P. W. Cheng and K. J. Holyoak's (1985) pragmatic reasoning theory was used to inform hypotheses relevant to the transfer of deductive reasoning rules during adolescence. In Experiments 1 and 2, participants were trained on either permission or causal versions of P. C. Wason's (1966) selection task, and transfer to arbitrary problems was examined. In Experiments 3 and 4, the effect of training on permission or arbitrary problems to target causal problems was examined. Transfer was more likely from source causal and arbitrary problems than from permission problems. Also, there were developmental progressions in source problem performance. The findings are discussed in terms of the domain specificity of the reasoning rules induced during acquisition and in terms of the effects of training on pragmatic reasoning schemata.