Human Evolutionary Genetics Research Paper

The field of human evolutionary genetics is built on the premise that one can infer evolutionary history through the analysis of human genetic variation in the context of relevant theory and nongenetic data. Patterns of genetic variation are considered to be the outcome of a set of processes that can be modeled, primarily according to population genetics and molecular evolutionary theory. The goal is to ‘undo’ the process, in a sense, in order to infer the past. The extent to which nongenetic data are incorporated, either explicitly or implicitly, cannot be overstated. Geographic, linguistic, archaeological, demographic, historical, and climatological data, along with findings from other areas in biology, are incorporated to varying degrees as human evolutionary geneticists draw conclusions from the genetic data.

1. Two Directions Within Human Evolutionary Genetics

The study of genetic variation provides two categories of insight into human history and evolution. Each category reflects an emphasis on one set of evolutionary processes. While the pattern of variation of some DNA segments may reflect both sets of processes, one set is often presumed to have predominated.

1.1 Phenotypic Evolution

A longstanding focus of human evolutionary genetics is phenotypic evolution, including changes in morphology, physiology, and behavior. By comparing patterns of genetic variation of humans and closely related species, for example, researchers can infer the processes that led to the current human phenotype (to the extent that this phenotype has an inherited genetic basis). Also included in this category is the study of the evolution of susceptibility to disease. To date, most findings regarding phenotypic evolution have been in this area. The more general question of what, at the DNA level, distinguishes humans from other species has only recently reached the forefront of research in human evolutionary genetics but promises to be an exciting research direction. Critical in this area of research is an understanding of the relationship between the variation at the DNA level and the variation at the phenotypic level.

1.2 Population Processes And Events

A second category, exemplified by the human– chimpanzee comparison above, includes population processes and events. Researchers attempt to infer, for example, population sizes of the past, rates of migration between groups, and the timing of population movements. This was a primary focus of the field during the latter half of the twentieth century, and remains so into this century. A central question has been the timing and location of the origin of anatomically modern humans.

2. ‘Population’ And ‘Group’

Essential to the practice of human evolutionary genetics are definitions of the terms ‘population’ and ‘group.’ The less precisely defined term, ‘group,’ is used here to mean any collection of individuals. In a theoretical framework, the term ‘population’ is defined very precisely, as a set of individuals constituting a mating pool. All individuals of the appropriate sex in the population are considered to be equally available as potential mates. Groups of humans rarely, if ever, fit this definition of a population. The boundary between one population and another is obscure. In practice, therefore, human evolutionary geneticists delineate populations along linguistic, geographic, sociopolitical, and/or cultural boundaries. A population might include, for example, all speakers of a particular Bantu language, all inhabitants of a river valley in Italy, or all members of a caste group in India.

3. ‘DNA Segment’

The term ‘DNA segment’ is defined here as any length of DNA that is passed intact (without recombination of paternal and maternal DNA) from parent to child. Under this definition, any single nucleotide constitutes a DNA segment, as does the entire non-recombining portion of the Y chromosome. The entire mitochondrial (mt) DNA genome can also be considered an independently evolving DNA segment. Essential to this definition is the consistency of history across the nucleotides of a DNA segment. That is, without recombination, all nucleotides of a given individual’s DNA segment have been inherited via the same set of ancestors (e.g., father, paternal grandfather, etc., for the non-recombining portion of the Y chromosome). Success in inferring human evolutionary history depends in part on the extent to which we can infer the history of a particular DNA segment. It also depends on the number of independently-evolving DNA segments studied, as discussed below.

While there are, as exemplified above, stretches of the human genomes (nuclear and mitochondrial) that fit the above definition of a DNA segment, most stretches of the nuclear genome have the potential to undergo recombination. When human evolutionary geneticists analyze DNA sequences such as the Human Leukocyte Antigen (HLA) region on chromosome 6, they must either assume that recombination has occurred rarely enough to be ignored, or bring recombination into their analytical framework.

4. Summarizing Genetic Variation

In summarizing genetic variation, initially researchers count the number of copies of each genetic variant in each group under study. For instance, frequencies of the variants A, B, and O (also called alleles) of the ABO blood group system, one of the earliest polymorphic (variable) genetic systems studied, have been estimated for thousands of human groups around the world.

