Stochastic computer simulations were used for quantifying the effect of selecting on prion protein (PrP) genotype on the risk of major outbreaks of classical scrapie and the rate of genetic progress in performance in commercial sheep populations already undergoing selection on performance. The risk of a major outbreak on a flock was measured by the basic reproduction ratio (R-0). The effectiveness of different PrP selection strategies for reducing the population risk was assessed by the percentage of flocks with R-0 < 1. When compared with the scenario where there was no selection on PrP genotype, selection against the VRQ allele had a minimal impact on genetic progress for performance traits. However, this strategy was not sufficient to eliminate the population risk after 15 years of selection when the initial frequency of the ARR allele was relatively low. More extreme PrP selection strategies aimed at increasing the frequency of the ARR allele and decreasing the frequency of the VRQ allele led to decreases in the rate of genetic progress for performance but reduced the population risk to very low values. The reduction in genetic progress was only large when the initial ARR frequency was low and, in general, the risk of major epidemics was very small when the frequency of this allele reached 0.7. (C) 2009 Elsevier B.V. All rights reserved.