Mitochondrial DNA Studies of Native Americans: Conceptions ...€¦ · Population Prehistory of the Americas JASON A. ESHLEMAN, RIPAN S. MALHI, AND DAVID GLENN SMITH For several reasons,

ARTICLES

Mitochondrial DNA Studies of Native Americans:Conceptions and Misconceptions of thePopulation Prehistory of the AmericasJASON A. ESHLEMAN, RIPAN S. MALHI, AND DAVID GLENN SMITH

For several reasons, mtDNA hasbeen regarded as particularly useful

for studying prehistory. The humanmitochondrion is an extra nuclear or-ganelle having DNA that exists as acircular molecule 16,569 base pairs inlength, in which all nucleotide posi-tions and coding loci are known.3 Be-cause this DNA is uniquely maternallyinherited and, unlike nuclear DNA,does not recombine, all changes inmtDNA sequence are the result of ac-cumulated mutations inherited frommother to daughter. In addition,mtDNA mutates an order of magni-tude faster than does nuclear DNA,with the control region mutating at aneven greater rate, making it particu-larly useful for analyses at shallowtime depths. Finally, mtDNA exists inhigh copy number in haploid condi-tion. Consequently, it is easily assayedin the laboratory and can be recoveredfrom prehistoric biological materialin sufficient quantities for amplifica-tion and analysis using the polymer-ase chain reaction.

HAPLOGROUPS ANDHAPLOTYPES

Early studies of Native AmericanmtDNA revealed four major clades, orhaplogroups, of haplotypes.4,5 Al-though they are broadly distributedthroughout the Americas,6 these fourhaplogroups exhibit significant re-

gional patterning among native popu-lations of North America. All fourhaplogroups are shared with Asianpopulations, confirming the conclu-sions of classical genetic studies thatthe first Americans migrated fromAsia across the Bering land bridge.4,5

Early analyses of restriction fragmentlength polymorphism in the entire mi-tochondria genome showed that thesefour major clades could be readily dis-tinguished by the gain or loss of oneor more restriction sites or by thepresence or absence of a 9 base-pairdeletion in the COII-tRNAlys inter-genic regions.5 Torroni and cowork-ers7 found that diagnostic mutationsin the CR accompanied the restrictionmarkers and the fragment deletionthat characterize the four haplo-groups, as is expected of a nonrecom-bining DNA molecule. Each haplo-group could be further divided intosubclades or discrete haplotypesbased on additional restriction frag-ment length polymorphisms or spe-cific CR mutations.

Although corresponding haplo-groups can be found in various Asianpopulations, only founding haplo-types of the New World are sharedbetween the two continents, againconfirming that the Americas wereinitially settled by a limited number offemale immigrants from Asia whosemtDNA underwent subsequent evolu-tion independent of its ancestral formin Asia.4 The fact that shared haplo-types on both sides of the Pacific areuncommon has generated consider-able debate as to the size and sourceof the ancestral population (or popu-lations), as well as the number ofwaves of migration that came out ofAsia. However, some haplogroupsshare more than one haplotype withAsia, and it is not clear whether thedivergence they represent occurred in

Jason A. Eshleman is a postdoctoral re-searcher at the University of California,Davis, where he completed his doctoralresearch examining mtDNA extractedfrom prehistoric California sites. He hascoauthored a number of publications ex-amining mtDNA diversity in the NewWorld. E-mail: [email protected] S. Malhi is a postdoctoral researcherin the Department of Human Genetics atUniversity of Michigan in Ann Arbor. He hasconducted his doctoral research on NativeAmerican migrations in the northeasternand southwestern United States with mod-ern mtDNA, as well as an analysis of an-cient DNA from the Columbia Plateau. Hehas recently authored (with Drs. Eshlemanand Smith) an analysis of mtDNA diversityacross North America.David Glenn Smith is a professor of anthro-pology at the University of California, Davis.He directs research in Native American ge-netic variation ancient mtDNA analysis aswell as non-human primate genetics. Un-der his direction, the laboratory of molecu-lar anthropology at UC-Davis has been in-vestigating human migrations with ancientDNA from the Great Basin, the ColumbiaPlateau, the Ohio River Valley, lower GreatLakes, Central California and Central Mex-ico.

Key words: mtDNA, Native Americans, migra-tions, ancient DNA

Evolutionary Anthropology 12:7–18 (2003)DOI 10.1002/evan.10048Published online in Wiley InterScience(www.interscience.wiley.com).

A decade ago, the first reviews of the collective mitochondrial DNA (mtDNA) data fromNative Americans concluded that the Americas were peopled through multiple migrationsfrom different Asian populations beginning more than 30,000 years ago.1 These reportsconfirmed multiple-wave hypotheses suggested earlier by other sources and rejected thedominant Clovis-first archeological paradigm. Consequently, it appeared that molecularbiology had made a significant contribution to the study of American prehistory. As Cann2

comments, the Americas held the greatest promise for genetics to help solve some of themysteries of prehistoric populations. In particular, mtDNA appeared to offer real potential asa means of better understanding ancient population movements. A decade later, none of theearly conclusions remain unequivocal. Nevertheless, in its maturity, the study of NativeAmerican mtDNA has produced a volume of reports that still illuminate the nature and timingof the first peopling and postcolonization population movements within the New World.

Evolutionary Anthropology 7

Asia or the New World. This has madeproblematic the use of mtDNA diver-sity to estimate the time of coloniza-tion, the size and source of the ances-tral population, and the number ofwaves of migration out of Asia.

WAVES OF MIGRATION

Although there has been little scien-tific controversy about the Asian ori-gins of Native American populations,contention surrounds the question ofthe number of waves of migration.The Americas are home to approxi-mately half of the world’s languagestocks.8 This extraordinary linguisticdiversity among the indigenousgroups of Native America suggests tomany comparative linguists that therewas either a single colonization sev-eral tens of thousands of years ago orthat there were multiple colonizationsby speakers of different unrelated lan-guage phyla.8 There is dispute, how-ever, about whether or not linguisticevidence supports an early8,9 or laterfirst occupation of the Americas.10

The pattern of language diversity hasbeen used to support the tripartite di-vision of Native American groupswidely popularized by Greenberg,Turner, and Zegura.11 Although thisdivision initially was suggested muchearlier,12 Greenberg, Turner, and Ze-gura11 proposed that the Amerind,Na-Dene, and Eskimo-Aleut exhibitparallel genetic and morphometricdifferences that indicate three sepa-rate migrations to the New World.Critics noted that the agreementamong linguistics, morphology andgenetics was not as consistent as ini-tially had been claimed and, indeed,the linguistic divisions themselveshave not held up to persistent scruti-ny.13 Nonetheless, the model hasstrongly influenced designs for re-search on Native American popula-tion genetics.

Early analyses of mtDNA indicatedthat the distribution of haplogroupsand the levels of sequence divergenceamong Greenberg, Turner, and Zegu-ra’s linguistic phyla were the result ofmultiple migrations. If the effectivefounding population was small, onlyone matriline would be likely to havesurvived each migration. Thus, mem-bers of different haplogroups that en-tered the New World at the same timeshould exhibit comparable levels of

within-haplogroup diversity. Torroniand colleagues7 noted that sequencediversity within haplogroups A, C, andD was substantially greater than thatin haplogroup B for the populationssampled. This, they argued, was evi-dence of a later migration of B matri-lines to the New World. The fact thathaplogroups A, C, and D are found inEastern Siberia, a likely staging pointfor any trans-Beringian migration,whereas haplogroup B is curiously ab-sent from the region, is consistentwith this argument. These two earlymigrations were argued to be inde-pendent of later migrations of Na-Dene. However, this model did not ad-dress the relationship of Eskimos toother American populations. Becausethe Eskimos represent the most recentarrivals in historical linguistic modelsof the settlement of the New World,the work of Torroni and coworkers7

implies that there were as many asfour independent migrations.

