Abstract Diamond mega-placers, defined as ≥ 50 million carats at ≥ 95% gem quality, are known only from along the coast of southwestern Africa, fringing the Kaapvaal craton, where two are recognized. One is associated with the Orange-Vaal dispersal, the other, to the south, has an uncertain origin. Placers are residual when left on the craton, transient when being eroded into the exit drainage, and terminal. Terminal placers, the final depositories of diamonds, have the greatest probability of being a mega-placer. There are four main groups of controls leading to the development of a mega-placer: the craton, the drainage, the nature of the environment at the terminus and the timing. Cratons, being buoyant, have a tendency to leak diamonds into surrounding basins; however, being incompressible they may have orogens converge onto them resulting in some lost sediment being returned as foreland basin fills. The craton size, its diamond-fertility and the retention of successive kimberlite intrusions that remain available to the final drainage, are significant to mega-placer development. Maximum potential recovery is achieved when the drainage delivering diamonds to the mega-placer is efficient, not preceded by older major drainages and focuses the supply to a limited area of the terminal placer. There should be sufficient energy in the terminal placer regime to ensure that sediment accompanying the diamonds is removed to areas away from the placer site. All conditions should be near contemporaneous and most were satisfied in the Orange-Vaal Rivers-Kaapvaal system and mega-placers were consequently generated.

This outline of the topographic evolution of Africa tied to the history of the African Surface illustrates how a unique geomorphic history over the past 180 million years reflects the continent's distinctive tectonics. The African Surface is a composite surface of continental extent that developed as a result of erosion following two episodes of the initiation of ocean floor accretion around Afro-Arabia ca. 180 Ma and 125 Ma, respectively. The distinctive tectonic history of the African continent since 180 Ma has been dominated by (1) roughly concentric accretion of ocean floor following those two episodes; (2) slow movement of the continent during the past 200 m.y. over one of Earth's two major large low shear wave velocity provinces (LLSVPs) immediately above the core-mantle boundary; (3) the eruption during the past 200 m.y. of deep mantle plumes that have generated large igneous provinces (LIPs) from the core-mantle boundary only at the edge of the African LLSVP; and (4) two episodes during which basin-and-swell topography developed and abundant intracontinental rifts and much intra-plate volcanism occurred. Those episodes can be attributed to shallow convection resulting from plate pinning, i.e., arrested continental motion, induced by the successive eruption of the Karroo and Afar plumes. Shallow convection during the second plate-pinning episode generated the basins and swells that dominate Africa's present relief. By the early Oligocene, Afro-Arabia was a low-elevation, low-relief land surface largely mantled by deeply weathered rock. When the Afar plume erupted ca. 31 Ma, this Oligocene land surface, defined here as the African Surface, started to be flexed upward on newly forming swells and to be buried in sedimentary basins both in the continental interior and at the continental margins. Today the African Surface has been stripped of its weathered cover and partly or completely eroded from some swells, but it also survives extensively in many areas where a lateritic or bauxitic cover has accordingly been preserved. Great Escarpments, which are best developed in the southern part of the continent, have formed on some swell flanks since the swells began to rise during the past 30 m.y. They separate the high ground on the new swells from low lying areas, and because they face the ocean at some distance from the African coastline, they mimic rift flank escarpments at younger passive margins. The youthful Great Escarpments have developed in places where the original rift flank uplifts formed at the time of continental breakup. Their appearance is therefore deceptive. The African Surface and its overlying bauxites and laterites embody a record of tectonic and environmental change, including episodes of partial flooding by the sea, during a 150-million-year long interval between 180 Ma and 30 Ma. Parts of African Surface history are well known for some areas and for some intervals. Analysis here attempts to integrate local histories and to work out how the surface of Afro-Arabia has evolved on the continental scale over the past ∼180 m.y. We hope that because major landscape development theories have been spawned in Africa, a review that embodies modern tectonic ideas may prove useful in re-evaluation of theory both for Africa itself and for other continents. We recognize that in a continental-scale synthesis such as this, smoothing of local disparities is inevitable. Our expectation is that the ambitious model constructed on the basis of our review will serve as a lightning rod for elaborating alternative views and stimulating future research.