WhyWe’re Still Arguing About the PleistoceneOccupation of the AmericasNICOLE M. WAGUESPACK

Although empirical issues surround the when, how, and who questions of New Worldcolonization, much of current debate hinges on theoretical problems because it hasbecome clear that our understanding of NewWorld colonization is not resolute.1 In fact,the central issues of debate have remained essentially unchanged for the last eightyyears. The now classic and probably incorrect story of NewWorld colonization begins inLate Pleistocene Siberia, with small a population of foragers migrating across Beringia(�13,500 calendar years before present (CYBP) (Box 1) through an ice-free corridor andtraveling through the interior of North America. Highmobility and rapid population growthspurred southward expansion into increasingly distant unoccupied regions, culminatingin the settlement of the Southern Cone of South America. Armed with the skills andweapons needed to maintain a megafauna-based subsistence strategy, early colonistsnecessarily had the adaptive flexibility to colonize a diverse array of Pleistocene land-scapes. For a time, this scenario seemed well substantiated. The earliest sites in SouthAmerica were younger than their northern counterparts, fluted artifacts were foundacross the Americas within a brief temporal window, and projectile points capable ofwounding elephant-sized prey were commonly found in association with proboscideanremains. The Bering Land Bridge connecting Asia to Alaska and an ice-free corridor pro-viding passage between the Pleistocene ice masses of Canada seemed to provide aclear route of entry for Clovis colonists. However, recent archeological, paleoenviron-mental, biological, and theoretical work largely questions the plausibility of these events.

The status of Clovis as a technologyrepresentative of the initial colonists isperhaps the most tenuous of assertions.Claims of pre-Clovis-aged sites have

been omnipresent in the Paleoindianarena and, until recently, have consis-tently been dismissed under professio-nal scrutiny. Plagued by poor chronolog-ical control, problematic site contexts,questionable recovery techniques, andsometimes a lack of unequivocal arti-facts, widely accepted evidence of a pre-13,500 CYBP occupation was absent.By the 1990s, however, pre-Clovisoccupation had gained considerablesupport, if not outright acceptance, bythe majority of archeologists.3

Monte Verde component II, dated tobetween �14,650–15,600 CYBP4 whilestill controversial,5 provided the first dis-tinctively non-Clovis evidence of occupa-tion, (a lithic assemblage comprised pri-marily of flake tools and two El Jobopoints). More recently, the Cactus Hilland Topper sites were also found to con-tain lithic assemblages possibly dating to�17,000–20,000 CYBP. Excavations atCactus Hill in Virginia revealed a lithic

assemblage dominated by small bladesand cores below a Clovis component in adune adjacent to the Nottaway River.Dates range from to 15,070 6 70 to16,9406 50 14CYBP.6,7 Two small penta-gular points, somewhat reminiscent ofClovis, that were produced on flakeblanks were also found. The oldest occu-pation layer, marked by an increase inphosphates and organics, is separatedfrom the Clovis level by several inches ofsand. Concerns have been raised regard-ing the discreteness and dating of thepre-Clovis occupation level, while geo-morphological analysis of the site depos-its has not been conclusive.7,8 Topper hasan assemblage of over one thousandsmall flakes and microblades, includingburins and burin spalls, found in a ter-race of the Savannah River in Georgia.9

As at Cactus Hill, the early material wasfound below a Clovis occupation level.Tentative dates of �16,000 14CYBP, aswell as claims in excess of 50,000 14CYBPhave been proposed.10Many ambiguitiesregarding site stratigraphy remain. Somearcheologists have referred to the artifactassemblage as a collection of ‘‘geofacts’’.Importantly, however, these sites and theenduring claims of a pre-14,000 CYBPcomponent at Meadowcroft, Pennsylva-nia,11 have provided the impetus for anadvanced suite of distinctions in the Clo-vis versus pre-Clovis debate.The ‘‘pre-Clovis’’ position of New

World colonization is more accuratelydescribed as a pre- or post-13,500,CYBP debate regardless of technologi-cal affinity. If sites such as Topper,Meadowcroft, and Monte Verde gainwidespread acceptance, it wouldimply that colonization occurred notonly earlier, but by peoples with dis-tinctively non-Clovis-like technolo-gies. If that is the case, Clovis could beconsidered a derivative of an earlierNewWorld technology. Consequently,pre-Clovis proponents have begun to

ARTICLES

Nicole M. Waguespack is an assistantprofessor of Anthropology at the Univer-sity of Wyoming. Her research concernsPaleoindian archeology, zooarcheology,and hunter-gatherer ecology; her pub-lished works concern forager subsistenceand labor strategies. She has participatedin field work at numerous Early Paleoin-dian localities throughout the UnitedStates and is currently investigating a Fol-som-aged campsite in Middle Park, Colo-rado. E-mail: nmwagues@uwyo.edu

Key words: Paleoindian; colonization; Clovis; pre-

Clovis

VVC 2007 Wiley-Liss, Inc.DOI 10.1002/evan.20124Published online in Wiley InterScience(www.interscience.wiley.com).

Evolutionary Anthropology 16:63–74 (2007)

question the temporal and morphologi-cal consistency of Clovis assemblages. Itis argued that regional variation in Clovisand other penecontemporaneous tech-nologies such as, Chesrow, Miller, andEl Jobo reflect unique technologicaladaptations as colonists settled into par-ticular environments.6 Alternatively,technologically distinct artifact assemb-lages in excess of the 13,500 CYBP dividecould be considered evidence of failedcolonization attempts.12 Moreover, thepossibility that pre-13,500 CYBP sitessimply reflect earlier variants of Clovistechnologymust also be considered.Pertaining to the latter possibility,

the Schaefer, Hebior, and Mud Lakesites in Wisconsin near the Laurentideice margin contain mammoth remains.Much like Clovis sites, the faunal re-mains have been interpreted as display-ing human modification (dismember-ment and butchery) of proboscideans.Radiocarbon dates on wood and bonecollagen range between 12,310 6 60and 13,500 14CYBP.13 Given the lack ofdiagnostic artifacts, the possibilityremains that these sites represent theactivities of Clovis peoples. Human-modified mammoth bones from the

Great Plains, dating to the Last GlacialMaximum (LGM), have also beenreported as evidence of early occupa-tion.14 If these sites represent mam-moth use comparable to that repre-sented in the Clovis archeological re-cord, they may extend the Clovis timeframe deeper into the Pleistocene.

The question of timing rests, unfortu-nately, on extremely variable and subjec-tive site reporting standards, often innonpeer-reviewedpopularmedia outlets.However, it must also be acknowledgedthat the chance of discovering the firstsite, or evenmultiple sites, created by theinitial colonists is extremely unlikely.While the dating of colonization eventsin other regions, such as Australia, havebeen rapidly pushed backward in time,pre-13,500 sites in the Americas remainrare and controversial.15 However,accepting the notion that Monte Verdeand the earliest dated Clovis site (theAubrey site in Texas, 13,350–13,500CYBP) do not respectively represent theearliest occupants of South and NorthAmerica renders a pre-13,500 coloniza-tion date certain.16 The question, ofcourse, is how much earlier? Relatedconcerns at the forefront of contempo-

rary Paleoindian studies are attempts todiscern cultural and biological relation-ships from technology, skeletal remains,and genetic variability; the evaluation ofpossible migration routes; the demo-graphic characteristics of colonists; andinterpretations of Early Paleoindian sub-sistence. In this paper, these generalthemes are reviewed individually, fol-lowed by descriptions of two compre-hensive scenarios that embody thespectrum of ideas concerning NewWorld colonization.

