Chapter 7 - Genetic Distance

    Populations that are reproductively isolated, usually because they are separated geographically, gradually become genetically different. The principal reason for the differences is that the selectors in different environments (or the selection pressures of those selectors) are different. Also, if a portion of a population moves to a different territory, or becomes isolated from the rest of the population due to waters rising, rivers shifting, glaciers and deserts forming, or other reasons and, if some of those isolated people just happen to be a little genetically different from the remainder of the population, which is probable, the entire isolated population is likely to become even more genetically different, which is called the “Founder Effect.” Chance mutations may also arise in one population that do not arise in another population, or only one of the populations may interbreed with a third population.
    “Genetic distance” is a way of numerically expressing how genetically different two individuals or two populations are. As explained in Chapter 3, everyone has the same genes, e.g., we all have a gene for eye color, but each gene comes in an average of 14 different A-C-G-T sequences, called “alleles.” To determine the genetic distance between two individuals, the number of alleles that differ between them can be counted; 1 for populations, the number of people in each population who have a particular allele is counted (preferably using a large number of alleles to increase precision), and the results are expressed mathematically. 2 If the other person is your identical twin, all of your alleles and your twin’s alleles will be the same and the genetic distance between you will be zero. 3 If the other person is your child, at least half will be the same.
Figure 7-1
(If his other parent has some of the same alleles that you do, more than half will be the same.) If a mating is incestuous, the number of the child’s alleles that are the same as a parent’s would be higher than if the parents were unrelated. The number of alleles in common is lower between cousins, still lower for people of your own ethny and race, 4 still lower for different races and, for different species, it continues to decrease as the age of the LCA between humans and the other species increases.
    If we plot your genetic distance (assuming you are Caucasian) from all the other people on the planet, it might look something like Figure 7-1. Figure 7-1 shows, very approximately, how genetic distance increases quickly as one moves away from one’s close relatives. Then a large increase in genetic distance occurs between you and Asians and a much larger increase between you and Africans. 5
    It is not yet possible to completely analyze the DNA of every person on the planet 6 and compare any person’s DNA to any other person’s DNA in order to determine how many alleles are identical, but there are some shortcuts that give approximately the same results. The genetic distance (the “variance,” FST) between people and populations can be calculated from DNA sampling. 7 By collecting DNA samples from individuals around the world and counting SNPs, scientists have determined the genetic distances between various populations, ethnies, races, and species. The numbers at the top of Figure 7-2 (Cavalli-Sforza, 1991) give the percentage genetic distances (multiplied by 10,000) between various human populations using a modified Nei method of calculating genetic distance.

Figure 7-2

    As to the three major races, Figure 7-2 shows that s-S Africans and everyone else are the most unrelated, and North Eurasians and Southeast Asians are the second most unrelated. Note that “Caucasoids” includes North Africans (i.e., around the Mediterranean Sea), S.W. Asians (Middle East), and Indians (from India). Also note that N.E. Asians and American Indians are more closely related to Caucasians than they are to Southeast Asians.

