The effect of gene dispersal on the dynamics and statics of gene substitution in plants

Abstract

A computer model has been developed to determine the effect of pollen and seed dispersal patterns on the probability of fixation and time to near-fixation of a rare advantageous gene in a plant population composed of 225 individuals. In the model, pollen and seed dispersal are according to four schedules (i) zero distribution, (ii) stepping-stone distribution, (iii) leptokurtic distribution, (iv) or random distribution. All combinations of pollen and seed dispersal schedules are treated. Relative fitnesses are assigned to three genotypes, AA, Aa, and aa, the latter always being the superior type. Relative fitness takes the form of competitive ability among the several seedlings germinating at each of the 225 safe sites in the population. The probability of fixation of a rare advantageous recessive gene (a = 0·013) is inversely correlated with neighbourhood size. Restricted pollen and seed movements also retard the tempo of gene substitution relative to that expected with panmixia. Decreasing the intensity of competition has the same effect. Gene frequency heterogeneity among subpopulations arises as a by-product of gene flow restriction, with graded patchworks of gene frequencies being most prominent when pollen and seed dispersal is narrow. We considered the consequences of three receipt patterns (edge, leptokurtic, and random) of extraneous genes via pollen from a population located in a prescribed direction from the recipient. If the gene is advantageous, the more restricted the distribution to one edge of the population, the slower the tempo of gene substitution. If the gene is detrimental, the equilibrium gene frequency is depressed by restricting the deposition of pollen to the edge. In both cases non-random receipt yields properties very different from that when random immigration is assumed. The literature indicates that gene flow within and between populations is non-random in plants. Our simulation demonstrates that abandoning the notion of panmixia will be a major step in understanding the dynamics and statistics of gene frequency change in plants.