Due to the pronounced regionalpatterning of mtDNA haplogroup andhaplotype frequency distributions inthe Americas, estimates of genetic di-versity are strongly influenced bysampling. Moreover, little is knownabout the actual number of initial col-onists or the population dynamics in-volved in colonizing a continent freeof other humans. Indeed, the degreeto which Native American popula-tions experienced an initial foundereffect or subsequent genetic bottle-neck is itself controversial.14

Horai and coworkers15 hypothe-sized that each haplogroup repre-sented an independent founding pop-ulation. This view is not widelysupported because a random selectionof even a small group of emigrantsfrom Eastern Siberia today wouldhave a high probability of includingmembers of haplogroups A, C, and D.While earlier studies of Native Amer-ican mtDNA seemed to support mul-tiple waves, in line with linguisticmodels of colonization, more recentstudies of mtDNA have supportedonly a single movement out ofAsia.16–19

Single-wave arguments principallyemerged from analyses of larger andmore diverse samples. The argumentshave followed two separate lines ofevidence, though they are quite com-patible. Early studies of Na-Dene pop-ulations suggested that they possessed

high frequencies of haplogroup A butlacked haplogroup B and exhibitedonly low frequencies of haplogroups Cand D, whereas Eskimo-Aleut popula-tions appeared to have high frequen-cies of haplogroups A and D butlacked haplogroups B and C. How-ever, on closer inspection of a largenumber of samples, Merriwether,Rothhammer, and Ferrell16 demon-strated that groups traditionally clas-sified as Eskimo and Na-Dene hadmeasurable frequencies of all fourhaplogroups when larger sampleswere assayed. Reasoning that it is un-likely that separate migrations fromAsia would have introduced exactlythe same four rare Asian types,20 Mer-riwether, Rothhammer, and Ferrell16

concluded that the Americas musthave been peopled from a singlesource. Postcolonization forces mightsubsequently have led to the regionalpatterning in the Americas that ap-peared to differentiate the three hy-pothesized linguistic phyla. Moreover,Lorenz and Smith17 showed thatwhen a larger, more regionally diversesample of haplogroup B was analyzed,within-haplogroup diversity was notless than that for haplogroups A, C,and D. Further consideration thatmore than a single founding haplo-type of one or more haplogroups sur-vives in modern Native American pop-ulations14,21 renders comparisons ofdiversity among the four haplogroupsmoot vis-a-vis their implications withrespect to the number of independentmigrations to the Americas.

The presence of all four haplo-groups in all three of Greenberg,Turner, and Zegura’s11 language phylacould, of course, be the result of ad-mixture after colonization. Neighbor-ing Algonquian and Athabaskan pop-ulations both have the rare mutationassociated with Albumin Naskapi(Al*Nas), possibly as a result of an-cient admixture, given that the Atha-baskan and Algonquian languages ex-hibit no evidence of a close or evenremote linguistic relationship.22 How-ever, haplogroup frequencies alonecannot distinguish between admix-ture and common ancestry. Makingthis distinction requires a more exten-sive analysis of the discrete haplo-types to determine if related types areshared only between neighboringpopulations, suggesting admixture.

Haplotype analyses are also consis-

8 Evolutionary Anthropology ARTICLES

tent with the single-migration hypoth-esis. The CR sequence of the great ma-jority of all Native American membersof haplogroup A, regardless of linguis-tic affiliation, shares a C3T transitionat np16111 that is not seen in anyAsian populations except a few inEastern Siberia, including the Chuk-chi.19 The predominance of thismarker in the Americas and its con-spicuous absence from Asia supportsthe view that this marker originated inBeringia soon after its settlement. Acharacteristic Native American formof haplogroup C that includes theC3T transition at np16325 and aform of haplogroup X that includesthe T3C transition at np16213 areboth widespread in the Americas andabsent from Asia, suggesting a Bering-ian source for, and a single origin of,those haplogroups as well. That a par-ticular marker is widespread in indi-viduals classified as Amerind, Eskimo,and Na-Dene but does not occur inany Asian source outside of easternSiberia suggests that speakers of allthree of the proposed divisions have acommon New World origin.19 More-over, Bonatto and Salzano19 reportedthat the diversity of haplogroup Aamong Greenberg, Turner, and Zegu-ra’s principle Native American lan-guage groupings was remarkably sim-ilar within each of the three linguisticphyla. In addition to the C3T transi-tion at np16111, the Chukchi ofNortheastern Siberia share a C3Ttransition at np16192 with many Na-Dene and Eskimo samples, suggestinga common ancestry for members ofboth language phyla.19,23 The Chukchimight be a rare Asian remnant of aBeringian population that separatedfrom all other Asian groups beforeemergence of the transition atnp16111 and, together with the Na-Dene and Eskimo, experienced a laterre-expansion during which the C3Ttransition at np16192 emerged, assuggested by Forster and coworkers.21

Population contraction among theremnant Beringians, who presumablywere isolated from the Amerind pop-ulations that had earlier moved southinto North and South America, mighthave resulted in the dramatic reduc-tion or extinction of members ofhaplogroups B, C, and D among theBeringians. This scenario is not con-sistent with separate migrations to theNew World for the Na-Dene and Eski-

mos, but only with a later reexpansionout of the north.

While Bonatto and Salzano couldnot estimate the sequence divergenceof haplogroups B, C, and D in the Na-Dene and Eskimo, in whom these hap-logroups are rare, they did assess therelative diversity of haplogroups A, B,C, and D in Amerinds. As in the anal-ysis of haplogroup distributions by

Merriwether, Rothhammer, and Fer-rell,16 larger samples of haplotypesshowed near-equal levels of sequencediversity in all four lineages.24 Lorenzand Smith17 found similar diversitylevels in four haplogroups when geo-graphically diverse samples were con-sidered. Although some authors ques-tion whether the equal diversityestimates are a result of more com-

plete sampling or different methods ofcalculating diversity,25 these resultsare consistent with a single wave ofmigration to the Americas. However,equal diversity within haplogroupswould be expected only if the numberof founding haplotypes is the same.Indeed, future clarification of this dis-pute will require that we consider notonly the level of diversity, but also thepopulation structure, which mightgive better clues to prehistoric ori-gins.26–28

WHERE DID THE FOUNDINGPOPULATIONS ORIGINATE?

Based on geographic proximityalone, Eastern Siberia stands as alikely candidate for the source of theNative American founder population.However, while haplogroups A, C, andD are all found in Eastern Siberia, B isconspicuously absent.29,30 While theabsence of haplogroup B from East-ern Siberia might suggest an addi-tional migration, presumably fromsouthern coastal Asia or South-Cen-tral China,31 where the 9 base pairdeletion is more common,29 it is alsopossible that haplogroup B waspresent in Eastern Siberia before theNew World was colonized but hassince become extinct there.30 If this isnot the case, Eastern Siberia is not alikely candidate for the source of asingle-wave migration. In fact, the mi-tochondrial lineages in Siberia them-selves appear to be derived from otherCentral East Asian populations. More-over, population histories within Si-beria probably have disrupted geneticpatterning since the Americas werefirst colonized.25,32 The restriction toeastern Siberia of some other Asianhaplogroups that exhibit low levels ofdiversity25 might suggest that most ofthe present populations of that regiondescend from a resettlement of east-ern Siberia after Beringia had alreadybeen settled.

