Constraints on the abundance of eclogite in the upper mantle

Abstract

Examination of heavy mineral concentrates from several well‐known eclogite‐dominated kimberlites (Roberts Victor, Bobbejaan, Zagadochnaya) has revealed that the upper mantle sampled by these kimberlites was dominantly peridotite, with only a small amount of eclogite (3–15% by volume). In the case of Group II (micaceous) kimberlites in southern Africa (e.g., Roberts Victor), it is proposed that garnet peridotites were metasomatized (phlogopitized) within the upper mantle prior to xenolith incorporation. This caused structural weakening, with the result that most of the garnet peridotites disaggregated during kimberlite ascent and eruption, leaving an eclogite‐dominated xenolith suite. The amount of eclogite in the upper 200 km of the subcontinental upper mantle is concluded to be perhaps <1% by volume overall. Support for models proposing a significant amount of eclogite at deeper levels or beneath ocean basins must be sought from evidence other than mantle‐derived xenoliths.