By John A. Tvedtnes
Part 3 of 3 – Comparing Living and Prehistoric Populations
Editor’s Note: This is the last of a three-part series that interprets the recent DNA data concerning peoples who settled the Western Hemisphere. Part 1 gives a background on the controversy, as well as thoughts on how the DNA data could confirm Book of Mormon accounts rather than disprove them. Part 2 covers differing interpretations of the DNA data that have been uncovered.
A number of scholars have objected to attempts to determine from the genetic makeup of modern populations who comprised the ancient inhabitants of a given region. There are many reasons for this, some of which we have discussed earlier (e.g., bottlenecks and the likelihood of some DNA markers disappearing over time). Theodor Schurr noted that
It’s also important to mention that the genetic composition of an ancient population may not be the same as the population currently occupying the same geographic region, because of migrations, genetic drift or other stochastic processes. For example, based on haplogroup frequencies, the ancient Stillwater Marsh population in the Great Basin region does not appear to be ancestral to any modern Amerindian population in the same area. On the other hand, ancient Eskimo and Aleut samples have nearly the same haplogroup frequencies as their modern descendents, and the same is true for the ancient Anasazi and Fremont cultures and the modern Pueblo Indian groups. 
The oldest complete human skeleton in Great Britain was found in Gough’s Cave in Cheddar Gorge, southwestern England. Dubbed “Cheddar Man,” the skeleton was dated to about 7150 B.C. In 1997, a research team headed by Bryan Sykes of the Oxford University Institute of Molecular Medicine, decided to perform mtDNA tests on residents of the nearby town of Cheddar to see if any of the modern inhabitants were related to Cheddar Man. Samples were collected from 15 students at the Kings of Wessex school and from four adults from old Cheddar families. To make up a sampling of 20 people, Adrian Targett, a teacher of modern history at the school, volunteered to give a sample. When the results were announced in March 1997, Targett was the only one whose mtDNA suggested that he and Cheddar Man had a common female ancestor.  He lives about half a mile from the site of the 1903 discovery.
According to Sykes, Targett’s mtDNA differs from that of Cheddar Man in only one base pair out of 300, which can be accounted for by mutation.  But this is far less mutation than has been noted in other mtDNA studies for a comparable period of time. Moreover, it is rather surprising that Cheddar Man has only one relative still living in the region. To be sure, the sampling (20 people) was relatively small, but it is quite comparable to many of the samplings done among Native Americans.
An American skeleton from approximately the same time as Cheddar Man was found in 1996 on the shore of the Columbia River near Kennewick, Washington. Anthropologists who examined and measured the skull noted that it showed strong Caucasoid traits, suggesting a European rather than an Asian origin. The Native American Protection and Repatriation Act of 1990 requires that skeletal remains found in the United States be returned to Native American tribes for reburial. Five tribes (Umatilla, Colville, Wanapum, Nez Perce, and Yakima) claimed the remains as their ancestor. Unfortunately, scientists were unable to recover DNA that could determine Kennewick Man’s true origins.  Controversy over the remains continues in both the court and the laboratory.
Another American skeleton from roughly the same period (7400 B.C.) was found in Spirit Cave, near Fallon, Nevada. Douglas Owsley, a forensic anthropologist at the National Museum of Natural History, noted that Spirit Cave Man is “most similar to the Ainu [a Caucasian people] from Japan and a medieval Norse population … I’m reluctant to say he’s a white guy, but he’s certainly very different from modern Asians and Native Americans.” 
Over the past few years, geoarchaeologist Silvia Gonzalez of the Liverpool John Moores University in England has examined 27 skeletons in the National Museum of Anthropology in Mexico City that were found in the region in 1959. She found that the skulls were dolichocephalic (long and thin) and did not match the brachycephalic (round and broad) skulls of Native Americans. 
Radiocarbon dating of four of the specimens at Oxford University indicated that they are older than other human remains found in the New World and that one of them, called “Peon Woman III,” dates to 12,700 B.P. Gonzalez noted that the morphology of the Mexico City skulls matches that of the Pericue Indians of Baja California, who became extinct in the 18th century, and whom early Spanish missionaries described as being racially and culturally different from other Native Americans. Gonzalez has noted that the skulls are like those of southern Asians, Australian Aborigines, and people of the South Pacific Rim and seem to also have a tie to the Ainu, a Caucasian group in Japan.