4.1 Within Group Genetic Diversity

Gene diversity (heterozygosity), defined as the chance that two DNA segments drawn randomly from a group are different, summarizes within-group variation and is estimated from allele frequencies. Other measures of diversity take into account additional information, such as the number of differences between two DNA segments (nucleotide diversity), or the total number of variable positions along a DNA segment (number of segregating sites), within a group. Molecular evolution and population genetic models allow the interpretation of such diversity estimates in terms of mutation rates, population size changes, natural selection, and migration.

4.2 Among And Between Group Diversity

Human evolutionary geneticists often condense information regarding the genetic variation among a set of human groups into a single measure, typically called FST (Hartl and Clark 2000). Estimates of FST and similar statistics lie behind the oft-mentioned finding that the greatest proportion of the total genetic variation in the human species lies within human groups rather than among them. This has been interpreted as evidence for a relative lack of isolation among human groups. While FST summarizes variation among groups, genetic distances summarize the genetic difference between a pair of populations (Hartl and Clark 2000). Relatively small genetic distances are considered to be evidence that two groups have been less isolated from one another than have other pairs of groups. Differences in population sizes, however, strongly influence estimates of genetic distance.

5. Patterns Of Variation For Different Segments Of The Human Genome Evolve Independently

At each segment of the human genome, variation is introduced through mutation and subsequently modified through natural selection, random genetic drift, and gene flow between groups. The current pattern of variation of each such segment tells a story. Closely linked segments (those near one another on a chromosome that undergoes recombination, or on the same non-recombining chromosome) are likely to tell a similar story, at least going back a short time. In general, however, segments of the human genome evolve independently and therefore have different histories. The most recent common ancestor (MRCA) of all contemporary representatives of a particular DNA segment (e.g., the non-recombining portion of the Y chromosome) existed in a particular place at a particular point in time. The MRCA of another, independently evolving DNA segment (e.g., the mitochondrial genome) is likely to have existed in quite a different place, at a different point in time.

In the absence of natural selection, the pattern of variation at a particular segment reflects population sizes and movements of people. Even if each DNA segment evolves independently and therefore has its own history, population size and migration influence the variation at all segments of the biparentally inherited parts of the genome similarly. Human evolutionary geneticists, therefore, combine the information from multiple DNA segments to draw conclusions regarding population processes. This consideration of multiple gene histories increases the robustness of conclusions (Mountain 1998).

An important consideration is the rate of mutation at each genomic location. DNA repair mechanisms that influence the rate of mutation differ between nuclear and mitochondrial DNA. Furthermore, DNA segments that include repeated elements (e.g., CACACACACACA) tend to mutate more rapidly than do nonrepetitive segments. Indeed, even within a short DNA segment of one hundred nucleotides, mutation rates may vary by an order of magnitude. The rate of mutation strongly influences the time depth over which a DNA segment is informative. A slowly mutating segment, for instance, will be uninformative over a short timescale. Mutation rates are very difficult to measure or estimate, primarily because mutation typically is rare (usually far fewer than 1 mutation per 10,000 nucleotide sites per generation).

With respect to gene flow, it is important to distinguish between interspecific and intraspecific questions regarding history and evolution. For questions regarding multiple species (e.g., the Hominoids), it is typically assumed that gene flow between species is essentially nonexistent. One consequence of a lack of exchange between groups is a greater chance of observing fixed genetic differences. That is, all individuals of one group are of one genetic type, while all individuals of another group are of a different type. Intraspecific questions must be approached differently in that gene flow may have occurred between groups below the species level. The current paucity of fixed differences between human populations reflects, among other factors, a great deal of genetic exchange between populations. There is a great deal of debate regarding the precise nature of genetic exchange between human populations. Migration models typically include a constant rate of gene flow over time. While such a model is likely to apply to human history at some level, more appropriate models include changing rates of gene flow and the possibility of rare but influential major migration events (Fix 1999).