EARLY PALEOINDIANTECHNOLOGICAL ANTECEDENTS

Explicit comparisons of Paleoin-dian and Eurasian technologiesappeared in the 1930s and haveremained a continuous topic ofresearch and debate. Two generalapproaches, both loosely based on bi-ological models of morphologicalchange, can be identified in the cur-rent literature. For simplicity, I referto these as the ‘‘clinal’’ and ‘‘type fos-sil’’ models of technological change.Interpretations of technological rela-tionships derived from clinal models

Box 1: Calibrated Radiocarbon Dates

Radiocarbon calibration curve (gray) for the late Pleistocene showing approximatecalibrated ages for the Clovis complex and the Monte Verde and Cactus Hill sites.Calibration curve reproduced fromOxCal v. 3.10.2

Due to more reliable radiocarboncalibration curves, those accustomedto seeing an 11,000 to 11,500 BP datefor the Clovis complex and NewWorldcolonization will need to calibrate theirmental timelines. Radiocarbon cali-bration by itself pushed New Worldcolonization beyond the 12,000 BPbarrier, which had held for so manyyears. A simple rule of thumb is thatall Pleistocene radiocarbon ages(14CYBP) are typically 1,500 to 4,000years younger than their equivalentcalendar ages (CYBP). The Pleisto-cene-Holocene boundary at 10,00014CYBP actually dates to approxi-mately 11,500 CYBP. The Clovis com-plex, dating to approximately 11,500to 10,800 14CBP represents 13,340–12,830 CYBP. Not only is the Clovisperiod approximately 2,000 years

older than originally believed but, dueto the steepness of the calibrationcurve, is actually shorter, having beencompressed from a length of 800 C14

years to 500 calendar years. Radiocar-bon calibration has not significantlychanged discussions of New Worldcolonization. However, it has important

implications regarding the demographyand genetics of Native American popu-lations since the time range of NewWorld prehistory has been expanded.

64 Waguespack ARTICLES

of variation rest on the assumptionthat technology changes in a predict-able and regular manner as a functionof geographic and temporal distancebetween biological populations ofshared ancestry. As applied to NewWorld colonization, the derived ex-pectation is that as one gets closer,both temporally and geographically,to the source of a Clovis or pre-Clovisprogenitor population, shared techno-logical characteristics will becomeincreasingly apparent.Most Clovis and pre-Clovis sites are

loosely linked by the production ofbifacial projectile points and/or theuse of flake and blade tools. However,the fluted point and its associatedtoolkit undeniably provide the firstwidespread, technologically consist-ent material cultural record present inunglaciated regions. The productionof large blades from conical and, lesscommonly, wedge-shaped cores isnow recognized as a relatively consist-ent, albeit regional, reduction strategyemployed by Clovis peoples.17,18

Assemblages from the SouthernPlains now provide a well-docu-mented record of retouched toolsmanufactured on large, longitudinallycurved blades with associated produc-tion debris. Whether Clovis is truly‘‘first’’ or not, the central questionremains: How can technologicalattributes be used to reconstruct thebiological or phylogenetic relation-ships between colonizers and their an-cestral populations?If colonization occurred via the

Bering land bridge and the earliestmaterials from Alaska are products ofthis event, then the Pleistocene arche-ological record of Alaska can beexpected, using a clinal approach tolithic variation, to display the strong-est Asian affinities of any New Worldtool industry. Assemblages from theTanana Basin of central Alaska fromthe sites of Swan Point, Mead, andBroken Mammoth consist of micro-blades, small burins, and scrapers.Dated from 13,400–14,000 CYBP, thematerial has broad morphologicalsimilarities to the wedge-shapedmicroblade core industries foundthroughout Siberia.19 In Alaska,microblade assemblages are followedby the Nenana complex, which maydate to as early as 13,400–13,900

CYBP, but is more consistently datedto �13,000 CYBP.20,21 Based onassemblages from sites such as DryCreek, Walker Road, Swan Point, andHealy Lake in the Nenana and Tananavalleys, the Nenana complex is pre-dominately characterized as a flakeand macroblade industry, althoughmicroblades are sometimes present.Characteristic of the complex arelarge bifacial knives, retouched bladetools, endscrapers, pieces esquilles,and projectile points with diagnosticteardrop to triangular shapes (i.e.,Chindadn points).

While the colonization of regionsadjacent to Beringia is sufficientlycomplex to render a complete reviewof the topic inappropriate for this pa-per, the basic spatio-temporal trendsrevealed by new sites and dates war-rants discussion due to its importanceto clinal arguments. The earliestUpper Paleolithic technologies ofSiberia appear �40,000 CYBP fromlocalities below 558 latitude.22 TheYana RHS site, situated 718 latitude,dated to �30,000 CYBP, is currentlythe oldest known site above the arcticcircle in Siberia. This site has a mixedassemblage of bifacial and unifaciallyworked flake tools, various bone tools,and no blades.23 Though Yana indi-cates the presence of humans in thehigh arctic before the LGM, it doesnot establish occupation of regionsnear the land bridge, since it sitsapproximately 1,500 km west of theBering Strait. Trends in the Eurasianradiocarbon record suggest that theoccupation of Siberia slowly grew inboth population and geographicextent from �40,000–22,000 CYBP,and that its growth slowed or declinedduring the LGM.24,25 Continuoushigh-latitude occupation of EasternEurasia in regions closest to Beringiaprobably did not occur until after�20,000 CYBP. Following post-LGMclimate amelioration, expansionnorthward and eastward restrains thetemporal window for establishing thepresence of viable populations forcontinued migration across theBering Strait.26

The earliest sites in Western Berin-gia, Berelekh (16,000–12,500 14CYBP),Ushki-1 (13,000–11,500 14CYBP), andUshki-5 (13,000–11,500 14CYBP), con-tain both blade and bifacially reduced

tool forms.26 As a result of recentwork, Ushki, previously reported asdating to 19,000–12,000 CYBP, is nowknown to be no older than 13,000CYBP.27 A microblade core, bifacialknives, large blades, and a ‘‘Chindadn-like’’ point on the surface20 at Bere-lekh, as well as the nonmicrobladecomponents at Ushki, which includelarge blades and leaf-shaped bifacialpoints, have possible parallels to Alas-kan Nenana assemblages. It has beenargued that a general trend of increas-ing reliance on bifacial reduction, cul-minating in Clovis projectile pointtechnology, signifies a trajectory oftechnological change originating inthe blade-dominated industries ofUpper Paleolithic Siberia. However,the simple fact remains that the ear-liest material from Alaska, consistinglargely of microblades and burins,bears little resemblance to Clovis tech-nology.The type-fossil approach relies

more heavily on tracking the move-ment of unique, presumably culturallyspecific, technological attributesthrough space and time. From thisperspective, it is assumed that discretecomponents of material culture, suchas specific tool morphologies and/orunique processes of reduction, aremore resistant to change than others.It is expected that at least somecomponent of a colonizing popula-tion’s material inventory will beretained through the migration pro-cess and reflect the population fromwhich it originated. Since few for-mally recognized and defined toolforms are present in Clovis assemb-lages, efforts to trace specific techno-logical attributes have focused almostexclusively on projectile-point mor-phology and the occurrence of basalthinning and fluting. Although numer-ous fluted points have been foundthroughout Alaska, nearly all arederived from surface sites or shallowdeposits of questionable stratigraphiccontext.28 Tentative dates on thesefinds attribute them to depositsyounger than 13,000 CYBP, consider-ably younger than Clovis point findsthroughout the interior United States.The Uptar site in Northeastern Sibe-ria, known only to predate 8,30014CYBP, produced a much celebrated‘‘fluted’’ point.29 This single artifact, a

ARTICLES Colonization of the Americas 65

somewhat crude bifacially shapedflake bears no resemblance to Clovisbiface morphology except for the pres-ence of a possible basal thinning flakeremoved from one surface.Since fluting, the most distinctive

feature of Clovis technology, currentlystands as the first clear technologicaldeparture from an Asian and/or Alas-kan substrate, some researchers havelooked elsewhere for type fossil mate-rial parallels. Bradley and Stanford30

have argued that the shared use ofoutre passe flaking (a method ofbiface thinning through controlled re-moval of flakes that span the entirewidth of the biface), the production ofbeveled bone or antler rods, and theuse of red ochre between Solutreanpeoples of Iberia and Clovis suggest a

European origin for New World colo-nizers. Interest in a Solutrean originhas received widespread attention inthe popular press, but is regardedwith extreme skepticism by archeolo-gists.31,32 Because the Solutrean, ter-minally dated to �19,500 CYBP inSouthern Spain and Portugal, pre-dates Clovis by at least 5,000 years,this scenario necessarily requires thatNew World colonization took placewell before Clovis.

