Soraker, J.; Ranke, P. S.; Wright, J.; Ringsby, T. H.; Saether, B.-E.; Jensen, H.; Araya-Ajoy, Y.

The causes of variation in multiple paternity (MP) in socially monogamous birds have received considerable attention. Traits like age and age-dependent morphology have been shown to be important for the distribution of MP. However, most studies fail to separate between the different processes underlying age-effects on MP, such as selective disappearance (that is, higher survival probability for certain phenotypes) versus individually plastic changes in morphology and mating behaviour with age. Using Bayesian multi-level path analysis on a long-term dataset from a house sparrow (Passer domesticus) metapopulation, we disentangle the effects of age on male and female MP, both independently from and through morphological traits. Age was a key determinant of MP for males when accounting for morphological traits, with males having MP more often as they get older (plastic component of age) and not due to disappearance of males unsuccessful to get MP (selective disappearance), challenging good genes explanations for MP. Conversely, selective disappearance explained the higher levels of MP in older females, suggesting that higher quality females have MP more often than lower quality females. We show how an appropriate statistical decomposition of age-components and age-dependent processes provides insights to the biological drivers of MP.