When natural selection has a stronger influence on genetic variation than random genetic drift, the pattern of variation within and among groups is affected. Natural selection has clearly played a strong role in generating human genetic variation, especially in constraining the observable variation. While an environmental factor such as average temperature may influence variation at a number of loci simultaneously, and in a similar direction, other factors may influence only one or two loci. This provides one way of distinguishing between a model of natural selection and a model involving population processes alone; the latter are expected to influence all loci, while environmental factors influence only a subset. Incorporating natural selection in addition to population processes into evolutionary models greatly increases the total number of models under consideration, so that demonstrating the impact of positive natural selection (i.e., new alleles arise within an individual and provide that individual with a selective advantage over other individuals) is a serious challenge. One thoroughly investigated case of natural selection influencing human genetic variation is that of the rise of lactose tolerance in adults (Durham 1991).

6. Development Of The Field

As currently practiced, research in human evolutionary genetics builds upon over a century of the development of a theoretical framework, of technical tools (data analysis and biological), and of collections of data. Prior to 1900 there was essentially no research in the realm of human evolutionary genetics. The latter half of the nineteenth century was critical, however, in that C. Darwin, G. Mendel, and their contemporaries provided the basis for a theoretical framework with an emphasis on variation and natural selection. This continued into the twentieth century with the classic work of G. H. Hardy and W. Weinberg in 1908 (Provine 1971). Independently, they described the simple but elegant relationship between the frequencies of an allele and the frequencies of resulting genotypes (in a population, with individuals mating at random with respect to genotype).

6.1 Development Of Tools And Theory: 1900–50

The first half of the twentieth century saw the development of some of the essential elements in the field—technical tools and population genetic theory. Serological (blood typing) techniques, developed by 1900, made possible the genotyping of the ABO locus and led to a decrease in the risk associated with blood transfusions. Initial studies of the ABO locus focused on the variation among United States’ soldiers. The resulting data are recognized as the first population data for what are now called ‘classical genetic markers’: Blood groups, and protein and enzyme polymorphisms. The discovery of blood groups and protein polymorphisms continued slowly during this period.

Other dramatic developments relevant to the field of human evolutionary genetics prior to 1950 lay in the area of population genetic theory, ‘the quantitative synthesis of Mendelism and Darwinism’ (Provine 1971, p. 129). As early as the 1920s Sewall Wright, Ronald A. Fisher, and J. B. S. Haldane were publishing in the field. Fisher’s The Genetical Theory of Natural Selection (1930), Wright’s ‘Evolution in Mendelian Populations’ (1931) and Haldane’s The Causes of E olution (1932) are part of a proliferation of theoretical work of this period. They were joined by Theodosius Dobzhansky, whose Genetics and the Origins of Species (1937) was just one of several major contributions. Together, these authors outlined the theory behind the roles of natural selection, mutation, migration, and random genetic drift in the genetic evolution of populations (see Provine 1971, for an accessible history of the origins of theoretical population genetics). These theoreticians continued contributing well into the second half of the century.

By 1950, a number of classical genetic markers had been studied across human populations (CavalliSforza and Bodmer 1971). Variation within the Rh genetic system, discovered in 1939, was considered in light of population genetic theory, as researchers attempted to explain the existence of alleles that are clearly deleterious to individuals under certain conditions. While studies often focused on the health impact of genetic variation, there was also an anthropological component: Researchers sought to measure and explain the genetic differences among human groups. The elevated frequency of Rh individuals among Basques, for instance, was initially seen as evidence that they were descendents of the earliest modern humans in Europe. In 1948, Linus Pauling and colleagues described the role of the different forms of hemoglobin in sickle cell anemia, developing a revolutionary technique called electrophoresis to distinguish the different forms. At about the same time, J. B. S. Haldane put forth a hypothesis to explain the high frequency of another hemoglobin disorder, alpha thalassemia. He suggested that the blood cells of individuals heterozygous (having two different alleles) at this locus were more resistant to attack by malarial protozoa. This hypothesis was subsequently extended to sickle-cell disease. The story of the advantage of the sickle cell allele in malarial regions remains one of the most convincing examples of the population genetic phenomenon known as heterozygote advantage. For a succinct summary of the research see Durham (1991).

