DESIGN OF SELECTION EXPERIMENTS FOR COMPARING ALTERNATIVE TESTING REGIMES

Selection response achieved in lines selected in different environments, for example feeding regimes, can be evaluated in two ways. Responses can be compared in a single environment, perhaps one of the test regimes or a commercial feeding regime. Alternatively, a test for interaction can be conducted in which both selected lines are tested on both selection regimes. The variances of direct and correlated responses are compared. It is shown that if there is a high genetic correlation (small interaction) in performance on the selection regimes, a high correlation due to drift is induced between the performances of a line on two environments. This enables a much more precise test to be conducted of interaction than of differences in response on a single environment. © 1978 The Genetical Society of Great Britain.