On Measuring Fitness Gain Curves in Plants

Abstract

The great variety of sexual systems possessed by flowering plants can be understood in terms of sex allocation theory, which assumes that individuals allocate resources to male vs. female reproductive structures so as to maximize total fitness. The relative shapes of the fitness gain curves for each sex can be used to predict and explain such allocation patterns. Several recent studies have attempted to measure the shapes of these gain curves. Typically, these studies have regressed estimates of male or female fitness against flower number for a set of plants of different size. Here I argue that this method is inappropriate, and that it can lead to incorrect conclusions about the true shapes of the gain curves if some third, causal factor, such as genotype, size, or environmental conditions, affects both allocation and the fitness returns from that allocation. I suggest that experimental manipulation of flower number is likely to be the most useful and easily interpretable technique for estimating the shapes of gain curves in plants.