Friday, November 30, 2007

Convergent evolution?

In an earlier post, I discussed how mtDNA evidence now shows that the Neanderthals ranged at least as far east as Lake Baikal. This finding is significant because there no longer seems to have been any geographical or ecological barrier to Neanderthal occupation throughout non-tropical Eurasia.

This point has been commented on by Michelle M. (Mica) Glantz, an associate professor in anthropology at Colorado State University. In an interview with anthropologist John Hawks, she argues that the European Neanderthals may simply have been one of several interbreeding Homo erectus populations that inhabited Eurasia. They may have been more Arctic-adapted and more specialized in various ways but they were not genetically isolated, at least not fully, from other Homo erectus populations.

If we keep moving the Neandertal boundary eastward, then wouldn't Neandertals cease being a recognizable entity that is really separate from other archaic groups in the Old World during the Middle Paleolithic? In other words, who isn't a Neandertal in this case? Certainly we do not have enough similarly aged specimens from China and other points east to make thorough comparisons, but really the specimens we do have are usually not included in any of our analyses that are concerned with European Neandertals. Exceptions to this, like Rosenberg et al. (2006) study of Jinniushan, show that Asian specimens often look like they are part of the same cline as European Neandertals.

This discovery is one of several challenges to the traditional view, which sees the Neanderthals as being not only distinct from other archaic humans but also intermediate on the line of descent from Homo erectus to modern humans. Indeed, the Neanderthals increasingly look like an evolutionary dead end. Modern humans seem to owe most, if not all, of their ancestry to a demic expansion that started in East Africa some 80,000 years ago and began to spread out of Africa some 50,000 years ago. There may have been some intermixture with archaic humans already present in Europe and Asia, but even this scenario is looking more and more problematic. We can now compare mtDNA from late European Neanderthals with mtDNA from early modern Europeans and there is no measurable gene flow from the former to the latter (Caramelli et al., 2003). Perhaps some minor intermixture did occur here and there, enough to provide the modern European gene pool with a few advantageous Neanderthal genes. It could not have been greater, however, than the surreptitious insertion of certain viral and bacterial genes into the human genome.

Why, then, did scientists place the Neanderthals above Homo erectus, even to the point of classifying them as a subspecies of Homo sapiens? Because Neanderthal brains were so big, like ours. This resemblance now appears to be just a case of convergent evolution. When humans first spread out of Africa, some 1.8 to 1.7 million years ago, their brains increased in size wherever they entered non-tropical environments, apparently because such environments were mentally more demanding (need to cope with seasonal variation in temperature, to identify and/or create shelters, to hunt over larger and riskier territory, etc.). This in situ evolution seems to have progressed the furthest in populations that we call ‘Neanderthal.’

Then, 50,000 years ago, humans again spread out of Africa, probably because a change in their neural wiring gave them an edge over archaic humans already established in Eurasia. This second wave continued to evolve as it spread into different environments with different adaptive landscapes—a subject that will be the focus of an upcoming PNAS article by Greg Cochran, Henry Harpending, John Hawks, Robert Moyzis, and Eric Wang.

The second wave out of Africa is called Homo sapiens whereas the first wave is called Homo erectus. But these are just names that simplify reality. There was and has been considerable variation and evolution within both ‘species.’


Caramelli, D., Laluez-Fox, C., Vernesi, C., Lari, M., Casoli, A., Mallegni, F., Chiarelli, B., Dupanloup, I., Bertranpetit, J., Barbujani, G., & Bertorelle, G. (2003). Evidence for a genetic discontinuity between Neandertals and 24,000-year-old anatomically modern Europeans. Proceedings of the National Academy of Sciences USA, 100, 6593-6597.

Rosenberg KR, Zuné L, Ruff CB. 2006. Body size, proportions, and encephalization in a Middle Pleistocene archaic human from northern China. Proceedings of the National Academy of Sciences USA, 103, 3552-3556. doi:10.1073/pnas.0508681103

Saturday, November 24, 2007

Sex linkage of human skin, hair, and eye color

Much of my writing has focused on sexual selection of women and how it may have structured certain pigmentary traits in our species—specifically by lightening skin color and by diversifying hair and eye color. Since most of the relevant alleles are not sex-linked, this selection would have spilled over on to men as well.