Subsequent DNA tests on the skeletal material confirmed that the Mexican finds were consistent with an Australian origin.  To Gonzalez and her colleagues, the evidence points to “several migration waves into the Americas at different times by different human groups” and that some of them crossed the Pacific in boats. Gonzalez believes that various groups of Old World peoples migrated to the Americas from northeast Siberia, the western Pacific, and even Europe.  DNA specialist Douglas C. Wallace also believes that the closest ties to Native American mtDNA are with the Polynesians,  and some scholars have noted cultural similarities between the two regions as well. 
Native American Objections
In recent years, some Native American groups have objected to the conclusions reached by geneticists, arguing that the research is faulty. Some of these groups fear that assigning an Asiatic origin to their ancestors will deprive them of their status as natives of the New World, which could impact legal decisions regarding the various tribes.
A paper, “Genetic Markers Not a Valid Test of Native Identity,” prepared for the Council for Responsible Genetics of the Indigenous People’s Council on Biocolonialism, lists reasons why the group rejects DNA claims for an Asian origin.  The authors write, “The most obvious problem is that being Native American is a question of politics and culture, not biology: one is Native American if one is recognized by a tribe as a member. And one is not necessarily a member of a tribe simply because one has Native American ancestors.”
Following a brief description of the mtDNA and Y chromosome markers used by geneticists, they note that “none of these markers is exclusive to Native American populations; all can be found in other populations around the world. They simply occur with higher frequency in Native American populations.” They further note that if “31 of your 32 great-great-great-grandparents” were Native Americans and the other was the ancestor whose mtDNA was passed to you but was not one of the five haplotypes (A-D, X), then an examination of your mtDNA would conclude that “you would not be identified as Indian.”
Conversely, if 31 of those great-great-great-grandparents were European but the one whose mtDNA you inherited fell into one of the five haplotypes, “you will have inherited a ‘Native American’ mtDNA marker.” The same holds true of Y chromosome studies and “there is a very high chance of someone having a significant amount of Native American ancestory [sic], and yet appearing to be a non-Native. All it takes is one non-Native person located in the proper position in a person’s ancestry.” The report adds that “Scientists have not tested all Native Americans, so they do not know for sure that Native Americans only have the markers they have identified, even when their maternal or paternal bloodline does not include a non-Indian.”
Ancient American DNA
While we await DNA research on skeletal remains such as Kennewick man, we can rely on a few other findings. The first study of mtDNA from ancestral Amerindian populations of South America was reported in 1996 and was based on mtDNA extracted from 18 skeletons of pre-Columbian Amerindians found in the Brazilian Amazon region and dated to 500-4,000 B.P. The sequencing of at least 354 bases permitted the identification of 13 haplotypes defined by variation of 26 nucleotide positions. Two of these haplotypes were shared by more than one sample, while the others were unique.
The researchers concluded that “if our sample is representative of Pre-Columbian South America, the percentage of haplotypes (39%) not belonging to the four [Asian] haplogroups described by Horai is much greater than in contemporary indigenous populations. This permits us to suggest that, in addition to the postulated bottleneck effect during the migration from Asia to the Americas, the depopulation effect started by European colonization in the 16th century contributed to the reduction in genetic variability of Amerindians.”  In other words, much of the mtDNA present in precolumbian South American populations seems to have been lost when Europeans began settling the New World, bringing deadly weapons and diseases to bear on Native Americans.
On the other hand, a 1997 study of mtDNA extracted from 60 skeletal remains representing four extinct human populations from the southernmost part of South America noted the “complete absence of two of the four mitochondrial haplotype groups present in contemporary Amerinds, namely A and B. In contrast, haplogroups C and D were found in all but one sample with frequencies of ~38% and 60%. These results, together with the decreasing incidence of group A in more southerly latitudes in the American continent and the absence of cluster B above 55o North in America and Asia, argue that the first settlers entering America 21 000-14 000 years ago already lacked both mtDNA lineages.” 
