Table1.PDF

<p>Glucocorticoid hormones are important regulators of metabolic processes, and of the behavioral and physiological responses to stressors. Within-population variation in circulating glucocorticoids has been linked with both reproductive success and survival, but the presence and direction of relationships vary. Although conceptual models suggest the potential for interactions between glucocorticoid secretion under acute stress and non-acute stress contexts to influence phenotype and fitness, very little is known about the presence or implications of such interactions. Here we use a large data set from breeding tree swallows (Tachycineta bicolor; n = 215 adults) to test the predictors of reproductive success and annual survival probability in females and males over a four-year period. Across years and life history stages, glucocorticoids predicted fitness in female tree swallows. Under challenging conditions, females that maintained low baseline corticosterone levels during incubation had higher reproductive success. During the nestling provisioning period, interactions between baseline glucocorticoids and the stress response suggest that females may face trade-offs in the regulation of glucocorticoids across contexts. Reproductive success was highest among females that maintained low baseline glucocorticoids coupled with a strong acute stress response, and among females with high baseline glucocorticoids and a weak acute stress response. Females with low or high glucocorticoid levels across contexts (baseline and acute stress) fledged fewer young. Glucocorticoid levels did not predict fledging success in males. None of the models of annual survival probability in females received strong support; model comparisons suggested weak negative effects of stress responsiveness during nestling provisioning, and positive effects of body condition and age, on survival. Male survival probability was not predicted by breeding phenotype. Within and across years the glucocorticoid stress response was individually repeatable; across-year repeatability was particularly high during the nestling provisioning period. Both measures of the acute stress response—circulating stress-induced corticosterone levels and their stress-induced increase—showed similar repeatability, but baseline corticosterone was not repeatable within or across years. Overall, our results suggest that taking into account the potential for individual differences in glucocorticoid trait expression in one context to influence optimal endocrine expression in other contexts could be important for understanding the evolution of endocrine systems.