurus lentiginosus, Falco sparverius, Accipiter cooperi, Charadrius vociferus, Calidris alba, Gallus gallus, Otus ken­nicottii, and Asio flammeus. With respect to 29 30 ORNITHOLOGICAL MONOGRAPHS NO. 56 numbers of identified specimens, ducks are the best-represented group of birds in the col­lection (2,028 specimens; 35.4%), followed by geese (1,825 specimens; 32.0%), cormorants (950 specimens; 16.6%), shorebirds (225 speci­mens; 3.9%), and murres (173 specimens; 3%). There can be little doubt that these bird materials owe their presence in the mound to the activities of human foragers. Not only is the mound clearly of anthropic origin (Uhle 1907, Schenck 1926, Broughton 1999), but stone-tool cut-marks and evidence of burning are pres­ent on many of the Emeryville bird specimens. Indeed, it is hard to imagine any nonhuman mechanism that could accumulate bird remains at this scale in this kind of context. Thus, many of the natural processes that plague human paleoecological analyses in other contexts, such as caves and rock shelters, are simply not involved here. The site also appears to provide evidence of human foraging activities throughout the annual cycle. Spring and summer occupation is clearly indicated by the abundance of fetal and newborn mule deer and elk (Broughton 1999), as well as cormorant chicks, nestlings, and juveniles (Table 3). Fall and winter occupation is indicated by the abundance of strictly winter- visitant avian taxa, such as all of the represented loons, grebes, and scolopacid shorebirds and the great majority of anatids (Grinnell and Wythe 1927, Small 1994). Importantly, those season­ally diagnostic specimens are well represented in all of the Emeryville strata (Table 2), which suggests that any trends present in bird use over time are not related to changes in seasonal occupation of the mound. In sum, the provenienced Emeryville avi­fauna represents a large, taxonomically diverse, well-stratified and well-dated sequence of human bird-exploitation over a period of nearly 2,000 years in the late Holocene. It thus provides a unique opportunity to investigate long-term human-avian paleoecological relationships, including the possible occurrence of resource depression. Archaeological Measures of Avian Resource Depression Previous analyses of late Holocene archaeo­logical faunas from California have docu­mented the occurrence of long-term resource depression for a wide variety of large-sized fish and mammal taxa. Quantitative trends in the relative abundances of large-sized or otherwise "profitable" prey resources and demographic indicators of harvest pressure have been the primary measures of prehistoric resource depression. Similar measures are used here to investigate the potential effects of human hunt­ing on avian populations of the San Francisco Bay area. Relative-abundance Indices The use of relative-abundance indices to mea­sure resource depression archaeologically is founded on logic from mathematical models of optimal foraging, especially the "prey model" (see Stephens and Krebs 1986 and references therein). That model focuses on how a forager should choose among a range of resources that vary in rate of energy earned for time spent in pursuing and processing (i.e. "handling") them. The model predicts that the most profitable or highest-ranked prey will be taken whenever they are encountered, whereas prey of lower rank may or may not be selected, depending on the abundance of the highest-ranked prey. As encounter rates of higher-ranked prey decrease, prey are added to the diet sequen­tially in order of decreasing rank (see Stephens and K