Geographic distribution of human skin colour: a selective compromise between natural selection and sexual selection?

Frost, P. 1994. "Geographic distribution of human skin colour: A selective compromise between natural selection and sexual selection?" Human Evolution, 9:141-153.

Human skin colour is widely believed to vary with the intensity and duration of sunlight. Thus, solar radiation at the equator has favoured individuals with enough melanin to protect them from sunburn, skin cancer, and vitamin D overproduction (Harrison, 1973). This selective force weakens away from the equator to the benefit of other ones, namely rickets and various symptoms of vitamin D deficiency, which tip the balance in favour of less-pigmented individuals (Loomis, 1970).

This explanation has been challenged by some authors (Baker, 1958; Beadle, 1977; Blum, 1961; Blum, 1969; Guthrie, 1970; Cavalli-Sforza & Bodmer, 197 1: 747; Deol, 1975; Holick et al, 1981; Diamond, 1988). Skin cancer usually occurs late in life and would be a weak agent of selection (Blum, 1961). Sunburn ages the skin but this effect, too, shows up later on; moreover, protection from sunburn relies as much on thickening of the corneum as on melanin (Blum, 1961; Blum, 1969).

Vitamin D in particular is a dubious agent of selection. Overproduction of it is a non-problem since the excess is converted to biologically inert products and then eliminated (Holick et al, 1981). Underproduction is largely a modern problem associated with poorly lit and overcrowded industrial slums (Robins, 1991). Beadle (1977) has shown that black skin can produce adequate levels of vitamin D as far north as the Arctic Circle, even if only 10.5% of the body surface is exposed. It is also worth noting that rickets has long been infrequent in North American Indians even though they are darker-skinned than Europeans and live at comparable latitudes (Wells, 1975).

Finally, any selection due to skin cancer, sunburn, and vitamin D deficiency would be partly offset by selection due to the body's heat load. Black skin absorbs about 30% more sunlight than does white skin and thus has a higher risk of overheating — a much more serious problem in the tropics (Baker, 1958; Blum, 1961; Blum, 1969).

Were sunlight the only agent of selection, skin colour would vary with latitude. Yet little north-south variation is evident in populations native to the New World (cf. Figure 1). It has been suggested that Arctic New World populations had no need to lose pigmentation because of the vitamin D they obtained from their marine diet (Loomis, 1967; Murray, 1934) . While this may hold true for the Inuit, it does not for the inland-dwelling Athapaskans who consume much less fish and are similarly pigmented.

There is another explanation why skin colour is so uniform in the New World: Modern humans have inhabited the New World for only 15-30,000 years, in comparison with up to one million years for the Old World, and so have not had enough time to differentiate in skin colour (Haldane, 1956; Brace, 1973: 346-347). Such an explanation is at odds with the "Out of Africa" model which traces the origins of modern humans to an African population that spread outwards and replaced the European neanderthals 30-45,000 years ago (Foley & Lahr, 1992; Vigilant et al, 1991). Depigmentation of these African emigrants in Europe must have happened over a time span comparable to that of human occupation of the New World. Why, then, did skin colour change in one case but not the other?

Even in the Old World the correlation between latitude and pigmentation leaves much to be desired (cf. Figure 2). There is at best a weak gradient in skin colour from northeast to southeast Asia (Brace, 1973: 346-348). Certainly, the inhabitants of Papua-New Guinea , just south of the equator, are very dark-skinned, but so were the now extinct natives of Tasmania at 45° S (Diamond, 1988). Only when one proceeds southwards from Europe to sub-Saharan Africa does a sharp gradient in skin colour emerge. Even there, it is known that as late as two thousand years ago most of southern and eastern Africa was home to yellow-skinned Khoisan peoples (Oliver, 1966). True "black" Africans were largely limited to West Africa.

Figure 2 - Old World distribution of human skin colour, c. 1492 AD. From Brace (1973)

In primates, subspecific variation in skin colour correlates poorly with exposure to sunlight (Hershkovitz, 1968). The same holds true for interspecific variation. Gorillas live in dense rain forests and are darkly pigmented whereas baboons and macaques live in savannas and are more lightly pigmented (Guthrie, 1970).

A selective compromise?

This poor fit between latitude and pigmentation led Guthrie (1970) to suggest that geographic variation in human skin colour represents a selective compromise between two counterbalancing forces: natural selection, which favours individuals less susceptible to skin cancer, sunburn, vitamin D deficiency, and overheating; and sexual selection, which favours those whose complexion enhances their ability to attract a mate.

