Substantial evidence now exists that variables measuring or correlated with human fertility outcomes have a heritable component. In this study, we define a series of age-sequenced fertility variables, and fit multivariate models to account for underlying shared genetic and environmental sources of variance. We make predictions based on a theory developed by Udry [(1996) Biosocial models of low-fertility societies. In: Casterline, JB, Lee RD, Foote KA (eds) Fertility in the United States: new patterns, new theories. The Population Council, New York] suggesting that biological/genetic motivations can be more easily realized and measured in settings in which fertility choices are available. Udry's theory, along with principles from molecular genetics and certain tenets of life history theory, allow us to make specific predictions about biometrical patterns across age. Consistent with predictions, our results suggest that there are different sources of genetic influence on fertility variance at early compared to later ages, but that there is only one source of shared environmental influence that occurs at early ages. These patterns are suggestive of the types of gene-gene and geneenvironment interactions for which we must account to better understand individual differences in fertility outcomes.
KeywordsFertility; Fisher's theorem; FTNS; Heritability; Shared Environment; Multivariate models; Phenotypic plasticity; Hox genes; Life history theory Fisher's (1930) Fundamental Theorem of Natural Selection (the FTNS) has been misinterpreted by many researchers for many years. The traditional interpretation is that fitness traits and behaviors strongly affected by natural selection will "lose" their genetic variance in the long run. The implication is that fertility and fertility precursors should have little or no genetic variance, and thus zero heritability. This inference is correct if natural selection is the only process at work. But it is not. The FTNS has been interpreted by many to mean that, by definition, for fertility (and other traits related to fitness), h 2 = 0. But it does not. Hughes and Burleson (2000) suggested a number of different processes that re-introduce genetic variance into fitness traits, even while natural selection is washing it out. These include NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript mutation (the most important), frequency-dependent selection, heterozygote advantage (overdominance), sexual antagonism, and environmental perturbations. Fisher (1930) clearly did not intend or participate in this misinterpretation of the FTNS. In the same source in which he presented the FTNS, he also discussed the role of contraception (an environmental perturbation) in re-introducing genetic variance into human fertility (see also Houle 1992, for additional and more modern discussion of "perturbing forces" from the environment). Further, Fisher presented empirical data suggesting significant heritability in family size among the British aristocracy, h 2 = 0.40. Eve...