Dating the Arrival of Man to America
2007
note: some
figures are missing
Beginning with 19th century archaeological finds
in the United States, the peopling of the Americas has been an issue of
interest for anthropologists and researchers. Putnam and Abbot's 1876 discovery
of lithic tools in the gravels of the Delaware River near Trenton, New Jersey;
or the 1844 excavation of “Dr. Dowler's Red Indian” from 16 feet below the
streets of New Orleans, Louisiana, seemed to indicate a considerable time depth
to the presence of Indians in the New World (Dillehay and Meltzer, 1991:1;
Neuman, 1984:17).
Thinkers as long ago as President Jefferson had suggested
that Native Americans may have entered North America via the area of the Bering
Strait (Milner, 2004:15). Beginning in the 1830s, the Frenchman Jacques Boucher
de Perthes undertook excavations in the Somme River Valley, the results of
which first demonstrated the indisputable association of the remains of extinct
large mammals with prehistoric tools made by man (Grayson, 1984:25). It was not
until 1932-3, with the discovery in the state of New Mexico of the remains of
Pleistocene megafauna in context with lithic tools , that physical evidence emerged consistent with an
ancient occupation of the Americas by members of the human race (Haynes,
2002:57).
With the advent of radiocarbon dating circa 1950, such
Paleoindian finds became datable in a non-relative manner, and it subsequently
became de rigueur in archaeology to place the entry of humans into the
Americas at about 12,000 to 13,000 years before present, just prior to the
beginning of the Holocene epoch. This was the beginning of the so-called Clovis
Era, named after the town of Clovis, New Mexico, proximate to the
archaeological site that yielded Paleoindian lithics and Pleistocene faunal
remains found in incontrovertible association.
Although radiocarbon dated Clovis sites proved that humans
were occupying the Americas by the end of the Pleistocene, it was impossible to
suggest when they had first arrived. There were surely no ancient
archaeological sites that yielded evidence of the evolution of Homo in
the Americas, but when exactly did anatomically modern humans get here?
Some of the past efforts aimed at dating human entry into the
New World included the use of glottochronology. Such models sought to use
lexicographies from various languages, comparing them with putative parental
forms from which subsequent daughter variants developed or evolved. This method
was well suited to dating the entry of humans into the Americas: groups who
migrated here from Asia would be removed from the parental language stock, so
that differentiation of their words would occur in an isolated manner,
analogous to genetic drift in physical anthropology. Linguists modeled the rate
of change of the lexicography as the loss of about 19 words per 1000 years of
separation from the parent language (Hoijer, 1956). In other words, after 1000 years, 81% of
tested lexemes should be cognate with the source tongue.
Since 1960 and
earlier, Joseph Greenberg had been attempting to use linguistics to date the
peopling of the Americas. This effort reached its acme with the 1986
publication of a paper which attempted to combine the efforts of linguistic,
dental, and genetic research (Greenberg, Turner, and Zegura, 1986). For better
or worse, this paper marked the beginning of the use of DNA analysis in efforts
to establish a chronology for the entry of humans into the Americas. Though
earlier papers had used genetic markers to show links among various native
American groups “no overall assessment of genetic relationships among North America,
South America, and Asia was undertaken [whereby] the conventional perspective
on the origin and evolution of aboriginal American diversity was challenged in
1986, when Greenberg, Turner, and Zegura published their three-migration
hypothesis” (Szathmary,1993:793-4).
Though this paper was
immediately (Campbell 1986) and is still criticized , particularly for its
reductionist categorization of linguistic groups, it is frequently cited in
works on DNA analysis in regards to the peopling of the Americas (e.g. Monsalve
et al 1999:2209). Even with such criticism, the model describing three waves of
migration into the Americas is still used in its basic form: a most ecent
influx of Aleut-Eskimo peoples, an earlier migration of the NaDene (the
Athapaskan speakers), and finally the most ancient entry of the group broadly
termed the Amerinds (by Greenberg,
Turner, and Zegura, 1986).
Haplotypes, Haplogroups, and Mitochondrial
DNA
As demonstrated by geneticists, the
chromosomes that encode for the replication of the cells in organisms are made
up strands of DNA. Along these strands, the diverse arrangements of the DNA
molecules form unique patterns that are specific to the design of the many
cells needed to sustain functioning life. Groups of these alleles occur
in particular arrangements at specific loci along the chromosomes. These
individual patterns can be used as markers to distinguish one organism from
another, analogous to the bar codes in use as price codes in markets today.
