Experimental corticosterone manipulation increases mature feather corticosterone content: implications for inferring avian stress history from feather analyses

Feathers incorporate circulating steroids during development. It is therefore assumed that the corticosterone (CORT) content of feathers (CORT f ) represents an integrated measure of plasma CORT over the moult period. We tested this assumption by quantifying CORT f in feathers of House Sparrows (Passer domesticus (Linnaeus, 1758)) that were plucked before and after experimental manipulation of circulating CORT. Two of the seven flight feathers collected from each bird were fully grown throughout the CORT-manipulation period. We found that CORT f of all seven feathers corresponded with plasma CORT levels of non-moulting reference House Sparrows given the same implants. Surprisingly, the CORT f of the two mature feathers was 4- to 10-fold higher than values measured in the new replacement feathers. Our results show that CORT f of mature feathers may be affected by circulating CORT outside the moulting period. The most plausible explanation for our results is that CORT was transferred onto feather surfaces externally, but the mechanisms involved remain to be identified. Researchers are encouraged to establish effective procedures, in terms of both solvent and duration, for removing surface residues without extracting CORT from within the feather. This will increase confidence when inferring moult-related stress status from feather analyses in future ecological studies.