Sexual Selection and Variation in Reproductive Success in a Long-Lived Organism

Abstract

Intensive observations of the reproduction of a population of bullfrogs during 1976-1978 indicated that variation in male yearly estimates of reproductive success (ERS) was 2-3 times greater than that of females. For both sexes, larger, older individuals produced more hatchlings than did smaller, younger individuals because of size-related advantages in competition for mates (males) or production of large clutches (females). Data on mating success alone explained most of the variation in hatchling production among males; data on mating success, clutch size and hatching success of embryos were necessary to explain an equivalent amount of variation in hatchling production among females. Use of mating success data to ascertain the amount of variation in RS [reproductive success] between the sexes in this species can be misleading. Computer simulations predicting variation in lifetime ERS were performed using measures of age-specific fecundity based on observed yearly ERS and estimates of age-specific survivorship based on mark-recaptures and estimates of migration. In contrast to the 2- to 3-fold difference observed in yearly ERS between the sexes, simulations predicted that males should be only slightly more variable in lifetime ERS than females. The discrepancy between observed yearly data and predicted lifetime values is consistent with the different life history trajectories of the sexes. Males achieve most of their RS only in the latter years of life; females accumulate RS during each year of adult life. The number of progeny sired by the most successful male during his lifetime may be similar to that produced by the most successful female during her lifetime. Adult survivorship plays a major role in determining variation in lifetime RS and tends to dampen the effect of sexual differences in variability in fecundity at each age. Yearly and lifetime RS data can be used to estimate what the current intensity of selection might be on the sexes or can be used retrospectively to interpret effects of past selection. Lifetime reproductive data are necessary to estimate selection intensity but yearly data may be nearly as good as lifetime data for retrospective analyses. A precise way to relate the amount of variation in yearly RS to that of lifetime RS is needed for future investigations of sexual selection.