Saturday, 29 September 2007

Sexual selection and Arctic environments

My 2006 paper is often criticized on one point. If Arctic environments did intensify sexual selection of women among ancestral modern humans, and if this selection did create inter alia the color traits of Europeans (diversification of eye and hair color, extreme depigmentation of skin color), then why are these traits absent among the native peoples of northern Asia and North America? Surely they too are products of Arctic environments.

Yes, they are. But it was not Arctic environments per se that intensified sexual selection of women. It was essentially two changes to the sexual division of labor that, among hunter-gatherers, generally correlate with distance from the equator. First, hunting distance increases with decreasing numbers of game animals per square kilometer, thereby increasing male mortality. Second, food gathering decreases with longer winters, thereby increasing women’s reliance on men for provisioning and increasing the costs of polygyny for men. It is this combination of higher male mortality and limited polygyny that intensifies sexual selection of women.

These points are made by Hoffecker (2002, pp. 7-8):

Hunter-gatherer diet is strongly influenced by latitude and temperature. To begin with, energy demands increase significantly in cold climates and caloric intake in arctic environments may be as much as 30 percent higher than it is in tropical regions. The percentage of meat and fish in the diet of recent hunter-gatherers increases as temperature, moisture, and primary productivity decline, and equals or exceeds 80 percent among most peoples who live in areas with an effective temperature of 10 degrees C or less. …

The high protein-fat diet and hunting and fishing subsistence of hunter-gatherers in northern environments has major implications for foraging strategy. Although cold maritime settings often provide rich concentrations of aquatic resources that require limited mobility, 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. Among peoples who rely primarily on nonaquatic foods, there is a correlation between temperature and the average distance of residential moves and a related correlation between the percentage of hunted food in the diet and territory size. Another consequence of low temperatures and a high meat diet is that males procure most or all food resources, generating a more pronounced sexual division of labor.

Hunting distance and male food provisioning of women seem to be at a maximum in a special kind of Arctic environment: ‘continental’ steppe-tundra, where almost all food is in the form of highly dispersed and mobile game animals. Today, this environment is confined to the northern fringes of continental Eurasia and North America. During the last ice age, it lay further south and covered more territory. This was especially so in Europe. The Scandinavian icecap had pushed the steppe-tundra zone far to the south and on to the broad plains stretching from southwestern France through northern Germany and into eastern Europe. This combination of treeless tundra and temperate latitudes created an environment quite unlike the northern barrens we know today. As Jochim (1983, p. 214) notes: “The low-latitude tundras and park-tundras of glacial Europe were richer than any modern northern counterparts.” Long intense sunlight favored a lush growth of mosses, lichens, and low shrubs that fed herds of large herbivores, mainly wild reindeer (a.k.a. caribou) but also mammoth, woolly rhinoceros, horse, bison, red deer, roe deer aurochs, ibex, chamois, saiga antelope, muskox, giant deer, wild ass, elk, and wild boar (Butzer, 1964, p. 138).

Though substantial, this kind of biomass is a volatile food source. Caribou herds in Alaska fluctuate considerably in any one area, in part because they cover long distances in the space of one year but also because they go through long-term demographic cycles of expansion and contraction (Burch, 1972). Among caribou-dependent Inuit, “at least 1 period of hunger or starvation is part of the normal annual cycle” (Burch, 1972, p. 350). In good times, caribou herds do provide a bountiful food source, but at the price of continual camp moves and extensive reconnoitering on foot. This is the real man-killer in Arctic groups that have not yet domesticated reindeer, as Krupnik (1985, p. 126) notes when explaining the Chukchi’s low ratio of men to women:

The herdsmen guarded the herd on foot. There were no herd dogs, and reindeer were not used for transport during the summer months, so that the men had to travel with the herds over the tundra with a minimum of portable possessions. All of this must have sharply intensified the physical burdens on adult, able-bodied men, and caused a higher mortality rate and consequently a proportional decrease of their numbers in the population.

Thus, on the steppe-tundra of the last ice age, the population of human hunters was probably as volatile as its resource base, all the more so if one also considers the climatic oscillations during this period. Moreover, because this population was dispersed over a wide area, its density was not necessarily high enough for long-term viability. Hoffecker (2002, pp. 8-10) writes:

The high mobility requirements of northern continental environments not only incur added time and energy costs, but also carry potential social and reproductive costs for dispersed populations in such environments. Populations must maintain a minimum threshold density in order to remain viable and avoid extinction, and it is estimated that the “minimum equilibrium size” for a mating network of modern hunter-gatherers is between 175 and 475 individuals. The degree of dispersal of these individuals across the landscape (typically grouped into bands containing roughly 25 individuals) cannot exceed their ability to sustain a social network through at least periodic contact and aggregation.

All of these factors hindered sustained human occupation. When temperatures fell during the last glacial maximum (19,000-18,000 BP), depopulation seems to have occurred throughout the steppe-tundra zone, but more so in some areas than in others. Least affected were the warmer and moister areas in Western Europe and the Carpathian basin, where continuous occupation is well attested throughout the glacial maximum (Hoffecker 2002, p. 194). Most affected were the colder and drier areas in northern Asia, where the steppe-tundra zone lay close to the Arctic Circle and far from the Atlantic. Goebel (1999) writes:

During the last glacial maximum (19 to 18 kya), Siberia was devoid of human populations, except perhaps in small refuges like the southern Yenisei or Transbaikal region. … The central Asian steppe also lacks archaeological sites spanning the last glacial maximum, suggesting that increased aridity, lower temperatures, and a lack of woody plants severely limited human settlement in this region as well.