Despite the availability of new ar-cheological data, it can be argued thatwe are no closer to solving the prob-lem of tracing migration through pre-historic material culture then we were80 or even 150 years ago. A staunchlyhistorical particularist approach tolithic technology as a means of iden-

tifying population movement is wellentrenched in New World coloniza-tion research. Yet the approach hasnot revealed any satisfying means ofreconstructing antecedent relation-ships. The inherent diversity withinthe technologies often attempted tobe linked (Box 2) clearly suggeststhat straightforward comparisonsare likely to reveal superficial, if notmeaningless, similarities. If, forinstance, we were able to comparelong lists of typological and techno-logical attributes between EarlyPaleoindian assemblages of theAmericas with Paleolithic assemb-lages of Europe and East Asia, whatwould be the significance of ob-served similarities and differences?An established, theoretically in-

Box 2: Variation in Early Paleoindian Projectile Point Technology

For decades, the fluted pointserved as an iconic representation ofuniformity in Late Pleistocene huntingtechnology of the Americas. How-ever, recent research suggests that amuch more diverse technology mayhave been present at the time of col-

onization or developed soon after-ward. Projectile points from CactusHill and Meadowcroft, possible pre-Clovis aged sites, Chindadn points ofthe Nenana industry, and Clovis andEl Jobo points bear little resemblanceto one another. Discerning techno-

logical parallels between thesediverse point types and the lithicindustries of Europe and Asia is ham-pered by poor chronological controland the persistent problem of linkingbiological and cultural affinities.

66 Waguespack ARTICLES

formed set of criteria is desperatelyneeded for distinguishing betweenwhich, if any, technological com-monalities are most likely to repre-sent antecedent relationships.

BIOLOGICAL PERSPECTIVES

Studies of modern human geneticsand prehistoric skeletal biology havethe potential to improve on the inad-equacies presented by the technologi-cal record of New World colonization,but few if any firm conclusions havebeen reached. Concerted efforts havebeen made to analyze all of the knownLate Pleistocene and Early Holocenehuman remains from the Americas, asample that includes material fromonly six sites older than �11,000CYBP in North America (Table 1). Thevast majority of these remains, includ-ing the famous Kennewick and SpiritCave finds,33 postdate colonization bythousands of years. As famous asthese finds have become, they simplywere not and should not be consid-ered ‘‘colonizers’’ by any stretch ofimagination. Although modern and,to a lesser extent, prehistoric geneticdata are available in increasing abun-dance, theoretical difficulties persistin linking large-scale geographic pat-terns of mitochondrial and y-chromo-some haplotype frequencies to the col-onization processes from which theyderived.Multiple osteometric analyses and

comparisons of material older than8,800 CYBP tend to converge on twopoints: 1) Early skeletal material fromNorth and South America is highly

variable and displays little internalmorphological consistency; 2) Earlyskeletal material bears little cranio-metric resemblance to any singlemodern population.34 The Buhl bur-ial, a single female dated to 10,67514CYBP, has been reported to displaydistinctively Polynesian affinities,35

while others find similarities withrecent Native Americans and Asianpopulations.36,37 The ‘‘Luzia’’ skullfrom Lapa Vermelha IV, a femaleapproximately 11,900–10,200 14CYBPfound in Brazil, is argued to havevague similarities with Australians,Africans, and Polynesians, but notmodern Asians.38 The Spirit Cave ma-terial, dated to �9,000 14CYBP, is alsoidentified as non-Asian and said to ex-hibit the strongest affinities with Ainuor Norse populations.37

These findings and the popularmedia’s fascination with publicizingfacial reconstructions which often uselight-colored clay ‘‘skin’’ and arbitraryeye and hair colors, have led to thewidespread characterization of earlyskeletal material as being decidedlynon-Native American in appearance.However, unless one is willing toassume the untenable position thatthe diverse array of Native peoples ofNorth and South America have notundergone any changes in their cra-niofacial morphology for the last8,000þ years, denial of Native ances-try is premature.39 As it currentlystands, the earliest individuals wehave look dissimilar to each other,and appear largely distinct fromrecent Native North and South Ameri-cans. Variation could be the result of

multiple founding populations34,40

and/or the effects of genetic driftamong small widely scattered popula-tions.41

The genetic record, based on re-gional and population frequencies ofmtDNA and Y haplotypes, does con-sistently indicate an Asian origin fornative New World populations. For abrief time, the identification ofmtDNA haplotype ‘‘X’’ fueled Euro-pean origin theorists. Haplotype X,found in low but significant frequen-cies among the Ojibwa and the Nuu-Chah-Nulth, is found more commonlyamong Europeans than Asians.42

Recent work indicates that the haplo-type is present, albeit in low frequen-cies, among peoples from the Altaimountain area and regions westward,as well as among prehistoric popula-tions of South and North America.43

As opposed to a Pleistocene Europeanpresence in the New World, haplotypeX likely represents the complex colo-nization history of eastern Eurasia.44

Based on mtDNA44,45 and y-chromo-some studies,44,46 posited geographicorigins of colonizing populations isgenerally placed in the area betweenthe Altai Mountains and Lake Baikalof Siberia. Disagreement primarilycenters on the number and timing ofprehistoric migration events neces-sary to account for modern genetic di-versity.Single migration events by peoples

carrying mtDNA haplotypes A-D aresupported by models that account forknown migration events of Nativepopulations throughout the Holo-cene.40 Others argue that mtDNA hap-

TABLE 1. Early Skeletal Finds From North America

SITE LOCATION

�AGE

C14 Calibrated

Anzick Montana 11,115 13,000Arlington Springs Santa Rosa Island, California 11,000 12,900Buhl Idaho 10,675 12,750Marmes Washington 10,500 12,500WarmMineral Springs Florida 10,250 12,000Gordon Creek Colorado 9,700 11,150Wilson Leonard Texas 9,600 10,900Harn Shelter Texas 9,560 10,900Grimes Shelter Nevada 9,470 10,700Spirit Cave Nevada 9,400 10,650Wizards Beach Nevada 9,250 10,450On-Your-Knees Cave Prince of Wales Island, Alaska 9,200 10,300Kennewick Washington 8,400 9,450

ARTICLES Colonization of the Americas 67

lotype B, haplotype X, or haplotypesA-D represent genetically distinctmigration events.43,44,47 Similarly,analyses of Y-chromosome haplotypesshow evidence frequencies that arearguably derived from either a singlePleistocene migration event48 or twoor more founding populations.46,49

Estimates of when populations carry-ing particular haplotypes first enteredthe Americas are highly variable.Estimates for the primary mtDNAgroups range between �35,000–20,000 CYBP; Y-chromosome haplo-type estimates range between�30,000–8,000 CYBP.43,44 Genetic

studies are currently plagued by equi-finality, as it has become clear thatmultiple scenarios of initial coloniza-tion and later population movementscan be devised to account for the mod-ern frequencies of American haplo-types.

ROUTES OF ENTRY INTO THEAMERICAS

Despite the longstanding appeal ofthe ice-free corridor (IFC) as a naturalfeature that effectively funneled earlycolonizers into the interior of North

America, the coastal-migration (CM)model has experienced a resurgenceof support.50 The IFC has clear impli-cations for the spatial distribution ofPleistocene-aged sites in North Amer-ica and later expansion into SouthAmerica. A key to the IFC model isthat the corridor existed and providedan ecologically viable habitat whenthe initial entrants sought passageinto the Americas. As originally con-ceived, the IFC formed through degla-ciation of the Cordilleran and Lauren-tide ice sheets in the Yukon andMcKenzie River Valleys (Box 3) dur-ing the Late Pleistocene.

Box 3: Competing Entry Routes

Researchers are divided on howthe initial colonists of the New Worldmigrated into unglaciated regions.Two options are possible, along thePacific coastline or through an ice-free passageway between the Lau-rentide and Cordilleran ice massesover present-day Alberta and Sas-katchewan. Although some research-ers have favored coastal models sincethe 1960s,51 they have remained less

popular than the Ice-Free Corridorroute until recently. Interpretationsconcerning the productivity of thecorridor vary greatly, depending onthe paleonvironmental indicatorsused, time periods examined, andthe assumptions made regardinghuman subsistence requirements.52

Estimates range from a relativelydiverse landscape capable of bothattracting and supporting human

populations to a wasteland virtuallyuninhabitable by humans. Both routesprovide clear predictions of wherethe earliest sites should be, depictedhere by the placement of ‘‘colonists’’through time either within the corridoror immediately adjacent regions or onthe coast. The earliest sites currentlyknown from these regions, as shown,provide little direct evidence to supporteither model.

