Context dependency of baseline glucocorticoids as indicators of individual quality in a capital breeder

Identifying markers of individual quality is a central goal of life-history theory and conservation biology. The 'corticosterone (CORT)-fitness hypothesis' postulates that low fitness signals impaired ability to cope with the environment, resulting in elevated baseline CORT levels. CORT can, however, be negatively, positively or neutrally related to fitness, depending on the context. In order to clarify this controversial issue, we elucidate the utility of using baseline CORT as a correlate of individual fitness in incubating female eiders across variable environments. An increase in serum CORT with decreasing body condition was evident in older, more experienced breeders, while increased clutch mass was associated with elevated serum CORT in females breeding late in the season. For faecal CORT, the expected negative association with body condition was observed only in early breeders. We found a strong increase in faecal CORT with increasing baseline body temperature, indicating the utility of body temperature as a complementary stress indicator. Females in good body condition had a lower baseline body temperature, but this effect was only observed on open islands, a harsher breeding habitat less buffered against weather variability. Females with higher reproductive investment also maintained a lower baseline body temperature. Nest success strongly decreased with increasing serum and faecal CORT concentrations, and individual stress hormone and body temperature profiles were repeatable over years. Although our data support the tenet that baseline CORT is negatively related to fitness, the complex context-dependent effects call for cautious interpretation of relationships between stress physiology and phenotypic quality.