To over-simplify, his argument is that it's due to sexual selection during the last ice age, with its maximum glaciation around ~20,000 BP.
1. Females in the northern latitudes, in contrast to their food-gathering counterparts in the tropics, would've been dependent on males for food.
2. Males, on the other hand, would've been in "short supply" because of the hazards of hunting in the cold over long distances. This is seen in arctic peoples of the modern age, when there was almost a 2:1 ratio of females to males.
3. Because the male can only support one female in those harsh conditions, there would've been sexual selection/competition in which the women who stand out the most (non-brown eyes, non-black hair) would've been selected. It's similar to how a Persian cat's blue eyes are a result of selective breeding by owners.
4. Female infanticide was practiced to reduce the female-to-male imbalance. This was enough for the arctic but the steppe would've produced more population and female infanticide wouldn't have compensated for the imbalance. *Personally, I also think the harsher Siberian conditions (physical demands on the appearance of men in their long exposures to cold during hunting) would've made male phenotype out compete any phenotype gotten through sexual selection through women. As for skin color, unlike Inuits and other Asian boreal populations who have a fish
based diet, Western Eurasians would have relied more on-land supply, so
to get enough vitamin D they would need to synthetize it from the Sun
Here's the complete section of the review by Peter Frost on Hoffecker's book [i]Desolate Landscapes: Ice Age Settlement in Eastern Europe[/i]: (For complete review: [url=http://human-nature.com/nibbs/02/frost.html]http://human-nature.com/nibbs/02/frost.html[/url])[/color][/b]
[size=3][b]Male Provisioning, Polygyny Constraints, and Sexual Selection[/b][/size]
[b]In any case, there was no replacement or intermixture [color=purple](between Neanderthals/Pre-Moderns and Moderns)[/color] on the loess-steppe. Modern humans were moving into an environment that no other Homo had successfully colonized. To overcome the challenges of the harsh Arctic climate, they created new forms of fuel, clothing, and shelter. To overcome the challenges of a different food supply, they reallocated the tasks of food procurement between men and women. This shift in food procurement is evident if we compare present-day hunter-gatherers from the Tropics and the Arctic. In the tropical zone, men hunt while women gather berries, fruits, roots, grubs, eggs, and other sessile food items, these tasks being more compatible with the demands of pregnancy, breast-feeding, and infant transport (Kelly 1995:268-269). Further north, food gathering is limited by the long winter, providing less than 10% of all food among hunter-gatherers above 60° N, as compared to 40-55% below 40° N (Martin 1974:16-18). The end point of this trend is Arctic tundra, where almost all of the available biomass is in the form of game animals. Such environments compel women to process food obtained through hunting instead of gathering food on their own.
Hoffecker discusses the implications (p. 8). First, “hunter-gatherers in northern continental environments who subsist on terrestrial mammals must forage across large areas in order to secure highly dispersed and mobile prey.” Second, “[a]nother consequence of low temperatures and a high meat diet is that males procure most or all food resources.”
This change in the sexual division of labour would have had demographic and, ultimately, evolutionary consequences. As hunters cover longer distances, they increase their risk of death from starvation, accidents, or inclement weather, a risk that is already high because they carry a minimum of supplies for sustenance and shelter. If we look at present-day Arctic groups with no herd dogs or domesticated reindeer (e.g., the Chukchi), male mortality rises sharply with hunting distance (Krupnik 1985). In addition, hunting is more hazardous in the Arctic because of the extreme weather and the relative absence of alternate food sources for hunting parties. There thus develops a male deficit in the sex ratio. Among 19th century Labrador Inuit, the 15+ age bracket had only 57 males for every 100 females (Scheffel 1984).
Few of the excess women, however, can become second wives. This is because of the high demands on male provisioning. In his review of Inuit mating systems, Kjellström (1973:118) concludes, "Since the duty of being a provider was more onerous for the man who had two or more wives, this meant that as a rule it was only the really able and skilful hunters and fishers who could manage this double duty." Together, these two factors-high male mortality and limited polygyny-skew the operational sex ratio towards a female surplus, thereby causing women to compete more intensely for mates. One result is an intensification of sexual selection.
Sexual selection is not unusual in itself, but it usually acts more on males than on females. All things being equal, more males than females are available for mating at any point in time, given the latter’s unavailability during pregnancy and early infant care. If, however, a population has more females than males and if polygyny is constrained, we have the conditions for a reversal of the usual pattern.
Nonetheless, this potential for sexual selection has to contend with a third factor. Female infanticide used to reduce the excess of women over men in Arctic hunter-gatherers living at high latitudes, like the Inuit of Canada’s High Arctic. The population density being very low, parents were reluctant to raise a daughter who would probably marry out of the band and not support them in their declining years, either directly or through her future husband (Smith and Smith 1994). In the Low Arctic, higher population density increased the likelihood of “son-in-law payback”; consequently, the Inuit practised much less female infanticide at lower latitudes (Schrire and Steiger 1974). If we extrapolate this trend to the loess-steppe, where high bioproductivity supported higher population densities than those of present-day Arctic groups, it is likely that female infanticide would not have “corrected” the excess of women over men.
If humans experienced intense sexual selection on the loess-steppe, were there evolutionary consequences? One may be the diverse eye and hair colours we see in a zone centred on the East Baltic and covering most of Eastern and Northern Europe. Within this zone, eyes may be not only brown but also blue, grey, hazel, or green, while hair may be not only black but also brown, flaxen, golden, or red (Beals and Hoijer 1965:212-214). In the absence of selection pressures, the current level of hair colour diversity would have taken 850,000 years to develop (Templeton 2002). One would have to conclude that the gene pool of Eastern and Northern Europe is derived mainly from the Neanderthals and even earlier Homo populations. Otherwise, some kind of selection must have favoured these colour polymorphisms.
Sexual selection has already been advanced as an explanation (Cavalli-Sforza et al. 1994:266). If faced with an abundance of equally suitable mates, the sex in short supply will select a mate that “stands out from the crowd” in having, for example, a rare and highly visible colour morph. Such selection is frequency-dependent, declining in strength as the rare morph becomes more common and tending towards an equilibrium that maximizes colour diversity (Endler 1980; Keegan-Rogers and Schultz 1988). Colour polymorphisms due to sexual selection have been reported from insects, fishes, and birds (Keegan-Rogers and Schultz 1988). There seem to be two preconditions: (1) a low level of predation, because brightly coloured traits facilitate prey detection; and (2) an absence of related species sharing the same geographic range, apparently because too much intraspecific variability interferes with species recognition (Endler 1980). Both preconditions were probably met by modern humans: predation was a minor cause of mortality and species recognition had become less problematical with the disappearance of the Neanderthals.[/b]