Figure 7-3

    Figure 7-3 is a graph that positions 42 human populations along two axes that measure differences between two highly variable sections of mtDNA. (Cavalli-Sforza, 1994, p. 82). The First PC (Principal Component) and Second PC divide the data into the two halves that have the greatest and second greatest variance, respectively (WikipediaPrincipal Components Analysis”); Africans are on one side of the two PC axes and everyone else is on the other side because Africans differ genetically the most from everyone else. Since some populations (Eurasians) have evolved more than others (Africans), the point where the First and Second PC axes cross is not necessarily at or close to the LCA for the populations on the graph.
Mongol: Nomadic people of Mongolia.
Tibetan: People of Tibet.
Eskimo: People inhabiting the Arctic coastal regions of North America,
     Greenland, and northeast Siberia.
Na-Dene: North American Indian language.
Uralic: Language family that comprises the Finno-Uric and Samoyedic
     subfamilies [named after the Ural Mountains].
North Turkic: Turkey.
Ainu: A separate indigenous people that live in Japan. [See p. 206].
South Dravidian: A language spoken by peoples in southern India and
     northern Sri Lanka.
Chukchi: Northeast Siberia.
Lapp: Nomadic herding people in northern Scandinavian countries.
Basque: A people inhabiting north central Spain [said to be the most
     homogeneous racial group found by Cavalli-Sforza, early
    &#8195Europeans, with their own unique language].
Sardinian: Sardinia, an island west of Italy.
Thai: A people of Thailand.
Polynesian: A division of Oceania including scattered islands of the central
    &#8195and southern Pacific Ocean roughly between New Zealand,
     Hawaii, and Easter Island.
Melanesian: Islands northeast of Australia and south of the equator.
Khmer: A people of Cambodia.
Micronesian: A division of Oceania in the western Pacific Ocean comprising
     islands east of the Philippines and north of the equator.
Malaysian: Southern Malay Peninsula and the northern part of the island of
Berber: North Africa.
San: Nomadic hunting people of southwest Africa.
Mbuti: African pygmies.
Bantu: linguistically related central and southern Africans.
Nilo-Saharan: linguistically related sub-Saharan Africans from Nigeria to Kenya
     regions of North America, Greenland and northeast Siberia.
Figure 7-4
Figure 7-5
    As you can see in Figure 7-3, Europeans are in the top right corner, Africans are in the lower right corner, 8 and Asians are on the left side. The Nguni, Sotho, and Tsonga are South Africans, the Blaka (Figure 7-4) are pygmies in Niger, and the Mbuti are pygmies in the NE Congo. Note that the center of the graph is relatively empty, even though it represents the average of these measurements. This is because, although all these populations were once a single population, they have been becoming increasingly genetically different, on their way to becoming different species.
    Figure 7-5 is a map from the same work and shows populations grouped according to genetic similarity. Africans are yellow, Caucasoids green, Mongoloids dark blue, and Australian Aborigines brownish-red. There is a Caucasoid component in the people of northern Africa, which does not show up well in the map. The map clearly shows that people who are genetically similar occupy the same geographical area, just as one would expect; 9 in other words, race is real.
    Figure 7-6 compares the genetic distance (numbers at the bottom) between African (blue in A and B and green in C) and European populations (red in A and B and yellow in C). 10 The vertical black lines at the top are the means and the horizontal black lines at the top are the standard deviations.

Figure 7-6
In Figure 7-6, note that when alleles that are common in Africa are compared to alleles that are common in Europe (graph C) the two populations can be separated with close to 100% accuracy. The means are farther apart and the genetic distances are greater in graph C. In graphs A and B the means are close together, the genetic distances are smaller, and there is much more overlapping because far fewer alleles that are unique to those populations were used in the comparison.
    Returning to numerical genetic distances, Cavalli-Sforza’s team (1994) compiled tables that give the genetic distance separating 2,000 different racial groups from each other. Table 7-1 gives the genetic distance (using the FST method of calculation) between a few selected populations in percent (multiplied by 10,000), e.g., Bantu-Australian aborigine FST = 0.3272%. 11

Table 7-1

Ban E.Af. W.Af.San Ind. N.E. Kor. S.C. Eng. Aus.
Bantu 0
E. Africa 658 0
W. Africa 188 697 0
San 94 776 885 0
India 2202 1078 1748 1246 0
Near East 1779 709 1454 880 229 0
Korea 2668 1475 1807 1950 681 933 0
S. China 2963 1664 1958 2231 847 983 498 0
English 2288 1163 1487 1197 280 236 982 1152 0
Australia 3272 2131 2694 2705 1176 1408 850 1081 1534 0
    Note that, of the Africans, the Bantu and San, who live in South Africa, are genetically close. The East Africans, who live in the Horn of Africa, where the Eurasians entered Africa, are closer to non-Africans than any other Africans and are the population that is the most genetically distant from other Africans. Also note that the most unrelated people are the Bantu and the Australian aborigines.
    Once numerical genetic distance data had been collected, it became possible to calculate other results, some of which are quite startling. For example, we all assume that a mother is more closely related to her own child than she is to anyone else’s child, but that is not always true. For most Asians, and a large (but less than half) percentage of white Europeans, a mulatto child with a Bantu African would be less closely related to them than a randomly-selected child of their own race! 12 The explanation for that strange result is simple – the isolation of the Bantus from the Eurasians has resulted in the two populations becoming so genetically different from each other that, because Eurasians have interbred among themselves for at least tens of thousands of years, the neighbor’s child has more alleles in common with the Eurasian than the Eurasian does to his or her own mulatto child. 13
    Compared to all the human genetic variation in the world, people in the same ethnic group can be almost as related to each other as a parent is to his child. (Salter, 2003, pp. 42, 67, 124, 327, 329). “… in most situations individuals have a larger genetic stake in their ethnic groups than in their families.” (Salter, 2003, p. 37). Thus, racism is in everyone’s genetic interest.
    Genetic distances are useful in trying to figure out man’s genetic tree, which shows how people evolved into their present populations. The less the genetic distance between populations, the more recently they were a single population or, at least, the more recently they interbred. A theory of human origins has to be consistent with, at least approximately, the genetic distances between different populations.
    The concept of genetic distance has, however, been distorted by the egalitarians to show that everyone is genetically about the same. 14 For example, in his January, 2000, State of the Union address, then President Bill Clinton stated, “We are all, regardless of race, 99.9 percent the same.” The implication is that the remaining 0.1% will produce only trivial differences and can be ignored, but “one-tenth of 1 percent of 3 billion is a heck of a large number -- 3 million nucleotide differences between two random genomes.” (Anthropologist John Hawks ). 15 On the other hand, …

