Clitoral variation says nothing about female orgasm.

In a recent article, Wallen and Lloyd (2008) conclude that clitoral structures are not under direct selection and state that the data on clitoral length variation instead support suggestions that female orgasm in humans is not adaptive; by adaptive they seem to mean subject to selection rather than only subject to natural selection. Their conclusions are reached because (1) in comparison to males, females orgasm less frequently during sex (the variation in female orgasm rates [or ease of orgasm] is very high) and (2) levels of phenotypic variation reveal the strength of selection on a character, and clitoral variation is large relative to penis or vaginal variation (as measured by the coefficients of variation [CV] in these traits). Point 1 is a statement based on previous work, whereas for point 2 the authors present and compare the CV for clitoral, vaginal, and penis length. The authors suggest that these two points, and particularly the second, support the notion that female orgasm is probably a pleiotropic effect of selection on male orgasm; there is strong selection on male orgasm and genetic correlations between the sexes mean that females orgasm as a by-product of selection on males. Here I argue that the authors’ conclusions are equivocal and possibly incorrect, based largely on misunderstandings of some fundamental evolutionary biology. Let us first consider the sexual dimorphism in ease or rate of orgasm. Although the authors do not present data directly related to this, they do regard the differences between the sexes as indicative of nonadaptive orgasm in females. Why does ease of orgasm say anything about selection or adaptive values? And more importantly, one specific sexual selection mechanism actually predicts that females should, unlike males, show enormous variability in orgasm frequency. Postcopulatory (cryptic) female choice is the postintromission version of classical Darwinian female choice (Eberhard 1996). Like classical choice, females are not expected to respond equally to all males; if they did, there would be no selection via female choice. Instead, females set hurdles for males to overcome. Males more able to meet female criteria have greater reproductive success, and although females may respond favorably to males who better stimulate them before copulation, they may also do so via postcopulatory (postintromission) routes. Cryptic female choice is potentially widespread (Eberhard 1996) and requires that females are not strictly genetically monogamous. This is certainly the case in humans, where paternity analysis suggests that a considerable number of offspring are sired by nonsocial mates (Simmons et al. 2004). Also, testis size variation across the great apes is consistent with selection via polyandry (Harcourt et al. 1981). Testis size reflects sperm competition risk (loosely, the likelihood that a female will mate with more than one male per reproductive event) (see, e.g., Hosken 1997) and humans have relative testis size between that of highly promiscuous chimpanzees and monogamous gorillas. Female orgasm could potentially then be part of an axis related to cryptic female choice of males. This would be the case if, for example, orgasm facilitated sperm transfer into the uterus, and if it facilitated ovulation, fertilization, or implantation. The same could be true if orgasm led to further copulations with the partner whose stimulations caused orgasm. Basically, if the reproductive success of males able to stimulate females to orgasm were higher, then female orgasm need not be merely a pleiotropic effect of male orgasm. Furthermore, if males better able to stimulate females were of higher genetic quality (good genes), then by exercising choice females would produce higher quality offspring, and so choice (orgasm) would be adaptive (i.e., it would be favored by natural selection and not just sexual selection). To reiterate, orgasm to be involved in cryptic female choice requires females to respond differentially to different males; hence, female orgasm should not be as frequent as male orgasm. This is not to say that female orgasm is related to any of the above, but merely to point out that the relative ease/rate/frequency of orgasm say nothing about selection on orgasm, and if female choice (based on orgasm) was, for example, for good genes, then orgasm could be adaptive. Quod erat demonstrandum. To the issue more directly related to the data the authors present. They argue as follows: If selection on a trait is strong, then there should be no genetic variation for the trait. If there is no genetic variation for the trait, then there should be no phenotypic variation in the trait. Hence, if a trait has high phenotypic variation, it is not under selection (or under weak selection). This line of reasoning is incorrect at a number of levels. The first part of the argument is based on Fisher’s (1930) fundamental theorem, which states ‘‘the rate of increase in fitness of any organism at any time is equal to its EVOLUTION & DEVELOPMENT 10:4, 393 –395 (2008)