Bioproductivity was clearly lower in northern Asia. This factor, on top of other factors affecting the entire steppe-tundra zone (volatile resource base, critically low density of human population), made any human presence vulnerable to extinction during periods of environmental stress. There likely were repeated cycles of colonization, extinction, and recolonization.

In conclusion, substantial and continuous human settlement seems to have been confined to the European end of the steppe-tundra zone. Only there did all conditions fall into place for sustained sexual selection of women.


Burch, Jr., E.S. (1972). The caribou/wild reindeer as a human resource. American Antiquity, 37, 339-368.

Butzer, K.W. (1964). Environment and Archaeology. Chicago: Aldine.

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

Goebel, T. (1999). Pleistocene human colonization of Siberia and peopling of the Americas: An ecological approach. Evolutionary Anthropology, 8, 208‑227.

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

Jochim, M.A. (1983). Palaeolithic cave art in ecological perspective. In Hunter-Gatherer Economy in Prehistory. A European Perspective. G. Bailey (Ed.). Cambridge: Cambridge University Press.

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


Sampiero said...

Great post Peter. I have nevertheless a couple of questions:
-maybe light skin, hair and eyes among Europeans could be related to the fact that Europe is the northernmost inhabited continent? After all, there is a skin and hair gradient in human populations. This gradient is very well correlated with UV exposure gradient on the globe.
"UV-friendly" white skin, hair and eyes would thus be favorably selected in the light-depleted continent...This would explain the extremely light skin, hair and eyes of people from the British Isles-Scandinavia-Baltic countries belt, that is the foggiest region in the world ? Light is so limiting in these places for vit-D synthesis that people use to have "luminotherapy".

-East Asian people look better adapted than Europeans to tundra and cold climate living. They are in general stockier than Europeans. They have rounder bodies, shorter limbs and more skin fat. They tend to be more K selected than Europeans (according to Rushton) which is useful if you do not want to "waste" your energy in raising extra- children you cannot feed.

PS: sorry for my english. French is my first language.

Peter Frost said...

Vous vous exprimez très bien en anglais. You essentially argue that Europe receives less solar radiation than does northern Asia and North America because European skies are more overcast. There was thus less selection for melanized skin among ancestral Europeans (to protect against skin cancer and sunburn) than among ancestral populations at similar latitudes in northern Asia and North America.

Actually, at any given latitude, solar UV radiation is just as intense at ground level in Europe as it is in northern Asia and North America. (Jablonski & Chaplin, 2000; see also charts on: At these latitudes, UV radiation is already weak. Only on the coastal fringe of northwestern Europe is the sky sufficiently overcast to have some effect on UV radiation.

You also argue that the genes for hair and eye color have some effect on skin color. Therefore, relaxation of selection for dark skin should allow defective alleles to proliferate at pigmentation loci, including those for hair color and eye color. Two papers, however, have shown that such a scenario would have needed close to a million years to produce the hair-color and eye-color variability that Europeans now display, with the redhead alleles alone being c. 80,000 years old (Harding et al., 2000; Templeton, 2002). Yet modern humans have been in Europe for only 35,000 years or so.

Instead of relaxed selection for dark skin, perhaps there was increased selection for light skin, notably to boost synthesis of vitamin D. This hypothesis solves the time problem but does not explain the increase in the number of MC1R and OCA2 alleles. Natural selection would have simply favored one allele at the expense of all others, i.e., whichever one optimally reduced skin pigmentation.

There are other problems with either hypothesis, or with any that attribute these color traits to weaker solar UV:

1) If we examine the many homozygous and heterozygous combinations of MC1R or OCA2 alleles, most have little visible effect on skin pigmentation, except for the ones that produce red hair or blue eyes (Duffy et al., 2004; Sturm & Frudakis, 2004).

2) If we consider the estimated time of origin of these color traits, at least two of them seem to have appeared long after modern humans had entered Europe's northern latitudes about 35,000 years ago. The whitening of European skin, through allelic changes at AIM1, is dated to about 11,000 years ago (Soejima et al., 2005). No less recent are allelic changes at other skin color loci and at the eye color gene OCA2 (Voight et al., 2006). Did natural selection wait over 20,000 years before acting?


Duffy, D.L., Box, N.F., Chen, W., Palmer, J.S., Montgomery, G.W., James, M.R., Hayward, N.K., Martin, N.G., & 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., & Rees, J.L. (2000). Evidence for variable selective pressures at MC1R. American Journal of Human Genetics, 66, 1351?1361.

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

Soejima, M., Tachida, H., Ishida, T., Sano, A., & Koda, Y. (2005). Evidence for recent positive selection at the human AIM1 locus in a European population. Molecular Biology and Evolution, 23, 179-188.

Sturm, R.A., & Frudakis, T.N. (2004). Eye colour: portals into pigmentation genes and ancestry. Trends in Genetics, 20, 327-332.

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

Voight, B.F., Kudaravalli, S, Wen, X, Pritchard, J.K. (2006). A map of recent positive selection in the human genome. PLoS Biology, 4(3), e72 doi:10.1371/journal.pbio.0040072

Sampiero said...

Merci beaucoup pour votre réponse!

Tod said...

The component of UV that produces vitamin D is UVB. Research is currently underway to establish how much UVB gets through Britains clouds. UVB has a lot more trouble getting though than the other UV(A). Even on a clear day it is only strong enough for vitamin D synthysis around 10am -2pm. A lot of good things are being found out about vitamin D hence the research.

Light hair would have to occur, if only with extreme rarity ,(as it does among austalian aborigines Baker 74) for it to be taken up intensified and made prevailent?

The genetic basis of skin pigmentation is a bit different in Asians compared to Europeans, or so I read somewere. A blonde Asian could be vanishingly rare?

Anyway, maybe light hair never put in an appearance with the relevant North Asians/Americans,so it couldn't be "taken up".