    The fact that the percentage difference between populations is small is not the whole story. Although some genes code for very specific traits that are not even easily detected, other genes, such as Hox genes, 19 can turn on or off large collections of genes and thereby have an immense effect on an individual’s traits. (Zimmer, 1996).
    Another distortion that has been repeated many times in the media is known as “Lewontin’s Fallacy.” (Edwards, 2003; Sarich, 2004, p. 169). Richard Lewontin stated, “nearly 85 per cent of humanity’s genetic diversity occurs among individuals within a single population.” 20 “In other words, two individuals are different because they are individuals, not because they belong to different races.” 21 Therefore, the egalitarians gleefully concluded (e.g., Zimmer, 2001, p. 81), that it is meaningless to classify people in races – biologically, there is no such thing as “race.” 22 Unfortunately, Lewontin made a statistical error because he was comparing differences in the alleles of single genes instead of groups of genes that are unique to each race. If you are told that Al has dark skin, Bob has very curly hair, Carl has short hair, Dave has black hair, Earl has long arms, Frank has a protruding jaw, Garth has a broad flat nose, and Harvey has small ears, you could not correctly identity the race of those people because those traits occasionally appear in people of all races. 23 Lewontin and the egalitarians would yell, “See, there is no such thing as race!” But suppose you are also told that those eight people are all the same person. Now you can easily correctly identify his race because having a collection of certain traits, or the alleles that code for those traits, is how we identify a race. (Figure 7-5). Some people become immortal for their discoveries, others for their mistakes. 24

Figure 7-7

    Similarities between the original languages spoken in different geographical areas coincide well with genetic similarities, 25 suggesting common ancestral populations. Figure 7-7 presents the results of an analysis of language similarities. In Figure 7-7, the small solid round circles are the locations of the Y chromosomes of populations relative to the two principal coordinate axes and the dotted ellipses enclose populations with similar languages. Note that language similarities coincide well (but not perfectly) with genetic similarities, as one would expect. The “Khoisan” cluster is the Bushmen and Hottentots (pp. 224-226), the “Niger-Congo” cluster is the western s-S Africans, the “Afro-Asiatic” cluster is the North Africans, Middle Easterners, and Sephardic and Ashkenazic Jews, and the “Indo-European” cluster is the people from India, the Australian aborigines, and the Europeans.

Chapter 8

Table of Contents


1. More accurately, the number of differences in the A-C-G-T bases on each allele (the number of SNPs) is counted. If the bases are different, but synonymous (see Appendix), that is still a SNP. However, SNPs are not the whole story. One SNP may make its allele 100% compatible with all the other alleles, while another SNP may make its allele incompatible; counting SNPs does not capture that information, which is relevant to the concept of “genetic distance.” Besides counting SNPs, the number of generations to an LCA could be counted; if you are Caucasian, there are more generations between your LCA with an African than between your LCA with another Caucasian. The number of paths of descent per generation (preferably weighted by relatedness) from you to your LCA with another person also provides an indication of genetic distance; if the other person is the same race as you, that number will be greater than 1, its magnitude increasing with the amount of inbreeding. All races are inbred, and inbreeding reduces the number of ancestors because more ancestors are the same individual, thereby increasing the number of paths of descent. (Sailer, S., “’Pedigree Collapse’ Due to Inbreeding,” iSteve Blog, March 17, 2006). Back

2.  The numerical result will depend upon the equations used, but the same relationships are obtained for the major methods. Back