When analyzing
mitochondrial (or “mt”) DNA, researchers use particular enzymes called
restriction endonucleases to cleave its molecular chain. These enzymes sever
the mtDNA at specific “recognition sites”, isolating for study what are known
as “restriction fragment length polymorphisms”, or RFLPs (Schurr 2002:62-64).
RFLP characteristics
for a given individual under study are what make up the so-called haplotype
for the individual. Like grocery store bar codes, these haplotypes are unique
and can be compared to those from other groups of individuals. Assemblages of
given haplotypes as defined by RFLP analyses are called collectively haplogroups,
or mtDNA lineages (Schurr, 2000:247).
Polymorphisms that
create unique haplotypes are found in both nuclear and mitochondrial DNA. There are however advantages to using
the mtDNA haplotypes as indices for tracking changes through time between given
generations. Mitochondrial DNA mutates up to ten times faster than nuclear DNA,
making it better suited to note changes on smaller time scales, such as those
applicable to understanding the peopling of the Americas (Crawford, 1998:135).
Also, mtDNA is easier to obtain from archaeological contexts and replicates
more easily in the laboratory (Eshleman et al, 2003:7). Mitochondrial DNA, as
the name implies, is from the mitochondria within cells of the body. It is in
the form of a circular molecule made up of 16, 569 nucleotide pairs. Due to the
mechanics of reproduction, mtDNA is passed on from mother to daughter with very
little genetic contribution from the father. Because the mtDNA does not undergo
genetic reshuffling, or meiosis, during reproduction, and because there is
minimal recombination, RFLPs are not
hypervariable. (Bailliet et al 1994:27). Any mutations, deletions, or
insertions of the mtDNA loop (Merriwether et al, 1995:412) are thought to be
predictable at given modeled rates through time. This rate is estimated to be
33% change per one million years, or 1% per 30,300 years (Crawford 1998:138;
Shields et al, 1993:552) (Stone and Stoneking (1998:1167) use a figure of 10.3%
per million years). Hence, if a mtDNA sample from an individual can be compared
to a putative parental component, not only can a proof of ancestry be obtained,
but a time interval of separation between the two can be estimated.
Native American Haplogroups
During the last 20 to
25 years, researchers have investigated the mtDNA from many Native American
tribal groups in the Americas. Efforts are made to obtain DNA samples from
individuals who are full bloods within their given ethnic group. Subsequent to
the European colonization of the New World, gene flow has occurred with the
effect of mixing Native American with both European and African genetic
material. A strong presence of the European H,
J, K or African L haplogroups in Native American test subjects
indicates genetic admixture, rendering such individuals non-ideal for studies
concerned with developing a chronology of the prehistoric peopling of the
Americas (Keyeux et al, 2002; Schurr, 2002; Torroni et al, 1993:581-582).
Gibbons (1993:312)
credits Douglas Wallace's 1985 work as pioneering the mitochondrial studies
that have focused on Native American prehistory (Wallace et al, 1985), while
Merriwether et al (1995:411) cite Wallace and Torroni (1992) as having
originated the now ubiquitous nomenclature for the founder haplogroups A, B, C,
and D categories used to describe Native American genetic phylogenies. These
and other molecular anthropologists have learned that virtually all Native
American mtDNA is in one or more of these haplogroups. Subsequent study has
shown that a fractional portion of genetic samples that have been tested, those
which were often labeled “other”, are actually a rare minority haplogroup
called type X (Merriwether et al 1995).
Once researchers have
collected a set of diverse and representative mtDNA samples from across the
globe, then there are problems as to how to interpret such data. First, there
is the matter of possible disruption and subsequent drift in the Native
American gene pool subsequent to colonial epidemics and conquest (Stone and
Stoneking 1998:1164). Secondly, there is the matter of mobility: it is
impossible to compare the migration of one population to another population
that may itself also be migrating. And thirdly, what may be the most
problematic, concerns the mutation rate models used in interpreting the mtDNA
data. The confidence intervals associated with these estimates are quite large
(Figure 1). The modeling ratios themselves are generated from linguistic and archaeological
data (Torroni et al, 1994:1161), and hence their accuracy can never supersede
those sources, and is probably biased toward correlating with them.
Nonetheless, the
geographic distribution patterns that emerge with worldwide mtDNA data are quite
compelling. The Native American mtDNA gene pool for North, Central, and South
America is made up of mixtures of the A, B, C, and D haplogroups, with less
than 5 to near zero percent across New World populations possessing mtDNA of
the fifth X lineage (Schurr, 2002; Torroni et al, 1994). This is also true of
the pre-Columbian genetic data from the 700 year old Oneota Norris Farm site in
Illinois (Stone and Stoneking 1998:1153), suggesting that colonial disruptions
did not erase the mtDNA signature from the Native American population.