Why should sexual selection target skin colour? First, human skin colour is sexually dimorphic: women are fairer in all human populations whereas men are browner and ruddier (Frost, 1988; Kalla, 1973; Mesa, 1983) — a result of differing melanin and haemoglobin levels in the skin's outer layers (Edwards & Duntley, 1939; Harrison, 1973; Van den Berghe & Frost, 1986). This sex difference persists even when one controls for lifestyle differences (Kalla & Tiwari, 1970) and appears to be due to hormonal and other innate factors (Edwards & Duntley, 1939; Edwards & Duntley, 1949).

Regardless of how this sex difference came about, it appears to have influenced the cultural construction of aesthetic norms. A cross-cultural survey of the Human Relations Area Files found that most traditional societies consider lighter complexions to be more feminine and darker complexions more masculine (Van den Berghe & Frost, 1986). In European societies, this sexual connotation was skin colour's primary meaning prior to the European expansion of the post-medieval era and the development of multiracial colonial contexts (Frost, 1990; Tegner, 1992). Even today, there seems to be significantly more preference by men for lighter female skin colour and by women for darker male skin colour (Feinman & Gill, 1978).

This cultural substrate may, under suitable conditions, have generated selection pressures to accentuate existing sexual characteristics, i.e. by favouring lighter-skinned, more "feminine-looking" women or darker-skinned, more "masculine-looking" men, depending on which sex had to compete for a mate. Skin colour being only mildly sex-linked, selection for pigmentary change in one sex should affect both (Guthrie, 1970). The whole population would lighten or darken in step with selection acting on females or males alone.

For sexual selection to have influenced the geographic distribution of human skin colour, at least some populations would have had to meet the operative conditions, i.e. too many of one sex seeking to mate with too few of the other sex. This situation can arise in two ways (Darwin, 1871:263-271; Emlen & Oring, 1977):

1. An unbalanced sex-ratio due to a higher mortality rate in males or females.

2. An unbalanced operational sex-ratio due to the effects of polygyny or polyandry, i.e. when a member of one sex mates with two or more of the opposite sex, a numerical imbalance results between those males and females still available for mating.

For sexual selection to have acted upon skin colour, either one of the above would have had to apply in some populations. These two cases will now be considered.

Sexual selection due to differing mortality rates

Early Egyptian art depicts white Europeans and black Africans as they look today (Vercoutter et al, 1976: 71). We should thus go further back for evidence of an unbalanced sex-ratio, particularly to the pre-Neolithic hunting and gathering societies that preceded the onset of agriculture beginning some 7,000 years ago (Ammerman & Cavalli-Sforza, 1971).

Such evidence may be found in contemporary hunter-gatherers. Among the !Kung of southern Africa, the sex ratio falls from 105 males per 100 females at birth to 91 males per 100 females at 10 years of age and over (Howell, 1979: 247). The Yanomamo of Amazonia show a similar reversal between birth and adulthood (Chagnon et al, 1979). This age effect seems even stronger in Arctic regions where high male mortality has been noted for the Labrador Inuit (Scheffel, 1984), the Caribou Inuit (Birket-Smith, 1929: 66-68), and the Netsilik Inuit (Balikci, 1967).

High male mortality is attributed to the sexual division of labour which assigns gathering to women and hunting to men. Howell (1979: 57) notes that "gathering, as opposed to hunting, does not seem to be a highly risky business". This difference in risk increases as one moves away from the equator, apparently because of the harsher climate. Hence the high mortality of Inuit males in northern Canada: "[H]unting accidents such as drowning and exposure constituted a major threat to men between approximately fifteen and fifty years of age" (Scheffel, 1984: 63) "The preponderance of adult women is generally explained by the higher death rate among men due to the natural hazards of hunting" (Weyer, 1969: 135-136). Balikci (1967: 623) concurs with Weyer and notes that "this statement is valid for the whole of the Eskimo area". He describes the causes of death:

Among the Netsilik, drowning, either during kayak-hunting or when crossing lakes or rivers, seems to have been the most frequent type of accident. Hunters were also devoured by polar bears. Starvation, too, seems to have found more victims among men. Of the 25 deaths from hunger counted by Rasmussen, 16 were men and only 9 women. Furthermore, men had a greater tendency to commit suicide than females. Finally, in all of the murder cases described by Steenhoven the victims were men.