6.2 Development Of The Field: 1950–75

The third quarter of the twentieth century saw rapid development in several dimensions of human evolutionary genetics. In the mid-1950s, Mourant published his first Distributions of Human Blood Groups (1954) with tables of population frequencies for nine blood groups and the sickle cell trait. In the mid 1960s, Harris, Hubby, and R. Lewontin (see Lewontin 1974, for references) discovered that a large fraction of proteins show variation within a species. This discovery led to a dramatic increase in the number of informative polymorphic systems typed across human populations: by the early 1970s, Cavalli-Sforza and Bodmer (1971) list 27 such systems. In 1972 Lewontin (1972) published a seminal article concluding that most human genetic variation is found within local human populations. Remarkably, today’s more sophisticated techniques and larger data sets consistently confirm this original finding.

A number of research groups carried out detailed studies of nonWestern populations of particular interest to anthropologists at the time. J. Neel and colleagues focused on South American populations, particularly the Yanomama of southern Venezuela, beginning in the 1960s. They were surprised by the extent of genetic differentiation between Yanomama villages, attributing this to the small size of groups founding the villages. For details of this and other regional studies conducted during the 1960s, see Crawford and Workman (1973).

During this period new numerical methods, phylogenetic methods, and computer simulation were introduced and were immediately applied to the burgeoning set of data on classical genetic markers. In the early 1960s, L. Cavalli-Sforza and A. Edwards developed several methods for generating trees from population genetic data, and applied these to data for five blood groups for 15 human populations from around the world. While their initial results are not entirely consistent with results from larger, subsequent data sets, many of the features of that initial tree can be found in recent analyses as well. One surprising finding was that genetic and anthropometric traits (physical dimensions) gave very different pictures of differentiation among human groups. While anthropometric traits indicated a similarity between Africans and Australians, genetic traits (blood groups) indicated then, and still do, that these geographically distant populations are genetically distant as well.

Meanwhile theoretical population geneticists, particularly P. A. P. Moran and his students, as well as M. Kimura, continued to extend and develop models relevant to the field of human evolutionary genetics (Ewens 1979). In the context of new understanding about DNA and mutation, Kimura (1969) argued that most mutations, or variants, found in living individuals provide neither a survival reproductive disadvantage nor advantage to the individuals whose genomes include these mutations. Instead, he claimed, these mutations are neutral, or nearly so. If most variation is neutral, then patterns of variation are to be interpreted in terms of population history rather than in terms of natural selection. During this period, statistical methods were developed to test the null hypothesis of no natural selection. The subsequent finding that the null hypothesis was difficult to reject (Lewontin 1974) had a direct impact on the field of human evolutionary genetics. It reinforced the idea held by some geneticists that random genetic drift (random changes in allele frequencies over time) plays an important role in microevolution (see Kimura 1983, for discussion of controversy). While drift was certainly recognized as a powerful evolutionary force prior to Kimura’s work, his ‘Neutral Theory’ of molecular evolution combined with the results of tests for natural selection led to a pendulum swing, with increased emphasis on drift (Weiss and Buchanan 2000). This was particularly the case for recently derived variation such as that found within the human species.

A number of research developments in biology, such as the discovery of the structure of DNA by J. Watson and F. Crick in 1953, had a significant impact on the field only much later, after DNA sequencing became sufficiently routine for population studies. The development of the clarifying concepts of cladistics by W. Hennig during the 1950s and 1960s had a much more immediate impact within systematics than within human evolutionary genetics. Nonetheless, Hennig’s work has clearly had an influence in recent years, as phylogenetic methods using parsimony criteria have been applied to human DNA sequence data. In 1960, L. Pauling and E. Zuckerkandl introduced the concept of mutations occurring regularly enough at a particular nucleotide site to be considered ‘clock-like.’ To the extent that evolution at the molecular level is clock-like, researchers can infer dates, at least relative dates, given the number of mutational differences between DNA sequences.

