Variance of lifetime reproductive success: estimation based on demographic data.

Variance of lifetime reproductive success is not a well-defined parameter; it is a function of the way a life cycle is defined. Therefore, comparisons of such variance across studies and taxa will only be valid if the life cycles employed are complete (i.e., an entire generation), precisely defined (e.g., egg to egg or adult to adult), and identical. Given a well-defined life cycle, we present methods for obtaining estimates of the variance of lifetime reproductive success among individuals using demographic data. The estimates are based on the assumptions of no selection and negligible covariance between life span and reproductive effort. The techniques can be used to investigate the effects of alternate life histories on the distribution of reproductive success in a population, as well as to obtain estimates of effective population size. They may also provide a null model for investigations of selection. Some theoretical and empirical results are presented. A method for obtaining confidence intervals for the estimates is described.