Sexual selection resulting from extrapair paternity in collared flycatchers

Extrapair paternity has been suggested to represent a potentially important source of sexual selection on male secondary sexual characters, particular ly in birds with predominantly socially monogamous mating systems. However, relatively few studies have demonstrated sexual selection within single sp ecies by this mechanism, and there have been few attempts to assess the imp ortance of extrapair paternity in relation to other mechanisms of sexual se lection. We report estimates of sexual selection gradients on male secondar y sexual plumage characters resulting from extrapair paternity in the colla red flycatcher, Ficedula albicollis, and compare the importance of this for m of sexual selection with that resulting from variation in mate fecundity. Microsatellite genotyping revealed that 15% of nestlings, distributed nonr andomly among 33% of broods (N = 79), were the result of extrapair copulati ons. Multivariate selection analyses revealed significant positive directio nal sexual selection on two uncorrelated secondary sexual characters in mal es (forehead and wing patch size) when fledgling number was used as the mea sure of fitness. When number of offspring recruiting to the breeding popula tion was used as the measure of male fitness, selection on these traits app eared to be directional and stabilizing, respectively. Pairwise comparisons of cuckolded and cuckolding males revealed that males that sired young thr ough extrapair copulations had wider forehead patches, and were paired to f emales that bred earlier, than the males that they cuckolded. Path analysis was used to partition selection on these traits into pathways via mate fec undity and sperm competition, and suggested that the sperm competition path way accounted for between 64 and 90% of the total sexual selection via the two paths. The selection revealed in these analyses is relatively weak in c omparison with many other measures of selection in natural populations. We offer some explanations for the relatively weak selection detected. (C) 199 9 The Association for the Study of Animal Behaviour.