The haplogroups C and
D are found to be distributed rather evenly throughout the New World (cf Schurr
2002; Merriwether et al 1995:421)(Figure 2). By comparison, the A and B
haplogroups are clinally distributed (Meriwether 1995:418 ). Haplogroup A is
concentrated toward northwestern Canada, while it is rare or absent in South
American populations. In North America, the Na-dene populations are
predominantly in the A haplogroup, interpreted as evidence of a separate
migration vent for that linguistic group (Torroni et al, 1993a:578).
Curiously, across the
New World, the distribution for
haplogroup B is opposite that of A.
Haplogroup B is concentrated especially among Andean coastal regions in
South America and is not common in North American populations (Figure 3). The
exception is the southwestern Pueblo Indians, who have significant percentages
of the B haplogroup mtDNA (Meriwether 1995:418; Schurr 2002:65).
The widespread
distribution across the New World of the C and D haplogroups is argued by some
as evidence for a single migration of humans into the Americas (Merriwether et
al, 1995, 1996; Schurr 2002:68). It is reasoned that the observed ubiquity of
these haplogroups could not be achieved by various smaller migrations separated
through time. The founder populations of any putative single migration would
have already possessed significant genetic diversity from their parental
populations in Siberia.
Three
of the four haplogroups that characterize New World populations, A, C, and D are found in northeast Asia and
Siberia, though they are by no means the majority groups present there. Of the
three, haplogroup C is the most common in Siberia, with eight of ten
individuals sampled falling in this category. Haplogroup D is also rather
common and widespread. Haplogroup A is limited to extreme northeast Asia and is
concentrated toward northeast Siberia, where it was found in 80% of the Asiatic
Eskimo people sampled. All Asiatic Eskimos sampled are in haplogroups A and D
exclusively (Torroni et al, 1993a, 1993b).
Haplogroup B is
conspicuously absent in all Siberian and northeast Asian groups tested. It is
strongly present in southeastern Asian populations including Samoans, Maoris,
and Taiwanese, and generally throughout the Pacific Islands. In the Americas, haplogroup B is concentrated
along the Pacific coast of the Americas, and it decreases in frequency clinally
from south to north in its distribution (Figures 2 and 3).
Prehistoric New World Founder Migrations
Though there is no
consensus amongst scientists today concerning the number of migrations that
peopled the New World. The basic foundational model is still that of Greenberg
et al (1986), the so-called three wave hypothesis. It holds that there was
a most recent influx of Aleut-Eskimo
peoples, which was preceded by an earlier migration of the NaDene (the
Athapaskan speakers), which itself was preceded by the migration of the group
broadly termed the Amerinds. This last group is that associated with the
initial peopling of the New World, viz: the Clovis or pre-Clovis occupation.
The predominance of
the A haplogroup among Na-Dene subjects supports a separate migration event for
this group, keeping in line with the three wave migration model. The A
haplogroup is not common in South America, perhaps also suggesting that the
migration that introduced this haplogroup was a separate and ostensibly later
event (Torroni et al, 1993a:563).
The ubiquity of the C
and D groups throughout the Americas earmarks them as representative of the
earliest “Amerind” migration. The relatively greater dispersal of these
haplogroups in the Americas suggest their greater time depth (Meriwether
1995:427).
The B haplogroup
presents another problem, outside of the original purview of the three wave
model. There are simply no populations in northern North American or in Siberia
that possess the B haplogroup mtDNA. The putative route used to colonize
prehistoric America , the greater Beringia area, does not amongst its modern
populations contain any trace of this founder haplogroup. A few researchers suggest that this migration
may have followed a coastal route paralleling the Bering land bridge, but this
does not explain why there are no populations of B haplogroup members along the
west coast of North America. As Cann (1994:10) puts it “A coastal route in
equilibrium along the entire Pacific Rim does not yet account for the
geographic gradient that is seen in B lineage frequencies, which are highest always
in the south.” Cann broaches the topic
of transoceanic travel having influenced the peopling of the Americas. “Pacific voyagers could have contributed this
lineage [B] separately to the Americas, without ever going through Beringia.”
The New World distribution of haplogroup B excludes the regions of northern
North America, and it is absent in Siberia. Furthermore, the B haplogroup is
found on the west coast of South America and it is strongly present in
Polynesia. Cann suggests that individuals from the Lapita cultural complex from
circa 6000 BP could account for such “Remote Oceanic lineages” (Cann, 1994:10,
Torroni 1993a:585).
