Although great progress in understanding the variation in plant mating systems has been made in the past century, most studies have focused on the role of the negative effects of inbreeding. Recently, a different approach has been developed, one that is based on a general analysis of plant reproduction and pollination biology. It assumes that rates of self-fertilization are determined, at least in part, by the relative amount of self and outcross pollen on stigmas--an assumption called the mass-action assumption. Preliminary work suggests that models incorporating the mass-action assumption may be able to explain the full range of observed plant mating systems, even without considering the effects of inbreeding. This preliminary work will be extended in two ways. First, the mathematical models will be extended to include (1) a more realistic representation of the spatial structure of plant populations and (2) inbreeding depression. Second, experimental tests of the model predictions will be made by taking advantage of a well studied mating system polymorphism in Senecio vulgaris (1) to make direct estimates of the key features of the model in natural and experimental populations and (2) to investigate the role of frequency-dependent pollen export in maintenance of the variations in mating system.
