Cost of reproduction and the evolution of deferred breeding in the western gull

Survival patterns in a population of western gulls (LOTUS ocddentaUs) of known age of first breeding, a, indicate a cost of reproduction related to the age of initial breeding. Among both sexes, birds that commenced breeding at the earliest ages (3 years in males and 4 years in females) had higher annual mortality than those that deferred breeding one or more years. In addition, females (but not males) evidenced a cumulative cost of reproduction: holding age constant, females with more annual breeding attempts demonstrated poorer survival. These patterns of a-specific survival were statistically significant after controlling for interannual variation in food availability and are not explained simply by variation in the intrinsic quality of individuals. To assess the effects of these sex-specific costs on fitness, we combined the observed survival patterns with data on prebreeding survivorship and a-spedfic reproductive success to »ttimgt«» rates of population growth and lifetime reproductive success for different ages at first reproduction. Males showed a dearly defined fitness optimum at a » 4 years, which coincided with the modal a for males in the population. Females showed no dear optimum, except that breeding at age 4 was suboptimaL hence females benefited from deferring breeding to ages 5-7 years. Observed age of first breeding also showed no dear mode for females, with slight peaks at ages 5 and 7. As a result, in both sexes, the fitness surface for a corresponded well with observed frequencies of a. We suggest that stabilizing selection has acted to shape the phenotypic distribution of a in males but, due to trade-offs between survival and early reproduction, stabilizing selection is weak or«bsent in female* Krj words: age, breeding experience, cost of reproduction, deferred breeding, gulls, reproductive success, survivorship. [Behav Eeol 8:140-147 (1997)1