The genesis of kimberlite: geochemical constraints

Deciphering the geochemical signatures of kimberlites, with the aim of constraining their source regions, has long been a “holy grail” in kimberlite studies. The complex, hybrid nature of kimberlites has contributed to the broad spectrum of petrogenetic models available that range from ultra-deep asthenospheric melts to more classical models that involve melting of the continental lithospheric mantle. Recently, models that involve both asthernospheric and lithospheric source components have evolved, with the lithosphere either acting as a “contaminant” e.g., Nowell et al. (2004) or as the sink for asthenosphere-derived metasomatic agents that later re-melt to yield kimberlite (le Roex et al., 2003; Becker and le Roex, 2006). These latter models show that the incompatible trace element patterns of S. African kimberlites can be generated by small degrees of melting (0.4 to 1.5%) of a metasomatised garnet lherzolite source that has been enriched by a melt already possessesing kimberlite-like trace element characteristics. These, along with all other models, are non-unique and blur the issue of the ultimate source of kimberlitic melts that move through the mantle and appear to be important in diamond genesis (Schrauder & Navon, 1996; Klein benDavid et al., this volume).