In Tacoma, the important dynamics of change were largely internal to the court's probation department while in Austin, the dynamics were systemic. In both sites, the existing distributions of interests and power among actors and institutions were critical to whether new programs were successfully implemented and in what form. At both the organizational and systemic levels, webs of interest, priorities, and power provided significant limits to planned change efforts. At the organizational level, planned changes threatened to decrease the control subordinates had previously exercised over their work, to increase the amount of work performed, and to eliminate the basic nature of their work--counseling. At the systemic level, the planned sentencing reform threatened prosecutorial hegemony over sentencing, which was reinforced by the interests of the court in maintaining fine revenue, of the judges in promoting career advancement, of the bar in protecting the sentence and plea negotiation process and minimizing clients' sentences, and of the probation department in maintaining its flow of revenue from its caseload. Because the reform failed to alter these incentives, its implementation was highly problematic. Findings, therefore, indicate that reformers must address resistance to redistribution of power among those affected by the change if the reform is to have its intended outcome. Eight notes and 31 references are listed. Tabular data and survey methodology are appended. (Author abstract modified)