A spatially explicit stochastic model demonstrates the feasibility of Wright's Shifting Balance Theory

Recently there has been a resurgence of theoretical papers exploring Wright 's Shifting Balance Theory (SBT) of evolution. The SET explains how traits which must pass through an adaptive valley may evolve in substructured popu lations. It has been suggested that Phase III of the SET (the spread of new advantageous traits through the populations) proceeds only under a very re stricted set of conditions. We show that Phase III can proceed under a much broader set of conditions in models that properly incorporate a key featur e of Wright's theory: local, random migration of discrete individuals.