Central East Asian populations doexhibit all four lineages common inNative American populations. Popula-tions in Tibet,33 Central China (desig-nated the Chinese Han),34 and Mon-golia20,35 carry detectable frequenciesof haplogroups A, B, C, and D. Merri-wether and coworkers35 and Kolman,Sambuughin, and Bermingham20

cited Mongolia as a likely source for asingle wave of migrations. Y-chromo-

That a particular markeris widespread inindividuals classified asAmerind, Eskimo, andNa-Dene but does notoccur in any Asiansource outside ofeastern Siberia suggeststhat speakers of all threeof the proposeddivisions have acommon New Worldorigin. Moreover,Bonatto and Salzanoreported that thediversity of haplogroupA among Greenberg,Turner, and Zegura’sprinciple NativeAmerican languagegroupings wasremarkably similar withineach of the threelinguistic phyla.

ARTICLES Evolutionary Anthropology 9

some analysis of aboriginal popula-tions in South-Central Siberia nearLake Baikal further support this as alikely staging ground for a Beringianmigration.36 Of course, just as the ab-sence of particular haplogroups fromSiberia does not mean that those hap-logroups were never present there, itis entirely possible that the presenceof markers elsewhere in Asia could bethe product of more recent populationmovements in Asia.

THE CURIOUS CASE OFHAPLOGROUP X, A FIFTH

FOUNDING LINEAGE

The identification of four haplo-groups found in Asia confirmed ear-lier evidence that Native Americanpopulations had Asian origins. Yet inseveral studies of modern NativeAmerican mtDNA, certain similar se-quences appeared that did not fall intoone of the four known lineages.7,14

Undoubtedly some of these repre-sented postcontact admixture. Wardand colleagues14 reported several se-quences sharing transitions atnp16223 and np16278 that, in theirphylogenetic analysis, did not clusterwith any other Native American types.Torroni and coworkers7 reported thata small number of haplotypes foundin the Ojibwa shared a DdeI site loss atnp1715, a marker also shared with alimited number of Europeans. Baillietand coworkers37 and Forster and co-workers21 suggested that the C3T atnp16278, coupled with the absence ofmutations marking haplogroups A, B,C, or D, constituted an additional hap-logroup.

Several lines of evidence now con-firm haplogroup X as a fifth foundinghaplogroup in the Americas. In Brownand colleagues’38 phylogenetic analy-sis, a larger sample of Native Ameri-can sequences from mtDNAs contain-ing the DdeI site loss at np1715 (aswell as an AccI site gain at np14465)and the transitions at np16223 andnp16278 formed a distinct clade. Al-though apparently sharing a matrilin-eal ancestor with the European hap-logroup X at some point deep in time,the Native American sequencesformed their own branches indepen-dent of European representatives ofhaplogroup X.

The distribution of haplogroup X is

also consistent with a pre-Columbiansource. Though presently thought tobe most common among speakers ofAlgonquian languages, haplogroup X,which reaches a frequency of 20% insome Algonquian populations, is geo-graphically widespread throughoutNorth America among groups sharingno close historic or linguistic ties.39

Sequences consistent with haplo-group X also have been reported fromancient human burials in South

America.40 If the presence of thesemtDNAs were the result of recent ad-mixture with modern Europeans,other European haplogroups shouldalso be found. However, among alleg-edly maternally full-blooded NativeAmericans, less than one-half percent(four individuals previously classifiedas “others” in a screening of morethan 800 individuals) were shown tobe members of haplogroup H,39 themost common mtDNA type in Eu-

rope, where its frequency approaches40%.41 In contrast to haplogroup H,haplogroup X, still the least commonNative American haplogroup (�3% ofsamples screened39), is relatively rarein Europe, where it accounts for onlyapproximately 3% of the samplesscreened.41 It is unlikely that admix-ture with Europeans could producethe wide distribution of haplogroup Xwithout also resulting in significantlydetectable levels of other, more com-mon European haplogroups. The highfrequency of haplogroup X among Al-gonquians and several other groupsalso indicates a prehistoric presencein the New World; as such presencereflects the result of common ances-try.39,42

Analysis of ancient DNA also dem-onstrates the presence of haplogroupX but, as yet, no other European hap-logroups in the New World before Eu-ropean contact. CR mutations atnp16223 and np16278 have been re-ported from two samples dating to4000 years BP and another sampledating to 1,000 years BP from lowlandSouth America.40 CR mutations con-sistent with haplogroup X also havebeen found in two individuals fromthe Norris Farms Oneota burials, a700-year-old cemetery in west-centralIllinois.18 Because the transitions atnp16223 and np16278 are also foundin several mtDNA types not associatedwith haplogroup X, sequence dataalone do not provide incontrovertibleevidence of the haplogroup. However,the Norris Farms sequences are virtu-ally identical to those of modern Al-gonquians from the Great Lakes re-gion confirmed to be members ofhaplogroup X.42 Malhi has also foundindividuals with both CR mutationsand the AccI site gain at np14465 inremains dated to 1340 � 40 years BPdiscovered near Vantage, Washing-ton, on the Columbia Plateau. Thehaplotype of this sample included theC3T transition at np16213 thatuniquely characterizes most CR se-quences of members of haplogroup Xfrom the New World.43

Although haplogroup X is now ac-cepted as a pre-Columbian NativeAmerican haplogroup, controversystill surrounds its origin. While theother Native American haplogroupsare found in Central East Asia, haplo-group X had not, until quite recently,been identified that far east,44 occur-

Y-chromosome analysisof aboriginalpopulations in South-Central Siberia nearLake Baikal furthersupport this as a likelystaging ground for aBeringian migration. Ofcourse, just as theabsence of particularhaplogroups fromSiberia does not meanthat those haplogroupswere never presentthere, it is entirelypossible that thepresence of markerselsewhere in Asia couldbe the product of morerecent populationmovements in Asia.

10 Evolutionary Anthropology ARTICLES

ring in highest frequencies in Europeand Western Asia.41 This has led to thehypothesis, fueled by morphometricstudies of the Kennewick Man re-mains in Washington State and otherPaleo-Indian remains,45,46 that therewas a prehistoric migration of Euro-peans to the New World. These re-mains, some dated to more than 9,000years BP, are morphologically distinctfrom most modern Native Americanand Central Eastern Asian popula-tions. The case for repatriation of theKennewick skeleton received consid-erable media attention, driven in largepart by popular accounts stating thatKennewick man has typically Cauca-soid features.47 Osteological analysesof these early Paleo-Indians actuallysuggest closer affinities to Polynesianpopulations and the Ainu of Japanthan to typical Europeans.48 They alsoindicate that, as a group, the earliestAmericans are also more varied thanmodern Native American groups.49,50

While this morphological heteroge-neity could reflect multiple origins, itmight also reflect a more generalizedadaptation of the earliest colonists be-fore the emergence of specialized ad-aptations reflected by later archaictraditions. Although DNA analysis didnot produce amplifiable ancient DNAfrom Kennewick Man,51–53 verifiableDNA from other Paleo-Indian samplesbelongs to typically Native Americanhaplogroups.54–56 Nevertheless, thepopular depiction of Kennewick Manas a pre-Columbian Caucasoid in theNew World, coupled with the discov-ery of haplogroup X as a foundingNative American lineage, fueled pre-mature speculation about early Euro-pean migrations to the New World.57

Genetic evidence does not supportsuch a migration. Furthermore, thelack of other more common Europeanhaplogroups (or other nonmitochon-drial genetic markers) in unadmixedNative Americans makes this scenariounlikely.