68 Waguespack ARTICLES

Assuming that the corridor wasopen and capable of providing at leastminimal food resources, the IFCmodel implies that the earliest siteswould be located within the corridorand the earliest dispersal of colonizerswould occur in the interior of NorthAmerica, just south of the ice sheets.Movement southward and outwardfrom the interior toward the coastswould follow. The initial Clovis findsof the 1930s and 1940s on the GreatPlains provided considerable impetusfor the IFC model. Recent argumentsfor general north-to-south temporaland technological trends in the earlyPaleoindian record53,54 have beenmarshaled to support this view. How-ever, no sites either predating or ofcomparable age to mid-continentalClovis sites have been found withinthe bounds of the corridor. Sites suchas Charlie Lake Cave,55 VermillionLakes,56 and Wally’s Beach57 all post-date 13,500 CYBP.Despite the clear gaps in the archeo-

logical record predicted by the IFCmodel, two key discoveries haverenewed interest in coastal models.First, the relatively widespread accep-tance of Monte Verde presents a conti-nental archeological record where anear-coastal South American site pre-dates all early interior sites. Second,recent geological work concerning thelocation and boundaries of continentalice masses has altered our understand-ing of the corridor. It is now generallyacknowledged that the Laurentide andCordilleran ice sheets coalesced duringWisconsin glaciation. Further, cosmo-genic dating of glacial erratics associ-ated with the Laurentide ice sheet,58 to-gether with other geologic evidence,imply that the corridor was not openbetween 26,000–14,000 CYBP, severelylimiting the temporal ‘‘freedom’’ the IFCmodel allowed for colonization. IfMonte Verde is as old is it appears to be,humans were south of glacial ice beforethe opening of the corridor.Key features of CM dispersal sce-

narios are that the northern coastlineregions of North America were rela-tively ice-free by at least 15,000CYBP, if not earlier, and that the out-ermost regions of the coast experi-enced only brief intermittent periodsof ice accumulation throughout thePleistocene.59,60 Considered a viable

route, most CM adherents furthersuggest that the initial colonizers fol-lowed the coast from initial entryinto North America all the way toSouth America, dispersing inlandonly after coastal environmentsaccrued unsustainable populationlevels. This initial colonization ofcoastally adapted foragers in bothNorth and South America was fol-lowed by eventual migration inland.Appealing features of the CM modelinclude a temporally flexible entrydate into the Americas and clear pre-dictions concerning where early sitesshould be found. If it is assumed thatcolonists followed the coast in orderto maintain occupation of familiarand preferred coastal habitats, it fol-lows that coastally situated sitesshould predate inland sites and pro-vide evidence of a coastal foraginglifestyle such as aquatically derivedfauna and associated technologies.Neither has been found. None of theearliest sites along the northern coastpredate 13,000 CYBP,61 with the pos-sible exception of the Manis Masto-don locality in Washington62 andDaisy Cave in California.63 Furthersouth, some South American local-ities64 and skeletal finds may provideClovis-age dates.

The paucity of early sites on the Pa-cific coast is generally explained as theproduct of submerged shorelines. Asthe ice masses melted in the LatePleistocene and Early Holocene, thePleistocene coastline and the sitespotentially situated on it were inun-dated. While that is a valid point, itremains difficult to accept the notionthat all of the presumably thousandsof early coastal sites were located onthe outermost regions of the coast andare now under water. In fact, the keyto arguments regarding the habitabil-ity of northern coastlines is the pres-ence of bears in Pleistocene depos-its.65 The presence of such large omni-vores suggests that humans could alsohave successfully foraged in suchenvironments. However, this also begsthe question of why ‘‘bear sites’’ arediscovered but human ones are not. Inaddition to the lack of sites, argu-ments based on optimal foraging anddemographic models have questionedthe plausibility of migrants maintain-ing a coastally limited dispersion to

South America before venturinginland.66

PLEISTOCENE MEGAFAUNA:PALEOINDIAN SUBSISTENCE AND

EXTINCTION

Few images of Pleistocene foragersin the Americas have captured theimagination like that of a small groupof spear-wielding mammoth hunters.While captivating, the certainty ofNew World colonists as big gamehunters has come under scrutiny. Therepeated association of Pleistocenemegafauna and artifacts establishedin the mid-twentieth century remainson solid empirical ground, as manyfauna bearing Clovis sites do containextinct species, particularly mam-moth. However, interpretations of thezooarcheological record vary widely.The debate largely centers on the sig-nificance, not the presence, of mega-fauna in the archeological record.Marking a departure from long-

standing interpretations of Clovis,recent arguments classify New Worldcolonists as generalized foragers whoused a diverse array of large and smallprey, plant resources and infrequentuse of megafauna. It is argued thattaphonomic and zooarcheologicalbiases favoring the preservation, dis-covery, and analysis of megafaunahave produced an archeological re-cord that erroneously over-representsthe contribution of large prey to theClovis diet.67 As a result of better pres-ervation, visibility, and attention byzooarcheologists and paleontologists,the record of megafauna predation isdeemed to present an inflated view ofmegafauna use. Assuming that a morecomprehensive Clovis zooarcheologi-cal record not associated with mega-fauna exists, but remains largelyundiscovered is problematic, but notimprobable. Widespread associationbetween Late Pleistocene artifacts andonly small game prey species and arti-facts or residues attesting to frequentplant use has yet to be discovered.However, new interest in the small-game assemblages associated withClovis artifacts has made it clear thata broad array of prey species wasused.68,69

ARTICLES Colonization of the Americas 69

In order to winnow the zooarcheo-logical record into a sample includingonly the faunal material with theclearest evidence of human modifica-tion, Grayson and Meltzer70 arguethat only 14 of 76 Early Paleoindiansites show subsistence use of extinctmegafauna. According to Graysonand Meltzer, this evidence does notbode well for the large-game specialistargument. Based on counts of preyspecies by body size class, it has beenargued that a diverse array of specieswere used and that smaller prey out-number megafauna in Clovis sites.68–70

From this perspective, the distinctionbetween specialized and generalizedhunting strategies depends exclusivelyon the use of small-bodied taxawhereby generalists use small gameresources and specialists do not. Indirect opposition to a pan-continentalsubsistence regime based on huntingPleistocene fauna, proponents of a gen-eralized food economy argue that whilenow-extinct mammoths and masto-dons may have been hunted occasion-ally, the vast majority of the diet camefrom a variety of floral and faunalresources.Other studies of subsistence have

reached different conclusions.71–73

Waguespack and Surovell73 argue thatjust as the hunting of megafauna byitself does not necessarily indicate aspecialist subsistence strategy, thehunting of small game does not neces-sarily indicate a generalist subsistencestrategy. Instead, the distinctionbetween specialist and generaliststrategies hinges on the relationshipbetween what could have been huntedand what was actually taken. Based onthe relative abundances of large andsmall game in Clovis faunal assemb-lages, they argue that Clovis hunter-gatherers were regularly ignoringopportunities to take small-bodied spe-cies, instead focusingmost of their sub-sistence effort on the largest availableprey. In contexts where human popula-tion densities are low and large gameare available (presumed conditionsduring colonization), it is proposedthat a specialized strategy incorporat-ing large high-ranked prey is the pre-dicted use of resources derived fromoptimal foraging models. They arguethat the traditional ‘‘Clovis as special-ist’’ argument not only still holds merit

but receives strong support from thearcheological record.

The role of early colonists in theextinction of Pleistocene megafaunaremains divided in accordance withsubsistence arguments. In addition toclimate change and human preda-tion,72,74–76 hypervirulent pathogenshave recently been invoked as causalagents.77 The disease argument sug-gests that during colonization multi-ple species of Pleistocene megafaunawere exposed to novel pathogens car-ried by humans or associated speciessuch as domesticated dogs.78 Exclud-ing the possibility that NewWorld col-onization occurred significantly ear-lier than is currently believed, Pleisto-cene extinctions are roughly coincidentwith the arrival of humans. Yet littleevidence of human predation onextinct species except mammoths,mastodons, and possibly camels andhorses has been found, promptingthe claim that the North Americanoverkill hypothesis has itself gonethe way of mammoth.79 The reorgan-ization of habitats,80 massive vegeta-tional changes,81 and the loss of key-stone species82 can cause extinctions,but why this process had such detri-mental effects during the Pleistocene-Holocene transition and not duringprevious glacial to interglacial transi-tions remains problematic. Alterna-tively, human predation, while provid-ing a clear mechanism for extinction,cannot be demonstrated to have occur-red for the vastmajority of extinct species.

