Recently I had some blood work done for a routine check up and it turned out that I had vitamin D deficiency. The doctor explained to me that this is common amongst darker skinned people who live at high latitudes, especially in areas where cloudiness is the norm. That would fit the bill for my own ecological surroundings for these past few years, but I never believed that I would be susceptible to vitamin D deficiency because I perceived myself as someone who got sufficient sun. To the left you see a photo of my foot. I normally wear sandals, so please note the contrast between my exposed and unexposed skin in terms of pallor. But I was obviously fooling myself, my vitamin D levels were half that of the lowest acceptable values. Needless to say I’m now taking supplements to bring my nutrition back into its natural balance. Now, I do find it interesting that of the few sources for vitamin D fish ranks high, and I do consume more fish than the typical American (who turns down salmon?). It also brought home to me the fact that biology and genetics aren’t just theoretical, they’re personal. On the flip side a friend of mine who was vacationing in the Maldives told me how he decided to go shirtless on the beach when he saw the locals swaggering about. My pale skinned friend was suitably fried pink despite his protective lotion.
I’ve talked a fair amount about the evolutionary genetics and history of skin color on this weblog. To review, it seems to be a polygenic trait whose variation world wide tends to be controlled by 4-6 loci of large effect. Not only do older quantitative genetic methods of inquiry come to this conclusion, but new data from genomics confirms this general picture. In terms of the history of the characteristic the general outline seems to be that our lineage started out light skinned, lost its fur a few million years ago and developed dark skin, and then diversified in complexion as we spread across the globe. I haven’t focused much on the details of why skin color varies because there are several competing hypotheses which some of you are likely familiar with, and I wanted to shine the spotlight on the power of convergent evolution and its relative speed. But sometimes to understand the big picture you need to focus on the details.
The chart to the left illustrates the two primary biochemical dynamics which serve as oppositional selective forces upon the color of human skin; which itself is just a reflection of melanin density and size. Though popular reports focus on the increased cancer rates of light skinned individuals, it is important to remember that cancers often occur late in life and so their reproductive impact is diminished. Nina Jablonski has put forward a more reproductively salient selective pressure: the interference with folic acid synthesis which occurs when excessive UV radiation penetrates deep into the dermis. The end result of this is reduced folate levels, which in pregnant females often causes neural tube abnormalities. Any impact on pregnancy success is an extremely powerful selective force. In this model the dark skin of humans naturally arose because women who were darker skinned carried more normal fetuses to term than those who were light skinned. In dark skinned populations the MC1R locus is extremely conserved, suggesting powerful selective constraints which prevent sequence variation. No matter the phylogenetic relationships between dark skinned populations there is a consensus sequence which seems to have been selected for deep in the human past which remains the norm across these groups (i.e., though Melanesians and Africans are both very distantly related their dark skin is the end product of the same genetic architecture). But counterbalancing the need to block sunlight due to reduction in folate levels is the fact that vitamin D synthesis requires a minimum level of radiation to be catalyzed. Reduced vitamin D levels not only result in bone deformations (i.e., rickets), but a heightened sensitivity to a host of diseases.
The map to the left is a rough sketch of variation in skin color throughout the world. It shouldn’t surprise anyone, there is the general trend of populations becoming lighter as one moves away from the tropics. In Jablonski’s model this is comprehensible insofar as one moves north the the UV radiation drops. This means that melanin pigement concentrations need not be as dense to protect the folic acid synthesis pathway from degradation. And, the reduced density of melanin maybe necessary for vitamin D levels to be maintained due to the essential role of radiation in catalyzing its production. In short, as one moves away from the tropics the selective constraint to maintain high folate levels by blocking UV via dark skin is removed, while a countervailing pressure to reduce the density of melanin so as to allow enough radiation to stimulate vitamin D synthesis emerges. But there’s a problem with this narrative: the lightest populations in the world are not from the farthest north. In Scandinavia the Sami are darker than the peoples to the south, and the Inuit are generally a light brown skinned people. It seems that the fairest peoples in the world reside around the shores of the Baltic sea, not in the circumpolar regions.
