For more than a century, biogeographers have sought to explain the large number of species found in Amazonian forests. The role of rivers as barriers to dispersal was recognized early on and this was the first evolutionary hypothesis to explain Amazonian diversity. Most of the recent debate on speciation in the Amazon has focused on the role of Pleistocene refugia. The methods of refuge biogeography helped shape early conservation priorities in Amazonia, although actual plans did not directly depend on the conceptual strengths or weaknesses of refugia biogeography. These approaches viewed people mostly as threats, though not always explicitly. Based on his work on primate distribution, Márcio Ayres formulated a synthetic speciation theory, the river-refuge hypothesis. The river-refuge model successfully resolved some of the historical and technical challenges of the earlier hypotheses. His work in várzea conservation, informed by this conceptual breakthrough, recognized that the maintenance of processes is at least as important as species numbers in prioritizing action. The work of Márcio Ayres broadened the scope of conservation in Amazonia by moving beyond the model of people-as-threats, and this was as great a conceptual contribution to conservation as anyone could make.