Krupnik (1985) attributes the hazards of Arctic hunting to a lifestyle requiring travel over long distances with a minimum of possessions. Mortality is high in Chukchi men because they follow reindeer herds on foot. Mortality is low in Nentsy men thanks to more recent cultural innovations: they cross the tundra on sleds and use herd dogs to keep reindeer nearby.

In non-Arctic environments, high male mortality does not leave a surplus of unattached females. Although "a relative scarcity of husbands, then, is a regular and expected part of the !Kung marriage system", this shortfall is offset by male polygyny (either in its serial or "true" form) so that almost all !Kung women eventually marry (Howell, 1979: 247-248). A !Kung male is not averse to taking a second wife: her gathering brings in more food than his hunting, thereby permitting him to expend less energy on sustenance (Lee, 1979: 205, 450-451).

In Arctic tundra environments, male mortality and male polygyny are in a less harmonious balance: more men die and fewer take second wives. Polygyny carries a higher cost because females procure less food than they do in tropical or temperate zones, this being a result of fewer opportunities for gathering on the tundra. "Eskimos do very little collecting of food" says Birket-Smith (1929: 133) who goes on to state that Caribou Inuit women gather only eggs (during two weeks of the summer), raw gadfly larvae, the root of a plant species, and some berries. Women's lesser role in food procurement is cited by the Inuit themselves when justifying female infanticide: "Informants agree on one of the reasons behind infanticide; namely that women do not hunt, they are not self-sufficient, and they are less independent than men" (Balikci, 1967: 621). Additional wives are thus a net drain on the family's food supply (Kjellstrom, 1973: 119):

The economic conditions for an existence in the Eskimo area imposed certain restrictions on the taking of more than one wife. To be able to provide for more than one wife, as has been mentioned earlier, a man had to be a very clever hunter or sealer in the traditional Eskimo communities.

Lower female participation in food procurement restrains male polygyny and prevents it from correcting an unbalanced operational sex ratio due to higher male mortality. In the North American Arctic, however, the higher male death rate appears to be offset by female infanticide (Balikci, 1967). In the Siberian Arctic, female infanticide is largely absent, but the tundra is confined to the northern fringe and any effects of sexual selection would have been diluted by gene-flow from boreal forest populations, e.g. the Nentsy and the Yakut, who are relative newcomers to the tundra (Krupnik, 1985).

More favourable conditions for sexual selection may have prevailed when Arctic conditions extended much further south. The last glacial maximum ca. 18,000 B.P. saw open tundra extend over the northern half of the American Great Plains (Frenzel, 1973: 144). Yet evidence of human habitation at that time is non-existent for the Great Plains and debatable for the New World in general. In Siberia, no single ecosystem predominated amid intersecting belts of forest-tundra mosaic, parkland , open scrub, forest-steppe, and true tundra (Butzer, 1971: 488-489). Only in Europe were there large expanses of continent-wide uninterrupted tundra (cf. Figure 3), on which migrating herds of reindeer and other large herbivores provided sustenance for nomadic human groups (Schild, 1976; Mellars, 1985; Soffer, 1985).

Figure 3 - Major vegetation zones in Europe at last glacial maximum (c. 18,000 BP) From Mellars (1985)

These reindeer-hunting nomads may have been exposed to a rupture in the equilibrium between high male mortality and male polygyny. Limited opportunities for food gathering by females would have discouraged polygynous males from taking additional mates. The "surplus" females would then have had to compete for a smaller number of unattached males, thereby favouring selection for accentuated female characteristics. Such conditions may have prevailed after 35-30,000 B.P., with the entry of modern humans into Europe, until about 10,000 B.P. when forest invaded the tundra in the wake of a warming climate (Schild, 1976). As the tundra retreated, reindeer hunting ceased but did not initially give way to more food gathering. Instead, there was greater exploitation of fish, seals, and whales (Price, 1991). Carbon isotope ratios are the same for 7,000-6,000 year old human remains from Denmark as for those of Greenland Inuit, whose diet is 70-95 % of marine origin (Tauber 1981).

Eventually, hunting of any sort fell off as agriculture spread out from the Middle East into Europe between 7,000 and 5,500 B.P. (Ammerman & Cavalli-Sforza, 1971). As male mortality fell with the decline of hunting and as female participation in food procurement rose with the expansion of agriculture, the driving force of sexual selection would have slackened and weakened in one population after another until it had ground to a halt.

In support of this model, a lesser role for women in food procurement is indicated by a study of human skeletal remains from Mesolithic Europe. Male remains, unlike female ones, showed signs of increased activity stress in individuals from higher social strata. The study's authors suggest that males earned higher status through success in food procurement, whereas females gained status by criteria unrelated to physical activity (Constandse-Westermann & Newell, 1989).

