55

Zooarcheologists specialize in old bones.Unlike paleontologists, who study fossil bones,and physical anthropologists, who study humanskeletal material, zooarcheologists study theosteological refuse of long-past meals. Our exper-tise is in identifying and analyzing discarded, usu-ally fragmentary, and often burnt skeletal remainsof mammals, birds, fish, and shellfish excavatedfrom archeological sites. During excavation, thesefragments are carefully retrieved, bagged, andlabeled with their exact site provenience, or placeof origin within the site, before being broughtback to the lab. With some collections numberingupwards of 10,000 specimens, analysis can takeseveral months or even years. Frequently the goalof zooarcheological or faunal analysis is to pro-vide a detailed picture of past human subsistencepractices.

Even before the first bag of catalogued bonesis opened, the zooarcheologist puts together a list,

inventorying all faunal species that are available inthe general site area, on either a year-round or aseasonal basis. However, sometimes after the anal-ysis has begun, there can be surprises. Sometimesthe fragments are “out of place,” or not what isexpected using modern faunal distribution maps.These fragments may represent species that oncelived in an area but are now extinct or no longerpresent within the region or species that werebrought into the site as the result of long-distancehunting forays or trade. In this review, the focuswill be on bones identified from archeological sitesthroughout Alaska that are “out of place“ geo-graphically. It highlights some of the Pleistocenemegafauna—the big game animals—hunted bythe earliest Alaskans, as well as some species ofsea mammals—walrus, ringed seal, and polarbear—hunted far outside their current ranges attimes when past climatic and ice conditions weremuch different than today.

Out of Place BonesBeyond the Study of Prehistoric Subsistence

This article was preparedby Becky M. Saleeby, an

archeologist for theNational Park Service.

This document has been archived.

56

Analysis of large zooarcheological collectionsis time-consuming, beginning with the sorting ofbones and bone fragments that are potentiallyidentifiable. Later, we tentatively identify skeletalelements and species based on drawings andphotographs in reference books and on simplepattern recognition. For example, the distal (lower)ends of the upper arm or forelimb of most mam-mals look similar, regardless of species. Subtledifferences in morphology, such as the angle on abony ridge or the shape of a particular ligamentattachment, may be all that separate fragments oftwo closely related species, so we turn to compar-ative collections of skeletal material for positivespecies identifications.

In Alaska, one collection of comparative faunalmaterial is housed at the Anthropology Labora-tory at the University of Alaska Anchorage. Overthe last several years, members of the AlaskaConsortium of Zooarcheologists (ACZ), which isa special interest group of the Alaska Anthropo-logical Association, have added many specimensof mammals, birds, and fish to broaden the existingcomparative collection. By virtue of state andfederal permits, we have been allowed to collectanimal carcasses for processing. Properly pre-pared as clean, white skeletons, they are acces-sioned into the growing inventory of modernspecimens used for comparative purposes byarcheologists throughout the state. National ParkService (NPS) archeologists have made frequentuse of these collections for identifying faunalremains from sites within Aniakchak, Bering LandBridge, Cape Krusenstern, and other NPS units inAlaska.

Humans, Bison, and ElkFor zooarcheologists working on collections

from early Alaskan sites dating between 10,000and 12,000 years ago, it is exciting to realize thatsome bone fragments do not match any modernspecies from the comparative collection. Thesesites represent the hunting and foraging camps ofpeople who ranged over the narrowing isthmus ofthe Bering Land Bridge (Beringia) at a time whenthe late Pleistocene environment was rapidlychanging. In general, faunal preservation at thesesites is so poor that bones are either absent or sodeteriorated that they cannot be identified. Fortu-

nately there are some exceptions, notably the DryCreek site in the Nenana River valley, adjacent toDenali National Park and Preserve.

Dry Creek is a multi-component site excavatedduring the 1970s by researchers from the Univer-sity of Alaska Fairbanks (UAF). It was heralded inarcheological circles not only for its 11,000-year-old dates,* but also for its preservation, albeitpoor, of faunal remains in association with ancientstone tools. R. Dale Guthrie, a Quaternary biolo-gist and paleontologist now retired from UAF,worked with the team of archeologists at the siteand identified fragments of Dall’s sheep, wapiti orelk, and bison in the small but significant faunalassemblage, composed mostly of teeth. Neitherwapiti nor bison are native to Alaska today,though some small herds have been reintroducedinto the state.