A 1996 study of mtDNA from the skeletal remains of 47 individuals from the Great Salt Lake Wetlands (A.D. 400-1000) and Levee (A.D. 1000-1350) phases of the Fremont culture, found haplogroups B, C, and D known from studies of modern Native American populations. The researchers reported that, “The most striking result is the absence of haplogroup A in this Fremont series, despite its predominance in contemporary Amerindian groups.” They also noted the presence of haplotype N, which had previously been “observed in some modern [Native American] populations and two other prehistoric samples.”  The total percentage of “other” mtDNA was the same (7%) as that for haplogroup D, which is generally present in higher proportions in modern Native Americans. The results of this study suggest that caution must be exercised when concluding that the mtDNA of ancient populations matched those of modern Native Americans living in the same geographical region.
One report noted that “While haplogroups B, C, and D have all been identified in Paleo-Indian skeletal remains, the oldest reported number of haplogroup A, the most common haplogroup in [modern] North America and the New World dates only to 4,504 + 105 years BP … In a preliminary restriction analysis of 18 samples dating to before 6,500 years of age, no members of haplogroup A were reported.” 
This demonstrates that studies of mtDNA in modern populations does not necessarily reflect the mtDNA of ancestral populations, since some lines can die out over time. One is tempted to note that the timeframe given by these researchers for the earliest occurrence of haplogroup A in Native American skeletal remains (ca. 2500 B.C.) roughly corresponds to the time when the Jaredites would have arrived in the New World (via Asia, if Nibley is right  ).
For mtDNA, only the most prevalent types survive for long. Low-level varieties disappear from small populations just as family names tend to disappear in populations that take their surnames from males.  In all likelihood, people from some migrations became mitochondrially or literally extinct.
Problems with DNA Population Studies
Despite its utility in population studies, a number of geneticists have found serious problems with using mtDNA to determine ethnic origins and population history. Some scholars have discussed the problems associated with the collection of DNA samples. One study warned,
Sampling strategy in Y-chromosome population studies has not generally been given sufficient attention. Whether collected samples truly reflect the structure of a population is difficult to assess. Since many populations may be both very heterogeneous and highly structured geographically, researchers must be careful when extrapolating conclusions based on an analysis of samples from a restricted area. Care must also be exercised in the interpretation of data from samples that are of unknown provenance and for which only a broad description or origin is available. 
A 2004 report indicates that, “The laboratory error in sequencing mtDNA is extremely low and most laboratories report an error rate of approximately 2% … Recently, however, reports have been published indicating that certain mtDNA sequences in widely used databases likely have a large number of errors (Bandelt et al. 2002). In order to reduce this error even further, many laboratories sequence the mtDNA of a sample twice.” 
A 1997 study by D. Andrew Merriwether and colleagues noted some of the difficulties in dealing with DNA recovered from skeletal remains. In a section entitled “Ancient DNA Research Problems and Prospects,” they note the advantage of working with ancient DNA is that one can “sample populations or cultures at different points in time to see how genetic variation changes in conjunction with observed changes in the archaeological and historical records and being able to test for the presence of specific genetic markers to detect the presence of specific lineages, genetic diseases, bacterial and viral infections, and domesticated crops and animals at specific times in the past.” They then list the “limitations” inherent in such studies:
1. Samples from a given site may span centuries of time. The ideal situation is that of the Oneota, where skeletal material from many individuals was restricted to “less than 100 years” during which the site was occupied.
2. “Especially with Native American variation, one needs to examine differences in gene frequency to compare populations, and this requires reasonably large sample sizes (40+).”
3. “To achieve large population samples, one must also overcome the generally less than 50% success rate in extracting amplifiable DNA from ancient samples. If you can only extract DNA from 33% of the samples, you would need to attempt 120 samples to get the suggested 40+ individuals.”
