Southern Africa's high Kalahari plateau and its flanking mountain ranges formed over an extended period of ~200 million years through vertical tectonic processes different from those at convergent plate boundaries. We refer to this as the Kalahari epeirogeny. Episodic uplift and erosion during the Kalahari epeirogeny is inferred from thermo-chronology and from stratigraphy of sediment accumulated around its continental margin. A total thickness of 2 to 7 km of rock (of which basalt was a major component) was eroded from the subcontinent's surface during two punctuated episodes of exhumation, in the early-Cretaceous and in the mid-Cretaceous. Major exhumation was over by the end of the Cretaceous, and the rate of erosion decreased by more than an order of magnitude to remove less than 1 km thickness of rock during the Cenozoic. Cosmogenic dating shows that erosion rates today are almost an order of magnitude less still.The cause for the Kalahari epeirogeny remains elusive, but three striking observations stand out: (i) major exhumation occurred during the late Mesozoic break-up of Gondwana; (ii) large scale basaltic magmatism closely match two accelerated episodes of exhumation: first along the west coast (~132 Ma, Etendeka large igneous province [LIP], associated with vast amounts of underplating along this entire passive margin), and the second flanking the sheared south coast margin (~90 Ma, Agulhas oceanic LIP). Yet exhumation that might have accompanied the vast Karoo LIP (~180 Ma) is not readily detected; (iii) the two main regional episodes of accelerated exhumation and coastal sediment accumulation are near synchronous with two regional 'spikes' of kimberlite intrusions: >450 kimberlites at ~90 Ma, and >200 kimberlites at ~120 Ma. Southern Africa is underlain by an anomalous warm region in the lowermost ~1500 km of the mantle that is linked to core to mantle heat loss. This warm region was inherited from a large Cretaceous thermo-chemical anomaly in the mantle created during long-lived subduction beneath Gondwana. Associated Mesozoic kimberlite genesis and basaltic magmatism may have been sufficient to cause volatile/heat-induced density changes in the lithospheric mantle of southern Africa to sustain the Kalahari plateau. In addition, such changes may have been induced by tectonic uplift and decompressional melting resulting from far-field collision processes between Africa and Europe, and/or final continental lithosphere decoupling between the Falkland plateau and southern Africa. CO 2 released into the ocean-atmosphere system during the Kalahari epeirogeny contributed to the global mid-Cretaceous hot-house conditions. However, because the CO 2 -consumption rate associated with basalt weathering is about eight times that of granite, atmospheric CO 2 was also efficiently sequestrated during the rapid erosion of Karoo basalt. Thus, the Kalahari epeirogeny also may have catalysed the onset of long-term global Cenozoic cooling.