This paper examines the level of protest activity by the unemployed in the United States between 1890 and 1940 as a test of the value of a political process model for explaining social movement activity. Data on protest events and elite attitudes towards the unemployed were collected from newspaper articles. Voting behavior was used as an indicator of contested elections and unemployment levels were reflected by available indicators. Consistent with previous research on lower-and working-class mobilization, a change in the political environment was key to the extensive protest by the unemployed in the 1930s. Toward the end of the 1920s, and especially in the early 1930s, elites were no longer simply making public statements about the problem of unemployment, but were also discussing the need for aid programs. In the context of this new political environment, elections were once again contested in the 1930s, and extensive protest began in 1930, even before unemployment hit its high point in 1933. Thus, it was not simply deprivation, but the changed political environment which legitimized the issue of unemployment and created prospects for reform, which in turn helped produce the massive protest of the 1930s.The primary goal of the research reported in this paper is to further understanding of the development, and lack of development, of protest by the unemployed in the United States between 1890 and 1940. The focus is on resource mobilization, and more specifically, on a political process perspective on the development of social movements, and the interaction between political environment and deprivation in the mobilization of social movement activities.We examine a specific condition (unemployment) likely to produce hardship, presuming that unemployment creates hardship for the vast majority of the unemployed. It could, of course, be argued that the pain of unemployment is eased somewhat in agricultural settings, or through the provision of extensive welfare benefits. For the period 1890 to 1940, however, neither of these factors had much effect on the impact of unemployment. By the 1890s, most of the labor force was urban (U.S. Bureau of Census 1975), and the best historical descriptions of unemployment during this time period do not credit welfare support or food resources from private lands as significant factors in reducing the impact of unemployment (Feder 1936; Garraty 1978.) Our approach overcomes some limitations of past research which typically focused on all collective action events in a specified time period, often involving diverse issues and interest groups (Kerbo and Shaffer 1986; Snyder 1975:275; Tilly 1984). We examine a specific source of hardship affecting a segment of the population which can be followed over an extended time period in order to understand the changing parameters associated with collective action.' * Correspondence to: Kerbo, Social Sciences Department, California Polytechnic State University, San Luis Obispo, CA 93407.1. In previous work we have argued for the use of "group an...