Friday, June 3, 2011

On the impossibility of blue eyes

Although blue eyes are more recessive than brown eyes, eye color does not follow a simple recessive/dominant mode of inheritance. There is a wide range of intermediate hues.

As discussed in my last post, one puzzle of human evolution is the diverse palette of European hair and eye colors. Although these two polymorphisms have largely developed at separate genes, they share a similar geographic range and similar conspicuous hues. They also appear on or near the face—the focus of human visual attention. Could a common selection pressure be responsible? And could it be sexual selection?

This topic came up a month ago on Steve Sailer’s blog, specifically the evolution of blue eyes. Greg Cochran pointed out that sexual selection could not be responsible because blue eyes are recessive:



First, an advantageous allele whose action is purely recessive is far more likely to be lost when new than a dominant allele with an equivalent advantage. Second, assuming that it is not lost and that the population mates randomly, it takes much longer to reach 50% frequency than a dominant allele. Third, if the population is spread out over space, the Fisher wave spreads far more slowly, something like 20 times more slowly.


Mutations are fairly common, but a potentially adaptive one—like an allele for blue eyes—is usually rare. In this case, the same rare allele must occur twice and come together in the same person before sexual selection can do its work. And this work would be lost in the next generation.

All of this assumes, of course, that blue eyes are recessive. Although eye color is polygenic, alleles at two STPs (rs12913832 and rs1129038) seem to account for most cases of blue eyes (Eiberg et al., 2008). In a Polish sample, 89% of the blue-eyed individuals had both copies of the ‘C’ allele at rs12913832 and no copies of the alternate ‘T’ allele (Branicki et al., 2009).

But the C allele is far from silent if only one copy is present, as seen in the same Polish sample. Among CT heterozygotes, 16% had blue or grey eyes, 10% green eyes, 47% hazel eyes, and only 27% brown eyes.

Although the C allele is relatively recessive for expression of blue eyes, it shows strong heterozygote effects for expression of green or hazel eyes:


Blue or grey-eyed individuals: 89% had both copies, 10% one copy, 9% no copies
Green-eyed individuals: 67% had both copies, 30% one copy, 2% no copies
Hazel-eyed individuals: 9% had both copies, 80% one copy, 11% no copies
Brown-eyed individuals: 0% had both copies, 84% one copy, 16% no copies

In short, the C allele is less dominant, but not truly recessive. Even in the heterozygous state, it usually produces hues that visibly diverge from the human norm of brown eyes.

Greg also forgets that evolution can reach an initially inaccessible state by passing through intermediate states. When the C allele first appeared, it produced only green or hazel eyes for sexual selection to act upon. As copies of this allele increased in the population, there was a corresponding increase in the probability of homozygotes that could produce blue eyes—which became a new target for sexual selection.

The selection here is not for a single color, be it blue, green, hazel, or brown, but rather for any colors that can catch attention by their novelty or brightness. The end result is more and more eye colors—a balanced polymorphism where sexual selection is always on the lookout for new and interesting hues. Needless to say, this outcome is possible only when the operational sex ratio is very lopsided, thus favoring the evolution of ‘eye candy’ among members of the sex in excess supply.

References

Branicki, W., U. Brudnik, and A. Wojas-Pelc. (2009). Interactions between HERC2, OCA2 and MC1R may influence human pigmentation phenotype, Annals of Human Genetics, 73,160–170.

Eiberg, H., J. Troelsen, M. Nielsen, A. Mikkelsen, J. Mengel-From, K.W. Kjaer, & L. Hansen. (2008). Blue eye color in humans may be caused by a perfectly associated founder mutation in a regulatory element located within the HERC2 gene inhibiting OCA2 expression, Human Genetics, 123, 177–187

Sailer, S. (2011). Old Blue Eyes, May 10
http://isteve.blogspot.com2011/05/old-blue-eyes.html

16 comments:

Ben10 said...

'Needless to say, this outcome is possible only when the operational sex ratio is very lopsided'

And this is explained by the very hard living conditions of the male hunters in the long distance traveled, OK. But this means that when these alleles combine to form a blue/pedomorphic phenotype associated with weaker male features, the carrier is now at decreased fitness. A pedomorphic blue-eyed kid (hunters bring their boys with them very early) should have a reduced survivalbility if he is not as tough as his brown eyed buddies.

You still havn't answer this question Peter and it is almost irrelevant to know whether the hunter choose pretty alleles in his wife for her beauty, or pretty alleles for the kids that she is going to give him, since the kids will have the alleles, want it or not.

To me, sexual selection occured as you describe, but in addition, the carrier of the selected alleles must have had a strong selective adavantage in the hard hunting conditions and despite the pedomorphic features. So that at the end of the hunting game, in 3 or 4 generations, all the hunters have these alleles.

Tod said...

Ben 10, I think there is something in what you say. If we consider the information about blue eyed men having reduced dominance and possibly masculinity then yes, under certain circumstances blue eyes would be selected against.

However I believe that would not apply to the steppe tundra environment when blue eyes originated. The steppe tundra inhabitant physiques did become much weaker than their massively built CroMagnon ancestors.

But remember it was long reconnaissance trips on foot to locate game that killed off men from exhaustion and starvation. In those circumstances being huge and muscular would be a serious liability (endurance athletes for example). If stamina was at a premium the more gracile physique of the blue eyed men could have paid off handsomely.

That may explain why blue eyes are so common; in the right environment there are positive advantages to being less 'tough'. The kind of selection Gregory Clark talks about could have easily resulted in selection for blue eyed men.