TIME, SPACE, AND DEMOGRAPHY

Although routes of colonizationwithin the New World may have takenthe form of a wave-like expansionacross space,75,83 a linear spreadthrough river valleys,84 or perhaps acombination of both,85 no clear spa-tio-temporal patterns of colonizationhave emerged from the archeologicalrecord. Most archeologists accept aninitial Beringian entry point, yet theoldest sites in the Americas are farfrom Beringia if purported pre-Cloviscontenders east of the Mississippi andin South America are reliably dated. IfClovis represents the colonizing popu-lation of North America, one mightexpect the greatest densities of Clovispoints to be in the American Great

Plains or Northwest.86 Despite theincreasing popularity of models ofcoastal colonization, they are dramat-ically weakened by the paucity of earlysites in the far west. Such contradic-tions between data and hypotheseshave served as the inspiration forwork attempting to understand whatthe archeological signature of a colo-nizing population might look likeunder various migration scenarios.Recent studies have begun to explorethe links between human demographyand the spatio-temporal dynamics ofNewWorld colonization.87

Advocates of a pre-Clovis entry dateoften refer to the Monte Verde site asnecessitating thousands of years ofoccupation before entry into SouthAmerica. It is also often assumed thatpopulations must achieve criticallyhigh densities and/or accumulatean adequate degree of landscapeknowledge before venturing into un-occupied, unfamiliar landscapes.88,89

While the motivation and prepared-ness of colonists to migrate throughand occupy an ecologically diversecontinent are important issues forascertaining the tempo of coloniza-tion, basic demographic factors mustfirst be explored.Notable colonization scenarios have

assumed high rates of mobility andfertility75,90 and have been criticizedfor relying on the seemingly implausi-ble combination of rapidly movingkid-toting foragers.91 Mathematicalmodels and comparisons of ethno-graphically derived mobility and fer-tility rates for hunting and gatheringpopulations have established thatrapid colonization is congruent withfrequent residential moves.92 Rates ofpopulation growth, upwards of 2.5%per year, can be and are maintainedby mobile foragers. These findings es-tablish that it is at least possible thatthe entirety of North and South Amer-ica could have been colonized bysmall initial populations in as little as1,000 to 2,000 years.92 Though demo-graphically plausible, verifying thatcolonization took place rapidly orslowly has proven extremely difficult.If we had the luxury of watching

the colonization of the New Worldhappen, we would see populationgrowth and migration decisionscombine, eventually causing people

70 Waguespack ARTICLES

to occupy all of North and South Amer-ica. But apparently our sampling of thearcheological record does not cleanlytranslate into cells on a roll of film thatcan be put into spatial and chronologi-cal order to make a coherent story.Continuing this analogy, in theoryradiocarbon dating could be used toarrange our cells (sites) in chronologi-cal order, but the inherent imprecisionof the radiocarbon method makes itpossible to order them only in a vagueway. Assuming we had a large enoughsample of dated sites, however, wemight expect a somewhat ordered colo-nization process to emerge from arche-ological sampling. Arguing from math-ematical traveling wave models, Hazel-wood and Steele93 have shown thatcolonizations characterized by long-distance dispersal, not necessarilycoupled with high population growth,may not produce easily detectablespace-time trends in the archeaologicalrecord. So while there is no disagree-ment that the ‘‘peopling’’ of the NewWorld involved population growth,how population growth translates intogeographic dispersal is not clear, buthinges on the issue of decision makingby colonizers.What are the rules and currencies

governing decisions about whether,when, and how far colonists shouldmove into unoccupied areas? Bea-ton94 proposed a distinction between‘‘transient explorers’’ and ‘‘estate set-tlers’’ as strategies of expansion forcolonizing populations, defining tran-sient explorers as following a patternof long-distance moves by smalldaughter colonies that maintain mini-mal social and biological contact withtheir progenitor group. In contrast,estate settlers are typified by themovement of larger groups settling‘‘closer to home’’ who sustain contactwith their friends and relatives. Demo-graphically, the transient-explorerstrategy is argued to be riskier, assmall groups in unfamiliar settingsare more likely to be susceptible tostochastic events, both genetic and ec-ological, while estate settlers follow amore conservative path, maintainingsocial and ecological familiarity.Though Beaton’s terms are not alwaysused, the strategies he outlined delin-eate the basic dichotomy between pro-ponents of a rapid, ecologically

diverse expansion front limited byonly a few key resources15,95,96 versusadvocates of slower selective migra-tion into favored habits.12,88

Would New World colonizers havebeen characterized by rapid or slowpopulation growth? How did theydecide to move into unfamiliar regions?Would these decisions be based on eco-nomic concerns, social concerns, orperhaps both? Would the unfamiliarityof unoccupied lands hinder coloniza-tion? How does colonization behaviortranslate into spatio-temporal archeo-logical residues? These are all issuesthat deserve further attention.

CONCLUSIONS

It is unfortunate that a review ofEarly Paleoindian archeology cannotbe organized around a laundry list ofimportant new sites, dates, and arti-facts. Contemporary research intoNew World colonization simply is nota topic that lends itself to answerssought only ‘‘in the ground.’’ Moreover,developments appear in the intellectualrealm at a faster pace than they do inthe empirical. It has been the theoreti-cal questioning of assumptions and be-havioral implications in light of advan-ces in our understanding of hunter-gatherer ecology, taphonomy, paleoen-vironments, human demography, andthe biases inherent in the archeologicalrecord that have pushed the field for-ward. Lest this review present only astring of plausible options regardingthe colonization process, akin to a‘‘choose-your-own-adventure’’ novel ofNewWorld prehistory, integratedmod-els have emerged that attempt to linkmany facets of colonizing events andbehaviors. I will outline two of these.

The Pre-Clovis, Coastal,Generalized Forager Model

Advocates of pre-Clovis-aged colo-nization (�15,000 CYBP and earlier)generally agree that particular sites inSouth America for example MonteVerde and/or some North Americansites such as Cactus Hill, Meadow-croft, and Topper necessitate a sub-stantially longer period of New Worldoccupation than shown by Clovis. Tosuccessfully navigate their way to

South America and eventually occupya diverse suite of Pleistocene environ-ments, it is postulated that a lengthyprocess involving the acquisition ofknowledge, technological adaptation,and population growth was prerequi-site. In order to minimize the con-straints imposed by unfamiliar land-scapes, it is proposed that the initialmigrants first moved down the coastwhere resource continuity could bemaintained. Inland occupations werethen established when coastal envi-ronments were fully populated or inparticularly favorable locales. Thecoastal model solves one particularlytroubling component of the pre-Clovisarcheological record. If it is acceptedthat New World prehistory extendssignificantly further back in time,thousands of years before Clovis, whyare there so few uncontested sitesdocumenting this occupation? Thecoastal route provides an obviousremedy; the earliest sites have simplybeen concealed by rising sea levels.Pre-Clovis scenarios also have

broad implications for subsistenceand other life-style characteristics.A lengthy period of co-existencebetween colonists and megafaunamight weaken human predation asthe primary factor in Pleistoceneextinction. Computer simulationsindicate that even minimal levels ofpredation over a brief period canresult in the extinction of extremelylarge prey with slow reproductiverates.74 Technologically, pre-Clovisproponents often emphasize the di-versity inherent in Pleistocene-agedassemblages throughout the Ameri-cas. Variation, considered to be sug-gestive of environmentally specificcultural and technological develop-ments that accrue over time, is consid-ered congruent with populations thatgrew slowly and ventured into unfa-miliar environments cautiously.