Haplogroup X might have origi-nated in Europe or West Asia. It isalso possible that this haplogroup wasonce present in a Central East Asianpopulation that gave rise to foundersof the New World and subsequentlyall but disappeared from Eastern Asia.Some of the mystery surroundinghaplogroup X seems to have beensolved by the recent detection of thishaplogroup, assessed by both restric-

tion fragment length polymorphismand corresponding CR mutations, inAltaian individuals of South-CentralSiberia.44 Haplogroup X is not ascommon in Native Americans as arethe other four haplogroups. Accord-ingly, there is no reason to believe thatit need ever have been common inAsia in order to achieve its presentdistribution in the Americas. It isnoteworthy that Y chromosome hap-lotype 1C is also found in Europe, theLake Baikal region, and the Ameri-cas.36 Haplogroup X might once havebeen present closer to Lake Baikal aswell, and spread both east to Europeand west to the Americas in the same

migrations that spread Y chromo-some haplotype 1C before becomingrarer still in the Baikal region.

The Altaians of South-Central Asiahave been identified as one populationoutside the Americas that contains allfive of the founding Native Americanhaplogroups. They may well be de-scendants of the same source popula-tion that lived in Southern Siberia be-fore both eastern Siberia and Beringiawere colonized. However, it is highlyunlikely that a single source popula-tion has remained stable and un-changed over many thousands ofyears. Indeed, there is substantial ev-idence to the contrary. Certainly pop-

ulation structure and gene pools inthe Americas appear to have been in-fluenced by population expansionsand intracontinental migrations sincesettlement.22,42,54,58 As a consequence,it seems unreasonable to identify anyextant group in the Old World, itself aproduct of many millennia of history,as the single parent population.

HOW OLD ARE THEFIRST AMERICANS?

A molecular clock, first suggestedby Zuckerkandl and Pauling in the1960s, has been employed to use mo-lecular divergence to date prehistoricevents. The rapid mutation rate andunilateral maternal descent of mtDNAappear to make it particularly usefulfor dating events in recent prehistory.Various estimates for the peopling ofthe New World drawn from mtDNAdata, summarized in Table 1, rangebroadly from about 11,000 to over40,000 years BP based on mtDNA di-vergences. This variation results inpart from sampling haplotypes usedto estimate diversity and in part fromvariation in the methods used to cal-culate molecular divergence. Whenthey are presented, the standard er-rors of these estimates and the rangeof the estimates themselves are ratherlarge, thus minimizing the utility ofsuch measures for evaluating differ-ent scenarios in prehistory. It shouldbe noted also that these dates do notnecessarily indicate the establishmentof any population in the New World,but only the separations between NewWorld and Old World lineages, whichmay well have begun in Asia.

Archeological evidence has estab-lished that humans were present inNorth America �11,500 years agowhen the widely distributed and well-dated Clovis culture appeared. Theearliest known human skeletal re-mains date to this time. Althoughthere has long been tantalizing evi-dence of earlier occupations, little of ithas withstood the close scrutiny nec-essary to establish an early humanpresence in the New World.59,60 Add-ing to the controversy surroundingpre-Clovis sites is the apparent lack ofa suitable migratory route out of Ber-ingia. While the Bering Land Bridgewas open throughout the last glacia-tion, an ice-free corridor between theLaurentide and Cordilleran ice

. . . the popular depictionof Kennewick Man as apre-ColumbianCaucasoid in the NewWorld, coupled with thediscovery of haplogroupX as a founding NativeAmerican lineage,fueled prematurespeculation about earlyEuropean migrations tothe New World. Geneticevidence does notsupport such amigration.

ARTICLES Evolutionary Anthropology 11

masses was apparently impassiblefrom 30,000 years BP until perhaps11,000 years BP.61 Thus, pre-Clovispeople would have to have migratedsouthward out of Beringia via the ice-free corridor before 30,000 years BP.

It now appears that a 12,500 year-old occupation level at Monte Verdein south-central Chile establishes thepresence of humans in the Americasbefore Clovis.62,63 This favors a migra-tory route to the New World otherthan the ice-free corridor. Recentanalysis indicates that a coastal pas-sage, open as early as 14,000 years BP,was a likely entry point to NorthAmerica.61 Similarities in mtDNAamong populations of the west coastof North America also appear to sup-port the hypothesis of population ex-pansion out of Beringia associatedwith gene flow along the west coast.64

On its own, this is not conclusive evi-dence of an early coastal entry intoNew World and, indeed, such an ex-pansion might have postdated the ear-liest migrations southward out of Ber-ingia. Nevertheless, this evidenceraises the possibility of such move-ment and should provide fuel for ar-cheological and linguistic investiga-tions of such a claim. In addition,genetic data indicate that populationsof North American brown bears firstreached the modern boundary of the

continental United States via coastalrefugia during the terminal Pleisto-cene rather than through the ice-freecorridor.65 This, at the very least,raises the possibility that humans alsocould have traveled south of the icesheets via a similar route at this time.Consequently, Monte Verde’s pre-Clo-vis occupation lends no support to thehypothesis that the Americas werecolonized as early as 30,000 years BP.

Unfortunately, this has not discour-aged attempts to seek molecular evi-dence of an early human presence inthe Americas. Great variation in diver-gence estimates among molecularstudies results from uncertainties re-garding the proper calibration of mu-tation and divergence rates, the errorestimates associated with these rates,and the events that genetic divergencemay actually reflect. Attempts to cali-brate a mitochondrial mutation ratehave employed divergences dates thatthemselves are uncertain, as is thecase with the Homo-Pan split, theemergence of modern Homo sapiens,and the linguistic break-up of theChibcha language phylum.15,66 Thecalculation of mutation rates hasachieved certain tautological qualitiesas well: Attempts to calculate a rate ofmtDNA sequence divergence used thepeopling of the Americas, assumed tohave occurred between 12,000 and

20,000 years BP as a benchmark at atime when relatively few Native Amer-ican samples had been analyzed.67

Other estimates are based not ondivergence from an Asian source, butinstead on the accumulated diversityin the New World using coalescenttheory.19,21,24 Such estimates exhibit avariability that in all probability is toolarge to be useful for selecting amongalternative hypotheses regarding theinitial peopling of the Americas.68

While the observed high mutationrates in mtDNA do make them partic-ularly useful for analyses at shallowtime depths such as those involvingthe evolution of human populations,such rapid rates of change can alsoadd to the error associated with usingmolecular divergence to date prehis-toric events.

It has been demonstrated that mo-lecular estimates do not always matchthe empirical data. A recent molecularanalysis of modern populations hy-pothesized the presence of haplo-group V in the Basque region of Spainby approximately 10,000 years BPbased on the diversity exhibited bymodern members of this haplo-group.69 However, subsequent analy-sis of 92 ancient human remains inthat region failed to confirm any signsof this haplogroup in the region asrecently as 4,000 years BP.70 This dis-crepancy between the molecular esti-mates and the molecular archeologi-cal record provides no confidence inmolecular estimates of the times ofpast events when those estimates arederived solely from studies of livingpopulations. Unlike the case of Ra-mapithecus, in which molecular datashowing a later divergence of humansand the African apes led paleontolo-gists to reconsider fossil evidence ofan early split for the human line, mo-lecular data suggesting an early occu-pation of the Americas have not led todiscoveries of an early occupation.However, molecular evidence is stillbest used as part of a holistic ap-proach to such inquiries alongsidetraditional archeological evidence ofhuman presence.

