Male phenotype and ejaculate quality in the zebra finch Taeniopygia guttata.

We tested the idea that female preference for relatively attractive extra-pair males arises because the morphological and behavioural features that females find attractive covary with ejaculate features: Sheldon's (Proc. R. Soc. Lond. B 257 25-30 (1994) phenotype-linked fertility insurance hypothesis. Two phenotypic traits that female zebra finches find attractive in males are song rate and symmetry of chest band plumage, but we found neither of these to be significantly related to any of the following ejaculate features: number of sperm, percentage of live sperm, absolute number of sperm, sperm length or sperm swimming velocity. Furthermore, and surprisingly, we did not find the predicted negative relationship between male song rate and fluctuating asymmetry of chest band plumage. Because most ejaculate features (except sperm numbers in rested males) show low levels of repeatability, it is unlikely that female zebra finches could reliably obtain a better quality ejaculate by choosing to copulate with a more attractive male. There was thus no evidence for the phenotype-linked fertility insurance hypothesis. Nor did we obtain evidence for the more general fertility insurance hypothesis: we found that female zebra finches paired to a vasectomized male, and hence receiving no sperm, were no more likely to seek an extra-pair copulation than females paired to an intact male.