The Proto-Clovis, Inland Route,Specialized Hunter Model

Arguments for a ‘‘Clovis first’’ modelare often caricatured as assuming thatcolonizers arrived in the Americas atexactly 13,500 CYBP, carrying Clovistechnology and quickly establishingpopulations with high growth rates

ARTICLES Colonization of the Americas 71

and long-distance mobility regimesfueled by megafauna hunting. Theseattributes are then dismissed asunsustainable and extravagant com-pared to ethnographically docu-mented hunter-gatherers. The classicmodel of colonization has become farmore sophisticated and remains aviable option. Supporters of Clovis ascolonizers generally accept that Clovistechnology did not spontaneously de-velop upon arrival into the NewWorld, and that, due to samplingissues, the earliest dated Clovis sitesare extremely unlikely to reveal theearliest Clovis inhabitants. ‘‘Clovisfirst’’ has come to mean that Clovispeoples were the first to colonize largeswaths of the continent. This claimacknowledges that the initial NewWorld migration certainly predatesClovis, but not significantly. It is fur-ther reasoned that Clovis technologydeveloped in North America, so theinitial migration event involved peo-ples with distinct, or Clovis-like, lithictool kits.The difference between pre-Clovis

and Clovis first centers around howmuch earlier than Clovis colonizationoccurred, and arguments aboutwhether Clovis peoples represent alater inland migration event or werepart of the initial colonization pro-cess. The longstanding Clovis versuspre-Clovis divide forces the debateinto semantically inaccurate terms,since it essentially boils down to ashort (hundreds upwards of 1,000years) period of proto-Clovis occupa-tion quickly developing into Clovis ora long (1,000þ years) period of occu-pation by populations with a decid-edly ‘‘pre-Clovis way of life’’. The for-mer position is frequently allied withan inland population spread south ofthe continental ice masses, rapidexpansion into diverse environments,and a hunting-based subsistenceeconomy.Based on optimal foraging models,

it is argued that selective predation onhigh-ranked resources, specificallymegafauna, is predicted in coloniza-tion settings where large game isabundant and human population den-sities are low. Lacking competinghuman populations, colonists are con-strued as having undergone rapidgrowth and expansion. Morphological

changes in Early Paleoindian projec-tile points and radiocarbon datesdrawn from a hemispheric samplehave identified a north-to-south gradi-ent of increasing technological diver-gence.53,54 In conjunction with inter-pretations emphasizing the rapid andwidespread appearance of Clovis tech-nology, a pan-continental subsistencestrategy based on hunting, and theclose parity in dates between acceptedNorth and South American sites,many argue that migration was nothabitat-dependent and that colonistsmigrated quickly outward from thepoint of entry in all available direc-tions.

Numerous variations on the twoscenarios have been proposed,87 butthese positions characterize the mainissues of recent debate. Currently, ourexpectations of the archeological re-cord and the reality of the record lackcongruence. Interpretive ambiguitiesand problems of equifinality continueto render integration of various topi-cal models and lines of evidence into acohesive model. Excavating more sitesand dating more deposits are simplynot enough to resolve these issues.How technology and skeletal mor-phology reflect population level rela-tionships, how our sampling of thearcheological record, coupled withradiocarbon accuracy and prehistorichuman behavior, affect our ability todetect initial colonization events, andour understanding of the zooarcheo-logical signature of prehistoric sub-sistence strategies and the human de-mographic response during coloniza-tion events are all theoretical topicsthat, with greater study, have thepotential to provide far more insightinto NewWorld colonization than doesany single site, assemblage, or date.

It may be tempting to conclude thatthe lack of resolution concerning thewhen, where, and why of colonizationindicates a lack of progress by Paleo-indian archeologists. To the contrary,the current state of debate is the prod-uct of a more sophisticated under-standing of the cultural, biological,and demographic nature of coloniza-tion. Our understanding of the pro-cess is likely to become even morecontentious and uncertain as the com-plex relationships among the migra-tion of people, culture, and biology

are refined. So why are we still argu-ing about the Pleistocene occupationof the Americas? As is always the case,the call for more reliable and consis-tently collected data is warranted.More importantly, theoretical con-structs capable of integrating suchdata are needed. Fortunately, as inter-pretations of New World colonizationbecome increasingly complex, theyare also to likely to provide an increas-ingly more accurate portrayal ofthe Pleistocene occupation of theAmericas.

ACKNOWLEDGMENTS

I would like to thank John Fleaglefor the opportunity to write thisreview—it is always motivating torealize just how little we know aboutthe past. Conversations with ToddSurovell, Robert Kelly, and the Uni-versity of Wyoming students whoparticipated in the seminar course‘‘Colonization of the Americas’’ duringthe Fall semester of 2005 provided anendless source of questions and opin-ions regarding the Pleistocene coloni-zation process and have significantlycontributed to my views on the sub-ject. David Meltzer, John Hoffecker,and two anonymous reviewers pro-vided insightful comments on an ear-lier draft of this manuscript.

REFERENCES

1 Meltzer DJ. 2005. The seventy-year itch: contro-versies over human antiquity and their resolution.J Anthropol Res 61:433–468.

2 Ramsey B. 2005. Ox Cal Program. Oxford:Oxford Radiocarbon Accelerator Unit.

3 Meltzer DJ, Grayson DK, Ardila G, Barker AW,Dincauze DF, Haynes CV, Men F, Nunez L, Stan-ford DJ. 1997. On the Pleistocene antiquity ofMonte Verde, Southern Chile. Am Antiquity 62:659–663.

4 Dillehay TD. 1997. Monte Verde: a Late Pleisto-cene settlement in Southern Chile. WashingtonDC: Smithsonian Series in Archaeological Inquiry.

5 Fiedel SJ. 1999. Artifact provenience at MonteVerde: confusion and contradictions. Sci Am Dis-covering Archaeol 1:1–12.

6 Adovasio JM, Pedler DR. 2004. Pre-Clovis sitesand their implications for human occupationbefore the Last Glacial Maximum. In: Madsen DB,editor. Entering America: northeast Asia andBeringia before the Last Glacial Maximum. SaltLake City: University of Utah Press. p 139–158.

7 Wagner DP, McAvoy JM. 2004. Pedoarchaeologyof Cactus Hill, a sandy Paleoindian site in southeast-ern Virginia, U.S.A. Geoarchaeology 19:297–322.

8 Perron JT, Mandryk CAS. 2000. Micromorphol-ogy of the Cactus Hill site, a potential pre-Clovissite in southeastern Virginia. Geological Society ofAmerica Annual Meeting. Nov. 9-18. Reno, Ne-vada.

72 Waguespack ARTICLES

9 Goodyear AC. 1999. Results of the 1999 AllendalePaleoindian expedition. Legacy 4:8–13.

10 Chandler JM. 2001. The Topper site: beyondClovis at Allendale. Mammoth Trumpet 16:10–15.

11 Adovasio JM, Pedler DR, Donahue J, Stucken-rath R. 1998. Two decades of debate on Meadow-croft Rockshelter. N Am Archaeol 19:317–341.

12 Meltzer DJ. 2004. Modeling the initial coloniza-tion of the Americas: issues of scale, demography,and landscape learning. In: Barton CM, Clark GA,Yesner DR, Pearson GA, editors. The settlement ofthe American continents. Tucson: University of Ar-izona Press. p 123–137.

13 Overstreet DF, Kolb MF. 2003. Pleistocenearchaeological sites with human-modified woollymammoth remains in southeastern Wisconsin,U.S.A. Geoarchaeology 18:91–114.

14 Holen SR. 2006. Taphonomy of two Last GlacialMaximum mammoth sites in the central GreatPlains of North America: a preliminary report on LaSena andLovewell. Quaternary Int 142–143:30–43.

15 Kelly RL. 2003. Maybe we do know when peo-ple came to North America. Quaternary Int 109–110:133–145.

16 Fiedel SJ. 1999. Older than we thought: impli-cations of corrected dates for Paleoindians. Am An-tiquity 64:95–115.

17 Collins MB. 1999. Clovis blade technology. Aus-tin: University of Texas Press.

18 Collins MB, Lohse JC. 2004. The nature of Clo-vis blades and blades cores. In: Madsen DB, editor.Entering America: northeast Asia and Beringiabefore the Last Glacial Maximum. Salt Lake City:University of Utah Press. p 159–183.