Guthrie’s paleoecological reconstruction of thesite allows us to imagine Beringian hunters livingin an interior Alaska landscape changing from drygrassland or steppe, which was once the dominantPleistocene habitat in Alaska. Today the environ-ment in the region is primarily boreal forest. Pale-

Zooarcheologist Bob Kopperl checks the differences inseal skulls during a workshop sponsored by the AlaskaConsortium of Zooarchaeologists at UAA in 1999.

*All the dates that appear in this article are uncalibrated.These are the dates listed in the originally published sitereports. Calibrated dates may be several hundred (ormore) years older than uncalibrated dates.

57

ontological specimens of mammoth, dating toabout 12,300 years ago, were found in surveysaround Dry Creek, but the bones of these behe-moths were not found at the site. Guthrie arguesthat although mammoths, horses, camels, saigaantelopes, lions, and other species may havealready become extinct in Alaska at the time whenthe lower two levels of Dry Creek were occupied,the regional extinction of wapiti and bison had notyet occurred. It is also interesting that the sheep,bison, and wapiti specimens from the site were aslarge as their Pleistocene forms, so Holocenedwarfing had apparently not yet begun.

A trio of early sites located on Shaw CreekFlats in the Tanana River valley—the BrokenMammoth site, the Mead site, and the Swan Pointsite—are also among the handful of early siteswith faunal preservation despite the fact that theirlowest occupations date to almost 12,000 yearsago. This preservation is due to the sites’ depositsof wind-blown glacial silt from the nearby flood-plain of the Tanana River. Similar in setting to theDry Creek site, they probably served not only ashunting overlooks for spotting and interceptinggame animals, but also as “spike camps” or pro-cessing stations for the meat and hide broughtback from kill sites. Best known of the three is theBroken Mammoth site, discovered in 1989 andrevisited almost annually for summer test excava-tions and field schools sponsored by the Univer-sity of Alaska Anchorage (UAA) and the AlaskaState Office of History and Archaeology (OHA).Charles Holmes, an archeologist with OHA, andDavid Yesner, associate anthropology professorand zooarcheologist at UAA, are the principalresearchers at the site.

David Yesner identified a wide range of water-fowl, small to medium-sized mammals, and fish,representing species still living in Alaska, from theearliest cultural layers at the multi-component Bro-ken Mammoth site. He also identified small num-bers of Dall’s sheep, caribou, and moose bones,along with much higher frequencies of long-horned steppe bison and elk or wapiti remains.Measurements of a bison horn core from Compo-nent 3 (second from lowest) are compatible withthis extinct species, and the site is outside therange for the wood bison, a northern species stillfound in free-ranging herds in Canada. Large-horned bison and elk were clearly the chief prey ofthe Broken Mammoth hunters, who also leftbehind an assortment of stone tools, mammothivory projectile points, toggles for clothing, andeyed bone needles.

The provocative name “Broken Mammoth”and the earliest dates that are at least 600 yearsyounger than those from the Dry Creek site begthe question, “where’s the mammoth?” Did humansand mammoths coexist in Alaska? Archeologicalevidence does prove their coexistence at severalsites in the “lower 48,” but the evidence in Alaskais still circumstantial. The name “Broken Mam-moth” actually comes from the numerous mam-moth tusk fragments uncovered during initial sitetesting. Similar fragments were also recoveredfrom the nearby Mead and Swan Point sites. Noother mammoth skeletal elements have beenrecovered from these sites. Yesner originally pos-tulated that the mammoth ivory, and possibly hideand meat, at the Shaw Creek sites may have beenobtained at kill sites located away from the bluff-top campsites and brought back for raw material.After many field seasons of excavation, the evi-dence now suggests to him that the ivory repre-sents scavenged material from the skeletons ofrecently extinct animals that was brought back tocamp specifically for tool production. Ongoinganalysis and dating of the specimens may yetbring to light indisputable evidence in supportof the possible coexistence of humans and mam-moths in Alaska.

