Parental investment in the Columbian ground squirrel: empirical tests of sex allocation models.

Parental allocation of resources into male or female offspring and differences in the balance of offspring sexes in natural populations are central research topics in evolutionary ecology. Fisher (Fisher, R. A. 1930. The genetical theory of natural selection, Clarendon Press, Oxford, UK) identified frequency-dependent selection as the mechanism responsible for an equal investment in the sexes of offspring at the end of parental care. Three main theories have been proposed for explaining departures from Fisherian sex ratios in light of variation in environmental (social) and individual (maternal condition) characteristics. The Trivers-Willard model (Trivers, R., and D. Willard. 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science 179:90-92) of male-biased sex allocation by mothers in the best body condition is based on the competitive ability of male offspring for future access to mates and thus superior reproduction. The local resource competition model is based on competitive interactions in matrilines, as occur in many mammal species, where producing sons reduces future intrasexual competition with daughters. A final model invokes advantages of maintaining matrilines for philopatric females, despite any increased competition among females. We used 29 yr of pedigree and demographic data to evaluate these hypotheses in the Colombian ground squirrel (Urocitellus columbianus), a semisocial species characterized by strong female philopatry. Overall, male offspring were heavier than female offspring at birth and at weaning, suggesting a higher production cost. With more local kin present, mothers in the best condition biased their offspring sex ratio in favor of males, and mothers in poor condition biased offspring sex ratio in favor of females. Without co-breeding close kin, the pattern was reversed, with mothers in the best condition producing more daughters, and mothers in poor condition producing more sons. Our results do not provide strong support for any of the single-factor models of allocation to the sexes of offspring, but rather suggest combined influences of relative maternal condition and matriline dominance on offspring sex ratio.