6.3 Development Of The Field: 1975–90

By the early 1980s, new techniques for assessing genetic variation supplemented serological and protein typing. Restriction enzymes were used to cut DNA segments at very precise locations. The sizes of the resulting DNA fragments varied depending on the underlying DNA sequence of an individual. Gel electrophoresis, introduced much earlier (1948), enabled researchers to compare and estimate fragment sizes. This approach led to an explosion in the number of known human polymorphisms during the 1980s. These were initially ascertained for medical genetic research (in the search for genes associated with disease), but were immediately recognized as valuable for human evolutionary genetics research. While thousands of polymorphisms were discovered, only a few hundred were actually studied in a geographically broad set of human populations.

During the 1980s DNA sequencing began replacing restriction analysis as the tool of choice within human evolutionary genetics. The initial focus was on the most highly variable regions of the mitochondrial genome, with its large number of copies per cell. The comparison of mtDNA sequences from numerous individuals throughout the world brought public attention to the field. The public was captivated by the notion of ‘the mother of us all,’ as the mtDNA ancestor was labeled. Many people were also surprised by the interpretations of the patterns of mtDNA variation: That the common ancestors of living members of the human species lived relatively recently (less than 200,000 years ago rather than 1–2 million years ago (Klein 1999)) and that these ancestors probably lived in Africa.

7. Human Evolutionary Genetics At The Turn Of The Millennium: Population Processes And Events

7.1 Human Genome Project

The rapid development of human evolutionary genetics over the final decade of the twentieth century owes a great deal to the Human Genome Project. Much of the technology developed under the auspices of that project (e.g., improved speed and accuracy of DNA sequencing methods) is currently applied in the field. In 1991, a group of human evolutionary geneticists led by L. Cavalli-Sforza proposed that the Human Genome Project be extended to include consideration of variation among individuals. Almost immediately, the proposal encountered criticism from groups concerned that geneticists might exploit individuals contributing DNA samples. Nonetheless, a subsequent Human Genome Project plan did include consideration of variation across individuals as a major goal. New methods of detecting variation among individuals were developed. The result is hundreds of thousands of nucleotide sites known to be polymorphic in humans across the entire human genome. Very few, however, have been studied in a broad set of human populations.

7.2 DNA Of Ancient Peoples

Under some conditions (e.g., low temperature, aridity) DNA persists relatively intact within bones and teeth for thousands, if not hundreds of thousands, of years (Wayne et al. 1999). During the 1980s human evolutionary geneticists began to take advantage of this preservation of genetic information of the past. DNA from the bones of Neanderthal specimens, for example, has been compared with that of living humans in order to address the question of whether Neanderthals and anatomically modern humans interbred on a large scale within the last 100,000 years. The DNA from early anatomically modern humans promises to provide even further insight into human evolution during this time period (Relethford 2001). With respect to more recent human evolution, ancient DNA of Native Americans living prior to 500 years ago has demonstrated no evidence of a dramatic loss of genetic variation in the intervening period, despite a dramatic reduction in the number of Native Americans (O’Rourke et al. 2000).

7.3 DNA Of Living Peoples: Inferences Across A Range Of Time Scales

While the DNA of ancient peoples has great potential to provide insight into human evolution, the DNA of extant peoples has been studied much more extensively (Cavalli-Sforza et al. 1994). The DNA of living people retains the signatures of events and processes at various time depths ranging from hundreds of thousands of years down to tens of years. Autosomal, Y chromosomal, and mtDNA of living people has shed light, for instance, on the peopling of Europe over the last 40,000 years. Researchers have estimated the relative contributions of Paleolithic and Neolithic Europeans to the present-day European gene pool. Independently evolving segments of the human genome are consistent in supporting the hypothesis of a relatively large contribution by Paleolithic Europeans to today’s European gene pool, while Neolithic peoples from the Near East appear to have contributed roughly a quarter of that gene pool (Lell and Wallace 2000).

Two examples illustrate the potential for making inferences on a shorter time scale (a few thousand years at most). The social structure of the Hindu caste system, for instance, has been investigated using a combination of Y chromosome and mtDNA. Human evolutionary geneticists have compared the relative rates of male and female movement between castes, concluding that the caste system is driven by the social mobility of women (Bamshad et al. 1998). As a second example, oral traditions have been examined in light of DNA variation. According to oral history, the Lemba, a Bantu-speaking people of southern Africa, descend from Jews who moved from the Middle East to Yemen and then to southern Africa within the last 2,700 years. The finding that Lemba Y chromosomes are similar to those common amongst Jews and rare amongst the Lemba’s African neighbors supports this tradition (Owens and King 1999).

