Search

Sexual selection on fitness-determining traits should theoretically erode genetic variance and lead to low heritability. However, many sexually selected traits maintain significant phenotypic and additive genetic variance, with explanations for this “lek paradox” including genic capture due to condition-dependence, and breaks on directional selection due to environmental sources of variance including maternal effects. Here we investigate genetic and environmental sources of variance in the intrasexually selected green badge of the sand lizard (Lacerta agilis). The badge functions as a cue to male fighting ability in this species, and male–male interactions determine mate acquisition. Using animal models on a pedigree including three generations of males measured over an extensive 9-year field study, we partition phenotypic variance in both badge size and body condition into additive genetic, maternal, and permanent environmental effects experienced by an individual over its lifespan. Heritability of badge size was 0.33 with a significant estimate of underlying additive genetic variance. Body condition was strongly environmentally determined in this species and did not show either significant additive genetic variance or heritability. Neither badge size nor body condition was responsive to maternal effects. We propose that the lack of additive genetic variance and heritability of body condition makes it unlikely that genic capture mechanisms maintain additive genetic variance for badge size. That said, genic capture was originally proposed for male traits under female choice, not agonistic selection. If developmental pathways generating variance in body condition, and/or the covarying secondary sex trait, differ between inter- and intrasexual selection, or the rate at which their additive genetic variance or covariance is depleted, future work may show whether genic capture is largely restricted to intersexual selection processes.

Vitellogenesis (‘yolking’ of follicles) is a bioenergetically costly stage of reproduction requiring enlargement of the liver to produce vitellogenin (VTG) yolk precursor proteins, which are transported and deposited at the ovary. VTG may, however, serve non-nutritive anti-oxidant functions, a hypothesis supported by empirical work on aging and other life-history transitions in several taxa. We test this hypothesis in female painted dragon lizards (Ctenophorus pictus) by examining covariation in VTG with the ovarian cycle, and relative to reactive oxygen species (ROS) including baseline superoxide (bSO). Plasma VTG decreased prior to ovulation, when VTG is deposited into follicles. VTG, however, remained elevated post-ovulation when no longer necessary for yolk provisioning and was unrelated to reproductive investment. Instead, VTG was strongly and positively predicted by prior bSO. ROS, in turn, was negatively predicted by prior VTG, while simultaneously sampled VTG was a positive predictor. These findings are consistent with the hypothesis that VTG functions as an anti-oxidant to counteract oxidative stress associated with vitellogenesis. The relationship between bSO and VTG was strongest in post-ovulatory females, indicating that its function may be largely anti-oxidant at this time. In conclusion, VTG may be under selection to offset oxidative costs of reproduction in egg-producing species.

While our understanding of male reproductive strategies is informed by extensive investigations into endocrine mechanisms, the proximate mechanisms by which females compete for mates and adjust reproduction to social environment remains enigmatic. We set out to uncover endocrine correlates of mate choice, social environment, and reproductive investment in female red-backed fairy-wrens Malurus melanocephalus. In this socially monogamous, yet highly sexually promiscuous species, females experience discrete variation in the phenotype of their mates, which vary in both plumage signals and level of paternal care, and in the composition of their breeding groups, which consist of either the pair alone or with an additional cooperative auxiliary; female investment varies according to these social parameters. We found that androgen, estrogen, and glucorticoid levels varied with reproductive stage, with highest androgen and estrogen concentrations during nest construction and highest corticosterone concentrations during the pre-breeding stage. These stage-dependent patterns did not vary with male phenotype or auxiliary presence, though androgen levels during pre-breeding mate selection were lower in females obtaining red/black mates than those obtaining brown mates. We found no evidence that androgen, estrogen, or corticosterone levels during the fertile period were related to extra-pair young (EPY) frequency. This study demonstrates clear changes in steroid levels with reproductive stage, though it found little support for variation with social environment. We suggest hormonal responsiveness to social factors may be physiologically constrained in ways that are bypassed through exogenous hormone manipulations.