ANCIENT DNA AND THEPEOPLING OF THE

NEW WORLD

The analysis of ancient DNA in theNew World has largely confirmed the

TABLE 1. Divergence Dates From mtDNA Data

StudyDivergence Range (Years Before Present; Errorand/or Confidence Intervals Where Reported)

Torroni and co-workers, 199466

A 25,862–34,091B 11,724–15,456C 33,105–43,636D 18,276–24,091

Schurr and co-workers, 199925

A 26,969–35,550B 13,483–17,773C 40,972–54,009D 19,483–25,682

Forster and co-workers, 199621

(haplogroups A, B, C, andD)

20,180 � 1,000

Horai and co-workers, 199315

(haplogroups A, B, C, andD)

14,000–21,000

Brown and co-workers, 199838

(haplogroup X only)12,00–17,000 or 23,000–36,000

Stone and Stoneking, 199818

A 19k (95% CI 12k–30k) or 37k (25k–57k)B 12k (8k–21k) or 25k (16k–41k)C 11k (6k–21k) or 22k (13k–40k)D 15k (9k–27k) or 31k (19k–51k)

12 Evolutionary Anthropology ARTICLES

findings of studies of modern DNAand other genetic polymorphism.Haplogroups A, B, C, and D have beenidentified through analyses of both re-striction fragment length polymor-phism and CR sequencing in manyprehistoric samples in both North andSouth America.18,40,54,58,71,72 Similari-ties in both haplogroup frequenciesand specific haplotypes from ancientDNA also indicate that, for the mostpart, European contact did not signif-icantly affect mtDNA diversity in theAmericas.18,54,72

The presence of haplogroup X hasbeen confirmed in prehistoric andprotohistoric burials on the ColumbiaPlateau,73 while sequence data sug-gest its presence in the prehistoric On-eota population18 and pre-ColumbianSouth America.40 Hauswirth and co-workers74 also reported haplogroup Xfrom Windover pond skeletons (7,000to 8,000 years BP), although other se-quences generated in the study sug-gest the possibility of contaminationin some samples.

While haplogroups B, C, and D haveall been identified in Paleo-Indianskeletal remains,54,55,56 the oldest re-ported member of haplogroup A, themost common haplogroup in NorthAmerica and the New World, datesonly to 4,504 � 105 years BP.75 How-ever, relatively few Paleo-Indian sam-ples have been analyzed and a major-ity of these have come from thewestern United States, where haplo-group A is rare in modern populationsexcept along the coast. In a prelimi-nary restriction analysis of 18 samplesdating to before 6,500 years of age, nomembers of haplogroup A were re-ported.56 The binomial probability ofidentifying no members of haplo-group A among 18 samples, given thepresent distribution of haplogroupswithin the continental United States,is 0.0017.

Finally, in almost all studies of an-cient Native American populations,individuals have been discovered whodo not appear to belong to one of thefive founding lineages. In many cases,this is undoubtedly a result of externalcontamination of samples lackingDNA or in which the DNA is inhibitedfrom amplifying using the polymerasechain reaction. Nonetheless, the pos-sibility remains that additional haplo-groups may be discovered by studiesof ancient DNA in the Americas. Sucha lineage may have either become ex-tinct or be a yet-undiscovered lineagepersisting at low levels in modernpopulations.

POSTCOLONIZATIONPREHISTORY WITHIN

THE AMERICAS

Studies of mtDNA diversity havedemonstrated that the multidisci-plinary approach that uses geneticdata in conjunction with cultural, ar-cheological, and linguistic patternscan provide significant insight intothe population prehistory of theAmericas (Table 2). Ward and co-workers76 examined haplotype diver-sity within three tribes from the Pa-cific Northwest region, the Nuu-Chah-Nulth, the Bella Coola, and the Haida.They concluded that these tribesshare a recent ancestry because thegenetic boundaries among NativeAmerican groups in this region do notcoincide with the boundaries de-scribed by Greenberg, Turner, and Ze-gura11 based on language differences.However, recent re-examination oflanguage stocks in North Americasuggests that the Haida probably donot belong to the Na-Dene languagefamily and therefore did not representa linguistic phylum that is separatefrom the one to which Nuu-Chah-

Nulth and Bella Coola belong.77

Therefore, sample design probably ac-counts for the lack of genetic correla-tion with language in this particularstudy.

On a continent-wide level, Lorenzand Smith6 demonstrated thatmtDNA haplogroup frequency distri-butions often do correlate with lan-guage or geography or both.6 In thenorth, correlations with geographywere high, as were both linguistic andgeographic correlations in westernNorth America. The significant levelsof correlation between language and

mtDNA haplogroup distributionamong native North Americans sug-gests that prehistoric populationmovements, especially in westernNorth America, were not negligibleevents. Lorenz and Smith6 also dem-onstrated that geographic regions ofNorth America exhibit differences inhaplogroup frequency distributions.Although haplogroup frequency dis-tributions vary significantly acrossNorth America, regional studies ofmtDNA diversity in the Northeast42

and the Southwest78 have confirmed apattern of within-region similarities

It now appears that a12,500 year-oldoccupation level atMonte Verde in south-central Chile establishesthe presence of humansin the Americas beforeClovis. This favors amigratory route to theNew World other thanthe ice-free corridor.Recent analysisindicates that a coastalpassage, open as earlyas 14,000 years BP, wasa likely entry point toNorth America.

TABLE 2. Applications of mtDNA Data to Native AmericanMigration Hypotheses

Hypotheses Regarding Migration in North America MtDNA Support

Athapaskan Migration to the Southwest Positive6,78

Neo-Eskimo Migration Positive87

Iroquian Migration Equivocal42

Numic Spread Positive54

Uto-Aztecan Migration and spread of MaizeAgriculture

Negative78

Penutian Migration into California Equivocal6,64

Proto-Algonquian Migration and spread Positive42,88

ARTICLES Evolutionary Anthropology 13

in haplogroup frequency distribu-tions. In the Southwest, these similar-ities cross-cut the boundaries amongdifferent unrelated languages, sug-gesting considerable admixture amongthem. These conclusions are consis-tent with the results of earlier studiesbased on both morphology (for exam-ple, dental variation79) and bloodgroup phenotypes.80,81 The Southeastregion of North America does not dis-play a homogeneous pattern of haplo-group frequency distributions, proba-bly due to genetic bottlenecks causedby the high impact of European con-tact in this region followed by geneticdrift.82

Studies of ancient mtDNA diversityin most regions of North America re-veal that Native American haplogroupfrequency distributions often exhibittemporal as well as regional continu-ity.72,83 In addition to regional studies,analyses of mtDNA have been used indirect tests of specific hypotheses ofpopulation movement proposed bytraditional North American prehisto-rians (archeologists and linguists) asshown in Table 2. Carlyle and col-leagues72 have demonstrated that thehaplogroup frequency distribution ofan ancient population that practicedthe Anasazi cultural tradition in theAmerican Southwest is not signifi-cantly different from that in modernPueblo populations. This study pro-vides biological as well as cultural

evidence of unbroken ancestor anddescendant relationships in the Amer-ican Southwest during the last twomillennia. Malhi73 has shown that

high frequencies of haplogroups Band D have been characteristic of pop-ulations of the Columbia Plateau forat least eight millennia. In contrast,Kaestle and Smith54 have demon-strated that ancient Western GreatBasin populations are statistically sig-nificantly different from modern pop-ulations in the same region, probablydue to a population spread of Numicspeakers into the Great Basin fromsouthern California approximately1,000 years BP.84

Recent regional studies of mtDNAdiversity within North America haveshown that detailed analyses of hap-lotypes can provide better evidence ofancient shared ancestry than do hap-logroup frequency distributions alone,which can be similar in two popula-tions due to chance alone. For exam-ple, Malhi, Schultz, and Smith42 haveprovided evidence from polymorphicsites in the control region of a morerecent shared ancestry among speak-ers of Iroquoian, Caddoan, and Si-ouan languages than between any ofthe three and speakers of Algonquianlanguages of Eastern North Ameri-

Figure 2. Electrophoretic gel showing PCR fragments amplified and digested to revealpolymorphic sites marking 5 known Native American founding haplogroups.

Figure 1. Map of human mitochondrion showing locations of the control region and ofpolymorphic sites marking 5 known Native American founding haplogroups.

