The benthos of Lake Sibayi, Lake Nhlange, Lake Shengesa and Lake Sifungwe has been sampled by means of a van Veen grab of bite area 0.225m². Except for Lake Shengesa, the substrates of the lakes have been investigated visually by SCUBA divers. Some quantitative sampling, using underwater techniques, has been carried out in Lake Sibayi. In general, the substrate of the lakes consisted of two types, either fine sand with an average particle size of 3.2 phi, or mud with a median phi value of less than 6. The sandy substrata was usually rich in fauna, and the mud usually poor. Lake Sibayi , which is freshwater (135 ppm Cl¯) has estuarine and freshwater species in the benthos. Some of the species, notably Grandidierella lignorum and Apseudes digitalis, show an uneven pattern of distribution with respect to depth. Lake Nhlange is a brackwater lake (3.4‰ salinity) and has mainly an estuarine fauna. Many of the species are different from those found in Lake Sibayi. There is some evidence that the benthic fauna of the lake has increased its range in depth in the years following a flood in January 1966. Only chironomid larvae were found in the benthos of Lake Shengesa. The greater part of the substrate of this lake consisted of a highly organic mud mainly derived from burnt grassland around the periphery of the lake. Lake Sifungwe, which is connected to Lake Nhlange, has a halocline at about 9 metres. The fauna of the benthos has more species than lake Nhlange. There is some evidence to suggest that LGC fauna invades deeper water in summer, and is driven into shallower water in winter because of the fluctuating level of anaerobic conditions below the discontinuity layer. The benthic fauna of the lakes has been compared with that of the estuaries reported on by the Cape Town Ecological Survey. experimental studies on G. lignorum, a species present in Lake Sibayi, suggested that the physiological responses of material from an estuary normally at 35‰ salinity (Kowie River estuary) and from lakes which are normally of low salinity (Groenvlei 2.5‰ salinity, Lake Sibayi) were not different. Studies on substrate choice, and pressure, indicated that these factors do not affect the distribution pattern of G. lignorum in the substrate. Studies on light orientation responses and the effects of rates of carbon dioxide increase of about 20 mm HgC0₂/hr or more, suggest that these may be the principle factors affecting the distribution of G. lignorum in the benthos of Lake Sibayi. Some supporting evidence for the presence of changing pC0₂ has been found. A discussion of the importance of dynamic events at the substrate/water interface is given.