Life History Evolution in Stochastic Environments: A Graphical Mean‐Variance Approach

Publisher Website
Google Scholar
Abstract
A small—variance approximation technique for examining the evolution of reproductive effort is developed, using a decomposition of fitness into arithmetic mean and variance components. Trait values often translate nonlinearly into fitness components such as offspring number or growth rate. A nonlinear relationship between phenotypic traits and fitness implies that the expected value of the fitness component will be a function of the arithmetic mean and variance in the traits influencing a given fitness component. The maximum expression of traits is often, however, a function of reproductive effort. Reproductive effort then influences the mean and variance in traits that affect fitness. Consequently, we can use these techniques to determine the optimal allocation of resources to current reproduction via their effects on the distribution of trait values. The optimal pattern of allocation is influenced by the quality and variability of the habitat and by the curvature of the trait—fitness relationship. The full range of adaptive responses, from semelparity to different degree of iteroparity, can be predicted from the model. We extend the model to cases where reproductive effort can be partitioned into different offspring types, for example, offspring produced through chasmogamous and cleistogamous flowers. When more than one type of offspring is produced, then reproductive risks can be spread across offspring types and the conditions favoring semelparity are less restrictive. Semelparity is especially favored if the offspring types exhibit negative covariance in their contributions to growth rate. This new technique appears generally consistent with traditional life history theory, but can lead to some new insights that result from the explicit partitioning of the effects of central tendency and variability in trait—fitness relationships.

This publication has 0 references indexed in Scilit: