Fisher's principle remains a plausible explanation for human sex ratio evolution. A Comment on: Zietsch et al. 2020

Zietsch et al. [1] present a study of the heritability of the human offspring sex ratio, based on data from the entire Swedish population born from 1932 onwards. Quantifying the association between individuals' offsprings' sex and their siblings' offsprings' sex, the authors find no genetic variation in the human offspring sex ratio. In colloquial terms, the authors show that female-biased or male-biased offspring sex ratios do not run in families, at least not in Sweden. The authors then draw the conclusion that their result of the negligible genetic contribution to sex ratio variation in a contemporary human population renders Fisher's principle untenable as a framework for understanding human offspring sex ratio. Fisher's principle is a theoretical argument put forward by Düsing [2] and Fisher [3] as an evolutionary explanation for equal investment in sons and daughters (see also [4,5]). Sex allocation theory has since developed in many directions [4,5], but most of its elaborations are built on the groundwork of Fisher's principle. A refutation of Fisher's principle (even if in just one species) would therefore be striking if true.

The central empirical finding of Zietsch et al. [1] is in itself fascinating, particularly given that observed variation in the majority of studied human traits does have a genetic component [6]. The subsequent rejection of Fisher's principle in humans is, however, unwarranted and based on a misunderstanding of what can be deduced based on a measurement of heritability. The authors imply that if Fisher's principle was the correct explanation for sex ratio evolution in the (possibly very distant) past, we should be able to currently observe genetic variation in the offspring sex ratio at evolutionary equilibrium. However, while genetic variation is needed for evolutionary change, genetic variation need not be observable at an evolutionary equilibrium where no further change is expected.

In the context of Fisher's principle, the heritability argument is subtle, and the principle is in fact theoretically consistent with both outcomes: high heritability at equilibrium and zero heritability at equilibrium. For example, it is compatible with a scenario where, at equilibrium, half of the mothers in the population are genetically predisposed to produce only daughters and the other half only sons [7,8 p. 81]. In such a situation, there would be substantial genetic variation in offspring sex ratio. But Fisher's principle is also compatible with e.g. XX/XY sex determination with unbiased Mendelian inheritance, where every mother produces daughters and sons with exactly equal probability at conception [7,8 p. 81]. The latter scenario requires no variation (genetic or otherwise) at equilibrium in the underlying tendency to produce daughters or sons, and this is fully consistent with Fisher's principle. As noted by Zietsch et al. [1], a further consequence of the frequency-dependent nature of Fisher's principle is that selection is strong when the population is far from equilibrium, but becomes neutral when the population as a whole is exactly at equilibrium (e.g. fig. 1 in [7]). Therefore, mutations for deviant sex ratios would not necessarily be quickly purged, and genetic variation could conceivably persist at equilibrium even in an XX/XY sex determination system. However the upshot is that overall, Fisher's principle by itself makes no prediction about heritability at equilibrium. This makes the empirical result of Zietsch et al. [1] no less important in other ways, and it is interesting to speculate on reasons for the lack of heritability. For example, a simple speculative explanation is that mutations causing deviations from fair meiosis and expected Mendelian ratios (and which have no other adverse effects) in an XX/XY sex determination system are rare. However, this question would remain even if it was indeed found that Fisher's principle did not apply to human sex ratio evolution.

In conclusion, the finding of negligible heritability in human offspring sex ratio [1] is interesting and important. However, the heritability of offspring sex ratio at equilibrium is not a falsifiable assumption of Fisher's principle as implied by Zietsch et al. [1], and their results should not be taken as evidence either for or against Fisher's principle in humans.

Data accessibility

This article has no additional data.

Competing interests

I declare I have no competing interests.

Funding

J.L. is funded by an Australian Research Council Discovery Early Career Research Award (project number DE180100526) from the Australian Government.