7.4 New Theoretical Tools

While much of the population genetic theory and many of the data analysis tools relevant to human evolutionary genetics were developed prior to the 1990s, valuable new tools have been introduced more recently. Among these are Markov Chain Monte Carlo (MCMC) methods (see, for example, Griffiths and Tavare 1995), and coalescent-based approaches (Kingman 1982). MCMC methods make possible estimation of effective population size, population growth rate, migration rate, and mutation rate. The coalescent approach to inference involves considering the history, or genealogy, of a sample of present-day genes. The approach has been most valuable in efforts to date events of gene histories such as the age of the MRCA of all human Y chromosomes (Thomson et al. 2000). Most applications of the genealogical approach have been in the context of lengthy, non-recombining segments of the genome (i.e., mtDNA and the nonrecombining portion of the Y chromosome). A major goal in the field today is to apply the same approach to a large number of segments of the autosomes and X chromosome.

8. Human Evolutionary Genetics At The Turn Of The Millennium: Phenotypic Evolution

For an unknown number of segments of the human genome, current variation among human beings or among primates reflects the impact of natural selection. In these cases, while population processes and events have played a role, differences in fitness between individuals with different alleles have had greater influence. The most readily detected evidence of selection is in the context of disease, as exemplified above in the case of sickle cell anemia, malaria, and the hemoglobin genes. The field of human evolutionary genetics overlaps with medical genetics in that researchers in both fields seek to understand the origins of any genetic variation that appears to influence disease. The history of the gene associated with cystic fibrosis, for example, has been explored extensively.

A current focus of human evolutionary genetics is phenotypic traits other than disease, such as locomotion, dentition, and skin color (Weiss and Buchanan 2000). While little is known about the genetic diversity underlying skin color, variation at one gene, the melanocortin-stimulating hormone receptor gene, is known to be associated with red hair and fair skin that burns easily. Genetic diversity at this locus is significantly higher than the average nucleotide diversity in human populations and is likely to be an adaptive response to variation in environmental factors such as the availability of sunlight (Owens and King 1999).

Recently, researchers within human evolutionary genetics have begun focusing their attention upon the question, ‘What makes us human?’ Primate evolutionary geneticist Morris Goodman (1999) and others have argued for a ‘Human Genome Evolution Project.’ The goal is to identify features unique to the human genome that influence such traits as cognition, the form of the hind limb, and susceptibility to disease. There are approximately 40 million nucleotide differences between the human and chimpanzee genomes amongst the roughly 2 ×10⁹ base pairs of non-repetitive DNA. While most of these DNA level differences appear to confer no difference at the phenotypic level, a subset clearly does. The challenge is to identify that subset.

Known differences between humans and chimpanzees include chromosomal inversions and fusions, gene duplications, and repetitive element insertions (Varki 2000). Changes in the regulation of gene expression, however, are expected to be a more common source of phenotypic change (Varki 2000, Weiss and Buchanan 2000). Goodman and colleagues have described one example, wherein γ globin genes went from being expressed only within the embryo to being expressed primarily during the fetal stage. This change correlates with an evolutionary trend towards prolonged fetal life in the anthropoid primates (Goodman 1999). The search for other such changes is likely to occupy human evolutionary geneticists for the foreseeable future.

9. The Human Evolution Puzzle: Genetics Pro Ides Some Of The Pieces

The study of human evolution takes place within a broad range of academic disciplines. These include the fields of paleoanthropology and archaeology (Klein 1999), as well as such seemingly disparate fields as linguistics, developmental biology (Weiss and Buchanan 2000) and paleoclimatology (Hewitt 2000). Only within the context of findings in these fields can the study of genetic variation provide insight into some of the many details that constitute the evolutionary history of the human species. Human evolutionary geneticists face an exciting challenge as they attempt to integrate their findings regarding human evolution with those of counterparts in these linked disciplines.

