This study tested opposing predictions made by Gamson's Minimum Resource Theory and equity theory concerning the type of coalitions that are formed when members of a triad expect that rewards will be correlated with amounts of individual resources. Results supported equity theory in that subjects, whose work on a pretask was used as the basis for assigning resources in a bargaining game, formed conservative (strong against weak) coalitions. When subjects either did not work or were paid prior to playing the game, results replicated past research in that there was a tendency to form (weak against strong) coalitions. Findings were discussed in terms of their implications for the generality of existing theories of coalition formation. The present study was an investigation of the role that the norm of equity plays in the formation of coalitions within triads. Past research on coalition formation (e.g., Kelley & Arrowood, 1960; Vinacke & Arkoff, 1957) to a large extent has focused on the effect of differences in apparent power. Typically, power is denned as resources that enable their owner to achieve some goal; the greater the resources, the more easily the goal is reached. For example, in one case of Caplow's (1956, 1968) Type 5 triad, resources are distributed in the ratios 4/3/2. Thus, the person with the greatest resources (4) will reach the goal first unless two participants pool their resources, that is, form a coalition. In Type 5 triads, any coalition will triumph, since any combination of two sets of resources (e.g., 2 + 3, 2 + 4, or 3+4) is greater than the resources of the isolate. In spite of this equality in potential effectiveness, subjects in past studies have tended to overchoose the 2-3 combination and form what Caplow (1968) has termed revolutionary coalitions. This coalition is because the two weaker parties join together to overthrow the person who originally