In most cases, if relevant, positive evidence raises and negative evidence lowers argument strength in induction. Previous research, however, has shown that it is possible to raise the argument strength of a single positive premise argument by introducing negative evidence (Heussen, et al., 2011). Here we test one possible mechanism for such an increase in argument strength. When people consider the positive premise they develop a set of hypotheses. Subsequently encountering a negative premise would render some of these hypotheses less likely and hence, if participants see the hypotheses as an exhaustive set, shift the probability to the remaining hypotheses to varying degrees. We test this idea by asking people to choose between four hierarchically nested conclusions across various conditions of evidence. The results are discussed in the light of models of induction.