*I am most grateful to L. Wesley Wager, Joan K. Jackson, and Herbert L. Costner for their invaluable help and advice. Also, I would like to thank Alan P. Bates for his constructive comments on an earlier draft of this paper. 1 Models of organizations have either tacitly assumed that there is consensus on roles in social systems or have failed to recognize the variability of consensus on the manner in which roles are to be performed. Gross, Mason, and McEachern in their study of a New England school system cogently demonstrate the conceptual utility of the role consensus variable. In their reformulation of the role concept, Gross and his colleagues challenge the postulate of role consensus which is explicit or implicit in much of the social science literature. Furthermore, they found that the length of interaction and the extent of homogeneity were related to role consensus within the school member group that they studied. However, their findings gave no support to the hypotheses that interaction and homogeneity were related to role consensus between school board members and school superintendents. Gross and his colleagues further hypothesized that role consensus determined the satisfaction of group members and also influenced the manner in which they evaluated one another. Satisfaction was found to be related to consensus within the school board member group, but was not found to be related to role consensus between school board members and school superintendents. superintendent's ratings of the school board were found to be related to consensus, but the school board's ratings of the superintendents were not related to consensus. Neal Gross, Ward S. Mason, and Alexander W. McEachern, Explorations in Role Analysis (New York: John Wiley and Sons, Inc., 1958). It has been only in recent years that social scientists have turned their attention to the role consensus variable. Reisman, for instance, utilizes the lack of consensus on role-expectations among incumbents of the same social position to classify his respondents into bureaucratic types. Leonard Reisman, A Study of Role Conception in Bureaucracy, Social Forces, 27: 305-10 (1949). Varying degrees of consensus on the definitions of the shop steward as held by foremen, shop stewards, and workers have been found to result in strains and stresses between these role groups by Jacobsen, Charters, and Lieberman. Eugene Jacobsen, W. W. Charters, Jr., and S. Lieberman, The Use of the Role Concept in the Study of Complex Organizations, Journal of Social Issues, 7: 18-27, (1951). It is believed that role consensus is related to certain basic group processes. small group studies by Bales and Slater indicate a relationship between role consensus and developmental trends in small groups. Bales and Slater found, for instance, that group development differed according to the degree of consensus