The above model also requires evidence of a surplus of unattached females. One piece of evidence may be the depigmentation of Europeans in comparison with populations indigenous to northern Asia and North America. As skin colour is only mildly sex-linked, selection for light-skinned, more "feminine-looking" women should have induced a general reduction of pigmentation in both sexes. Another piece of evidence may be the incidence of Pernicious Anaemia. Characterized by pallor, this illness is largely genetic in origin and especially occurs in people of Scandinavian, English, and Irish ancestry, among whom it strikes twice as many women as men (Wintrobe et al, 1981, p.563). Given that this illness tends to produce its harmful effects later in life , the pallor it causes may have bestowed a selective advantage on women with some genetic susceptibility.

Other evidence may lie in cultural traits. Patterns of behaviour become stereotyped over time, with the result that their ritualized vestiges can persist much longer than the conditions that created them. A surplus of unattached females should be associated with a pattern of specialization in communal rather than family-oriented niches, e.g. shamanism, maintenance of base-camp dwellings, and tending of communal fires. Another pattern should be taboos that would have come to define this caste of unmarried women, e.g. virginity as a mark of caste membership, immunity from harm for fear of their shamanistic power.

Shamanism is strongly linked in early European traditions to women, especially virgins. This linkage is weaker in Siberian cultures, where female shamans predominate but are nonetheless married, and virtually unknown to the native peoples of North America, among whom most shamans are married men (Czaplicka, 1969: 243-255; Saladin d'Anglure, 1988; Hallowell, 1971: 19-22). The oldest sources from Greco-Roman, Germanic, and Slavic culture areas show an overwhelming preponderance of women among seers, witches, sibyls, oracles, and soothsayers (Baroja 1964: 24-57). Gimbutas (1982) and Dexter (1985) have argued that virgin females in early Europe were seen as "storehouses" of fertile energy and thereby possessed of extraordinary power. Thus, at the dawn of the Christian era the geographer Pomponius Mela mentioned nine virgin priestesses on an island off Brittany who knew the future and gave oracular responses to sailors who consulted them (Chadwick, 1966: 79). The first-century historian Cornelius Tacitus described a virgin prophetess among the Bructeri in present-day Germany, saying that this tribe "regards many women as endowed with prophetic powers and, as the superstition grows, attributes divinity to them" (Tacitus Histories 4:61). A similar caste of prophetesses, called dryades, existed among the Gauls (Chadwick, 1966: 80-81).

Single women also figured in what appear to be ritualized communal activities. The first-century geographer Strabo described a community of women who inhabited an island at the mouth of the Loire where "no man sets foot." (Geography 4.4.6) A sacred rite required them to unroof the temple and roof it again before sunset, a rite which Lefèvre (1900: 93) interpreted as recalling an age when women daily removed their hut's thatched roof to air the smoke-filled interior.

Another pattern links female virginity to the tending of communal fires. In both Roman and Greek mythology a virgin goddess, Vesta or Hestia, guards the communal hearth. The cult of Vesta required that the sacred fire of Rome be tended by a caste of virgin women — the Vestals. There is general agreement that this cult constituted an archaic element of Roman religion; the word Vesta, itself an archaism, appears to have come down unchanged from proto-Indo-European (6,000 B.P.), suggesting ritualization at an early date (Dumézil, 1970: 311-324). The idea that a celibate female must guard the hearth still survives in European folklore, the most familiar example being Cinderella — an unmarried woman whose name came from her having to sleep by the cinders of the fireplace.

Sexual selection due to polygyny

The second model of sexual selection depends not on an unbalanced sex ratio per se but rather on an unbalanced operational sex ratio, i.e. when a disproportionate number of one sex is taken by either polygynous males or polyandrous females. Polyandry being extremely rare in our species (cf. Table 1), sexual selection more likely occurs in polygynous societies where too few females are available for too many males. This will not happen, however, if polygyny is infrequent enough and male mortality high enough to balance each other out, the result being the equilibrium previously described among the !Kung. This equilibrium is disrupted only in highly polygynous societies where many men, usually younger and less dominant ones, stay single (van den Berghe 1979: 65-66).

TABLE 1 - Distribution of Marriage Forms by Region (as a percentage of HRAF samples given between brackets)

Source: (Bourguignon and Greenbaum, 1973:51)

¹. Less than 40% of wives estimated to be polygynous.