Yet, if this selection acted primarily on women, shouldn’t it have tended to favor sex-linked alleles that confine these pigmentary traits to females? All things being equal, wouldn’t such alleles have come to replace those that are not sex-linked? Indeed, Mother Nature loves organisms that don’t waste their energy on things they don’t need. We see this principle in the loss of pigmentation by organisms that live solely in dark caves. It’s not because albino skin is now more useful. It’s because pigmented skin is now useless and may be dispensed with.

Human skin color does show sex linkage. From puberty on, women are lighter-skinned than men in all human populations. This sexual dimorphism seems to be greater in populations that are medium in skin color, perhaps because floor and ceiling effects constrain its expression in populations that are either very dark or very light-skinned (Frost, 2006, pp. 54-60; Frost, 2007; Jablonski & Chaplin, 2000; Madrigal & Kelly, 2006). In women, lightness of skin correlates with thickness of subcutaneous fat, apparently because of a common hormonal causation and not because of a mechanical effect of fat on skin color (Mazess, 1967). It also correlates with digit ratio, which in turn correlates with prenatal estrogenization (Manning et al., 2004). It is this exposure to estrogen before birth that seems to “program” the lightening of female skin after puberty.

Hair color too shows some sex linkage. Hair is darker in girls than in boys before puberty and then lighter afterwards (Keiter, 1952; Leguebe & Twiesselmann, 1976; Olivier, 1960, p. 74; Steggerda, 1941). In a still unpublished British study, digit ratios were found to be higher in blond participants than in darker-haired ones. This finding, if true, suggests increased prenatal estrogenization among people with blond hair.

For eye color, we have no studies that track variation by sex and age. A study of Icelander and Dutch adults found green eyes to be much more prevalent in women than in men (by at least a factor of two). Blue eyes were less prevalent and brown eyes somewhat more prevalent. The participants, however, seem to have been very heterogeneous for age. Many had been recruited for a prostate cancer study among the men or for a breast cancer study among the women (Patrick et al., 2007). Razib discusses this topic on ‘Brown eyed girl’ at GNXP. In the above unpublished British study, digit ratios were found to be higher in light-eyed participants than in brown-eyed ones. This finding, if true, suggests increased prenatal estrogenization among people with non-brown eyes.


Frost, P. (2007). Comment on Human skin-color sexual dimorphism: A test of the sexual selection hypothesis, American Journal of Physical Anthropology, 133, 779-781.

Frost, P. (2006). European hair and eye color - A case of frequency-dependent sexual selection? Evolution and Human Behavior, 27, 85-103.

Frost, P. (2005). Fair Women, Dark Men. The Forgotten Roots of Color Prejudice. Cybereditions: Christchurch (New Zealand).

Jablonski, N.G., and Chaplin, G. (2000). The evolution of human skin coloration. Journal of Human Evolution, 39, 57-106.

Keiter, F. (1952). Über ͈Nachdunkeln” und Vererbung der Haarfarben. Z. Morph. Anthrop. 44, 115-126.

Leguebe, A., & Twiesselmann, F. (1976). Variations de la couleur des cheveux avec l’âge. Z. Morph. Anthrop. 67, 168-180.

Madrigal, L, and Kelly, W. (2006). Human skin-color sexual dimorphism: A test of the sexual selection hypothesis. American Journal of Physical Anthropology, 132, 470-482.

Manning, J.T., Bundred, P.E., and Mather, F.M. (2004). Second to fourth digit ratio, sexual selection, and skin colour. Evolution and Human Behavior, 25, 38-50.

Mazess, R.B. (1967). Skin color in Bahamian Negroes. Human Biology, 39, 145‑154.

Olivier, G. (1960). Pratique anthropologique. Paris: Vigot Frères.

Patrick, S., D.F. Gudbjartsson, S.N. Stacey, A. Helgason,T. Rafnar, K.P Magnusson, A. Manolescu, A. Karason, A. Palsson, G. Thorleifsson, M. Jakobsdottir, S. Steinberg, S. Pálsson, F. Jonasson, B. Sigurgeirsson, K. Thorisdottir, R. Ragnarsson, K.R. Benediktsdottir, K.K. Aben, L.A. Kiemeney, J.H. Olafsson, J. Gulcher, A. Kong, U. Thorsteinsdottir, and K. Stefansson. (2007). Genetic determinants of hair, eye and skin pigmentation in Europeans. Nature Genetics Published online: 21 October 2007 doi:10.1038/ng.2007.13

Steggerda, M. (1941). Change in hair color with age. Journal of Heredity, 32, 402-403.

