A review and classification of South African platinoid-bearing ores leads to a documentation of the main problems of the geochemistry of the platinum group metals.The question of the lattice substitution of base metal cations by the platinum metals is investigated but is not supported. In spite of the apparent feasibility of this process on theoretical grounds, the authors point to the fact that all available evidence does not corroborate either an oxyphile or lithophile tendency for the platinum metals. Furthermore, studies on chrome spinels, in which platinoids become concentrated under certain conditions in igneous rocks, indicate rather that the platinoids exist as discrete minerals or compounds contained as minute inclusions, probably down to the colloidal scale, within the chromite.The most preferential collection of the platinoids is found to occur within base metal sulfides, but the siderophile tendency of platinum and the chalcophile behavior of palladium affect their separation in the sulfide environment of igneous rocks. This noncoherence of the principal platinoid metals constitutes the main theme of the paper and is related to ascending iron enrichment and acidification within igneous masses, termed the Bushveld and Sudbury trends, respectively. Any mobilization of primary ores or protores, regardless of the agency, leads to the palladium enrichment of the Sudbury trend.The behavior of the platinoids in the minor exogenic cycle is examined. Some of the main conclusions reached include the fact that palladium suffers depletion during weathering and transport. The other platinoids are reduced to the metals which aggregate and accrete during weathering. Transport of the metals into fresh-water placers leads to platinum enrichment, but marine placers show concentrations of osmium and iridium due to selective leaching of the other platinoids by sea water. The leached metals, including palladium, report in deep-water sediments.