². 40% or more wives estimated to be polygynous.

Sexual selection under such conditions can cause male characteristics to be accentuated, either by female choice or through male-male competition. Males who look more masculine would be better at impressing potential mates and at intimidating rivals.

What evidence do we have for male-targeted sexual selection in highly polygynous societies? One piece of evidence may be the much darker pigmentation of sub-Saharan Africans in comparison with inhabitants of other tropical regions. Africa has most of the world's highly polygynous societies (i.e. 40% or more of wives in polygynous unions, cf. Table 1). The reasons for this are unclear, although African women do contribute much more to food procurement and thereby make polygyny an attractive option for African men (van den Berghe, 1979: 65). The resulting male-male competition may have favoured darker complexion as a male characteristic and, since this trait is only mildly sex-linked, have led to a general darkening of the entire population.

Male-male competition may have accentuated other male characteristics. In this respect, African Americans show greater sexual dimorphism than do Euro-Americans for such traits as weight, chest size, arm girth, and leg girth, all of which would be favoured by competition between males (Wolff & Steggerda, 1943; Todd & Lindala, 1928). Stature too should be favoured although efforts to correlate its sexual dimorphism with polygyny have failed, mainly because variation is small and easily overwhelmed by noise in the data (Alexander et al, 1979; Gaulin & Boster, 1985). Using a world-wide sample of societies also makes it harder to isolate the effects of sexual selection, since the widely differing ecological conditions greatly multiply the selection pressures to control for.

Stature and skin colour being only partly sex-linked, it may be more worthwhile to see whether both sexes manifest greater body size and darker pigmentation in polygynous societies. Such a study should use a population sample that maximizes variance in polygyny while minimizing variance in ecological variables that have a selective influence on stature. Since high rates of polygyny are limited largely to sub-Saharan Africa and New Guinea, these regions would be the best candidates for study.

Polygyny does seem to correlate in sub-Saharan Africa with darker skin colour and increased body size. Khoisan peoples and pygmies stand out as a cluster of weakly polygynous societies in contrast to the generalized polygyny of this region (Bourguignon & Greenbaum, 1973: 171; Cavalli-Sforza, 1986a). They are also lighter in skin colour and shorter in stature (Schapera, 1930: 51-54, 60; Weiner et al, 1964; Cavalli-Sforza, 1986b). The literature does not clearly explain these physical differences although adaptation to the shade of rain forests is said to account for pygmy stature and skin colour (Turnbull, 1986: 121). Yet shortness and light colour also characterize the Khoisan peoples who live in open desert (Schapera, 1930).

TABLE 2 - Polygyny, Stature, and Skin Reflectance in Eastern Pygmies, Western Pygmies, and Bantu (Means ± S.E.)

Notes:

¹. Proportion of polygynists among married males.

². Percentage of light (green filter) reflected by the skin of the upper inner arm.

³. Central African Republic

Sources: (Cavalli-Sforza, 1986a & 1986b)

Cavalli-Sforza (1986a & 1986b) provides data on the interrelationships between polygyny, skin colour, and stature in pygmy groups and neighbouring Bantu (cf. Table 2). Polygyny becomes more frequent, skin reflectance decreases, and stature increases as one considers in turn the Eastern Pygmies, the Western Pygmies, and the Bantu. All three populations have long been in relative genetic isolation from each other (Bernini, 1986). Linguistic reconstruction of proto-Bantu, spoken 3,000 years ago, has uncovered a specific term for "taking a second wife", indicating that then, as now, Bantu peoples had a high incidence of polygyny (Polome, 1977). The three populations in question may thus manifest the cumulative selective effects of polygyny on skin colour and stature.

Some authors, such as van den Berghe and Frost (1986) , have proposed that sexual selection of women can still continue even in a polygynous context. Females would compete not for males in general (who are in excess supply) but rather for a minority of high-status males. Even so, such "two-way" selection may not necessarily favour lighter skin in contexts where women procure much more food than men do, as in Africa. The Ibo of Nigeria, for instance, exhibit some ambivalence towards different shades of female skin colour, these ranging from black to lighter, "yellow" hues (Ardener, 1954: 72):

In the choice of a wife, yellow-skinned girls are regarded as beauties, and, other things being equal, they command higher bride prices. On the other hand it is generally held, especially by dark-complexioned persons, that yellow-skinned people are not as strong as the dark and do not live as long. A 'black' girl is said to be a harder worker.