Friday, November 16, 2007

Natural selection in proto-industrial Europe

Do human populations vary statistically in their mental abilities and predispositions? And if so, why? Such questions have lately refocused from the ‘macro’ to the ‘micro’ level—from differences that may have developed in prehistory between large continental populations to those that may have arisen in historic times between much smaller groups.

In part, the focus is now on the higher intellectual performance of Ashkenazi Jews (Cochran et al., 2006; Murray, 2007). It is also on the attitudinal and behavioural changes that ultimately sparked England’s Industrial Revolution. In a newly released book, Gregory Clark (2007) argues that natural selection gradually raised the English population to a threshold that made this economic sea-change possible, specifically by selecting for middle-class values of non-violence, thrift and foresight. The question remains, of course, as to why this threshold was first reached in England.

This new focus reflects a growing recognition that natural selection does not need aeons of time to change a population significantly (Eberle et al., 2006; Harpending & Cochran, 2002; Voight et al., 2006; Wang et al., 2006). Change can occur over a dozen generations, certainly during the last six millennia of history. This possibility is all the likelier given that these millennia have seen humans specialize in a wide range of occupations, some more mentally demanding than others. Charles Murray, for one, has argued that selection for intelligence was historically weaker in farming and stronger in sales, finance and trade.

He may be right. But the reason, I believe, lies not so much in the occupation itself as in its relations of production.

In the Middle Ages and earlier, farmers had little scope for economic achievement and just as little for the intelligence that contributes to achievement. Most farmers were peasants who produced enough for themselves, plus a surplus for the landowner. A peasant could produce a larger surplus, but what then? Sell it on the local market? The possibilities there were slim because most people grew their own food. Sell it on several markets both near and far? That would mean dealing with a lot of surly highwaymen. And what would stop the landowner from seizing the entire surplus? After all, it was his land and his peasant.

The situation changes when farmers own their land and sell their produce over a wide geographical area. Consider the "Yankee" farmers who spread westward out of New England in the 18th and 19th centuries. They contributed very disproportionately to American inventiveness, literature, education and philanthropy. Although they lived primarily from farming, they did not at all have the characteristics we associate with the word "peasant".

Conversely, trade and finance have not always been synonymous with high achievement. In the Middle Ages, the slow growth economy allowed little room for business expansion within one's immediate locality, and expansion further afield was hindered by brigandage and bad roads. Furthermore, the static economic environment created few novel situations that required true intelligence. How strong is selection for intelligence among people who deal with the same clients, perform the same transactions and charge the same prices year in and year out?

This point has a bearing on the reported IQ differences between Ashkenazi and Sephardic Jews. Charles Murray, like others, believes that the Ashkenazim were more strongly selected for intelligence because more of them worked in sales, finance and trade during the Middle Ages. Now, we have no good data on the occupations of medieval Ashkenazim and Sephardim. But the earliest censuses (18th century for Polish Jews and 19th century for Algerian Jews) show little difference, with the bulk of both groups working in crafts.

There was, however, one major demographic difference. While the Sephardim grew slowly in numbers up to the 20th century, the Ashkenazim increased from about 500,000 in 1650 to 10 million in 1900. The same period saw strong population growth among Europeans in general. This boom used to be attributed to falling death rates alone, but demographers now recognize that rising birth rates were also key, in some countries more so. England, in particular, saw a rise in fertility that contributed two and a half times as much to the increase in growth rates as did the fall in mortality, largely through a younger age of first marriage. This was how England overtook France in total population.

The baby boom was particularly strong among one class of people: semi-rural artisans who produced for the larger, more elastic markets that developed with the expanding network of roads, canals and, later, railways. Their family workshops were the main means for mass-producing textiles, light metalwork, pottery, leather goods and wood furnishings before the advent of factory capitalism. Unlike the craft guilds of earlier periods, they operated in a dynamic economic environment that had few controls over prices, markets or entry into the workforce. "They were not specialized craftsmen in life-trades with skills developed through long years of apprenticeship; they were semi-skilled family labour teams which set up in a line of business very quickly, adapting to shifts in market demand" (Seccombe, 1992, p. 182). Their workforce was their household. In more successful households, the workers would marry earlier and have as many children as possible. In less successful ones, they would postpone marriage or never marry.

