We used the "morphology-performance-fitness" paradigm (Arnold, 1983) as our framework to investigate endocrine control of performance and fitness in Sceloporus undulatus (Eastern Fence Lizard). Focusing on males, we used the "natural experiments" of seasonal, sexual, and developmental variation in growth and in exercise endurance to identify testosterone and corticosterone as potential modulators of performance and related traits of interest. We followed with experimental manipulations of testosterone to investigate functional relationships, both in the laboratory and in the field. Further, we used focal observations and demographic studies, coupled with genetic determination of paternity, to test associations between performance and fitness, measured as reproductive success. We found that in males, endurance and plasma concentrations of testosterone and corticosterone are at their peaks in the spring breeding season, when lizards are most actively engaged in patrolling home ranges and in reproductive behavior. At that time, plasma concentrations of testosterone are correlated with body size; plasma concentrations of corticosterone and parameters of home range, including area and the number of overlapped females, are correlated with home-range overlap between males and females. During prereproductive development, males (but not females) experience a maturational increase in plasma testosterone. At about the same time, they become more active, expand their home ranges, and grow less quickly than do females, suggesting a trade-off in the allocation of energy, mediated by testosterone. Experimentally, testosterone has positive effects on fitness by stimulating endurance and reproductive activity and increasing home-range area, but it exacts costs in fitness by increasing ectoparasitism, decreasing growth, and decreasing survivorship. We found evidence of selection on body size, endurance, and home-range size (and thus access to potential mates). Despite having positive effects on performance traits, plasma concentrations of testosterone were not correlated with number of offspring sired by males. However, we found a strong correlation between the level of plasma corticosterone and the number of offspring sired. We also found evidence of size-assortative mating, indicating that for males, both the number and the size (and thus, fecundity) of their mates increase with body size. Our studies exemplify the power of natural history combined with experimental endocrine manipulations to identify hormonal regulators of performance and linkages to fitness. Furthermore, our results illustrate ecological and evolutionary significance of individual variation in endocrine traits.