Hotly debated since the topic was proposeddecades ago is whether the large-scale die-off ofNorth American megafauna at the end of the Pleis-tocene (approximately 10,000 years ago) was thedirect result of over-predation by human hunters.The issue of whether the extinction was caused byhumans, environmental change, or a combinationof factors has not been resolved. An accumulatingbody of zooarcheological evidence indicates that

Lorraine Alfsen uncoversa mammoth bone point

from the lowest com-ponent of the Broken

Mammoth site duringexcavation in 2000.

58

for some species it may have not have been abruptas previously thought, particularly for the bison.The persistence of bison in Alaska and Canadavirtually throughout the Holocene is documentedin a recent study by Fairbanks researcher RobertStephenson and his colleagues, in which they pro-vide a long list of radiocarbon-dated paleontologi-cal and zooarcheological specimens. They alsopresent oral narratives of Athapaskan elders livingon the upper Yukon and Tanana Rivers that sug-gest that bison may have been sufficiently abun-dant to be a resource of some importance asrecently as 200–300 years ago. Their zooarcheo-logical evidence in Alaska consists of a bison tibiafragment from the Delta River Overlook site, datedat about 2,200 years ago, and a bison foot bone inprobable association with the Killik River site, datedat about 2,300 years ago. The latter is located inGates of the Arctic National Park and Preserve.Bison bones are also present in the upper compo-nent (about 2,000 years ago) of the Broken Mam-moth site. The Gerstle River quarry site and theSilver Fox site, both in the Tanana Valley, provideevidence for the persistence of elk until about thissame time period, suggesting that east-centralAlaska may have served as a refugium for thesespecies. Refugia are areas of relatively unalteredenvironment inhabited by relic forms of plants andanimals during periods of climatic change, such asoccurred at the end of the Pleistocene.

Scarcity of Moose in theZooarcheological Record

Eventually, bison and elk did become extinct inAlaska, while other large mammals, such as cari-bou, moose, and Dall’s sheep, survived. Fromethnographic and historic records we know thatmoose and caribou were the primary big gamespecies hunted by the interior Alaska Athapas-kans, but was this also the case prehistorically?The flip side to the presence of geographically outof place bones in zooarcheological assemblagesis the absence or scarcity of a key species, suchas moose, which we would expect to find in abun-dance, given present-day distributions. Todaymoose populations exist almost throughoutAlaska, with the exception of islands in the south-east and in the Aleutians.

My experience in identifying moose and cari-bou comes from the analysis of zooarcheologicalcollections from a very large sample of siteslocated on the Susitna River in south-central

Alaska. Seventy-eight of these sites producedbone and resulted in a huge collection of almost143,000 specimens, ranging from minute fire-whitened fragments to complete unburned largemammal bones. Moose bones were only foundat nine of the sites, including one paleontologicalsite where five mandibles of late Pleistocenemoose were recovered. The other eight sites wereyounger than 600 A.D. Even within the subsampleof late prehistoric sites, fully 93% of the largemammal remains were identified as caribou; theremainder were moose and Dall’s sheep.

David Yesner undertook a much more extensivesurvey of the occurrence of moose in the prehis-toric archeological record of the Alaskan sub-Arctic some years ago. Questioning whether theapparently heavy reliance on moose by Athapas-kans in ethnographic accounts was an accurateportrayal of their subsistence prehistorically, heturned to published accounts from 19 sites or siteclusters from a vast area of interior Alaska andwestern Canada. Yesner’s overall impression fromlooking at these data was that moose appear onlyrarely in any of these assemblages until quiterecently, perhaps within the last 400 years or so.He suggests that climatic and vegetational changes,fire, and natural population cycles have all beenfactors in this apparent scarcity of moose in theregion during most of the prehistoric period. Aswith the findings for the Susitna River sites dis-cussed above, his study indicated that the speciesof primary importance for prehistoric populationsin northern interior Alaska was caribou.

