In spite of the massive literature on the distribution of power at the national level, little attention has been devoted to the public's perception of the shape of the power structure. This study utilizes a 1974 national survey to test the Form—Rytina hypothesis that the level of support for the perceived distribution of power is related to the magnitude of the rewards obtained from the stratification system. The findings do not uphold the research hypothesis but are quite compatible with a prediction derived from balance theory: in a democracy, perceived group influence is inversely related to the rater's affect towards the group stimulus. The analysis also reveals that both group membership and affective orientation towards the group make independent contributions to the perception of group influence. Emphasis on the primacy of social class as the pivotal determinant exaggerates the importance of instrumental rewards and ignores other important lines of cleavage.