4. “When attempting to amplify ancient human DNA, one must be aware of the possibility or even probability of modern DNA contamination when humans are performing the labwork and excavations … Contamination can be avoided, but it requires extreme caution.” “Some questions may never be addressable by ancient DNA owing to this size reduction. Nonetheless, we would encourage archaeologists and anthropologists to consider the use of ancient DNA as a powerful tool for examining human prehistory.”
5. “Owing to the level of damage to ancient DNA, it is usually not possible to amplify larger genes from ancient remains. Most ancient DNA fragments are less than 300 nts [Nts+ neucleotides] in length, thus requiring multiple amplifications and typings to construct higher resolution haplotypes. Additionally, one must be cautious about which samples should be included in analyses of gene frequencies, as the inability to amplify some sites from ancient DNA may bias the observed distribution of haplotypes.” 
The problem with undersampling of both ancient and modern mtDNA has been noted by other researchers. Malhi and Eshleman noted that, “542 tribes are currently federally recognized in the United States. Our database contains sequences from approximately 80 tribes, representing 15% of the tribes in North America. Ward et al. (1993) shows that 40 individuals need to be sampled within a tribe to capture 50-80% of the lineages in that tribe. Our database has tribal sample sizes ranging from one to over 150, with an average of approximately 25 individuals per tribe. Data from 25 individuals would capture anywhere from 40-60% of the lineages in these tribes.”  The same researchers note that, “approximately 3% of tribes in North America” are represented in Y-chromosome data. 
Though confident of their results, Kaestle and Smith note that, “Due to sampling error, it is possible that a sample size as small as 39 individuals provides an inaccurate estimate of the frequencies of the five haplogroups in the ancient inhabitants of Western Nevada, even if those inhabitants were both ancestral and genetically identical to the modern inhabitants.”  They further wrote,
Nineteen of the 21 ancient Pyramid Lake [Nevada] samples could be assigned to one of the five modern Native American haologroups. One sample does not belong to any of these haplogroups, and could represent either a sixth Native American haplogroup, a recent mutation causing the loss of the restriction site used to identify its true haplogroup, or contamination of the bone sample. The remaining sample did not amplify well and could not be placed unambiguously into one haplogroup; both samples were excluded from the analysis. The only sample possessing the C haplogroup markers was also excluded from the analysis because it is more than 3,000 years older than any of the other samples studied. 
The problem of contamination of ancient DNA (aDNA) is discussed at length in an article published in 2000. The likelihood of contamination by modern DNA is higher for ancient skeletal remains because the bones will have been handled by a number of people, from the archaeologist excavating the site to the one recording the finds and packaging them for shipment to the anthropologists and students who examine them in the classroom or lab. By the time the geneticists receive the material, the ungloved hands of those who touched it may have left modern DNA on the samples.
Concerned about “the inability to amplify a significant number of [ancient] samples and the contamination of samples with modern DNA,” the authors “analyzed five well-preserved skeletal specimens from the western United States dating from 800-1600 A.D.” Levels of contamination in the ancient DNA (aDNA) ranged from 0 to 100%, “as determined by the presence or absence of New World-specific mitochondrial markers.” I.e., those samples that did not have one of the haplogroups already determined to be “Native American” were considered to be contaminated, so even if these samples represent different Old World mtDNA haplogroups, that information will not show up in published results. “Only the determination of DNA sequence from a cloned amplification product clearly revealed the presence of both ancient DNA and contaminating DNA in the same extract.” They conclude, “Our results demonstrate that the analysis of aDNA is still an excruciatingly slow and meticulous process.”
The researchers note that “Careful selection of polymorphic markers capable of discriminating between ancient DNA and probable DNA contamination is critical. Research strategies must be designed with a goal of identifying all DNA contaminants in order to differentiate convincingly between contamination and endogamous DNA.” 
Recent discoveries suggest that there may be other problems with mtDNA studies. For example, researchers have documented rare cases of paternal mtDNA inheritance.  Normally, the production of sperm cells results in the cell being stripped of its mitochondria, but on rare occasions this does not occur. How much impact this will have on the mtDNA studies performed to date has yet to be determined. R. Sanders Williams, dean of Duke University’s school of medicine, thinks that it is a serious issue. He said, “Even a single validated example of paternal mtDNA transmission suggests that the interpretation of inheritance patterns in other kindreds thought to have mitochondrial disease should not be based on the dogmatic assumption of absolute maternal inheritance of mtDNA.” 