Cave Sites and Bear BonesNot all the faunal collections that zooarcheolo-

gists identify come from unequivocally culturalcontexts. This is particularly true of cave sites,where the refuse from early human occupation canbe difficult to differentiate from the refuse leftbehind by other species of predators. Work doneat Trail Creek caves, on the Seward Peninulawithin Bering Land Bridge National Park, providesan excellent example of the type of meticulousanalysis needed to unravel the complexities ofbone deposition within cave sites. Quaternarygeologist David Hopkins and Danish archeologistHelge Larsen were the first to test and excavateseveral of the twelve caves on Trail Creek in thelate 1940s. From two of the caves, Larsen and hiscrew recovered artifacts of ancient caribouhunters, as well as those of the historic Inupiat.The oldest tools date back 8,000 years or older.

59

He also reported thousands of bone fragments ofextinct and extant species, including bison, horse,and mammoth dating back about 15,800 years ago.For decades after the original excavations, ques-tions remained about the possible association ofhuman artifacts with the bones of Pleistocenemegafauna.

Within the faunal assemblages were brokencanine teeth from several levels of two Trail Creekcaves. Larsen identified them as dog teeth. Theirsize and the fact that they were broken led him tobelieve that they had been purposely knocked outby humans to prevent the dogs from chewing onskins or on tethering lines. Archeologists E. JamesDixon and George Smith, both formerly of theUniversity of Alaska Museum in Fairbanks, recog-nized that if this identification were accurate, thesewould be among the oldest specimens of domesti-cated dog in the world. Animal domestication isknown archeologically from Old World sites, sothese teeth were clearly out of place in an Alaskanassemblage. Dixon and Smith compared thecanines with the permanent teeth of a variety ofmammals that could be expected in a cave deposit,but they found no morphological and size matchesuntil they compared the specimens with the decid-uous dentition of brown bears. These teeth exfoli-ate during the second winter of hibernation. Theirpresence in the faunal assemblage from Trail Creekand other caves sites of similar age was thusattributed to a very long history of bear denning,rather than dog domestication.

Subsequent zooarcheological work at TrailCreek caves has shown that brown bears wereresponsible for more than simply hibernating anddropping their deciduous teeth. According to DaleVinson, who methodically analyzed the bonesfrom two of the Trail Creek caves tested in 1985 bythe NPS, disturbance within the layers of the cavedeposits was probably due to bear denning activi-ties. Although not completely ruling out the possi-bility that early Alaskans brought in and modifiedthe bones of Pleistocene mammals found in thecaves, Vinson made a strong case for non-humanscavengers and carnivores being responsible forthe bone breakage and cut marks he documented.

Polar Bear, Walrus,and Ringed Seal

Exactly when the ancient caribou hunters ofnorthern Alaska began to dwell along the coastand hunt for sea mammals is not known for

certain. Some of the earliest evidence for seamammal hunting on the northwest coast of Alaskais represented by only a few charred fragmentsof seal bones in a hearth at the earliest culturallevel at the Iyatayet site on Norton Sound. Thecharacteristic Denbigh Flint complex tools at thislevel date to approximately 5,500–4, 000 ago. Themakers of these tools are thought to be the ances-tors of the present-day Inupiat of northern Alaska.J. Louis Giddings, who excavated at Iyatayet inthe late 1940s and early 1950s, identified bonesfrom an upper, 2,500-year-old level of the site(Norton culture) as predominantly “small seal.”He also identified bearded seal, walrus, and belugain this Norton assemblage, along with a smallnumber of caribou bones.

Since Giddings’ pioneering archeological field-work in northwest Alaska, our knowledge of theprehistoric cultures has increased enormously, inpart because of the fieldwork and research carriedout by the National Park Service in Bering LandBridge National Preserve (BELA). As the result ofsurveys and excavations in BELA by archeolo-gists Jeanne Schaaf and Roger Harritt, we nowhave extensive faunal collections from BELA sitesat Cape Espenberg, the Ikpek Lagoon area, andthe mouth of the Kitluk River. Besides the smallringed seal that appeared in collections made byGiddings, the spotted seal and the ribbon sealhave been identified at BELA sites. Bearded sealsor ugruk, walruses, belugas, and polar bears alsooccur in the assemblages. These species all thrivealong the far northern coastline, locked during thewinter in shore-fast ice. They are not out of placegeographically but fit well within current distribu-tions of sea mammals north of Bristol Bay.