14 Evolutionary Anthropology ARTICLES

ca.42,85 Weiss and Smith82 have shownshared mutations in the control re-gion that suggest shared ancestryamong speakers of the Muskogeanlanguages in the Southeast, eventhough haplogroup frequency distri-butions among these groups are sig-nificantly different. Thus, while ge-netic boundaries do not alwayscoincide with boundaries based onthe distribution of languages and cul-ture, the latter provide hypothesesabout prehistory that can be testedusing modern and prehistoric popula-tions. It is important to note that hy-potheses based on genetic evidencemust be consistent with evidence de-rived from historical linguistic and ar-cheological studies.

CONCLUSIONS

It is not surprising that mitochon-drial DNA has largely confirmed thefindings of classical genetic markersregarding genetic relationships amongNative American tribal groups and yethas not conclusively resolved ragingdebates regarding number of migra-tions, source populations, and the

timing of these migrations.80,81 Thisdoes not undermine the utility of ge-netic data, and mtDNA in particular,for future research. While our knowl-edge of the mtDNA diversity amongmany tribal and language groups re-mains limited, the growing mtDNAdatabases both within and outside theAmericas offer a wonderful compara-tive tool. However, it is important toremember that mtDNA is but onemarker, and one that is solely mater-nally inherited, and is unlikely to an-swer all questions regarding the ori-gins of Native Americans.86 WhileY-chromosome markers have beenemployed to address the peopling ofthe Americas, they have not yet beenspecifically used to address postcolo-nization events. Like mtDNA, Y-chro-mosome data have not on their ownconclusively answered questions re-garding either source populationswithin Asia or the number of migra-tions out of Asia into the New World.Clearly, nuclear markers from morepopulations should be examined toprovide additional data relevant tothese controversies, even though it isunlikely that additional data willsignificantly simplify what is a convo-luted and complex scenario of migra-

Figure 4. Alternative routes from Asia into the New World. Dotted lines represents a possiblepath through the ice free corridor as Codilleran and Laurentide ice sheets separated. Solidline of arrows represents a costal route around edge using ice-free refugia along Pacificcoast.

Figure 3. Regional distribution of mtDNA haplogroup frequencies in North America.Adapted from Lorenz and Smith, 1996.6

ARTICLES Evolutionary Anthropology 15

tions and postmigrational evolution-ary forces.

Although it is possible that nucleargenes may someday be more easilyrecovered from ancient human re-mains, as of now population-levelstudies of single-copy genes remainprohibitively difficult with ancientDNA. The scale of the questions mostreadily addressed by ancient mtDNAis different from that of questions ad-dressed by the earliest mtDNA stud-ies. While a decade ago research fo-cused on the Asian affinities andprinciple migrations to the NewWorld, little light was shed upon, norinterested exhibited in, postcoloniza-tion movements and interactions. Thesampling necessary for addressingcontinent-wide phenomena is differ-ent from that necessary for character-izing the relationships between localgroups. Yet the population dynamicsat local levels have contributed to thecurrent genetic diversity witnessed ona continental scale. Identification ofsuch population dynamics can con-tribute significantly to our under-standing of the broader questions per-taining to the settlement of the NewWorld. Further, while critics of an-cient DNA may charge that small sam-ples limit the power of the conclu-sions drawn, ancient DNA remainsthe only direct way of detecting tem-poral change in the genetic composi-tion of a population. MtDNA shouldbe seen as another tool with which toformulate and test hypotheses aboutdemic events such as migrations, ex-pansions, and continuity through

time. Together with historical linguis-tics, ethnographic comparisons, andarcheological investigations, mtDNAretains real power to illuminate pre-history.

REFERENCES

1 Wallace DC, Torroni A. 1992. American Indianprehistory as written in the mitochondrial DNA:a review. Hum Biol 64:403–416.2 Cann RL. 2001. Genetic clues to dispersal inhuman populations: retracing the past from thepresent. Science 291:1742–1748.3 Anderson S, Bankier AT, Barrell BG, de BruijnMHL, Coulson AR, Drouin J, Eperon IC, NierlichB, Roe A, Sanger F, Schrier PH, Smith AJH,Staden R, Young C. 1981. Sequence and organi-zation of the human mitochondrial genome. Na-ture 290:457–465.4 Wallace DC, Garrison K, Knowler WC. 1985.Dramatic founder effects in Amerindian mito-chondrial DNA species. Am J Phys Anthropol68:149–156.5 Schurr TG, Ballinger SW, Gan YY, Hodge JA,Merriwether DA, Lawrence DN, Knowler WC,Weiss KM, Wallace DC. 1990. Amerindian mito-chondrial DNAs have rare Asian mutations athigh frequencies, suggesting they derived fromfour primary maternal lineages. Am J HumGenet 46:613–623.6 Lorenz JG, Smith DG. 1996. Distribution offour founding mtDNA haplogroups among Na-tive North Americans. Am J Phys Anthropol 101:307–323.7 Torroni A, Schurr TG, Cabell MF, Brown MD,Neel JV, Larsen M, Smith DG, Vullo CM, WallaceDC. 1993. Asian affinities and continental radia-tion of the four founding Native AmericanmtDNAs. Am J Hum Genet 53:563–590.8 Nichols J. 1990. Linguistic diversity and the 1stsettlement of the New World. Language 66:475–521.9 Rogers RA, Martin LD, Nicklas TD. 1990. Ice-Age geography and the distribution of nativeNorth American languages. J Biogeogr 17:131–143.10 Nettle D. 1998. Explaining global patterns oflanguage diversity. J Anthropol Archaeol 17:354–374.

11 Greenberg JH, Turner CGI, Zegura SL. 1986.The settlement of the Americas: a comparison oflinguistic, dental and genetic evidence. Curr An-thropol 27:477–497.12 Sapir E. 1929. Central and North AmericanIndian languages. London: Encyclopedia Britan-nica Company, Ltd. Encyclopedia Britannica14th ed, vol 5. p 138–141.13 Bateman R, Goddard I, Ogrady R, Funk VA,Mooi R, Kress WJ, Cannell P. 1990. Speaking offorked tongues: the feasibility of reconciling hu-man phylogeny and the history of language. CurrAnthropol 31:1–24.14 Ward RH, Frazier BL, Dew-Jager K, Paabo S.1991. Extensive mitochondrial diversity within asingle Amerindian tribe. Proc Nat Acad Sci USA88:8720–8724.15 Horai S, Kondo R, Nakagawa-Hattori Y, Ha-yashi S, Sonoda S, Tajima K. 1993. Peopling ofthe Americas founded by four major lineages ofmitochondrial DNA. Mol Biol Evol 10:23–47.16 Merriwether DA, Rothhammer F, Ferrell RE.1995. Distribution of the four founding lineagehaplotypes in native Americans suggests a singlewave of migration for the New World. Am J PhysAnthropol 98:411–430.17 Lorenz JG, Smith DG. 1997. Distribution ofsequence variation in the mtDNA control regionof native North Americans. Hum Biol 69:749–776.18 Stone AC, Stoneking M. 1998. mtDNA analy-sis of a prehistoric Oneota population: implica-tions for the peopling of the New World. Am JHum Genet 62:1153–1170.19 Bonatto SL, Salzano FM. 1997. A single andearly migration for the peopling of the Americassupported by mitochondrial DNA sequence data.Proc Nat Acad Sci USA 94:1866–1871.20 Kolman CJ, Sambuughin N, Bermingham E.1996. Mitochondrial DNA analysis of mongolianpopulations and implications for the origin ofNew World founders. Genetics 142:1321–1334.21 Forster P, Harding R, Torroni A, Bandelt HJ.1996. Origin and evolution of native AmericanmDNA variation: a reappraisal. Am J Hum Genet59:935–945.22 Smith DG, Lorenz J, Rolfs BK, Bettinger RL,Green B, Eshleman J, Schultz B, Malhi R. 2000.Implications of the distribution of AlbuminNaskapi and Albumin Mexico for new world pre-history. Am J Phys Anthropol 111:557–572.23 Starikovskaya YB, Sukernik RI, Schurr TG,Kogelnik AM, Wallace DC. 1998. mtDNA diver-sity in Chukchi and Siberian Eskimos: implica-tions for the genetic history of ancient Beringiaand the peopling of the New World. Am J HumGenet 63:1473–1491.24 Bonatto SL, Salzano FM. 1997. Diversity ofage of the four major mtDNA haplogroups, andtheir implications for the peopling of the NewWorld. Am J Hum Genet 61:1413–1423.25 Schurr TG, Sukernik RI, Starikovskaya YB,Wallace DC. 1999. Mitochondrial DNA variationin Koryaks and Itel’men: population replacementin the Okhotsk Sea-Bering Sea region during theNeolithic. Am J Phys Anthropol 108:1–39.26 Excoffier L, Langaney A. 1989. Origin anddifferentiation of human mitochondrial DNA.Am J Hum Genet 44:73–85.27 Szathmary EJE. 1993. mtDNA and the peo-pling of the Americas (editorial). Am J HumGenet 53:793–799.28 Templeton AR. 1998. Human races: a geneticand evolutionary perspective. Am Anthropol 100:632–650.29 Shields GF, Hecker K, Voevoda MI, Reed JK.1992. Absence of the Asian-specific region V mi-tochondrial marker in native Beringians. Am JHum Genet 50:758–765.30 Torroni A, Sukernik RI, Schurr TG, Starik-