In Western Europe, these cottage industries were located in areas like Ulster, Lancashire, Yorkshire, Brittany, Flanders, Alsace, Westphalia, Saxony, the Zurich uplands, the Piedmont and Lombardy. In Eastern Europe, they were concentrated among Ashkenazi Jews. Selection for intelligence among the Ashkenazim may thus have been part of a larger European-wide selection for intelligence among cottage industry workers. These entrepreneurial artisans had optimal conditions for selection: 1) tight linkage between success on an intelligence-demanding task and economic achievement; 2) broad scope for economic achievement; 3) tight linkage between economic achievement and reproductive success; and 4) broad scope for reproductive success. Such artisans were a minority in Western Europe. Among the Ashkenazim, they appear to have been the majority.

In the late 19th century, cottage industries gave way to factories and the tight linkage between economic achievement and reproductive success came undone. Entrepreneurs could now expand production by hiring more workers. Henry Ford, for instance, produced millions of his Model T but had only one child.

Thus, it is not the type of occupation that drives selection for intelligence. It is the relations of production. In particular, do people own their means of production? Do they operate in a large, elastic market that rewards progressively higher levels of ability with commensurate increases in production? Finally, do they meet the increased demand for labour by increasing the size of their families?


Clark, G. (2007). A Farewell to Alms: A Brief Economic History of the World. Princeton (NJ): Princeton University Press.

Cochran, G., Hardy, J., & Harpending, H. (2006). Natural history of Ashkenazi intelligence. Journal of Biosocial Science, 38, 659-693.

Eberle, M.A., M.J. Rieder, L. Kruglyak, D.A. Nickerson. (2006). Allele frequency matching between SNPs reveals an excess of linkage disequilibrium in genic regions of the human genome. PLoS Genet 2(9), e142

Harpending, H., & G. Cochran, G. (2002). In our genes. Proceedings of the National Academy of Sciences, 99(1), 10-12.

Murray, C. (2007). Jewish Genius. Commentary, April.

Seccombe, W. 1992. A Millennium of Family Change. Feudalism to Capitalism in Northwestern Europe, London: Verso.

Voight B.F., Kudaravalli, S., Wen, X., &Pritchard, J.K. (2006). A map of recent positive selection in the human genome. PLoS Biol, 4(3), e72

Wang, E.T., Kodama, G., Baldi, P., & Moyzis, R.K. (2006). Global landscape of recent inferred Darwinian selection for Homo sapiens, Proc. Natl. Acad. Sci USA, 103, 135-140.

Friday, November 9, 2007

Neanderthal Redheads?

Recently, a Spanish team has analyzed the bones of two Neanderthals and recovered the MC1R gene—the one that controls hair color and, in the case of red hair, lightens skin color. Their findings? First, both of the Neanderthals had an MC1R variant unlike any that now exist in modern humans. This pours cold water on the idea that the unique hair colors of modern Europeans are due to Neanderthal intermixture. Second, the MC1R variant looks like a ‘loss-of-function’ allele, much like the ones that now produce red hair.

According to the lead author, “In Neanderthals, there was probably the whole range of hair colour we see today in modern European populations, from dark to blond right through to red” (Rincon, 2007). Well, maybe yes, but maybe no. Europeans have a diverse palette of hair colors because their MC1R gene has at least 11 functionally different alleles—and not because a new allele replaced the original African one. In fact, the ‘African’ allele is still common in Europe.

Will the Neanderthals be shown to have many MC1R variants, just like modern Europeans? The lead author seems to think so, as does Dr Clive Finlayson, director of the Gibraltar Museum, who says that Neanderthal and modern European hair color may reflect a common “propensity towards the reduction of melanin in populations away from the tropics. … a good example of parallel, or convergent evolution - a similar evolutionary response to the same situation” (Rincon, 2007). Melanin was reduced because there was less natural selection for dark skin (i.e., for protection against solar UV) and more for light skin (i.e., for vitamin D synthesis).

Neither selective pressure, however, would have diversified hair color, at least not that of modern Europeans:

1. When there is less selection for dark skin, ‘loss-of-function’ variants will proliferate at any gene associated with skin color. But such proliferation needs about a million years to produce the hair-color variability that Europeans now display, including c. 80,000 years to produce just the current prevalence of red hair (Harding et al., 2000; Templeton, 2002).

2. When there is more selection for light skin, the original allele at any one gene will be replaced by an allele that optimally reduces skin pigmentation. But the overall number of alleles will remain the same.