19 Hoffecker JF, Elias SA. 2003. Environment andarchaeology in Beringia. Evol Anthropol 12:34–39.

20 Hamilton TD, Goebel T. 1999. The late Pleisto-cene peopling of Alaska. In: Bonnichsen R, Turn-mire KL, editors. Ice age peoples of North Amer-ica. Corvallis: Center for the Study of the FirstAmericans. p 156–199.

21 Hoffecker JF. 2001. Late Pleistocene and earlyHolocone sites in the Nenana River valley, centralAlaska. Arctic Anthropol 38:139–153.

22 Madsen DB, editor. 2004. Entering America:Northeast Asia and Beringia before the Last GlacialMaximum. Salt Lake City: University of Utah Press.

23 Pitulko VV, Nikolsky PA, Girya EY, BasilyanAE, Tumskoy VE, Koulakov SA, Astakhov SN, Pav-lova EY, Anisimov MA. 2004. The Yana RHS site:humans in the arctic before the Last Glacial Maxi-mum. Science 303:52–56.

24 Brantingham PJ, Kerry KW, Krivoshapkin AI,Kuzmin YV. 2004. Time-space dynamics in theearly Upper Paleolithic of Northeast Asia. In: Mad-sen DB, editor. Entering America: northeast Asiaand Beringia before the Last Glacial Maximum.Salt Lake City: University of Utah Press. p 255–284.

25 Kuzmin YV, Keates SG. 2005. Dates are notjust data: Paleolithic settlement patterns in Siberiaderived from radiocarbon records. Am Antiquity70:766–772.

26 Goebel T. 2004. The search for a Clovis progeni-tor in sub-arctic Siberia. In: Madsen DB, editor.Entering America: northeast Asia and Beringiabefore the Last Glacial Maximum. Salt Lake City:University of Utah Press. p 311–356.

27 Goebel T, Waters MR, Dikova M. 2003. Thearchaeology of Ushki Lake, Kamchatka, and thePleistocene peopling of the Americas. Science301:501–505.

28 Bever MR. 2001. An overview of Alaskan LatePleistocene archaeology: historical themes andcurrent perspectives. J World Prehist 15:125–191.

29 King ML, Slobodin SB. 1996. A fluted pointfrom the Uptar site, Northeastern Siberia. Science273:634–636.

30 Bradley B, Stanford D. 2004. The North Atlan-tic ice-edge corridor: a possible Palaeolithic routeto the NewWorld. World Archaeol 36:459–578.

31 Straus LG. 2000. Solutrean settlement of NorthAmerica? a review of reality. Am Antiquity 65:219–226.

32 Straus LG, Meltzer DJ, Goebel T. 2005. Ice Age

Atlantis? Exploring the Solutrean-Clovis ‘‘con-nection.’’ World Archaeol 37:507–532.

33 Chatters JC. 2001. Ancient encounters: Kenne-wick Man and the First Americans. New York:Simon & Schuster.

34 Brace CL, Nelson AR, Qifeng P. 2004. Peoplingof the NewWorld: a comparative craniofacial view.In: Barton CM, Clark GA, Yesner DR, Pearson GA,editors. The settlement of the American continents.Tucson, University of Arizona Press. p 28–38.

35 Neves WA, Blum M. 2000. The Buhl burial: acomment on Green et al. Am Antiquity 65:191–193.

36 Green TJ, Cochran B, Fenton T, Woods JC, Tit-mus G, Tieszen L, Davis MA, Miller S. 1998. TheBuhl burial: a Paleoindian woman from southernIdaho. Am Antiquity 63:437–456.

37 Jantz RL, Owsley D. 2001. Variation amongearly North American crania. Am J Phys Anthropol114:146–155.

38 Neves WA, Powell JF, Prous A, Pucciarelli HM.1998. Lapa Vermelha IV, Hominid 1: morphologi-cal affinities of the earliest known American. Am JPhys Anthropol 26:169.

39 Fiedel SJ. 2004. The Kennewick follies: ‘‘new’’theories about the peopling of the Americas.J Anthropol Res 60:75–110.

40 Mulligan CJ, Hunley K, Cole S, Long JC. 2004.Population genetics, history, and health patternsin Native Americans. Ann Rev Genomics HumGenet 5:295–315.

41 Steele DG, Powell JF. 2002. Facing the past: aview of the North American human fossil record.In: Jablonski NG, editor. The first Americans: thePleistocene colonization of the New World. SanFrancisco: University of California Press. p 93–122.

42 Brown MD, Schurr TG, Torroni A, Bandelt HJ,Allen JC, Schurr TG, Scozzari R, Cruciani F, Wal-lace DC. 1998. Haplogroup X: an ancient linkbetween Europe/Western Asia and North America?Am J HumGenet 63:1852–1861.

43 Eshleman JA, Malhi RS, Smith DG. 2003. Mito-chondrial DNA studies of Native Americans: con-ceptions and misconceptions of the population pre-history of the Americas. Evol Anthropol 12:7–18.

44 Schurr TG. 2004. Molecular genetic diversity inSiberians and Native Americans suggests an earlycolonization of the New World. In: Madsen DB,editor. Entering America: northeast Asia andBeringia before the Last Glacial Maximum. SaltLake City: University of Utah Press. p 187–238.

45 Merriwether DA. 2002. A mitochondral perspec-tive on the peopling of the New World. In: Jablon-ski NG, editor. The first Americans: the Pleistocenecolonization of the NewWorld. San Francisco: Uni-versity of California Press. p 295–310.

46 Karafet TM, Zegura SL, Posukh O, Osipova L,Bergen A, Long J, Goldman D, Klitz W, HariharaS, Knijff P, Wiebe V, Griffiths RC, Templeton AR,Hammer MF. 1999. Ancestral Asian source(s) ofNew World Y chromosome founder haplotypes.Am J HumGenet 64:817–831.

47 Starikovskaya YB, Sukernik RI, Schurr TG,Kogelnik AM, Wallace DC. 1998. mtDNA diversityin Chuckchi and Siberian Eskimos: implicationsfor the genetic history of ancient Beringia and thepeopling of the New World. Am J Hum Genet63:1473–1491.

48 Bianchi NO, Catanesi CI, Bailliet G, Martinez-Marignac VL, Bravi CM, Vidal-Rioja LB, Herrera

RJ, Lopez-Camelo JS. 1998. Characterization ofancestral and derived Y chromosome haplotypesof New World Native populations. Am J HumGenet 63:1862–1871.

49 Lell JT, Sukernik RI, Starikovskaya YB, Bing S,Jin L, Schurr TG, Underhill PA, Wallace DC. 2002.The dual origin and Siberian affinities of NativeAmerican Y chromosomes. Am J Hum Genet70:192–207.

50 Mandryk CAS. 2004. Invented traditions and

the ultimate origin myth: in the beginning. . .there

was an ice free-corridor. In: Barton CM, Clark GA,

Yesner DR, Pearson GA, editors. The settlement of

the American continents. Tucson: University of Ar-

izona Press. p 113–122.

51 Fladmark KR. 1979. Routes: alternative migra-tion corridors for Early Man in North America.Am Antiquity 44:55–69.

52 Mandryk CAS, Josenhans H, Fedje D, Mat-thewes R. 2001. Late Quaternary paleoenviron-ments of northwest North America: implicationsfor inland vs. coastal migration routes. Quat SciRev 20:301–314.

53 Morrow JE, Morrow TA. 1999. Geographic var-iation in fluted projectile points: a hemisphericperspective. Am Antiquity 64:215–231.

54 Pearson GA. 2004. Pan-American Paleoindiandispersals and the origins of Fishtail projectilepoints as seen through the lithic raw-materialreduction strategies and tool-manufacturing tech-niques at the Guardirıa site. In: Barton CM, ClarkGA, Yesner DR, Pearson GA, editors. The settle-ment of the American continents. Tucson: Univer-sity of Arizona Press. p 85–102.

55 Driver JC, Handly M, Fladmark KR, NelsonDE, Sullivan GM, Preston R. 1996. Stratigraphy,radiocarbon dating, and culture history of CharlieLake Cave, British Columbia. Arctic 49:265–277.

56 Fedje DW, White JM, Wilson MC, Nelson DE,Vogel JS, Southon JR. 1995. Vermilion Lakes site:adaptations and environments in the CanadianRockies during the Latest Pleistocene and EarlyHolocene. Am Antiquity 60:81–108.

