A theoretical framework is developed to identify the major contextual variables influencing bail setting in the two cities. The basic perspective of the framework is that bail setting is a form of choice behavior which reflects the interplay of political and organizational incentives and constraints on the willingness and ability of court officials to sanction defendants. Study data were collected through observations of the bail-setting processes, conversations with the officials making these decisions and with other participants, published and unpublished reports, newspapers, and two samples of defendants processed by the two courts. These samples were part of a larger study in which the author participated. The hypothesis that bail reform would explain the policy differences between Detroit and Baltimore was not supported by the data. An alternative explanation based on the characteristics of the defendants, their prior criminal activity, formal charges, and evidence was also not supported. Bail-setting decisions were found to flow from the responses of court officials to the political risks they faced in releasing or detaining felony defendants, the resources available to support various bail decisions, and their vulnerability to sanctions for making certain choices. Baltimore's fuller complement of reform measures failed to produce a less punitive bail policy than in Detroit, where reform was less complete. The lessons these cities provide regarding the feasibility and effectiveness of various reform strategies are discussed. Appended are research methods of the study, examples of preliminary and final bail-setting observations forms, and a sample of the initial Baltimore interview form for commissioners. Tabular data and a bibliography of about 50 citations are provided.