3. Although identical twins have the same alleles, their environment may have altered the expression of those alleles in a way that is heritable so, in that case, one might say that they differ genetically. Also, a process called “random monoallelic expression” causes individual cells to switch off an allele received from one of the parents. (Gimelbrant, 2007). Back

4. “[O]n average, people are as closely related to other members of their subracial "ethnic" group (e.g., Japanese or Italian) versus the rest of the world as they are related to their grandchildren or nephews and nieces versus the rest of their ethnic group.” (Sailer, 2007a). A race is “a partly inbred extended family.” (Sailer, 2002). A race is “a group of persons related by common descent or heredity.” (Webster’s College Dictionary, Random House). Back

5. Within the last 60,000 yrs, the genetic distance between the races has increased due to their more rapid evolution in different directions. (Hawks, 2007; Barreiro, 2008). Back

6. The complete genomes of 2 Caucasians, 1 Asian, and 1 African (Nigerian) have now been sequenced, but only the two Caucasian sequences have been released to the public. (“Illumina unveils genome sequence of African male,” Nature News, Feb. 13, 2008). Back

7. (Salter, 2003; the mathematics of doing this will be omitted). Genetic distance data can be mitochondrial or autosomal; it is not always clear which are being used, but the mitochondrial values are much higher. (John Goodwin, "The Race FAQ"). Back

8. The genetic difference between Africans and Europeans is so distinct that the proportion of European admixture in African Americans can be determined with a margin of error of only 0.02. (Destro-Bisol, 1999). Back

9. This is to be expected because people in the same geographical area face the same selectors and share alleles due to interbreeding. “Racial categorizations have never been based on skin pigment, but on indigenous continent of origin.” (Risch, 2002). Back

10. (Witherspoon, 2007; graph A compares individual Africans to individual Europeans, graph B compares each individual to the centroid of its population, and graph C compares alleles common in Africa to alleles common in Europe; also see "Italians," excerpted from Rosenberg, 2005). Back

11. Taken from (Salter, 2003, p. 64, based on Cavalli-Sforza, 1994). Comparisons can be made between populations, such as that the South Chinese are about six times as closely related to the Koreans as they are to the Bantu (2963/498 = ~6). Back

12. The statement will therefore be true of any population where the genetic distance, “FST,” between it and Bantus is greater than 0.25%; even if the “FST” of the population is less than 0.25%, the statement will still be true of a percentage of the population, which will increase with its “FST” to the Bantus. (Salter, 2003, pp. 38, 45, 46, 64). Relatedness, r, = (½)n, where “n” is the number of generations between two related people. (Salter, 2003, p. 38). For a parent and his child, n=1 so r = ½. Kinship, f = r/2 (Salter, p. 45), so your kinship to your child is ¼. The local kinship coefficient, fo = FST + (1 – FST)[ –1/(2N – 1)], where “FST” is the genetic distance or variance and “N” is the number of people in the population. (Salter, p. 46). If the population, N, is large, then – 1/(2N – 1) will be close to zero and fo ≈ FST. Back

13. In fact, people tend to choose mates who look like their parent of the opposite sex, thereby ensuring that their children will have more of their alleles and that favorable traits will be passed on to their own children. (Bereczkei, 2004). Back

14. Craig Ventor, the “star” of the Human Genome Project, reported the 99.9% figure in 2001, but now admits that it is wrong and the true figure is over 7 times greater. (World Science, “Finding said to show ‘race isn’t real’ scrapped,” Sept. 3, 2007). Back

15. (Tang, 2005) showed that self-described race coincides almost perfectly with genetically-identified race. (Korbel, 2007) found that rearrangement of large chunks of DNA made the differences 2 to 5 times larger than the widely-quoted 0.1%. In addition, large strings of DNA are duplicated, missing, or inverted, and that may be even more important for explaining racial differences. (Lucito, 2003; Eichler, 2006; Nguyen, 2006; Redon, 2006). When those differences are included, people can differ genetically by at least 12%. (Redon, 2006; Komura, 2006). In addition to racial differences in alleles, there are also racial differences in the expression of those alleles. (Spielman, 2007). “The genetic differences between continentally defined groups are sufficiently large that one can accurately predict ancestral continent of origin using only a minute, randomly selected fraction of the genetic variation present in the human genome.” (Allocco, 2007; also see Newsome, M., “The Inconvenient Science of Racial DNA Profiling,” Wired, Oct. 5, 2007). Back