Figure 5. World wide distribution of Haplogroup X frequencies.39,41,89

16 Evolutionary Anthropology ARTICLES

ovskaya YB, Cabell MF, Crawford MH, ComuzzieAG, Wallace DC. 1993. mtDNA variation of ab-original Siberians reveals distinct genetic affini-ties with Native Americans. Am J Hum Genet53:591–608.

31 Yao YG, Watkins WAS, Zhang YP. 2000. Evo-lutionary history of the mtDNA 9-bp deletion inChinese populations and its relevance to the peo-pling of east and southeast Asia. Hum Genet107:504–512.

32 Crawford MH, Williams JT, Duggirala R.1997. Genetic structure of the indigenous popu-lations of Siberia. Am J Phys Anthropol 104:177–192.

33 Torroni A, Miller JA, Moore LG, Zamudio S,Zhuang J, Droma T, Wallace DC. 1994. Mito-chondrial DNA analysis in Tibet: implications forthe origin of the Tibetan population and its ad-aptation to high altitude. Am J Phys Anthropol93:189–199.

34 Ballinger SW, Schurr TG, Torroni A, Gan YY,Hodge JA, Hassan K, Chen KH, Wallace DC.1992. Southeast Asian mitochondrial DNA anal-ysis reveals genetic continuity of ancient Mongol-oid migrations. Genetics 130:139–152.

35 Merriwether DA, Hall WW, Vahlne A, FerrellRE. 1996. MtDNA variation indicates Mongoliamay have been the source for the founding pop-ulation for the New World. Am J Hum Genet59:204–212.

36 Karafet TM, Zegura SL, Posukh O, Osipova L,Bergen A, Long J, Goldman D, Klitz W, HariharaS, de Knijff P, Wiebe V, Griffiths RC, TempletonAR, Hammer MF. 1999. Ancestral Asian sourc-e(s) of New World Y-chromosome founder hap-lotypes. Am J Hum Genet 64:817–831.

37 Bailliet G, Rothhammer F, Carnese FR, BraviCM, Bianchi NO. 1994. Founder mitochondrialhaplotypes in Amerindian populations. Am JHum Genet 55:27–33.

38 Brown MD, Hosseini SH, Torroni A, BandeltH-J, Allen JC, Schurr TG, Scozzari R, Cruciani F,Wallace DC. 1998. mtDNA haplogroup X: an an-cient link between Europe/Western Asia andNorth America? Am J Hum Genet 63:1852–1861.

39 Smith DG, Malhi RS, Eshleman J, Lorenz JG,Kaestle FA. 1999. Distribution of mtDNA haplo-group X among Native North Americans. Am JPhys Anthropol 110:271–284.

40 Ribeiro-Dos-Santos AKC, Santos SEB,Machado AL, Guapindaia V, Zago MA. 1996.Heterogeneity of mitochondrial DNA haplotypesin Pre-Columbian natives of the Amazon region.Am J Phys Anthropol 101:29–37.

41 Torroni A, Huoponen K, Francalacci P,Petrozzi M, Morelli L, Scozzari R, Obinu D,Savontaus ML, Wallace DC. 1996. Classificationof European mtDNAs from an analysis of threeEuropean populations. Genetics 144:1835–1850.

42 Malhi RS, Schultz BA, Smith DG. 2001. Dis-tribution of mitochondrial DNA lineages amongnative American tribes of northeastern NorthAmerica. Hum Biol 73:17–55.

43 Malhi RS, Smith DG. 2002. Brief communi-cation: haplogroup X confirmed in prehistoricNorth America. Am J Phys Anthropol 119:84–86.

44 Derenko MV, Grzybowski T, Malyarchuk BA,Czarny J, Miscicka-Sliwka D, Zakharov IA. 2001.The presence of Mitochondrial haplogroup X inAltaians from South Siberia. Am J Hum Genet69:237–241.

45 Jantz RL, Owsley DW. 2001. Variation amongearly North American crania. Am J Phys An-thropol 114:146–155.

46 Steele DG, Powell JF. 1997. Biological affini-ties of the oldest recovered North Americans: theSpirit Cave and Pyramid Lakes data. Am J PhysAnthropol Supplement 24:218.

