GENETIC ESTIMATES OF ANNUAL AND LIFETIME REPRODUCTIVE SUCCESS IN MALERED-WINGED BLACKBIRDS

DNA-based methods of determining parentage in birds have revealed that apparent reproductive success can be very different from true reprodu ctive success. Our main goal in this study was to provide estimates of true annual and lifetime reproductive success of male Red-winged Blac kbirds (Agelaius phoeniceus) to determine the importance of extra-pair mating to reproductive success. Males that were more successful at ex tra-pair mating, both annually and over their lifetimes, were also mor e successful on their own territories, so extra-pair success was not a chieved at the expense of within-pair success. Relative to within-pair success, extra-pair success accounted for only one-fifth as much vari ance in total success (annual and lifetime). For some individuals, how ever, extra-pair success was a substantial part of their total success . Standardized variance in true annual success was 16% higher overall than the variance in apparent success, although the effect was inconsi stent from year to year. Standardized variance in lifetime success inc reased 39% due to extra-pair paternity. Thus, extra-pair mating modest ly increased the opportunity for sexual selection. Males with the high est annual success on their own territories mated with more females on their territories, had fewer of those females mating with other males , and lost fewer nests to predation. Annual extra-pair success was hig hly correlated with the number of extra-pair mates. The number of year s a male bred and the number of females with which he mated contribute d significantly to the variance in lifetime success, both within pairs and extra-pair. Reproductive success in the first breeding season was significantly positively correlated with lifetime success. Positive c orrelations between true and apparent reproductive success and between true mating success and harem size suggest that conventional (i.e., n ongenetic) estimates of success provide reasonable estimates of popula tion patterns, despite being unreliable for estimating individual succ ess. Many of the ecological factors that affected lifetime success (e. g., territory acquisition, longevity) of males and females appeared si milar between our Canadian population and a population in Washington s tate, USA. However, variance in male success appears to be much higher in Washington, suggesting that exploiting this geographical variation might be a productive way to study sexual selection in this species.