AN ESTIMATE OF THE RELATIVE OPPORTUNITIES FOR NATURAL AND SEXUAL SELECTION IN A POPULATION OF MILKWEED BEETLES.

Abstract

In a recent paper Wade and Arnold (1980) provide an elegant framework for the empirical study of sexual selection. They point out that an evolutionary response to sexual selection is dependent on both the intensity of selection and the heritability of fitness. While the heritability of fitness or its components would be very difficult to measure in most field populations, the parameters necessary for an estimation of the intensity of selection can often be obtained. Their paper focuses on measures of the intensity of, or opportunity for, sexual selection on males and shows this intensity to be equal to the variance in number of mates per individual divided by the square of the mean number of mates. Though variation among individuals in mating success is an obvious requirement for the operation of sexual selection, surprisingly few studies have succeeded in documenting the extent of such individual differences in natural populations. While empirical studies of sexual selection have been quite successful in documenting those morphological or behavioral characteristics that determine mating success in a variety of species, actual measures of variance in mating success or the intensity of sexual selection in natural populations are rarer. This is particularly true when mating success is considered over the lifetime of the organism rather than a single breeding season. Considerations of variance in mating success measured over the lifetime of the study organisms must take into account both the intraand intersexual interactions commonly associated with sexual selection and any variation in survivorship during the reproductive period that is usually associated with natural selection. Because the expression of an individual's mating ability is conditional on surviving to a given mating opportunity, a long-lived individual could accrue more mates than a shorter-lived competitor simply because he experiences more opportunities for encountering mates. Thus, the intensity of both natural and sexual selection must be considered when one uses the total number of mates acquired in a lifetime as a measure of fitness. The question then arises as to whether differences between males in their ability to find and court females (mating efficiency) or differences between males in survivorship after onset of reproduction contribute more to the total variance in the fitness of males. In understanding the evolution of life-history or mating strategies it would also be important to know if there are any phenotypic or genetic correlations between these two components of variance; such correlations have been identified by Arnold and Wade (unpubl.) as cointensities or the covariance between fitnesses at sequential episodes of selection. In order to estimate variance among individuals in number of mates one must be able to recognize individuals, identify their whereabouts during the mating season or seasons, and identify mating events. These criteria are most often met in vertebrate mating systems and accurate measures of mating success have been made for several such systems (e.g., Howard, 1979; Payne, 1979; Downhower and Brown, 1980). Few studies follow individuals through more than one breeding season, however, and thus measures of variance in lifetime mating success or fitness are virtually non-existent. Studies of invertebrates rarely examine