One Ibo man is cited as believing that "the preference for yellow girls was greater nowadays than in his youth. He thought that the reason for this was that people formerly looked for strength rather than beauty and tended to marry black girls" (Ardener, 1954: 72). Similar sentiments have been reported from a polygynous tribe in Papua-New Guinea. When choosing a mate, men looked for a "capability and willingness to work hard" and "if two available women, one dark, one light, possessed equal qualities, the dark one would have preference" (Macintosh, 1959: 281).

The belief that darker individuals are stronger may have arisen from an association of ideas between strength and darker pigmentation as masculine traits. In any case, this belief would influence not only male mate-choice but also female mate-choice and, ultimately, male-male competition for females. Lighter-skinned men would internalize the perception of themselves as being weaker and thus tend to defer to darker-skinned rivals in competition for mates.

Conclusion

Most human populations show a weak gradient in skin colour from the Arctic to the equator. The extremes of depigmentation in Europe and hyperpigmentation in sub-Saharan Africa stand out as anomalies which resist explanation in terms of natural selection alone. Sexual selection may account for much of this unexplained variation.

Macintosh (1959) tried to test this hypothesis in the field. He attempted an inter-generational study in the highlands of Papua-New Guinea where skin colour varies greatly and where the topography hindered gene-flow between populations. He gave up after numerous attempts to reconstruct genealogies more than three generations in depth with reference to skin colour.

Are there other lines of enquiry? If the Human Genome Project does locate the genes for skin colour, it should be possible, by comparing different populations, to determine when and over how long a period of time depigmentation took place in Europe and hyperpigmentation in sub-Saharan Africa.

References

Alexander, R.D., Hoogland, J.L., Howard, R.D., Noonan, K.M. & Sherman, P.W., 1979. Sexual dimorphisms and breeding systems in Pinnipeds, Ungulates, Primates, and Humans. In (N.A. Chagnon and W. Irons, eds.). Evolutionary Biology and Human Social Behavior - An Anthropological Perspective, pp. 402-435. Duxbury Press.

Ammerman, A.J. & Cavalli-Sforza, L.L., 1971. Measuring the rate of spread of early farming in Europe. Man, 6:674-688.

Ardener, E.W., 1954. Some Ibo attitudes to skin pigmentation. Man, 54:71-73.

Baker, P.T., 1958. Racial differences in heat tolerance. American Journal of Physical Anthropology, n.s. 16:287-305.

Balikci, A., 1967. Female infanticide on the Arctic coast. Man, 2:615-625.

Baroja, J.C., 1964. The World of Witches, trans. by N. Glendinning, Weidenfeld and Nicolson.

Beadle, P.C., 1977. The epidermal biosynthesis of Cholecalciferol (Vitamin D3). Photochemistry and Photobiology, 25:519-527.

Bernini, L.F., 1986. Hemoglobins, haptoglobin, and transferrins. In (L.L. Cavalli-Sforza, ed.). African Pygmies, Academic Press, pp. 231-245.

Birket-Smith, K. 1929. The Caribou Eskimos. Material and social life and their cultural position. Report of the 5th Thule Expedition 1921-24, Vol. 5. Copenhagen: Nordisk Forlag.

Blum, H.F., 1961. Does the melanin pigment of human skin have adaptive value? Quarterly Review of Biology, 36:50-63.

_____, 1969. Is sunlight a factor in the geographical distribution of human skin color? The Geographical Review, 59:557-581.

Bourguignon, E. & Greenbaum, L.S., 1973. Diversity and Homogeneity in World Societies, HRAF Press.

Brace, C.L., 1973. A nonracial approach towards the understanding of human diversity. In (C.L. Brace, and J. Metress eds.). Man in Evolutionary Perspective. Wiley.

Butzer, K.W., 1971. Environment and Archeology. An Ecological Approach, Aldine Atherton.

Cavalli-Sforza, L.L., 1986a. Demographic data. In (L.L. Cavalli-Sforza ed.). African Pygmies, pp. 23-44. Academic Press.

_____, 1986b. Anthropometric data. In (L.L. Cavalli-Sforza ed.). African Pygmies, pp. 81-93. Academic Press.

_____ & Bodmer, W.F., 1971. The Genetics of Human Populations, W.H. Freeman and Co.

Chadwick, N.K., 1966. The Druids, University of Wales Press.

Chagnon, N.A., Flinn, M.V. & Melancon, T.F., 1979. Sex-ratio variation among the Yanomamo Indians. In (N.A. Chagnon and W. Irons eds.). Evolutionary Biology and Human Social Behavior -An Anthropological Perspective, pp. 290-320. Duxbury Press.

