The story about HERC2 & OCA2 is getting a lot of press; that is, the genetics behind how people have blue eyes. But see this in ScienceNow:
There are still large questions, though. Why did blue eyes persist? Scientists say it is difficult to see how eye color would have an environmental advantage, as skin color does. Some theories suggest that women may have played a role in driving the selection. Perhaps, Kayser says, "the females thought it more exciting to have a male with blue eyes."
I already posted this before: the SNPs which are used to predict blue eyes also track skin color variation. In other words, pleiotropy. This shouldn't be a surprise, OCA2 is a pigmentation locus which in many cases doesn't exhibit tissue specific expression patterns; its name derives from the fact that some forms of albinism are associated with mutants on it. In any case, some concrete data about skin color and the OCA2 SNPs can be found in a previous paper from a research group behind one of the current publications, A Three-Single-Nucleotide Polymorphism Haplotype in Intron 1 of OCA2 Explains Most Human Eye-Color Variation. Look at table 1, and you find these data:
|Genotype||Fair skin||Medium skin||Olive skin|
Let's do something with the numbers. Give fair skin a trait value of 1, medium skin 2, and olive skin 3. Then generate an average value for each genotype by weighting appropriately, and divide by the number associated with heterozygotes. This is what I get:
Looks additive for skin color, doesn't it? Since blue eyes as a trait seems to exhibit strong recessivity HERC2/OCA2 derived variants are unlikely to have initially been selected for that phenotype. It could be something besides skin color, but that is the most plausible abduction at this point from where I stand (we know that selection was powerful on the locus).