57 Kooyman B, Hills LV, McNeil P, Tolman S.2006. Late Pleistocene hunting at the Wally’s Beachsite (DhPg-8), Canada. AmAntiquity 71: 101–122.

58 Jackson LE and Wilson MC. 2004. The ice-freecorridor revisited. Geotimes 49:16–19.

59 Fedje DW and Christensen T. 1999. Modelingpaleoshorelines and locating Early Holocenecoastal sites in Haida Gwaii. Am Antiquity 64:635–652.

60 Fedje DW, Mackie Q, Dixon EJ, Heaton TH.2004. Late Wisconsin environments and archaeo-logical visibility on the northern Northwest Coast.In: Madsen DB, editor. Entering America: north-east Asia and Beringia before the Last Glacial Max-imum. Salt Lake City: University of Utah Press. p97–138.

61 Ames KM. 2003. The Northwest Coast. EvolAnthropol 12:19–33.

62 Gustafson CE, Gilbow D, Daugherty RD. 1979.The Manis Mastodon site: early man on the Olym-pic Peninsula. Can J Archaeol 3:157–164.

63 Erlandson JM, Kennett DJ, Ingram BL, GuthrieDA, Morris DP, Tveskov MA, West GJ, Walker PL.1996. The archaeological and paleontological chro-nology for Daisy Cave (CA-SMI-261), San MiguelIsland, California. Radiocarbon 38:355–375.

64 Sandweisss DH, McInnis H, Burger RL, CanoA, Ojeda B, Paredes R, Sandweiss M, GlascockMD. 1998. Quebrada Jaguay: early South Ameri-can maritime adaptations. Science 281:1830–1832.

65 Heaton TH, Talbot SL, Shields GF. 1996. An IceAge refugium for large mammals in the AlexanderArchipelago, Southeastern Alaska. Quat Res 46:186–192.

ARTICLES Colonization of the Americas 73

66 Surovell TA. 2003. Simulating coastal migra-tion in New World colonization. Curr Anthropol484:580–591.

67 Meltzer DJ. 1993. Is there a Clovis adaptation?In: soffer O, Praslov ND, editors. From Kostenkito Clovis: Upper Paleolithic Paleo-Indian adapta-tions. NewYork: Plenum Press. p 293–310.

68 Cannon MD, Meltzer DJ. 2004. Early Paleoin-dian foraging: examining the faunal evidence formegafaunal specialization and regional variabilityin prey choice. Q Sci Rev 23:1955–1987.

69 Byers DA, Ugan A. 2005. Should we expectlarge game specialization in the Late Pleistocene?An optimal foraging perspective on Early Paleoin-dian prey choice. J Archaeol Sci 32:1624–1640.

70 Grayson DK, Meltzer DJ. 2002. Clovis huntingand large mammal extinction: critical review of theevidence. J World Prehist 16:313–359.

71 Barton CM, Schmich S, James SR. 2004.The ecology of human colonization in pristinelandscapes. In: Barton CM, Clark GA, Yesner DR,Pearson GA, editors. The settlement of the Ameri-can continents. Tucson: University of ArizonaPress. p 138–161.

72 Haynes G. 2002. The catastrophic extinction ofNorth American mammoths and mastodonts.World Archaeol 33:391–416.

73 Waguespack NM, Surovell TA. 2003. Clovishunting strategies, or how tomake out on plentifulresources. Am Antiquity 68:333–352.

74 Alroy J. 2001. Amultispecies overkill simulationof the end-Pleistocene megafaunal mass extinc-tion. Science 292:1893–1896.

75 Martin PS. 1973. The discovery of America. Sci-ence 179:69–74.

76 Surovell TA, Waguespack NM, Brantingham PJ.2005. Global archaeological evidence for probosci-dean overkill. Proc Natl Acad Sci 102:6231–6236.

77 MacPhee RDE, Marx PA. 1997. The 40,000-yearplague:humans,hyperdiseaseandfirst contact extinc-tions. In: Goodman S, Patterson B, editors. Natural

change and human impact inMadagascar. Washing-ton,DC:Smithsonian InstitutionPress. p 169–217.

78 Fiedel SJ. 2005. Man’s best friend—mammoth’sworst enemy? A speculative essay on the role ofdogs in Paleoindian colonization and megafaunalextinction. World Archaeol 37:11–25.

79 Grayson DK, Meltzer DJ. 2003. A requiem forNorth American overkill. J Arch Sci 30:585–593.

80 Guthrie RD. 1984. Mosaics, allelochemics, andnutrients: an ecological theory of Late Pleistocenemegafaunal extinctions. In: Martin PS, Klein RG,editors. Quaternary extinctions: a prehistoric revo-lution. p 259–298.

81 Graham RW, Lundelius EL. 1984. Coevolution-ary disequilibrium and Pleistocene extinctions. In:Martin PS, Klein RG, editors. Quaternary extinc-tions: a prehistoric revolution. p 223–249.

82 Owen-Smith N. 1987. Pleistocene extinctions:the pivotal role of megaherbivores. Paleobiology13:351–362.

83 Young DA, Bettinger RL. 1995. Simulating theglobal human expansion in the Late Pleistocene. JArcheol Sci 22:89–92.

84 AndersonDG, Gillam JC. 2000. Paleoindian col-onization of the Americas: implications from anexamination of physiography, demography, andartifact distributions. Am Antiquity 65:43–66.

85 Bettinger RL, Young DA. 2004. Hunter-gath-erer population expansion in North Asia and theNew World. In: Madsen DB, editor. EnteringAmerica: northeast Asia and Beringia before theLast Glacial Maximum. Salt Lake City: Universityof Utah Press. p 239–251.

86 Anderson DG, Faugt M. 2000. Paleoindian arti-fact distribution: evidence and implications. Antiq-uity 74:507–513.

87 Barton CM, Clark GA, Yesner DR, Pearson GA,

editors. 2004. The settlement of the American con-

tinents. Tucson: University of Arizona Press.

88 MacDonald DH. 2004. Early Paleoindians asestate settlers. In: Barton CM, Clark GA, Yesner

DR, Pearson GA, editors. The settlement of theAmerican continents. Tucson: University of Ari-zona Press. p 173–176.

89 Webb E, Rindos D. 1997. The mode andtempo of the initial colonization of emptylandmasses: Sahul and the Americas com-pared. In: Barton CM, Clark GA, editors. Re-discovering Darwin: evolutionary theory inarchaeological explanation. Washington, DC:Archaeological Papers of the American Anthro-pological Association. p 233–250.

90 Kelly RL, Todd LC. 1988. Coming into thecountry: Early Paleoindian hunting and mobility.Am Antiquity 53:231–244.

91 Bryan AL, Gruhn R. 2003. Some difficulties inmodeling the original peopling of the Americas.Quaternary Int 109–110:175–180.

92 Surovell TA. 2000. Early Paleoindian women,children, mobility, and fertility. Am Antiquity65:493–509.

93 Hazelwood L, Steele J. 2004. Spatial dynamicsof human dispersals: constraints on modeling andarchaeological validation. J Archeol Sci 31:669–679.

94 Beaton JM. 1991. Colonizing continents: someproblems from Australia and the Americas. In: Dil-lehay TD, Meltzer DJ, editors. The first Americans:search and research. Boca Raton: CRC Press. p209–230.

95 Fiedel SJ. 2004. Rapid migrations by Arctic

hunting peoples: Clovis and Thule. In: Barton

CM, Clark GA, Yesner DR, Pearson GA, edi-

tors. The settlement of the American conti-

nents. Tucson: University of Arizona Press. p

79–84.

96 Haynes G. 2002. The early settlement ofNorth America. Cambridge: Cambridge Uni-versity Press.

VVC 2007 Wiley-Liss, Inc.

Forthcoming Articles

• Towards a Map of Capuchin Monkeys’ Tool UseEduardo B. Ottoni and Patricia Izar

• Primate group size and socioecological models:Do folivores really play by different rules?Tamaini V. Snaith and Colin A. Chapman

• Alternative Statistical Approaches to the Use of Dataas Evidence in Human Behavioral EcologyMary C. Towner and Barney Luttbeg

74 Waguespack ARTICLES