Sexually selected traits and adult survival: A meta-analysis

Traits correlated with mak mating success are likely to be subject to sexua l selection. Sexually selected characters are thought to be costly to devel op and maintain. If males do not vary their investment in sexual traits in relation to their ability to bear the costs, there should be a negative rel ationship between male longevity or survival and the expression of sexual t raits. In particular, a negative relationship is predicted by pureFisherian models for the evolution of sexual ornaments. The same should also be true for traits that evolve via pleiotropy (e.g., due to,sensory exploitation o r bias) with no subsequent evolution of condition dependent modification. W e collected information on the relationship between traits correlated with male mating rate and estimates of adult male survivorship or life span. In total we obtained 122 samples from 69 studies of 40 species of bird, spider , insect, and fish. In a mete-analysis we calculated the average sample siz e weighted correlation between trait expression and adult survival Analyses at the level of samples, studies, and species revealed significant positiv e relationships (r = 0. 08, 0. 10, and 0. 13, respectively; all P < 0.001). The unweighted correlation at the species level was r = 0.24. In general, males with larger ornaments or weapons, greater body size, or higher rates of courtship showed greater survivorship or longevity. This finding is inco nsistent with pureFisherian model or other models that do not incorporate c ondition or quality dependent trait expression. It suggests that male inves tment in sexually selected traits is not fixed but varies in relation to th e ability to pay the underlying costs of expressing these characters. Hence , many secondary sexual characters are likely to be condition dependent In their expression.