Participatory approaches to watershed management, including stakeholder involvement initiatives, have increasingly been advocated as a way to improve management of water resources. These approaches seem to hold particular promise for complex issues, including management of interstate water bodies. Individual initiatives, however, vary widely and this variation has implications for both decision-making processes and results.This paper presents case studies of watershed planning in three interstate basins: the Delaware Bay, Lake Champlain, and the Merrimack River watersheds. In each of these, recent initiatives involved an array of stakeholders in substantive roles in development and implementation of watershed plans.Planning for the Delaware Bay was initiated under the federal National Estuary Program. In 1988, USEPA and three states convened a Management Conference charged with development of Comprehensive Conservation and Management Plan, a process that was ultimately completed in 1996. Watershed planning for Lake Champlain was initiated in 1990 by an act of the U.S. Congress. This act established a Management Conference that completed a comprehensive plan for the U.S. portion of the basin in 1996. For the Merrimack River, USEPA initiated a bi-state planning effort in 1992, convening a Management Committee that completed a comprehensive watershed plan in 1997.Case studies of these stakeholder involvement initiatives were designed to evaluate three factors: 1) the structures for participation in watershed planning, 2) the effectiveness of decision processes, and 3) the changes in policy and management programs directed as a result of decision making. Data sources included in-depth interviews with key participants and staff, a mail survey, and documentation of decision making (e.g., draft and final management plans).Results highlight two aspects of the structure for participation as important in the design and evaluation of stakeholder involvement processes: the breadth of participation and the degree of authority. Results also indicate that different structures may promote different types of results.In these cases, broader participation was associated with more effective decision processes. That is, broader participation should not be assumed to be an obstacle to effective decision making. Rather, it may lead to decision processes that are more effective from competing dimensions, that meet more of the competing demands on environmental decisions.