Determinants of age-dependent change in a secondary sexual character

Many secondary sexual characters vary in a systematic way with the age of individuals, with young and old individuals displaying at lower levels than individuals of intermediate age. Analyses quantifying the within-individual and among-individual components of phenotypic variation can help partition effects of phenotypic plasticity and selective mortality. We analysed phenotypic variation in the expression of a secondary sexual character, tail length, in male and female barn swallows Hirundo rustica from four European populations studied during 11-26 years, using linear mixed effect models to describe age-related expression. Tail length increased from yearlings to intermediate aged birds with a subsequent decrease at old age. In males, this age-related pattern was because of both within-subject and between-subject effects, with no difference among populations. Males having longer lifespan had shorter tails when young than those having shorter lifespan. Females showed similar patterns of age-related variation as males, with no difference among populations. The major difference between sexes was that the between-subject effects (i.e. disappearance effects or selection) were much more important for males compared to females for which lifetime variation in tail length was mainly because of a within-subject effect (i.e., a plastic response). These findings suggest that whereas males trade greater expression of the secondary sexual character at young age against longevity, that was not the case for females. This is consistent with tail length being more costly in males than in females, with the cost of long tails potentially being offset by elevated mating success, whereas that is not the case in females.