47 Preston D. 1997. The lost man. The NewYorker, June 16, 1997 p. 70.

48 Powell JF, Rose JC. 1999. Report on the non-destructive examination, description, and analy-sis of the human remains from Columbia Park,Kennewick, Washington: report on the osteolog-ical assessment of the “Kennewick Man” skeleton(CENWW.97.Kennewick). Chapter 2: NationalPark Service.49 Powell JF, Neves WA. 1999. Craniofacial mor-phology of the first Americans: pattern and pro-cess in the peopling of the New World. YearbookPhys Anthropol 110, Supp 29:153–188.50 Neves WA, Munford D, do Carmo Zanini M,Pucciarelli HM. 1999. Cranial morphologicalvariation in South America and the colonizationof the New World: towards a four migrationmodel? Ciencia e Cultura (Sao Paulo) 51:151–165.51 Smith DG, Malhi R, Eshleman JA, Kaestle FA.2000. Report on DNA analysis of the remains of“Kennewick Man” from Columbia Park, Wash-ington: National Park Service.52 Kaestle FA. 2000. Report on DNA analysis ofthe remains of “Kennewick Man” from ColumbiaPark, Washington: Nation Park Service.53 Merriwether DA, Cabana GS. 2000. Ken-newick Man ancient DNA analysis: final reportsubmitted to the Department of the Interior: Na-tional Park Service.54 Kaestle FA, Smith DG. 2001. Ancient mito-chondrial DNA evidence for prehistoric popula-tion movement: the Numbic expansion. Am JPhys Anthropol 115:1–12.55 Stone AC, Stoneking M. 1996. Genetic analy-ses of an 8000 year-old Native American skele-ton. Ancient Biomolecules 1:83–87.56 Smith DG, Malhi RS, Eshleman JA, KaestleFA. 2002. Mitochondrial DNA haplogroups of Pa-leoamericans in North America. In: BonnichsenR, Gruhn R, Steele DG, Stanford D, Lepper B,Warren CN editors. Beyond Clovis: where wehave been, where we are, and where we are go-ing. College Station, TX: Texas A & M UniversityPress.57 Morell V. 1998. Human genetics: genes maylink ancient Eurasians, Native Americans. Sci-ence 280:520–520.58 Parr RL, Carlyle SW, O’Rourke DH. 1996. An-cient DNA analysis of Fremont Amerindians ofthe Great Salt Lake Wetlands. Am J Phys An-thropol 99:507–518.59 Meltzer DJ. 1993. Pleistocene peopling of theAmericas. Evol Anthropol 1:157–169.60 Dixon EJ. 2001. Human colonization of theAmericas: timing, technology and process. Qua-ternary Sci Rev 20:277–299.61 Mandryk CAS, Josenhans H, Fedje DW,Mathewes RW. 2001. Late Quaternary paleoen-vironments of Northwestern North America: im-plications for inland versus coastal migrationroutes. Quaternary Sci Rev 20:301–314.62 Dillehay TD. 1997. Monte Verde: a Late Pleis-tocene settlement in Chile, vol. 2: The archaeo-logical context. Washington, D.C.: SmithsonianInstitution Press.63 Dillehay TD. 1999. The Late Pleistocene cul-tures of South America. Evol Anthropol 7:206–216.64 Eshleman JA. 2002. Mitochondrial DNA andprehistoric population movements in WesternNorth America. University of California, Davis.Doctoral dissertation, Dept. of Anthropology.65 Leonard JA, Wayne RK, Cooper A. 2000. Pop-ulation genetics of ice age brown bears. Proc NatAcad Sci USA 97:1651–1654.66 Torroni A, Neel JV, Barrantes R, Schurr TG,Wallace DC. 1994. Mitochondrial DNA “clock”for the Amerinds and its implications for timingtheir entry into North America. Proc Nat AcadSci USA 91:1158–1162.67 Stoneking M, Bhatia K, Wilson AC. 1986.

Rate of sequence divergence estimated from re-striction maps of mitochondrial DNAs fromPapua New Guinea Cold Spring Harbor SympQuant Biol 51(Part I):433–439.

68 Tavare S, Balding DJ, Griffiths RC, DonnellyP. 1997. Inferring coalescence times from DNAsequence data. Genetics 145:505–518.

69 Torroni A, Bandelt HJ, D’Urbano L, LahermoP, Moral P, Sellitto D, Rengo C, Forster P, Savon-taus ML, Bonne-Tamir B, Scozzari R. 1998.mtDNA analysis reveals a major late Paleolithicpopulation expansion from southwestern tonortheastern Europe. Am J Hum Genet 62:1137–1152.

70 Izagirre N, de la Rua C. 1999. An mtDNAanalysis in ancient Basque populations: implica-tions for haplogroup V as a marker for a majorPaleolithic expansion from Southwestern Eu-rope. Am J Hum Genet 65:199–207.

71 Lalueza C, Perez-Perez A, Prats E, CornudellaL, Turbon D. 1997. Lack of founding Amerindianmitochondrial DNA lineages in extinct aborigi-nes from Tierra del Fuego-Patagonia. Hum MolGenet 6:41–46.

72 Carlyle SW, Parr RL, Hayes MG, O’RourkeDH. 2000. Context of maternal lineages in theGreater Southwest. Am J Phys Anthropol 113:85–101.

73 Malhi RS. 2001. Investigating prehistoricpopulation movements in North America withancient and modern mtDNA. University of Cali-fornia, Davis. Doctoral dissertation, Dept. of An-thropology.

74 Hauswirth WW, Dickel CD, Rowold DJ,Hauswirth MA. 1994. Inter- and intrapopulationstudies of ancient humans. Experientia (Basel)50:585–591.

75 Kaestle FA. 1998. Molecular evidence for pre-historic Native American population movement:the numic expansion. University of California,Davis. Doctoral dissertation, Dept. of Anthropol-ogy.

76 Ward RH, Redd A, Valencia D, Frazier B,Paabo S. 1993. Genetic and linguistic differenti-ation in the Americas. Proc Nat Acad Sci USA90:10663–10667.

77 Campbell I. 1997. American Indian languag-es: the historical linguistics of Native America.New York: Oxford University Press.

78 Malhi RS, Mortensen HM, Eshleman JA,Lorenz JG, Kaestle FA, Johnson JR, GorodezkyC, Smith DG. 2002. Native American mtDNA inthe American Southwest. Am J Phys Anthropol,in press.

79 Turner CGI. 1993. Southwest Indian teeth:Southwest Indians: Prehistory through dentition.Res Exploration 9:32–53.

80 Brown KS, Hanna BL, Dahlberg AA, Strand-skov HH. 1958. The distribution of blood groupalleles among Indians of Southwest North Amer-ica. Am J Hum Genet 10:175–195.

81 Spuhler JN. 1979. Genetic distances, treesand maps of North American Indians. In: Laugh-lin WAS, Harper AB, editors. The first Americans:origins, affinities, and adaptations. New York:Gustav Fischer. p 135–184.

82 Weiss DA, Smith DG. n.d. Evidence for a re-cent mtDNA bottleneck among Native Americansfrom the Southeastern United States. Submitted.Am J Phys Anthropol.

83 O’Rourke DH, Hayes MG, Carlyle SW. 2000.Spatial and temporal stability of mtDNA haplo-group frequencies in native North America. HumBiol 72:15–34.

84 Bettinger RL, Baumhoff MA. 1982. The Nu-mic spread: Great Basin cultures in competition.Am Antiquity 47:485–503.

85 Chafe WL. 1976. The Caddoan, Iroquoian andSiouan languages. The Hague: Mouton.

ARTICLES Evolutionary Anthropology 17

86 Szathmary EJE. 1993. Genetics of aboriginalNorth Americans. Evol Anthropol 1:202–220.

87 Hayes MG, O’Rourke DH. 2001. Genetic sig-natures of migrations and population replace-ments during human colonization of the NorthAmerican Arctic. Am J Hum Genet 69:179.

88 Schultz BA, Malhi RS, Smith DG. 2001. Examin-

ing the Proto-Algonquian migration: analysis ofmtDNA. Papers of the 32nd Algonquian Conference.Manitoba: University of Manitoba Press. p 470–492.

89 Reidla M, Kivisild T, Metspalu E, Kaldma K,Tambets K, Tolk H-V, Parik J, Bermisheva M,Zhadanov S, Pennarun E, Gubina M, Gol-ubenko M, Damba L, Fedorova S, Gusar V, Ge-

berhiwot T, Moisan JP, Khunsnutdinova E, Os-sipova L, Stepanov V, Torroni A, Rickards O,DeStefano GF, Papiha S, Beckman L, Rudan P,Angnou NP, Koziel S, Usanga E, Villems R. 2002.Early postglacial spread of haplogroup X fromEurasia to America. In Preparation.

© 2003 Wiley-Liss, Inc.

Forthcoming Special IssueCultural Evolution in Primates

Edited by Charles Janson and Eric Alden Smith

Contributions Include:• An Overview of Primate Traditions

Dorothy Fragaszy

• Traditions in MonkeysSusan Perry and Joseph H. Manson

• Cultural PanthropologyAndrew Whiten

• Is Culture a Golden Barrier Between Human and Chimpanzee?Christophe Boesch

• The Adaptive Nature of CultureMichael Alvard

• Cultural Evolution and the New Old ArchaeologyMarta Mirazon Lahr and Robert Foley

18 Evolutionary Anthropology ARTICLES