Inheritance of eye color doesn’t follow a simple Mendelian model. Although the blue-eye allele (C) is less dominant than the brown-eye allele (T), CT heterozygotes aren’t necessarily brown-eyed and CC homozygotes aren’t necessarily blue-eyed. Even TT homozygotes are sometimes blue-eyed. There is also a sex difference, with women having a more diverse palette of eye colors. (source)
Most humans have black hair and brown eyes. Europeans display a much more diverse range of hues, their hair being also brown, flaxen, golden, or red and their eyes being also blue, gray, hazel, or green.
This diversification has gone farther in European women than in European men. According to a twin study, women have a more diverse palette of hair colors, with a greater prevalence of lighter shades, particularly red hair (Shekar et al., 2008).
Women also have a more diverse palette of eye colors, according to a recent study of six SNPs associated with eye color. When both copies of the so-called blue-eye allele are present, the resulting phenotype is more variable in women than in men (Martinez-Cadenas et al., 2013). This translates into a greater range of female eye colors in regions, like northern and eastern Europe, where blue eyes are the single most common phenotype (Walsh et al.,2012). As the study’s authors observe, “in populations with very high blue-eye frequency, such as Iceland or Holland, females show greater proportion of green eyes at the expense of blue eyes” (Martinez-Cadenas et al., 2013). The authors also confirmed earlier findings that eye color doesn’t follow a simple Mendelian model. Although the blue-eye allele (C) is less dominant than the brown-eye allele (T), CT heterozygotes aren’t necessarily brown-eyed and CC homozygotes aren’t necessarily blue-eyed. Even TT homozygotes are sometimes blue-eyed.
Thus, both hair color and eye color tend to be more diverse in women. There is, however, one difference. With hair color, the “derived” alleles are more fully expressed in women than in men. With eye color, they are less fully expressed. This seems to argue against the theory that hair and eye color diversified through a process of sexual selection that acted more strongly on women than on men. Since blue eyes are the derived phenotype, they should be more fully expressed in women because the female sex is the main target of this selection pressure. Yet the reverse is actually true.
The reason may be physiological. It seems easier to produce new eye colors by modifying the way the blue-eye allele is expressed than by simply creating new alleles. Thus, to produce a range of hues that extends beyond brown and blue, the so-called blue-eye genotype must be more common than the actual incidence of blue eyes. As a result, “more females bearing the ‘blue eye genotype’ (HERC2/OCA2 CC homozygous genotype) end up having green or intermediate eyes” (Martinez-Cadenas et al., 2013).
Other recent studies
A recent paper has confirmed that European eye color diversified through some kind of selection pressure, and not random factors like genetic drift or founder effects. Blue-eye alleles show a very strong signal of selection (Donnelly et al., 2012). Another study, however, has failed to find any preference for blue eyes over other colors, an indication that all eye colors are at selective equilibrium, at least for the German population under study. This finding may be related to the already high frequency of blue eyes in that population:
Perhaps the frequency of eye colors plays a role. In most countries, blue eyes are less prevalent than other eye colors and may have the image of something special and more valuable. If this assumption is true, brown eyes should be preferred in countries where the majority of the population has blue eyes. (Gründl et al, 2012).
In this case, sexual selection is frequency-dependent, shifting to whichever eye color is least frequent. Eventually, an equilibrium is reached where color novelty is in balance with other characteristics, such as color brightness, that may increase sexual attractiveness.
This last finding shows the opportunistic nature of sexual selection. When too many of one sex have to compete for mating opportunities with too few of the other sex, there will be selection for any traits that increase mating success. In many cases, these traits will hyperstimulate a mental algorithm that is used for sex recognition. In other cases, hyperstimulation will simply involve use of bright or novel colors that can better engage visual attention and remain longer in memory.
Donnelly, M.P., P. Paschou, E. Grigorenko, D. Gurwitz, C. Barta, R-B. Lu, O.V. Zhukova, J.-J. Kim, M. Siniscalco, M. New, H. Li, S.L.B. Kajuna, V.G. Manolopoulos,
W.C. Speed, A.J. Pakstis, J.R. Kidd, and K.K. Kidd. (2012). A global view of the OCA2-HERC2 region and pigmentation, Human Genetics, 131, 683–696.http://europepmc.org/articles/PMC3325407
Gründl, M., S. Knoll, M. Eisenmann-Klein, and L. Prantl. (2012). The blue-eyes stereotype: Do eye color, pupil diameter, and scleral color affect attractiveness? Aesthetic Plastic Surgery, 36, 234–240.http://link.springer.com/article/10.1007/s00266-011-9793-x#page-1
Martinez-Cadenas, C., M. Pena-Chilet, M. Ibarrola-Villava, and G. Ribas. (2013). Gender is a major factor explaining discrepancies in eye colour prediction based on HERC2/OCA2 genotype and the IrisPlex model, Forensic Science International: Genetics, 7, 453–460.http://europepmc.org/abstract/MED/23601698
Shekar, S.N., D.L. Duffy, T. Frudakis, G.W. Montgomery, M.R. James, R.A. Sturm, & N.G. Martin. (2008). Spectrophotometric methods for quantifying pigmentation in human hair—Influence of MC1R genotype and environment, Photochemistry and Photobiology, 84, 719–726.https://genepi.qimr.edu.au/contents/p/staff/shekaretal2008b1.pdf
Walsh, S., A. Wollstein, F. Liu, U. Chakravarthy, M. Rahu, J.H. Seland, G. Soubrane, L. Tomazzoli, F. Topouzis, J.R. Vingerling, J. Vioque, A.E. Fletcher, K.N. Ballantyne, and M. Kayser. (2012). DNA-based eye colour prediction across Europe with the IrisPlex system, Forensic Science International: Genetics, 6, 330–340.http://www.fsigenetics.com/article/S1872-4973(11)00144-X/abstract