The fact that I eat a relatively rich diet and do expose myself to the sun when I can, but still exhibit vitamin D deficiency, brought home to me the dependence upon nutritional parameters as well as skin color. In terms of nutrition the last 10,000 years has not, on average, been “good” for humanity. Though agricultural peoples are efficient at reproduction and natural increase because of the high yields they can extract from the land via intensive farming, they have generally had to deal with the trade off that their diet became reliant on starches which were poor in many vital nutrients as well as proteins. Human have become smaller over the last 10,000 years, and the contrast between the relatively healthy skeletons of hunter-gatherers and the physiological stress exhibited by the farmers which succeeded them have long been noted. Recently work in genomics also suggests genes implicated in various metabolic functions have been under powerful selection over the past 10,000 years as the agricultural lifestyle has spread. The deleterious consequences of switching many non-agricultural populations to the starch rich diet are well known (obesity, diabetes, etc.). Selection happens, and it seems likely that a genetic revolution was ushered in by the radically altered nutritional universe of the farmer. Which brings me to Europeans and why they might be so light. Frank W. Sweet published an essay in 2002 which offered that the feasibility of a farming lifestyle at very high latitudes in Europe due to peculiar climatological conditions served to drive Europeans to develop light skins over the last 10,000 years. In short, Sweet argues that the diets of pre-farming peoples were richer in meats and fish which provided sufficient Vitamin D so that skin color was likely light brown as opposed to pink. But with the spread of agriculture Vitamin D disappeared from the diets of northern European peoples and so only by reducing their melanin levels could they produce sufficient amounts of this nutrient to keep at bay the deleterious consequences of deficiencies. This explains why the Sami, who never adopted agriculture, remained darker. One could hypothesize that the relative swarthiness of groups like the Welsh might be due to greater reliance on fish and game as opposed to agriculture, but the point is not to explain every last detail but to clarify the overall trend.
Sweet’s essay was written in 2002. In 2005 a gene, SLC24A5, was implicated in explaining a large proportion (25-38%) of the between population difference in skin color for Europeans and Africans. It seems that on this locus the two populations were disjoint, they exhibited no substantial overlap. In European it seems that 6 to 10 thousand years ago a new variant arose which subsequently swept to fixation. In the model above it seems likely that the mutation was just there at “the right place and right time.” Interestingly in East Asians SLC24A5 exhibits the same sequence as it does in Africans. But, it seems that other loci are responsible for the lightening of the skin of East Asians recently as well, though not to the same extent as Europeans. The reason for this is likely the fact that temperate East Asia as at a far lower latitude than Europe.
Of course, there are other anomalies. Sweet points out that South Americans are far paler than they should be if Old World populations are to be any judge. His explanation is simple: light skin evolves quickly via loss of function mutations while the original settlers of the Americas did not carry the fully complement of alleles necessary for black skin. Though Sweet doesn’t say it in so many words he is basically suggesting that gain of function mutations are rare relative to loss of function. There are two parameters which might explain the relative lightness or darkness (and overall homogeneity) of New World populations. First, settlement was relatively recent, so there hasn’t been enough time for evolution to work. Of course the copious evidence of recent human evolution suggests that this isn’t really a major issue. Rather, the bigger problem was likely that extant genetic variation was reduced by a population bottleneck. Over time this variation would be replenished by mutation, but this is a slow process. Tropical populations of the New World are not only lighter than one would expect, they also display a far stockier and “northern” build than they should if form followed from function. This suggests evolutionary lag on many characteristics due to the low genetic diversity combined with the particular time frame over which evolution could occur (but even here, do note that there is a skin color gradation from the tropics to the temperate zones in the New World among indigenous peoples).
I don’t expect this model to explain all the details. Some theorists have offered other hypotheses. For example, Peter Frost puts the focus on sexual selection. Peter’s thesis is basically that particular ecological conditions in Ice Age Europe, the low-latitude continental tundra, resulted in specific social patterns which forced females to compete for the attention of males (there was a shortage of males, but the constraints of the ecosystem limited mating patterns to monogamy). All things equal I tend to favor more ecological resolutions to questions of phenotypic evolution than sexual selective ones because the evolutionary dynamics of the latter seem to be so chaotic and difficult to characterize. Judith Rich Harris has a different spin insofar as she believes that mothers selected lighter skinned infants because they found them beautiful. I am generally skeptical of this explanation simply because beauty does not emerge from a vacuum. Of course these factors are not eliminated by the selective forces sketched out above, and it seems likely that evolutionary forces are multivalent (Frost in particular focuses upon hair color as opposed to skin color for his frequency dependent sexual selection model).
Overall, this a good time to be interested in questions about normal human phenotypic variation. The tools are manifold and triangulating to the most plausible explanation is far easier than in the past simply because various methods are on hand. Me, I’m taking my Vitamin D pills, and perhaps I’ll get me some cod oil.