Dating Human Migrations with mtDNA
Mitochondrial DNA can
be used to estimate the timing of migrations of humans into the Americas.
Changes between New World and Siberian mtDNA samples can be compared, and the
quantity of change in the mtDNA can be noted. A rate of change through time
must be assumed to convert such data into years BP. In actual practice, a range
of conversion ratios are examined. For instance, Szathmary (1993:797) refers
to “using the more commonly employed
mutation rate of 2%-4% / million years”.
As noted earlier, other researchers mention mutation rates of 33%
per one million years (Crawford 1998:138; Shields et al, 1993:552),
while Stone and Stoneking (1998:1167) use a figure of 10.3% per million years.
The confidence
intervals on these data span a considerable duration (Figures 1 and 4), and
assumptions or overly confident assertions based on this range of uncertainty
have been criticized (Gibbons 1993). Using the combined Native American mtDNA
data for haplogroups A, C, and D, Torroni et al (1994:1162) obtain an averaged entry date with a range from
25,707 to 33,939 YBP. They calculate the
entry date separately for haplogroup B as it genetically appears to be a later,
more recent migration (Figure 1). There is some ambiguity as to the actual
geographic location of founder groups during the genetic bottlenecks or
fissionings accounted for by these data. It is possible that some bands
separated from parent populations while still in Siberia, so that haplotype
derived dates may refer to pre-migration events that occurred in Siberia,
rather than to the Beringian migration itself. Some data run counter to this
notion and suggest that there is sufficient variation between the Siberian and
American gene pools to confirm their independent development over a prolonged
period (Torroni et al, 1994:604).
Also perhaps
indicative of time depth in the Americans is what Torroni et al (1993a:584)
refer to as the “tribalization process”.
The researchers found that individual tribes have mutations in their
mtDNA that are unique to only their group.
These are called “private
polymorphisms”, and according to the authors
they indicate that “there was an early and rapid population radiation
followed by tribal isolation and localized differentiation [implying] that
tribal linguistic boundaries have been quite effective barriers to gene flow”.
Summary
Genetics is still a
relatively new technique for unraveling the history of the human species. Its
use in demonstrating a chronology for the peopling of the Americas will likely
prove invaluable in the long run. The data are still not robust enough however to
convince all of the “Clovis first” camp of any migration earlier than that
suggested by the lithics record. This is with good reason. Contemporary
articles will print widely divergent figures based on essentially the same
mtDNA data. For instance, Figure 4 shows estimated migration / entry dates for
the four main founder haplogroups from Stone and Stoneking (1998). They
differ appreciably from those in Figure 1 from Torroni et al (1994). Haplogroup
B is the most recent set of dates in Figure 1, whereas haplogroup C migration
is the most recent in Figure 4. Such divergent results cannot inspire great
confidence.
However, in general
the data do seem to confirm the long held idea that Native American ancestors
emigrated to the New World from northeast Asia. Also, there is a notable lack
of European or African mtDNA in the American samples. The unusual geographic
distribution of the haplogroup B suggests that the Beringia route was probably
not the only port of entry for the ancient colonists to the New World. The data
for haplogroup B seem to suggest some sort of transoceanic contact, but that
this occurred more recently than the primary and oldest Amerind migrations.
The model posited by
Greenberg et al (1986) which seeks to tie linguistic groups to mtDNA haplogroups
seems to be at least partially confirmed. This is in regard to the strong
association of the haplogroup A with the Na-Dene peoples, speakers of the
Athapaskan dialects. The strong clinal positive correlation of the A haplogroup
with the northern latitudes of North America also suggests a discrete and more
recent entry for the NaDene from northeast Asia.
Figure 1. Estimated dates of haplogroup separation from Siberian
source. Note the wide confidence intervals. Based on a mutation rate of 0.029-0.022%
per 10,000 years, from Torroni et al (1994:1161, fig. 4).
Figure 3. Presence of haplogroup B indicated by black coloring in bars.
From Torroni et al
(1993b:604, fig 5).
Figure 4. Estimated dates for the
four founder haplogroups entering the New World.
Two different rates of
mutation are employed.
From Stone and
Stoneking (1998:1166, table 6).
LITERATURE CITED
Campbell L.1986. Comments. Cur
Anthropol 27(5):488.
Cann RL. 1994. mtDNA and Native
Americans: a southern perspective. Am J Hum Genet 55:7-11.
