GENE FLOW ACCELERATES LOCAL ADAPTATION AMONG FINITE POPULATIONS - SIMULATING THE EVOLUTION OF INSECTICIDE RESISTANCE

Stochastic computer simulation of the evolution of insecticide resistance among finite, subdivided populations showed that interactions among gene flow, population size, initial genetic variation, and environmental heterogeneity influence adaptation to local selection caused by insecticide treatments. When all subpopulations were exposed to pesticide selection, gene flow had little effect unless the allele providing resistance was rare. When rare, the resistance allele was lost in many of the finite subpopulations. Higher rates of gene flow increased the overall rate at which resistance evolved by spreading the resistance allele more rapidly from subpopulations where it persisted. In heterogeneous environments (mixtures of treated and untreated fields), intermediate levels of gene flow produced the most rapid rates of local adaptation. At higher rates of gene flow, migration of susceptible alleles into treated subpopulations delayed the evolution of resistance. At low rates of gene flow, resistance evolution was delayed by the low rate of movement of resistance alleles between subpopulations. Although high gene flow (> 10% per generation) retarded resistance evolution in treated subpopulations, it greatly increased the resistance allele frequency in the untreated subpopulations. Results from finite population models suggest that in many field situations, gene flow may speed evolution of insecticide resistance. Previous results from infinite, continent-island population models had supported the opposite conclusion. These qualitatively different conclusions illustrate the importance of finite population size, particularly when lack of genetic variation constrains evolutionary adaptation.