It is not uncommon to find that newly introduced, pest resistant crop varieties soon lose their resistance. This study examines some possible problems in present resistance screening programs which may lead to development of such temporarily resistant varieties. A procedure is suggested which may improve our ability to predict whether a variety's resistance will be more or less permanent in the field. This procedure is used to predict future resistance in a model system involving resistant ‘Marketmore 70’ cucumber and the twospotted spider mite, Tetranychus urticae .