South of Bristol Bay in Shelikof Strait and theGulf of Alaska, an entirely different suite of seamammals is usually found within faunal assem-blages, even at sites dating back earlier than 6,000years ago. The harbor seal is the only seal speciesof the genus Phoca (as opposed to fur seals in thegenus Callorhinus) that currently inhabits Alas-kan waters south of the Alaska Peninsula. Othercommonly identified species are the sea otter,Steller sea lion, fur seal, and two species of por-poises. Again, these are species that would beexpected in the region. Clearly out of place insouthern coastal assemblages are the bones ofthe ice-loving polar bear, ringed seal, and walrus,so their presence in the faunal assemblage fromthe Margaret Bay site on Unalaska Bay in theAleutians was a surprise to zooarcheologist BrianDavis.

60

The Margaret Bay site was noted by zoologyprofessor Alvin Cahn, who was a Lt. Commanderin the U.S. Naval Reserve stationed in Dutch Har-bor in the early 1940s. Archeologists, working atthe site in later decades, recognized the impor-tance of this stratified (or many-layered) site, butit was not until excavations in 1996-97 by RichardKnecht of the Museum of the Aleutians that adense shell midden with an abundance of animalbones was encountered and excavated. Thismidden was radiocarbon dated at 4,700–4,100years before present. Brian Davis analyzed over5,000 mammalian specimens from the midden,using the comparative collections housed at theUniversity of Alaska Museum in Fairbanks, andhe made some unexpected identifications. Harborseal bones accounted from almost 50% of theidentified specimens, but the ringed seal was alsoabundant at the site, contributing about 11% ofthe total bone count. Davis’s most exciting findswere the mandible, forelimb, and hindlimb of apolar bear. The bones of this species are very rare,even within its current range on Alaska’s far

northern coastline. A few specimens of walruswere also found within the Margaret Bay assem-blage. The age of this midden is congruent withthe Neoglacial, a cooling period of glacial advanceidentified between 5,000 and 3,500 years ago in theAleutians. The effect that these climatic condi-tions, and the resulting geographically displacedspecies, had on the hunting techniques and cul-ture of the prehistoric Aleuts will be a subject ofarcheological study for many years.

We are undoubtedly in for more faunal sur-prises and out of place bones when identificationand analysis of the enormous Mink Island sitecollection are completed. The Mink Island site,located in Amalik Bay off the coast of KatmaiNational Park and Preserve, was excavated byJeanne Schaaf and her NPS crews in 1997–2000.It has two main components: the upper one datingto 370–2,010 years before present, and the lowerone dating to 5,000–7,300 years old, making it oneof the oldest known sites along the south-centralcoastline of Alaska. Well-preserved bones recov-ered from both components are currently under

An extensive zooarcheo-logical collection was

recovered from the MinkIsland site on the coast of

Katmai National Parkand Preserve. Archeolo-

gists built this domestructure to protect fragile

site stratigraphy andartifacts during

excavation. Brown bears(see center of photo) were

frequent visitorsat the site.

61

analysis by zooarcheologist Maribeth Murray atthe University of Alaska Fairbanks. According toMurray and her colleague, S. Craig Gerlach, a neo-natal walrus mandible was identified in the uppersite component. According to modern speciesdistributions, walrus are usually considered out ofplace in the Shelikof Strait region. The verdict onwhether polar bear specimens are present amongthe Mink Island bear bones awaits Murray’s finalidentification and analysis. These bones may dateto a glacial period known as the Little Ice Age(1300–1850 A.D.), a global phenomenon of lowtemperatures that dramatically affected culturesaround the world.

