When A causes B and B causes C, under what conditions is A a good explanation for the occurrence of C? We propose that distal causes are only perceived to be explanatory if the causal mechanism is insensitive to inessential variations of boundary conditions. In two experiments, subjects first observed deterministic A->B->C relationships in a single exemplar of an unknown kind. They judged A to be crucial for C by default. However, when they subsequently learned that the causal mechanism fails to generate the A->C dependency in other exemplars of the same kind, subjects devalued A as a crucial explanation for C even within the first exemplar. We relate these findings to the idea that good explanations pick out portable dependency relations, and that sensitive causes fail to meet this requirement.