Freshwaters contribute 15 percent of the world's reported fish catch, or about 10.1 million tonnes in 2006, most of which comes from tropical systems. The true contribution of tropical inland fisheries is likely to be higher, as less than half of the inland capture production is actually reported. While reservoir fisheries are already an essential component of this production, the potential of most of them may even exceed their current catch levels. Opportunities exist to increase productivity, provided that environmentally and socially sustainable management systems can be adopted. To realize this untapped potential, it is necessary to improve understanding of the processes influencing reservoir productivity in such a way as to involve both biological principles and stakeholder participation, as each reservoir has different properties and different research and management institutions. Seen in isolation, catch and productivity data of individual reservoirs may be difficult to interpret. The present technical paper attempts to address this issue by reviewing the knowledge accumulated in reservoirs in some very different tropical river basins: the Indus and Ganges/Brahmaputra Basin in India, the Nile River Basin in Eastern Africa and the Volta River Basin in West Africa. In particular, it focuses on many of the reservoirs of northern India and Pakistan in the Indus and Ganges systems, Lake Nasser in the Nile River and Lake Volta in the Volta River. Information collated from grey and published literature on the three basins is synthesized and standardized with reference to wider knowledge and up-to-date information on tropical reservoir fisheries. A considerable quantity of data and information were collected on many aspects of the systems of the three reservoirs, including hydrological, biophysical and limnological features, primary production, and fish and fisheries data. This information was condensed and synthesized with the aim of providing a baseline against which the ecological changes that have taken place since impoundment can be described and analysed. Efforts are made to explain changes in fish catch in relation to climatic variations, ecological succession and fishing effort. The review shows that biological data and information are generally available. However, as is also common elsewhere, all three cases suffer from the general tendency to isolate and compartmentalize research into separate disciplines. Usually, there is very limited cross-disciplinary flow of information or recognition of how results of various disciplines can contribute to a more comprehensive understanding of the behaviour of fish populations, human communities and ecosystems and the productive activities that depend on them. This uniform tendency severely hampered the identification of relevant management actions. A more pragmatic and holistic understanding of reservoir ecosystems is needed in order to guide the choice of indicators and the development of monitoring systems that can inform management of changes in reservoir productivity and, hence, the potential catch. The next step would be to devise a hierarchy of indicators describing the different ecological and economic processes influencing fisheries catches and to organize monitoring systems around those indicators. Only by combining information across sectoral disciplines will it be possible to reach a better understanding of the processes that drive fish stocks, fisheries and reservoir productivity.