EFFECT OF SPATIALLY RESTRICTED POLLEN FLOW ON SPATIAL GENETIC-STRUCTURE OF AN ANIMAL-POLLINATED ALLOGAMOUS PLANT-POPULATION

The consequences of population genetic structure under different degrees of restriction of pollen flow in an animal-pollinated allogamous plant species were investigated through computer simulations. A single locus with two alleles in a population of self-incompatible, bisexual diploid plants was considered. The formula f(is-proportional-to)e(-Kd) was adopted to describe pollen flow patterns where f is the probability of mating, d is the distance from the pollen source, and K is a constant. Values of K estimated by fitting the formula to data in the literature for different plant species varied from 0.076 to 0.787 for gene flow through pollen and from 0.652 to 2.805 for pollinator movement. Starting with randomly distributed genotypes with equal allele frequencies, simulation runs up to the 200th generation were made for different values of K in the range from 0.06 to 3.00. Even with the highest empirically based value of K(= 3.00), the decrease in heterozygosity and development of patch structure in our simulations were less than those of the 'relaxed' model by Turner et al. (1982) and were similar to the 'neighborhood' model by Sokal & Wartenberg (1983) and Epperson (1990). Local differentiation was very sensitive to the degree of gene dispersal. We conclude that one cannot ascribe any local differentiation observed solely to restriction of gene flow unless one explicitly examines whether the degree of restriction is sufficiently severe to produce the differentiation.