Because most mtDNA testing looks only at specific portions of the mtDNA molecule, Baldelt et al. have written, “Unfortunately, we now know that coding region data and their analysis are not without problems. To obtain and report reasonably correct sequences does not seem to be a trivial task, and to discriminate between Asian and Native American mtDNA ancestries may be more complex than previously thought.” 
They note that, “While errors in the scoring of basal polymorphisms could cause problems in the phylogenetic analysis of the mtDNA sequences, and eventually in the identifications of founder mtDNAs, errors in detecting private mutations would lead to biased estimation times for founder events. Phantom mutations (Bandelt et al. 2002) would inflate the age estimates.” They warn that, “With the larger and larger mtDNA sequence sets that are being analyzed, it is inevitable that the number of errors will also increase and that these errors will confound analyses.” 
Critics of the Book of Mormon make it appear that the ancestry of Native Americans has been proven to be uniquely Asian, which is not true. There are evidences of ancestry from non-Asians as well. Moreover, as one study indicated, “similar haplotypes exhibited in Asia and America could be due to convergence rather than common ancestry,” i.e., their similarity came about by happenstance. 
Most of the mtDNA studies of Native Americans have been done with a view to determining when their Asian ancestors migrated to the New World. It is generally assumed that mtDNA mutates more rapidly than nuclear DNA and a mutation constant has been used (2-4% per million years, which is hardly a precise figure over such a long period). That constant makes it possible to suggest a timeline for the original founder group to have crossed the Bering Strait from Siberia into Alaska. However, the DNA experts are not agreed on the matter, whether there were one or more migrations and the timing of each. So while the first migration is placed ca. 30,000 years B.P. or earlier, there may or may not have been subsequent migrations according to the data.
Is there a constant rate of mutation for mtDNA? (Is anything constant in nature?) It is instructive to learn how the mutation rate was determined. The first method used was to examine the mtDNA of related individuals of known female ancestry to determine how many mutations have occurred in each line (i.e., where each individual differs from his maternal relatives).  The problem with this is that the studies could only examine short-term changes because the ancestry of the individuals could not be traced back more than a few generations.
Another potential problem is the arbitrary assignment of 20 years per generation on average, which may or may not be accurate. Being off by just a year or two in each generation can make a big difference when one is dealing with tens of thousands of years. Moreover, the calculation varies considerably if one postulates that more than a single individual carrying a given haplotype arrived in the New World in the same wave of immigration. Coalescence times that assume a single founder are much older than those that assume two or more such founders (e.g., siblings), in the same way that using a 2% mutation rate per million years will give an older date than were one to use the 4% rate.
The other way to calculate mutation rates is to examine the mtDNA of animal species that have a much shorter reproductive rate than humans, such as mollusks. This assumes that the mtDNA of all creatures mutates at the same rate, which may not be so. If the time between generations (including gestation periods) differs between species, could not the mutational rate be different as well?
The whole situation is reminiscent of linguist Morris Swadesh’s method of calculating when related languages began to differ one from the other. He termed the procedure glottochronology (the term lexicostatistics is sometimes used). Swadesh compared English with related Germanic tongues, of which Frisian seems to be the closest, and, knowing the approximate time the Angles and Saxons came from medieval Germany to the British Isles, he calculated a rate of change in a core lexicon.
He began with a list of 100 words that he claimed changed at a slower rate than other words in languages, then later expanded it to 200. Following language classifications determined by other linguists, he ultimately applied his formula to most of the world’s languages or language families and suggested dates for their separation from parent languages shared with other related tongues. As it turns out, the formula worked for the Germanic languages on which it was based, but was ultimately rejected when it came to other language families. Swadesh’s assumption that languages change at a constant rate is no less “scientific” than the assumption that all mtDNA mutates at a constant rate.