Constandse-Westermann, T.S. & Newell, R.R., 1989. Limb lateralization and social stratification in Western European Mesolithic societies. In (I. Hershkovitz, ed.). People and Culture in Change, pp. 405-433. BAR International Series 508 (ii).

Czaplicka, M.A., 1969. Aboriginal Siberia. A Study in Social Anthropology, Clarendon.

Darwin, C., 1871. The Descent of Man and Selection in Relation to Sex, John Murray.

Deol, M.S., 1975. Racial differences in pigmentation and natural selection. Annals of Human Genetics, Lond., 38:501-503.

Dexter, M.R., 1985. Indo-European reflection of virginity and autonomy. Mankind Quarterly, 26:57-74.

Diamond, J., 1988. Survival of the sexiest. Discover, (May) pp. 74-81.

Dumézil, G., 1970. Archaic Roman Religion, University of Chicago Press.

Edwards, E.A. & Duntley, S.Q., 1939. The pigments and color of living human skin. American Journal of Anatomy 65:1-33.

_____ & _____, 1949. Cutaneous vascular changes in women in reference to the menstrual cycle and ovariectomy. American Journal of Obstetrics and Gynecology 57:501-509.

Emlen, S.T. & Oring, L.W., 1977. Ecology, sexual selection and the evolution of mating systems. Science, 197(4300):215-223.

Feinman, S. & Gill, G.W., 1978. Sex differences in physical attractiveness preferences. Journal of Social Psychology 105:43-52.

Foley, R.A. & Lahr, M.M., 1992. Beyond "out of Africa": reassessing the origins of Homo Sapiens. Journal of Human Evolution, 22:523-529.

Frenzel, B., 1973. Climatic Fluctuations of the Ice Age, The Press of Case Western Reserve University.

Frost, P., 1990. Fair women, dark men: the forgotten roots of colour prejudice. History of European Ideas 12:669-679.

_____, 1988. Human skin color: a possible relationship between its sexual dimorphism and its social perception. Perspectives in Biology and Medicine, 32:38-58.

Gaulin, S. & Boster, J., 1985. Cross-cultural differences in sexual dimorphism. Is there any variance to be explained? Ethology and Sociobiology, 6:219-225.

Gimbutas, M.A., 1982. The Goddesses and Gods of Old Europe, 6,500-3,500 BC, myths and cult images. University of California.

Guthrie, R.D., 1970. Evolution of human threat display organs. Evolutionary Biology, 4:257-301.

Haldane, J.B.S., 1956. The argument from animals to men: an examination of its validity for anthropology. Journal of the Royal Anthropological Institute, 86(2):1-14.

Hallowell, A.I., 1971. The Role of Conjuring in Saulteaux Society, Publications of the Philadelphia Anthropological Society, Vol. 2, Octagon.

Harrison, G.A., 1973. Differences in human pigmentation: measurement, geographic variation and causes. Journal of Investigative Dermatology, 60:418-426.

Hershkovitz, P., 1968. Metachromism or the principle of evolutionary change in mammalian tegumentary colors". Evolution, 22:556-575.

Holick, M.F., MacLaughlin, J.A. & Doppelt, S.H., 1981. Regulation of cutaneous previtamin D₃ photosynthesis in Man: skin pigment is not an essential regulator. Science, 211:590-593.

Howell, N., 1979. Demography of the Dobe !Kung, Academic Press.

Kjellstrom, R., 1973. Eskimo Marriage - An Account of Traditional Eskimo Courtship and Marriage, Nordiska Museets Handlingar 80.

Kalla, A.K., 1973. Ageing and sex differences in human pigmentation. Zeitschrift für Morphologie und Anthropologie 65:29-33.

Kalla, A.K. & Tiwari, S.C., 1970. Sex differences in skin colour in Man. Acta Geneticae Medicae et Gemellologiae 19:472-476.

Krupnik, I.I., 1985. The male-female ratio in certain traditional populations of the Siberian Arctic. Inuit Studies, 9:115-140.

Lee, R.B., 1979. The !Kung San. Men, Women, and Work in a Foraging Society, Cambridge University Press.

Lefèvre, A., 1900. Les Gaulois - origines et croyances, Librairie C. Reinwald.

Loomis, W.F., 1970. Rickets. Scientific American, 223: 77-91.

