Testing for individual variation in breeding success

I measured the breeding success of Northern Fulmars (Fulmarus glacialis) in 6 consecutive years at a colony in the Semidi Islands, western Gulf of Alaska (56°N, 156°W). I tabulated the frequency distribution of the number of years successful for a sample of 224 nest sites at which breeding occurred in all 6 yr. A consistently higher probability of success occurred in some sites than in others (Fig. 1a). The same over-representation of extreme values was present in a subsample of 137 sites in which one or both members of the pair was individually known from plumage differences and no changes of identity were detected between years (Fig. 1b). More pairs were consistently successful, or consistently unsuccessful, than expected on the null hypothesis that all pairs had the same probability of breeding success in a given year.