The calculations for genetic separation of New World peoples from their Old World counterparts were met with skepticism by archaeologists, based on radiocarbon-datable findings at American prehistoric archaeological sites. The calculations made from mtDNA place the migrations some 15,000 years earlier than the archaeological evidence suggests. One genetic study noted,
Great variation in divergence estimates among molecular studies results from uncertainties regarding the proper calibration of mutation and divergence rates, the error estimates associated with these rates, and the events that genetic divergence may actually reflect. Attempts to calibrate a mitochondrial mutation rate have employed divergences data that themselves are uncertain. 
A Summary of Problems with Native American DNA Research
An excellent summary of problems with DNA studies is presented by Peter N. Jones of the Buu Institute,  in his “American Indian Demographic History and Cultural Affiliation: A Discussion of Certain Limitations on the Use of mtDNA and Y Chromosome Testing.”  Jones explains,
This paper examines six weaknesses inherent in current uses of genetic anthropology that attempt to resolve questions of demographic history and prehistoric cultural affiliation: 1) interpretation of coalescent times as times of origin; 2) the current uses of haplogroups; 3) sample sizes; 4) use of language groups to define population groups; 5) use of contemporary American Indian reservations to infer prehistoric tribal history; and 6) a combination of these to determine American Indian population history, historic migrations, and demographic history. 
Jones noted that, “Tracing the coalescent times leads to one ancestor of a unilineally transmitted set of markers, but the descendents of the original mtDNA will have had haplotype frequencies that differed among themselves, resulting in a biased sample of the total historic population when using coalescent times. This is so because working back in time does not allow one to take into account the various branches of diversity that the historic population had, but only can detect the lineal history of the specific marker being coalesced.” 
Jones lists several incorrect assumptions that lie behind the population studies: 1) there is a constant rate of genetic mutation, 2) Native American populations were isolated one from another, and 3) the history of particular gene systems is the history of the specific populations in which they are found. 
Jones challenges the isolationist view of precolumbian Native Americans, noting that, prior to the establishment of reservations (beginning in the 1850s), “many American Indian groups were highly mobile autonomous entities, covering large areas of land” and that “many American Indians practiced a high degree of spousal exchange and intergroup marriage among other groups in order to solidify trade arrangements and political alliances.” 
Jones also finds fault with the sampling methods used for the DNA research, saying that, “Most studies have not used random samples, but instead have used convenience samples obtained from diabetic studies, rheumatic studies, and AIDS studies, as well as other studies … As Donnelly and Tavare (1995: 418) point out, ‘In practice, genetic data are typically obtained from convenience samples rather than proper random samples. There is an obvious danger that such data may contain individuals who share relatively too much ancestry on the relevant timescales.'” 
Since there are genetic predispositions to some diseases, such as diabetes, there is more likelihood of these non-random samples coming from individuals who share a common lineage, to the exclusion of other lineages. Jones explains,
Convenience samples means that the blood samples or genetic material were not collected by the investigating scientist, but instead through third parties. Many of these third parties initially acquired the blood or genetic material for other reasons, such as diabetes testing. A review of the literature has revealed that over a hundred institutions have allowed these scientists access to American Indian blood, a lot of the time without the individual who gave the blood having any knowledge of this. Though there are many problems with this in and of itself, the point that is important in the present discussion is that the scientists have no means of verifying the actual tribal affiliation of the blood sample they are using. For example, when an individual goes in for diabetes testing, they designate themselves and their tribal affiliation, though there is no guarantee that this designation is correct, nor is there any knowledge of that individual’s family genetic history. This fact could greatly mislead the scientists into concluding various tribal haplotype frequencies that may not be correct. 
According to Jones, “most studies rely on the idea that American Indians came over in small groups … If this is the case, coalescence times will be shorter because smaller populations in the past are more likely to share ancestors and thus lead to an accelerated time of origin for American Indians and thus not truthfully demonstrating the occupational time depth American Indians have [been] in the Americas. 