Moreover, if we examine the many homozygous and heterozygous combinations of hair color (MC1R) alleles, most have little visible effect on skin pigmentation, except for the ones that produce red hair (Duffy et al., 2004). It is difficult to see how either relaxed selection for dark skin or increased selection for light skin could have given rise to most of these alleles, especially over such a short span of evolutionary time.

It was probably another selective pressure that diversified European hair color. Sexual selection is especially likely because it is known to produce bright color traits, especially polymorphic ones.

Sexual selection is also indicated by the geographic distribution of hair-color diversity. During the last ice age, particularly on the steppe-tundra of northern and eastern Europe, a unique demographic environment intensified sexual selection of women by reducing the supply of men and by limiting polygyny. Hunters die in proportion to the distances they cover, and hunting distance was at a maximum on these open plains with their dispersed and highly mobile herds. Since women depended on men for sustenance, there being little food to gather on the tundra, only the ablest hunter could provide for multiple wives. Result: too many women competing for too few men. In a saturated, competitive market, success hinges on visual merchandising; therefore, a woman with a bright, novel color would have attracted attention and edged out otherwise equal but bland rivals. It is this sexual selection for both brightness and novelty that may have multiplied the number of hair color alleles among early modern Europeans.

This demographic environment was unknown to the Neanderthals. Unlike early modern Europeans, they never colonized the steppe-tundra of northern and eastern Europe under ice age conditions:

When temperatures declined during the Early Pleniglacial (OIS 4), the Neanderthals were apparently unable to cope with periglacial loess-steppe environments on the East European Plain. Much of the latter seems to have been abandoned by Neanderthals at this time, although some areas (notably the southwest regions) were reoccupied during the milder Middle Pleniglacial (OIS 3). By contrast, modern humans successfully colonized the periglacial loess-steppe during the terminal phases of OIS 3 and the subsequent Last Glacial Maximum (OIS 2). (Hoffecker, 2002, p. 136)

The Neanderthals also seem to have been characterized by relatively limited movements and small territories in comparison to recent hunter-gatherers in northern latitudes. (Hoffecker, 2002, p. 135).

These archaic humans seem to have occupied a niche that excluded hunting of herds over expanses of steppe-tundra. Among the many environments that modern humans colonized, this seems to be the one that most intensified sexual selection of women—by reducing both the supply of men and their demand for mates.

So I doubt that the Neanderthals had a wide range of hair colors. They probably all sported the same reddish coat. Coat? Yes, you read right. They were as furry as bears. How else did they survive in subzero temperatures without tailored clothing? (Hoffecker, 2002, pp. 107, 109, 135, 252). Moreover, both needles and the human body louse (which lives in clothing) seem to date back no earlier than 50,000 years ago, i.e., the transition from Neanderthals to modern humans (Harris, 2006).


Duffy, D.L., Box, N.F., Chen, W., Palmer, J.S., Montgomery, G.W., James, M.R., Hayward, N.K., Martin, N.G., and Sturm, R.A. (2004). Interactive effects of MC1R and OCA2 on melanoma risk phenotypes. Human Molecular Genetics, 13, 447-461.

Harding, R.M., Healy, E., Ray, A.J., Ellis, N.S., Flanagan, N., Todd, C., Dixon, C., Sajantila, A., Jackson, I.J., Birch?Machin, M.A., and Rees, J.L. (2000). Evidence for variable selective pressures at MC1R. American Journal of Human Genetics, 66, 1351?1361.

Harris, J. R. (2006). Parental selection: A third selection process in the evolution of human hairlessness and skin color. Medical Hypotheses, 66, 1053-1059.

Hoffecker, J.F. (2002). Desolate Landscapes. Ice-Age Settlement in Eastern Europe. New Brunswick: Rutgers University Press.

Laleuza-Fox, C., Römpler, H., Caramelli, D., Stäubert, C., Catalano, G., Hughes, D., Rohland, N., Pilli, E., Longo, L., Condemi, S., de la Rasilla, M., Fortea, J., Rosas, A., Stoneking, M., Schöneberg, T., Bertranpetit, J., Hofreiter, M. (2007). A Melanocortin 1 receptor allele suggests varying pigmentation among Neanderthals. Science. doi:10.1126/science.1147417.

Rincon, P. (2007). Neanderthals 'were flame-haired'. October 25, 2007.

Templeton, A.R. (2002). Out of Africa again and again. Nature, 416, 45-51.