Temporal overlap and repeatability of feather corticosterone levels: practical considerations for use as a biomarker

The measurement of corticosterone (CORT) levels in feathers has recently become an appealing tool for the conservation toolbox, potentially providing a non-invasive, integrated measure of stress activity throughout the time of feather growth. However, because the mechanism of CORT deposition, storage and stability in feathers is not fully understood, it is unclear how reliable this measure may be, especially when there is an extended interval between growth and feather collection. We compared CORT levels of naturally grown feathers from tree swallows (Tachycineta bicolor) that were moulted and regrown concurrently and therefore expected to have similar CORT levels. Specifically, we compared the same feather from the left and right wing (moulted symmetrically) and different types of feathers (wing, back and tail) expected to have been moulted within the same time period. We found that larger, heavier feathers held more CORT per unit length. In addition, we found a lack of concordance in CORT levels both within the same feather type and between different feather types, even after taking into account differences in feather density. Our results indicate that naturally grown feathers may not consistently provide an indication of stress status. Additionally, conflict in results may arise depending on the feather assayed, and total feather volume may be an important consideration when interpreting feather CORT levels. Future work is necessary to determine explicitly the mechanisms of CORT deposition, the effects of environmental exposure and feather wear on the permanence of the feather CORT signal, and the influence of responses to wild stressors on feather CORT levels, before feather CORT can be implemented effectively as a tool for ecological and conservation applications.