Crawford MH. 1998. The origins
of Native Americans: evidence from anthropological genetics. Cambridge:
University Press. 308p.
Dillehay TD, Meltzer DJ.
1991.The first Americans: search and research.
Boca Raton, Florida: CRC Press. 309 p.
Eshleman JA, Malhi RS, Smith DG. 2003. Mitochondrial DNA
Studies of Native Americans:
conceptions and misconceptions of the
population prehistory of the Americas. Evol Anthropol 12:7–18.
Gibbons A. 1993. Geneticists
Trace the DNA Trail of the First Americans. Science 259(5093):312-313.
Grayson DK. 1984. Nineteenth
century explanations of Pleistocene extinctions: a review and analysis. In
Quaternary extinctions, edited by Martin PS and Klein RG, p.5-39. Tucson:
University of Arizona Press. 892 p.
Greenberg JH, Turner CG, Zegura
SL. 1986. The settlement of the Americas: a comparison of the linguistic,
dental, and genetic evidence. Cur Anthropol 27(5):477- 488.
Haynes G. 2002. The early
settlement of North America: the Clovis era. Cambridge: University Press. 345
p.
Hoijer, H. 1956. The chronology
of the Athapaskan languages. Int J Am Ling 22(4):219-232.
Keyeux G, Rodasv C, Gelvez N,
Carter D. 2002. Possible Migration
Routes into South America Deduced from Mitochondrial DNA Studies in
Colombian Amerindian Populations. Hum Biol 74(2):211-233.
Merriwether DA,Ferrell RE.1996.
The four founding lineage hypothesis for the new world: a
critical reevaluation. Mole Phylogen and Evol 5(1):241-246.
Merriwether DA, Rothhammer F,.
1995. Distribution of the four founding lineage haplotypes in Native America
suggests a single wave of migration for the New World. Am J Phys Anthropol
98:411-430.
Milner GR. 2004. The
moundbuilders: ancient peoples of eastern North America. London: Thames and
Hudson. 224p.
Monsalve MV, Helgason A, Devine
DV. 1999. Languages, geography, and HLA haplotypes in Native American and Asian
populations. Proc R Soc Lond 266:2209-2216.
Neuman RW. 1984. An
introduction to Louisiana archaeology. Baton Rouge: LSU Press. 310p.
Shields GF, Schmiechen AM, Frazier BL, Redd A, Voevoda
M, Reed JK, Wardt RH. 1993. mtDNA
Sequences Suggest a Recent Evolutionary Divergence for Beringian and Northern
North American Populations. Am. J Hum Genet 53:549-562.
Schurr TG. 2002. The molecular
anthropological perspective on the peopling of the Americas. Athena Rev
3(2):62-75.
Schurr TG. 2000. Mitochondrial
DNA and the peopling of the New World. Am Scien 88(3):246-253.
Szathmary EJE. 1993. mtDNA and
the Peopling of the Americas. Am. J Hum Genet 53:793-799.
Torroni A, Neel JV, Barrantes
R, Schurr TG, Wallace DC. 1994. Mitochondrial DNA "clock" for the
Amerinds and its implications for timing their entry into North America. Proc.
Natl. Acad. Sci. USA
Vol. 91, pp. 1158-1162.
Torroni A, Schurr TG, Cabell
MF, Brown MD, Neel JV, Larsen M, Smith
DG, Vullo CM, Wallace DC. 1993.Asian
Affinities and Continental Radiation of the Four Founding Native American
mtDNAs. Am. J. Hum. Genet. 53:563-590.
Torroni A, Sukernik R,
Schurr TG, Starikovskaya YB, Cabell MF,
Crawford MH, Comuzzie AG, Wallace DC. 1993.mtDNA Variation of Aboriginal
Siberians Reveals Distinct Genetic Affinities with Native Americans. Am J Hum
Genet 53:591-608.
Wallace DC, Garrison K, Knowler
WC.1985. Dramatic founder effects in Amerindian mitochondrial DNAs. Amer J Phys
Anthropol 68:149-155.
Wallace DC, Torroni A. 1992.
American Indian prehistory as written in the mitochondrial DNA: a review. Hum
Biol 64:403-416.
The author grants permission to
reproduce text, tables, maps, or images included herein, provided that the
author is cited as Turner, Christopher S.,
year of article, name of article, conference event and date if
applicable to paper, page, and source, and provided that use of any text,
tables, maps, or images included herein is for non-commercial, academic
purposes.
No comments:
Post a Comment