Walrus ivory artifacts have been found at sitesfarther east, in Prince William Sound, even moreremoved from the present-day species range.Zooarcheologist Linda Yarborough, who exca-vated the Palutat site, reports that ivory togglesand projectile points found at the site possiblydate from between 2,000 and 1,400 years ago. Therewere no other skeletal elements of walrus identi-fied at the site. Yarborough is unsure whether theivory tusks were brought to the area in trade andthe artifacts manufactured on-site, or whether theseice-adapted creatures were hunted nearby during aperiod of glacial advance in Prince William Sound.

Cultural FactorsNatural environmental conditions affecting

past animal distributions explain the presence ofsome bones that appear to be out of place, butcultural factors are also important to consider. Inanalyzing the faunal collection from the proto-historic (about 1850 A.D.) Kitluk River site a fewyears ago with my colleague, Angela Demma, wecame across a specimen that we simply could notidentify. It took several visits to wildlife biologistsin Anchorage before we were satisfied with apositive identification. The specimen was a horncore of a Dall’s sheep, certainly not something weexpected to find on the coastal margin of theSeward Peninsula, far from any mountainoushabitat. We interpreted the horn core as either atrade item or a remnant brought back from a dis-tant hunting trip, possibly far to the north in thehilly country around Cape Lisburne or the BairdMountains north of Kotzebue.

Trade between coastal and inland people, par-ticularly of caribou antlers and walrus ivory, is welldocumented in the ethnographic literature of theArctic and appears to have deep roots in the past.Anthropologist Otto Geist studied the Siberian

Yupik people of St. Lawrence Island in the BeringSea during the 1930s and excavated their ancientsites, including the Kukulik Mound. He reportedfinding tool handles and scratchers fashionedfrom caribou or reindeer antlers deep within themound. Caribou are not native to St. LawrenceIsland, and reindeer were introduced as late as1900. It therefore appears that the antlers fromwhich these tools were fashioned must be prehis-toric trade items the ancient St. Lawrence Island-ers received from mainland caribou hunters. Largetrade fairs, such as one held every summer atSheshalik, near Kotzebue, in the 1800s, may havebeen the source of such trade goods.

Otto Geist also reported that the people of St.Lawrence Island spoke of hunting “the real walruswithout tusks” in the past. Geist conjectured thatthey were referring to Steller’s sea cow, an extinctrelative of the manatee, hunted to extinction byRussian fur traders in the late 1700s and early1800s in the Bering Sea and the Aleutians. Onlyrecently have the bones of this species turned upin zooarcheological assemblages. Debbie Corbett,archeologist with the U.S. Fish and Wildlife Ser-vice, excavated a few fragments of what shebelieved to be sea cow bone from 1,000-year-oldsites on Buldir Island in the western Aleutians.These bones, probably ribs, are very dense anddistinctly different from bones of other sea mam-mals and walrus ivory. Corbett believes that theancient Aleuts not only hunted these creatures fortheir meat but may have also made artifacts fromtheir bones.

Out of place bones tease our imagination,whether they come from archeological contexts orfrom more recent surface finds. Notable in myexperience is a foot bone brought to my office byDale Vinson of Lake Clark–Katmai National Parkand Preserve. Vinson’s expertise as a zooarcheolo-gist was called into play when, surveying on TakliIsland, he stumbling upon an unusual bone herecognized as an animal not indigenous to thearea. It was, in fact, part of a horse skeleton. Witha bit of historic sleuthing, he was able to shedsome light on this out of place bone. As the storygoes, a bay gelding was the only horse thatsurvived a shipping mishap in Amalik Bay on theKatmai coastline in 1956. The horse continued tosurvive in the hostile environment for the next 18years and was known as a living legend to localfisherman. This bone, the subject of much discus-sion, is now properly accessioned as a historicspecimen in the NPS collections at the Lake Clark–Katmai Study Center in Anchorage.

62

Zooarcheology andBiogeographic History

Although bettering our knowledge of prehis-toric subsistence is often the rationale in zooar-cheological analyses, the bones themselves some-times force us to go beyond subsistence in ourinterpretations. Some bones simply cannot beidentified on the basis of present-day animal dis-tributions. Extinctions, shifts in range, trade, andlong-distance hunting are all possible factors forexplaining bones that appear to be out of place.These specimens challenge our assumptions andremind us that past landscapes were different thanthose of today and that cultural patterns were notwhat we might expect them to be. The integrationof a wide variety of data—geological, biological,ethnographic, and historic—has proven success-ful for zooarcheologists. Now it’s time to turn thetables and convince wildlife biologists that zooar-cheological data can benefit them by providingthe element of great time depth to their studies ofspecies that may be threatened or endangered.

