A simulation model of the population dynamics and genetics of western corn rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae), was created to evaluate the use of refuges in the management of resistance to transgenic insecticidal corn, Zea mays L., expressing one or two toxin traits. Hypothetical scenarios and a case study of a corn hybrid pyramided with existing toxins are simulated. In the hypothetical situations, results demonstrated that evolution is generally delayed by pyramids compared with deployment of a single-toxin corn hybrid. However, soil insecticide use in the refuge reduced this delay and quickened the evolution of resistance. Results were sensitive to the degree of male beetle dispersal before mating and to the effectiveness of both toxins in the pyramid. Resistance evolved faster as fecundity increased for survivors of insecticidal corn. Thus, effects on fecundity must be measured to predict which resistance management plans will work well. Evolution of resistance also occurred faster if the survival rate due to exposure to the two toxins was not calculated by multiplication of two independent survival rates (one for each insect gene) but was equivalent to the minimum of the two. Furthermore, when single-trait and pyramided corn hybrids were planted within rootworm-dispersal distance of each other, the toxin traits lost efficacy more quickly than they did in scenarios without single-trait corn. For the case study involving transgenic corn expressing Cry34/35Ab1 and Cry3Bb1, the pyramid delayed evolution longer than a single trait corn hybrid and longer than a sequence of toxins based on at least one resistance-allele frequency remaining below 50%. Results are discussed within the context of a changing transgenic corn marketplace and the landscape dynamics of resistance management.