An account is given of an ecological investigation of freshwater habitats near Ibadan, Nigeria, which was designed to identify and evaluate key factors that might be manipulated to achieve control of the snail hosts of schistosomiasis and fascioliasis. Statistical analyses of data from 24 stations from both lentic and lotic habitats showed that there were statistically significant tendencies for the following snail species to coexist: at the macro-distribution level Lanistes libycus and Bulinus (P.) globosus, in lotic habitats, Lymnaea natalensis, Bulinus forskali, Biomphalaria pfeifferi and Bulinus rohlfsi in lentic habitats; at the micro-distributional level, B. rohlfsi, B. pfeifferi, B. forskali and Anisus coretus. Several species of aquatic macrophytes, including species that occur at water margins (Acroceras and Commelina spp.) and those that occur in more persistent aquatic habitats (Nymphaea and Ceratophyllum spp.) also coexist to a statistically significant extent both with each other and with snail species in the same microhabitats. However, Salvinia proved exceptional as it was negatively associated with four of the snail species. Persistent snail refugia, including areas at stream origins, backwaters and the deeper bays or inlets in lentic habitats, were characterized by the presence of particular macrophyte species. The reasons for the associations between snails and macrophytes are discussed and it is argued that they may be mutualistic. The snails proved tolerant to a wide range of chemical conditions with the exception of low oxygen concentration. However, the absence of aquatic macrophytes and lentic refugia in head waters and the presence of fast flowing water and high temperatures in shallow water are harmful. Although the indigenous snails, L. libycus and Aplexa waterloti, proved to be efficient predators on the eggs and juveniles of snail hosts, other laboratory and field evidence appear to preclude the possibility that they may be useful as biological control agents. In contrast, there is evidence that Melanoides tuberculata might exclude the snail hosts from eroding substrates in lentic habitats. The following integrated control measures in lotic habitats are discussed: (a) judicious drainage and canalization of marshy areas at stream origins, combined with replacement of key marginal macrophytes; (b) dredging and straightening of stream courses; (c) replacement of key macrophytes along the banks; (d) introduction of M. tuberculata if not present. In lentic habitats the following control measures are discussed: (a) prevention of immigration of both macrophytes and snails from feeder streams; (b) control of subaquatic emergent plant species by using grazing domestic animals or competitor species; (c) maximizing wind-generated surface currents to exclude floating plants; (d) eradication of submerged rooting plants by herbivorous fish or invertebrates; (e) the management of such water bodies as either plankton - herbivorous fish or plankton - herbivorous fish - carnivorous fish systems; (f) introduction of efficient, detritivorous non-host molluscs or fish to feed on bottom deposits.