Max said...

This may be a stupid question but I'll post it anyway: why didn't black African women select for blue and green eyed males? Do the alleles responsible for eye color also affect skin coloring, so that their selection for darker skin won out? Do women have a different reaction to non-brown colored eyes in mates?

Beyond Anon said...

Mark said:

This may be a stupid question but I'll post it anyway: why didn't black African women select for blue and green eyed males?


Since sub-Saharan African women could provision their own children they are under different selection pressure. They are certainly under less pressure to attract mates who will stick around and help provision their children.

You also have it ass-backward. It is the males in Peter's model who are doing the selection, and thus different eye-color serves to attract attention to them from males who will stick around and provision their children.

However, I am waiting for Greg Cochran's comments.

Max said...

Beyond Anon,

Maybe I wasn't clear. My question is, why didn't black African women, who had the privilege of selecting their mates, choose men with non-brown eyes like European men, who had the privilege of selecting their mates, chose women with non-brown eyes? Do men and women look for different things in their mates, or did other factors in the African environment preclude the spread of the pertinent alleles?

Tod said...

I could understand the skepticism about sexual selection for something like dry earwax, but blue eyes ? Google blue contact lenses, the market is 99% female.

Peter Frost said...

Ben10,

First, I have doubts about the study I reviewed in my last post. It looks like blue eyes interfere with the masculinization of facial structure. But I would like to see replication in another human population (with no possibility of ethnic confounding).

Second, even if the above finding is true, it doesn't follow that blue-eyed males are a bunch of sissies. Presumably, natural selection would intervene to minimize any deleterious effects.

I read somewhere that average life expectancy was in the mid-30s during the Paleolithic. If true, there would have been strong natural selection to limit feminization to superficial features. Blue-eyed males aren't necessarily Pillsbury doughboys.

Mark,

There is a qualitative difference between female-female competition for mates and male-male competition for mates. The latter tends to involve males fighting for access to females. The selection pressure thus favors physical robustness, i.e., denser bones and increased muscle mass, as well as certain personality traits (verbal bombast, "presence", ability to intimidate, etc.)

Anonymous said...

RE:Blue Eyes,

Isn't there a relationship between marksmanship and blue eyes? I seem to recall studies that show that elite snipers/marksmen tend to be disproportionately blue-eyed.

Ben10 said...

"Blue-eyed males aren't necessarily Pillsbury doughboys"

Of course not, but pedomorphism suggests delayed development, or decreased masculinization, or more delicate features, etc., after all, that's the definition of a pedomorph.
My point was actually that the pedomorphic hunters beneficiated from some, as yet unknown selective advantage.

Too bad we can't say yet the skin, hair and eyes phenotypes of those paleolithic people. Until then, we will never know for sure.

DK said...

Why can't the gamut of colors be explained simply by relaxation of selection? It's not like everything in life has to be "good for something".

Peter Frost said...

DK,

It couldn't be relaxation of selection because the time depth is too shallow. Modern humans entered northern Europe about 30,000 years ago. The time depth for relaxation of selection to produce current MC1R diversity is close to a million years.

Ben10 said...

I know that lamarckism has been disproved as an effective mechanism for Evolution (because, so far, no mechanism has been found that could explain a transfer of information from the individual to its gametes), but lets say that in its explanetory form, Lamarckism can still be usefull.
In our case, Lamarck would say that women AND men of postglacial europe were trying to get fairer with more colored hues, maybe initially by dyeing their hair or painting their skin, using earth pigment, and by doing so, their descendant developped and amplified a hardwired, genetically encoded 'organ and/or function' (since 'the function creates the organ') of chromatic skin, hair and eyes variation.

We already know that Cro magnons used white, ochre, yellow and black pigments in their cave paintings. What are the chances they never used them on their body? If they did, could that make a difference in your theory, Peter?

Katydid said...

Could blue-eyed men have been selected for alcohol tolerance?

http://www.unt.edu/rss/class/mike/5700/eyecoloralcohol.pdf

Hollywood Viking movies would suggest that this is possible :-)

Katydid said...

ugh, the link got munged. Try:

http://www.unt.edu/rss/class/mike/5700
/eyecoloralcohol.pdf

Anonymous said...

"A pedomorphic blue-eyed kid (hunters bring their boys with them very early) should have a reduced survivalbility if he is not as tough as his brown eyed buddies."

1) Milk. Classical writers mention the northern barbarians were bigger and taller. If so the lactose tolerance and subsequent size could have compensated for softer features.

2) Co-operativeness. Some of the traits that are most beneficial at individual or small-scale violence are *handicaps* at larger-scale violence. Machismo is a benefit for warriors but a disbenefit for soldiers.

3) Tattoos. If male facial features softened as a result of this process artificial cultural masculinizations can be developed.

Anonymous said...

All these ideas about sexual selection directed towards alternate sexes as having a lot to do with racial differences in this blog are really interesting... just a couple of questions:
a)Is not the yellow race too somewhat effeminized, though more so in anatomy and physiology, and in these even more so than the white race? why this difference, with white effeminization directed towards colors, while asian towards getting smaller, a more fragile built, less facial hair, and lower T levels? also, why full-blood amerindian males have no facial hair, while at the same time being more robust (and darker?) than asians proper? why these different modes of effeminization?
b) Are hoe-farming papuans darker and otherwise more masculine (and more polygamous) than native australians? and why are the later so dark anyways?