Proximate developmental mediators of sexual dimorphism in size: case studies from squamate reptiles.

Sexual dimorphism in size (sexual size dimorphism; SSD) is nearly ubiquitous, but the relative importance of genetic versus environmental control of SSD is not known for most species. We investigated proximate determinants of SSD in several species of squamate reptiles, including three species of Sceloporus lizards and the diamond-backed rattlesnake (Crotalus atrox). In natural populations of these species, SSD is caused by sexual differences in age-specific growth. Males and females, however, may often share similar potentials for growth: growth is strongly responsive to the availability of food, and sexual differences in growth can be greatly suppressed or completely absent under common environmental conditions in the laboratory. Sexually divergent growth is expressed in natural environments because of inherent ecological differences between males and females and because of potential epigenetic effects of sex-specific growth regulators. In field-active Sceloporus, sexual differences in growth rate are associated with sexual divergence in plasma testosterone. Experiments confirm that testosterone inhibits growth in species in which females are larger (for example, S. undulatus and S. virgatus) and stimulates growth in those in which males are larger (for example, S. jarrovii). Interestingly, however, sexual divergence in plasma testosterone is not accompanied by divergence in growth in S. jarrovii or in male-larger C. atrox in the laboratory. Furthermore, experimental effects of castration and testosterone replacement on growth are not evident in captive S. jarrovii, possibly because growth effects of testosterone are superseded by an abundant, high-quality diet. In female-larger S. undulatus, growth may be traded-off against testosterone-induced reproductive costs of activity. In male-larger species, costs of reproduction in terms of growth are suggested by supplemental feeding of reproductive female C. atrox in their natural environment and by experimental manipulation of reproductive cost in female S. jarrovii. Growth costs of reproduction, however, do not contribute substantially to the development of SSD in male-larger S. jarrovii. We conclude that the energetic costs of testosterone-induced, male reproductive behavior may contribute substantially to the development of SSD in some female-larger species. However, despite strong evidence that reproductive investment exacts a substantial cost in growth, we do not support the reproductive cost hypothesis as a general explanation of SSD in male-larger species.