“Furthermore,” Jones writes, “departures from random mating due to inbreeding, assortative mating, or population stratification can lead to non-random association between genotypes and further complicate the interpretation of the data and coalescent times.”  He cites examples of Native American groups that have very selective mating practices, drawing attention to the fact that
It is evident that neither American Indians nor specific American Indian groups were ever isolated populations and that the history of a contemporary group’s genes are not a specific history of that American Indian population. Therefore, using gene coalescent times as possible times of origins for American Indians can lead to spurious conclusions, for there is no evidence that American Indians were ever: 1) part of a neutral system that can be timed like a regularly clicking clock, 2) were isolated from each other or from Asian populations, and 3) that the current genes systems found in a particular population fully represent the diversity and history of that population. 
Noting the studies that have identified five mtDNA haplogroups (A, B, C, D, X) among Native Americans, Jones writes that, “One of the current limitations with the uses of haplogroups for inferring American Indian cultural affiliation is that there is the possibility of discovering new haplotypes as more tribes are studied and techniques developed.”  Current testing is done for markers of the five specific lineage markers, which means that “it is likely that other haplotypes will go undetected, resulting in spurious conclusions from simplified haplotype frequencies.” 
Jones writes that, “Another problem with the current sample sizes being used is the actual numbers of individuals tested to infer the genetic makeup of the entire population. Typically, sample sizes range between four and 30 individuals per tribal population; this is insufficient to detect little more than the most common haplotypes in each population.”  Even studies that purport to use a larger sample base are flawed because they merely borrow published data on smaller samplings to make up the numbers. “The largest study to date on American Indians dealt with 2,198 males from 60 global populations, including 20 American Indian groups … this study relied on large amounts of data gathered from previously published reports, and thus could not correct for those sample sizes.”  “Many authors have tested only a small set of markers on one gene (univariate) for their studies, combining their data with those of others to result in several sets of markers to arrive at their multivariate analysis. Not only have limited numbers of markers been studied and subsequently combined with other studies (which was noted above), but the mutation rate for insertions and deletions on those markers is unknown.” 
Jones also notes the tendency among some geneticists to conclude that “some American Indian tribes recently moved into a geographical area despite contrary evidence from oral history and archaeology.”  The current distribution of many North American native groups on reservations also skews the data because, “Presently, studies concerning American Indian cultural affiliation and demographic history test individuals from a reservation and combine their allele frequencies to arrive at the haplotype makeup of that population. Therefore, the researchers are using contemporary American Indian reservation demographics to arrive at a population that they infer back into prehistory. However, one of the primary problems with this method is that most contemporary American Indian reservations are not made up of a single group, but consist of different groups of American Indians that prior to being forced onto reservations were autonomous groups.”  He cites several examples of such misguided categorization.
Jones also argues that the original basis for determining Native American populations, based on language families, is faulty.  “Not only have sample sizes of groups or tribes being tested been inadequate, but most studies have relied on the use of controversial linguistic phyla in order to place their data into objective, quantifiable groups. However, as several papers have pointed out, not only do the correspondences between languages and populations differ, but there is also no agreed-upon set of linguistic phyla for American Indians.”
Citing other authorities, he concludes that, “Languages do not change at specific rates and therefore using contemporary linguistic phyla to extrapolate prehistoric population groups is ill-founded.”  “To use current American Indian languages as a baseline for prehistoric American Indian genetic affiliations and population groups seems presumptuous. Until linguistic specialists agree upon the classifications of American Indian languages, they should not be used as a means of inferring and objectifying prehistoric population groups.”  Similarly, “Mitochondrial DNA or Y chromosome lineages are not human populations.” 
In his conclusion, Jones writes,
Although mtDNA and Y chromosome studies can provide insights on America Indian origins and prehistoric relationships, they should be used with caution. Mitochondrial DNA and Y chromosome studies are in their infancy. Because of the various limitations listed above, as well as a lack of correlation between anthropological genetic data, archaeological data, ethnographic data, and oral tradition, these studies should be viewed as inchoate and requiring further investigation and support from the other fields of anthropology … The mtDNA and Y chromosome data for American Indians, as well as many other regions throughout the world, have serious limitations. However, because of the claimed authoritative validity of these studies there is great danger that they will convince nonspecialists of the validity of the hypothesized associations between American Indian groups.