Bibliography:

1. Bamshad M J, Watkins W S, Dixon M E, Jorde L B, Rao B B, Naidu J M, Ravi Prasad B V, Rasanayagam A, Hammer M F 1998 Female gene flow stratifies Hindu castes. Nature 395: 651–52
2. Cavalli-Sforza L L, Bodmer W 1971 The Genetics of Human Populations. W H Freeman, San Francisco, CA
3. Cavalli-Sforza L L, Piazza A, Menozzi P 1994 History and Geography of Human Genes. Princeton University Press, Princeton, NJ
4. Crawford M H, Workman P L (eds.) 1973 Methods and Theories of Anthropological Genetics. University of New Mexico Press, Albuquerque, NM
5. Dobzhansky T 1937 Genetics and the Origins of Species. Columbia University Press, New York
6. Durham W H 1991 Coevolution. Stanford University Press, Stanford, CA
7. Ewens W J 1979 Mathematical Population Genetics. SpringerVerlag, New York
8. Fisher R A 1930 The Genetical Theory of Natural Selection. Clarendon Press, Oxford, UK
9. Fix A 1999 Migration and Colonization in Human Microevolution. Cambridge University Press, Cambridge, UK
10. Goodman M 1999 The genomic record of humankind’s evolutionary roots. American Journal of Human Genetics 64: 31–9
11. Griffiths R C, Tavare S 1994 Simulating probability-distributions in the coalescent. Theoretical Population Biology 46: 131–59
12. Haldane J B S 1932 The Causes of Evolution. Harper & Row, New York
13. Hartl D L, Clark A G 2000 Principles of Population Genetics. Sinauer Associates, Sunderland, MA
14. Hewitt G 2000 The genetic legacy of the quaternary ice ages. Nature 405: 907–13
15. Kimura M 1969 The rate of molecular evolution considered from the standpoint of population genetics. Proceedings of the National Academy of Sciences, USA 63: 1181–8
16. Kimura M 1983 The Neutral Theory of Molecular Evolution. Cambridge University Press, Cambridge, UK
17. Kingman J F C 1982 The coalescent. Stochastic Processes and their Applications 13: 235–48
18. Klein R G 1999 The Human Career: Human Biological and Cultural Origins. University of Chicago Press, Chicago, IL
19. Lell J T, Wallace D C 2000 The peopling of Europe from maternal and paternal perspectives. American Journal of Human Genetics 67: 1376–81
20. Lewontin R 1972 The apportionment of human diversity. Evolutionary Biology 6: 381–98
21. Lewontin R 1974 The Genetic Basis of Evolutionary Change. Columbia University Press, New York
22. Mountain J 1998 Molecular evolution and modern human origins. Evolutionary Anthropology 7: 21–37
23. Mourant A E 1954 The Distribution of Human Blood Groups. Blackwell, Oxford, UK
24. O’Rourke D H, Hayes M G, Carlyle S W 2000 Spatial and temporal stability of mtDNA haplogroup frequencies in native North America. Human Biology 72: 15–34
25. Owens K, King M-C 2000 Genomic views of human history. Science 286: 451–3
26. Provine W B 1971 The Origins of Theoretical Population Genetics. University of Chicago Press, Chicago, IL
27. Relethford J H 2001 Ancient DNA and the origin of modern humans. Proceedings of the National Academy of Sciences, USA 98: 390–1
28. Simpson G G 1944 Tempo and Mode of Evolution. Columbia University Press, New York
29. Thomson R, Pritchard J K, Shen P, Oefner P, Feldman M W 2000 Recent common ancestry of human Y chromosomes: Evidence from DNA sequence data. Proceedings of the National Academy of Sciences, USA 97: 7360–5
30. Varki A 2000 A chimpanzee genome project is a biomedical imperative. Genome Research 10: 1065–70
31. Wayne R K, Leonard J A, Cooper A 1999 Full of sound and fury: The recent history of ancient DNA. Annual Reviews of Ecology and Systematics 30: 457–77
32. Weiss K M, Buchanan A V 2000 Rediscovering Darwin after a Darwinian century. Evolutionary Anthropology 9: 187–200
33. Wright S 1931 Evolution in Mendelian Populations. Genetics 16: 97–159