Bouguer gravity anomaly and topography data have been used to estimate the effective elastic thickness, Te, of the African lithosphere. The highest values ( Te > 100km) correlate with relatively old Archaean cratonic regions and the lowest (0 < Te < 10) with younger late Palaeozoic fold-belts and Mesozoic/Tertiary rift systems. No simple relationship exists, however, between Te and the age of the lithosphere at the time of loading. Cratons (e.g., West Africa) and fold-belts (e.g., Atlas) of similar thermal age display both high and low Te values. A good correlation exists between Te and the present-day surface heat flow over Africa: regions of low Te correlate with high heat flow whereas high Te regions have low heat flow. Thus, continental Te depends more on the present-day geotherm and, hence composition, than on the cooling history of the lithosphere. The Te and heat-flow data over the West Africa craton, Kaapvaal-Zimbabwe craton, Damara fold-belt, Niger basin, and the Central African Plateau, for example, can be explained by a thermal and mechanical model in which Te is given approximately by the depth to the 400°C isotherm. We have used the Te structure of Africa to isolate that part of the gravity field which cannot be accounted for by flexure and must be caused by other processes. The resulting anomalies, which have amplitudes of up to ±25 mGal and wavelengths of about 2000 km, correlate with regional changes in topography. There is a suggestion of a lineation of the long-wavelength gravity field in the direction of absolute African plate motion which, if correct, implies that even beneath the slow moving plates there may be some alignment of motions in the underlying manle.