_____, 1967. Skin-pigment regulation of vitamin-D biosynthesis in Man. Science 157:501-506.

Loring, B.C. & Metress, J., 1973. Man in Evolutionary Perspective, John Wiley and Sons.

Macintosh, N.W.G., 1959. A preliminary note on skin colour in the western highland natives of New Guinea. Oceania, 30:279-293.

Mellars, P.A., 1985. The ecological basis of social complexity in the Upper Paleolithic of southwestern France. In (T.D. Price and J.A. Brown eds.). Prehistoric Hunter-Gatherers. The Emergence of Cultural Complexity, pp. 271-297. Academic Press.

Mesa, M.S., 1983. Analyse de la variabilité de la pigmentation de la peau durant la croissance. Bulletin et mémoires de la société d'anthropologie de Paris 13(10):49-60.

Murray, F.G., 1934. Pigmentation, sunlight, and nutritional disease. American Anthropologist 36:438-445.

Oliver, R., 1966. The problem of the Bantu expansion. Journal of African History, 7:361-376.

Polome, E.C., 1977. The reconstruction of Proto-Bantu culture from the lexicon. In (L. Bouquiaux ed.). L'Expansion bantoue, 2:779-791. Centre national de la recherche scientifique.

Price, T.D., 1981. The Mesolithic of northern Europe. Annual Review of Anthropology, 20:211-233.

Robins, A.H., 1991. Biological Perspectives on Human Pigmentation, Cambridge Studies in Biological Anthropology 7, Cambridge University Press.

Saladin d'Anglure, B., 1988. Penser le "féminin" chamanique. Recherches amérindiennes au Québec, 18(2-3):19-50.

Schapera, I., 1930. The Khoisan Peoples of South Africa. Bushmen and Hottentots, Routledge and Kegan Paul.

Scheffel, D. 1984. From polygyny to cousin marriage? Acculturation and marriage in 19th century Labrador Inuit society. Inuit Studies 8:61-75.

Schild, R., 1976. The final paleolithic settlements of the European Plain. Scientific American, 234(2):88-99.

Soffer, O., 1985. Patterns of intensification as seen from the Upper Paleolithic of the Central Russian Plain. In (T.D. Price and J.A. Brown eds.). Prehistoric Hunter-Gatherers. The Emergence of Cultural Complexity, pp. 235-270. Academic Press.

Strabo, 1923. The Geography of Strabo, Loeb Classical Library, William Heinemann.

Tacitus, 1969. The Histories, Loeb Classical Library, William Heinemann.

Tauber, H., 1981. ¹³C evidence for dietary habits of prehistoric man in Denmark. Nature, 292:332-333.

Tegner, E., 1992. Sex differences in skin pigmentation illustrated in art. The American Journal of Dermatopathology 14:283-287.

Todd, T.W. & Lindala, A., 1928. Dimensions of the body: Whites and American Negroes of both sexes. American Journal of Physical Anthropology 12:35-101.

Turnbull, C.M., 1986. Survival factors among Mbuti and other hunters of the equatorial African rain forest. In (L.L. Cavalli-Sforza ed.). African Pygmies, pp. 103-123. Academic Press.

van den Berghe, P.L., 1979. Human Family Systems. An Evolutionary View, Elsevier.

_____ & Frost, P., 1986. Skin color preference, sexual dimorphism, and sexual selection: a case of gene-culture co-evolution ? Ethnic and Racial Studies, 9:87-113.

Vercoutter, J., Leclant, J., Snowden Jr., F.M. & Desanges, J., 1976. The Image of the Black in Western Art, vol. 1, Menil.

Vigilant, L., Stoneking, M., Harpending, H., Hawkes, K. & Wilson, A.C., 1991. African populations and the evolution of human mitochondrial DNA. Science, 253:1503-1507.

Weiner, J.S., Harrison, G.A., Singer, R., Harris R & Jopp, W., 1964. Skin colour in southern Africa. Human Biology, 36:294-307.

Wells, C., 1975. Prehistoric and historical changes in nutritional diseases and associated conditions. Progress in Food and Nutrition Science, 1:729-779.

Weyer, E.M. 1932. The Eskimos. New Haven, Conn.: Yale University Press.

Wintrobe. M.M., Lee, G.R., Boggs, D.R., Bithell, T.C., Foerster, J., Athens, J.W. & Lukens, J.N., 1981. Clinical Hematology, Len and Febiger.

Wolff, G. & Steggerda, M., 1943. Female-male index of body build in Negroes and Whites: An interpretation of anatomical sex differences. Human Biology 15:127-152.