When one looks carefully at the various DNA population studies of Native Americans (of which only a small percentage have been cited herein), one sees that the supposed “evidence” cited by Book of Mormon critics evaporates. The process of detecting genetic markers and sequencing Native American DNA involves techniques designed only to detect DNA of Asian origin, something that the studies have done. Native American DNA that do not fit the testing for Asian DNA have typically been labeled “other” until tests are devised to look for other markers. This is what happened with mtDNA haplogroup X, which some geneticists hold to be Asian, while others argue for a European origin. It is also found in the Near East, as are other DNA haplogroups already detected in Native American samples. The HLA and Y-chromosome evidence points to the ancient Near East and Jews in particular as a possible source for some Native American DNA.
None of this proves the Book of Mormon is true, but neither does it prove that it is untrue. Even if no Israelite DNA were found among Native Americans, this would be insufficient to invalidate the Book of Mormon, since no human being has all of the DNA of his/her ancestors, and one can even be descended from someone whose DNA is no longer in the gene pool of a given population. In sum, the DNA issue, insofar as Lamanites and other descendants of Lehi are concerned, is another red herring.
I sincerely hope that any who are troubled by the unfounded reports that DNA studies disprove the Book of Mormon will at least read this article to see what some of the studies actually say. And yes, as Moroni 10:3-5 suggests, one can also seek a spiritual witness from God.
Appendix: Latter-day Saint Responses to Critics on the DNA Question
Over the past several years, several Latter-day Saints have responded to critics who appeal to DNA studies to discredit the Book of Mormon. A few of these responses deal with the question of DNA itself, while others point out that the critics who use the DNA argument are relying on popular opinions about the Book of Mormon, its geography, and people, rather than on the text itself.
John L. Sorenson, “The Problematic Role of DNA Testing in Unraveling Human History,” Journal of Book of Mormon Studies 9/2 (2000): 66-74, at
http://farms.byu.edu/jbms/pdf/9_2_2000_11.pdf (pdf format) or
a=jbms/9_2_2000_11.inc&x=2 (html format)
John L. Sorenson and Matthew Roper, “before DNA,” Journal of Book of
Mormon Studies 12/1 (2003), posted at
Michael F. Whiting, “DNA and the Book of Mormon: A Phylogenetic Perspective,” Journal of Book of Mormon Studies 12/1 (2003), posted at
John M. Butler, “A Few Thoughts From a Believing DNA Scientist,” Journal of Book of Mormon Studies 12/1 (2003), posted at
D. Jeffrey Meldrum and Trent D. Stephens, “Who Are the Children of Lehi?” Journal of Book of Mormon Studies 12/1 (2003), posted at
Daniel C. Peterson, “Prolegomena to the DNA Articles,” FARMS Review 15/2 (2003), posted at
David A. McClellan, “Detecting Lehi’s Genetic Signature: Possible, Probable, or Not?” FARMS Review 15/2 (2003), posted at
Matthew Roper, “Nephi’s Neighbors: Book of Mormon Peoples and Pre-Columbian Populations,” FARMS Review 15/2 (2003), posted at
Matthew Roper, “Swimming in the Gene Pool: Israelite Kinship Relations, Genes, and Genealogy,” FARMS Review 15/2 (2003), posted at
Brian D. Stubbs, “Elusive Israel and the Numerical Dynamics of Population Mixing,” FARMS Review 15/2 (2003), posted at
FAIR Web Site
Kevin L. Barney, “A Brief Review of Murphy and Southerton’s “Galileo Event’,” http://www.fairlds.org/apol/bom/bom08.html
Doug Fabrizio Interview of Terryl L. Givens, Thomas Murphy, and Scott Woodward, “KUER Radio West Science & Foundations of the Book of Mormon,”
Brant Gardner, “The Tempest in a Teapot DNA Studies and the Book of Mormon,” http://www.fairlds.org/apol/bom/bom07.html, also at
Cooper Johnson, “DNA and the Book of Mormon,”
Allen Wyatt, “Motivation, Behavior, and Dissension,”
“Dr. Scott Woodward, DNA and the Book of Mormon,”