Table 1. Summary of conclusions. Polycentricity Participation Experimentation Bioregional Approach Theoretically sound? Yes, but there are concerns about transaction costs and accountability. Yes. Yes, but experiments as a research methodology yield limited reflexivity. Management as experimentation is only supported if the experiments are small in scale. Yes. Preference is shifting from unitary river-basin organizations as a means of implementing a river-basin management approach to collaborative, polycentric arrangements. Feasible? Fact of life, but the value of diversity is not always understood. Depends on the willingness and ability of authorities to organize participation, and of potential participants to become involved. Experiments are often difficult to implement because of societal resistance (equity concerns). The political landscape is often transformed by the time the experiment is finished. Unitary river-basin organizations are very rare; collaborations at the river-basin scale are increasingly popular. Effective? Not much can be said because of the lack of monitoring data and the attribution problem. Can contribute to the quality and legitimacy of decisions, but the connection to the formal decision process needs to be clearly specified. Not much known, but in other policy domains, experiments are often watered down to “pilots.” Not much can be said because of the lack of monitoring data and the attribution problem.