Listed in the 2002 program for the 67th annualmeeting of the Society for American Archaeologywas a symposium entitled “Zooarchaeology’sContribution to Conservation Biology.” Includedwere papers addressing the interface betweenarcheological perspectives and wildlife manage-ment of elk in Washington, black bears in Minne-sota, pronghorn antelopes in Wyoming, fresh-water fish in Virginia and North Carolina, andothers. Perhaps the paper most relevant for Alas-kan wildlife managers was the one presented byMichael Etnier on seal remains from the Ozette sitein western Washington. He documented the differ-ences between prehistoric and modern abundanceand migration patterns of six North Pacific seamammal species and discussed both anthropo-genic and natural catalysts for behavior change.Work such as Etnier’s may be the wave of thefuture for wildlife managers who want to expandthe narrow time range of their studies—just a fewdecades or less—to centuries or even millennia bylooking into the zooarcheologists’ bags of bones.

Suggestions forFurther ReadingDavis, B.L. (2001) Sea mammal hunting and the

Neoglacial: An archaeofaunal study of environ-mental change and subsistence technology at

Margaret Bay, Unalaska. In Archaeology in theAleut Zone of Alaska (D. E. Dumond, ed.). Uni-versity of Oregon Anthropological Papers 58,Eugene.

Dixon, E.J., and G.S. Smith (1986) Broken caninesfrom Alaskan cave deposits: Re-evaluating evi-dence for domesticated dog and early humansin Alaska. American Antiquity, vol. 51, no. 2, p.341–351.

Geist, O.W., and F.G. Rainey (1936) Archaeologicalexcavations at Kukulik, St. Lawrence Island,Alaska. Miscellaneous Publications of theUniversity of Alaska, Vol II. Published by theDepartment of the Interior, Washington, D.C.

Giddings, J.L. (1964) The Archeology of CapeDenbigh. Brown University Press, Providence,Rhode Island.

Guthrie, R.D. (1983) Paleoecology of the site andits implications for early hunters. Chapter Six inDry Creek, Archeology and Paleoecology of aLate Pleistocene Alaskan Hunting Camp (W.R.Powers, R.D. Guthrie, and J.F. Hoffecker, ed.).National Park Service.

Larsen, H. (1968) Trail creek, Final report on theexcavation of two caves on Seward Peninsula,Alaska. Acta Arctica Fasc., XV, Copenhagen.

Saleeby, B., and A. Demma (2001) Protohistoricfauna from the Kitluk River site on the SewardPeninsula, Alaska. In People and Wildlife inNorthern North America, Essays in Honor ofR. Dale Guthrie (S.C. Gerlach and M.S. Murray,ed.). BAR International Series 944, HadrianBooks, Oxford.

Stephenson, R.O., S.C. Gerlach, R.D. Guthrie, C.R.Harrington, R.O. Mills, and G. Hare (2001)Wood bison in late Holocene Alaska and adja-cent Canada: Paleontological, archaeologicaland historical records. In People and Wildlifein Northern North America, Essays in Honorof R. Dale Guthrie (S.C. Gerlach and M.S.Murray, ed.). BAR International Series 944,Hadrian Books, Oxford.

Vinson, D.M. (1993) Taphonomic analysis offaunal remains from Trail Creek caves, SewardPeninsula, Alaska. M.A. thesis, University ofAlaska, Fairbanks.

Yesner, D.R. (1989) Moose hunters of the borealforest? A re-examination of subsistence pat-terns in the western subarctic. Arctic, vol. 42,no. 2, p. 97–108.

Yesner, D.R. (2001) Human dispersal into interiorAlaska: Antecedent conditions, mode of colo-nization, and adaptations. Quaternary ScienceReviews, vol. 20, p. 315–327.