A ptilochronological study of carry-over effects of conditions during wintering on breeding performance in the barn swallow Hirundo rustica.

The ecological conditions that a bird experiences during any stage of its life cycle may have consequences that become manifested at later life stages, and these 'carry-over effects' may be major components of variance in individual performance. Condition-dependent feather growth rate, as assessed by growth bars width (GBW), provides a unique, though largely under-exploited tool to investigate carry-over effects of ecological conditions and individual physiological state during molt. In this study of breeding barn swallows Hirundo rustica, which undergo a single complete annual molt of tail and wing feathers during wintering in sub-Saharan Africa, we first show that old (≥ 2 yr) females have larger GBW than old males and yearlings. GBW was smaller with larger infestations by common ectoparasites of barn swallows, independent of the swallows' age or sex, and larger GBW was associated with a higher index of body condition during the breeding season in males. Larger GBW predicted higher seasonal reproductive output of older males, but not reproductive output of younger males or females of any age class, with this higher reproductive output of older males mediated by higher offspring fledging success. However, no relationship with GBW was observed for seasonal reproductive output of males in the spring preceding the winter when the feathers were grown. Hence, this study suggests that the analysis of the rate of feathers growth ('ptilochronology') has a larger potential to serve as a powerful tool in the study of carry-over effects than has been appreciated to date. Specifically, the present results support the idea that conditions experienced during wintering in Africa and proximately reflected by GBW have carry-over effects on body condition and breeding success. These effects are sex and age specific, being more pronounced in older males, possibly as a consequence of differences in annual time routines and susceptibility to extrinsic factors among sex and age classes. [ABSTRACT FROM AUTHOR]