Collective induction is the cooperative search for descriptive, predictive, and explanatory generalizations, rules, and principles. Intergroup collective induction occurs when groups work separately on a common problem but exchange hypotheses and evidence. Two separated four-person groups exchanged hypotheses and evidence on each trial in the solution of the same rule induction problem (Exchange Condition) or solved the problem independently (No Exchange Condition). A theory of intragroup collective induction (Laughlin & Hollingshead, Organizational Behavior and Human Decision Processes, 1995) predicted the probabilities of group hypotheses for distributions of member hypotheses better than five other plausible models, although all six models could be rejected. Aggregating over 400 groups in the current and three previous experiments, the predictions of the theory could not be rejected whereas the other five models could confidently be rejected. A formal analysis distinguishes unidirectional and bidirectional instigative and supportive influence by correct, plausible, and nonplausible hypotheses. There was weak evidence in the Exchange Condition for instigative influence by correct hypotheses when one group received the correct hypothesis from the other, and strong evidence for mutually supportive influence when both groups simultaneously proposed the correct hypothesis and received it from the other group.