16. Breeds of dogs are vastly more different in appearance than races of people, yet they are so genetically similar that until 2003 geneticists could not distinguish between them using DNA. (Sarich, 2004, p. 185). Back

17. Since behavioral changes drive genetic changes (Chap. 4, Rule 12), one can expect behavior to be vital to reproductive success and therefore to be largely genetically controlled. Back

18. Entine, J., “Demystifying Genetics: What Sydney Can Teach Us About Science,” San Francisco Examiner, Sept. 20, 2000). (“Tiny genetic differences have huge consequences,”, Jan. 19, 2008). Back

19. Hox genes are highly conserved, i.e., they don’t mutate much. “It is mind-boggling to realize that, for all intents and purposes, many differences between a fruit fly and a human may lie pretty much in where and when certain homeobox genes are activated.” (Schwartz, 1999, p. 13). “Geneticists believe that just one regulatory gene, the testis determining factor on the Y chromosome, is responsible for all sex differences.” (Salter, 2003, p. 90). Back

20. “Evidence from the analysis of genetics (e.g., DNA) indicates that most physical variation, about 94%, lies within so-called racial groups. Conventional geographic ‘racial’ groupings differ from one another only in about 6% of their genes.” American Anthropological Association Statement on “Race.” Similarly, "Greater mtDNA differences appeared within the single breeds of Doberman pinscher or poodle than between dogs and wolves." (The 85% truism, Evo and Proud, Jan. 4, 2008). Back

21. In a 1972 paper, "The apportionment of human diversity," and again in a 1974 book, The Genetic Basis of Evolutionary Change. Back

22. The popular science magazine, Discover, published (Jan., 2004, No. 25) an article, “Our Genes Prove It: We Are Family,” which asserted “Humans are all so closely related that our entire population shows less genetic diversity than that of a small group of chimpanzees,” a version of Lewontin’s Fallacy. Also see (Jared Diamond, “Race Without Color,” Discover, Nov., 1994). New Scientist (Buchanan, M., "Are we born prejudiced?" Mar. 17-23, 2007) informs us that “… what we recognize as racial markers are biologically next to meaningless,” and Scientific American ( Dec. 2003), published “Does Race Exist?” which denied that genetic information can be used to distinguish human groups that have a common heritage and assign individuals to those groups, even though for about $100 you can have a DNA test done that will do exactly that, though they will tell you it is the “geographical area” your ancestors came from, not your racial makeup; the origin of Europeans can sometimes be determined from DNA to within a few hundred kilometers. None of these magazines apologized to their readers for misleading them. “Repeatable, independent academic research has established that with 100 genetic markers, it is possible to sort people whose known ancestors are from Africa, Europe, Asia, or the Americas with almost 100 percent accuracy.” (Sarich, 2004, p. 21; also, Witherspoon, 2007). Other scientists determined the continent people came with “perfect intercontinental differentiation” using only 14 SNPs; only 50 SNPs were needed to assign people to 9 different populations. (Paschou, 2007). Indeed, in some cases, "DNA could reveal your surname" and, if you are European, your geographic origin "within a few hundred kilometers" of where you were born. (Novembre, 2008 Back

23. See (Witherspoon, 2001, 2007) for a detailed explanation of Lewontin's Fallacy. Actually, for some traits, such as Gm blood type, you could fairly accurately determine a person’s race. A person who is fb1b3 is almost certainly white or who is ab1b3 is almost certainly s-S African. Back

24. To be fair, Lewontin has made important contributions to biology, e.g., the mathematics of population genetics. On the other hand, he has also denied that humans have genetic interests in their ethnies, again revealing his allegiance to politics over science. (Dobzhansky et al., ed., Evolutionary Biology, 1972, Vol. 6., pp. 381-98). Here is another example of Lewontin’s Fallacy by a group that should know better: “Evidence from the analysis of genetics (e.g., DNA) indicates that most physical variation, about 94%, lies within so-called racial groups. Conventional geographic "racial" groupings differ from one another only in about 6% of their genes. This means that there is greater variation within "racial" groups than between them.” American Anthropological Association Statement on "Race" (May 17, 1998). Back

25. (Poloni, 1997). “Mex” is Mexican Indians, “Pol” is Polynesians, “Bas” is Basque, and “Chi” is Chinese. Lack of data prevented inclusion of much of Asia in the graph. Back