The Selective Interactions of Dispersal, Dormancy, and Seed Size as Adaptations for Reducing Risk in Variable Environments

Abstract

Seed size, dormancy, and dispersal share three population-dynamic functions in temporally and spatially varying environments: risk reduction or bet hedging, escape from crowding, and escape from sib competition. Here we developed a model to explore the ways they may interact to reduce risk. We have shown that the risk-reducing properties of these three seed traits evolve only in response to global temporal variance. Thus, to understand how selection impinges on the seed traits, creating fitness interactions, we must understand the factors contributing to global temporal variance and how they are mitigated by the various seed traits. Since the traits interact to reduce variance, arbitrarily fixing any one trait at different values alters the fitness-maximizing values of the others, resulting in trade-offs among traits. We explore how changes in the number of independent environmental patches, the probability of favorable conditions, the radius of dispersal, and spatial and temporal autocorrelation of environmental conditions alter selection on the interacting syndrome of seed traits. We discuss the implications of these fitness interactions for our understanding of each of the seed traits in isolation, as well as for our understanding of seeds as reproductive structures integrating coadapted functions.