Boreholes have intersected the interval hosting the Pyroxenite Marker at the base of the Upper Main zone on the farm Moorddrift in the northern Bushveld Complex. The concentrations of the major and compatible trace elements in the rocks are in the range of other Pyroxenite Marker intervals elsewhere in the Bushveld Complex (MgO up to 22 wt %, Cr up to 300 ppm). Over a stratigraphic height of some 300 m the interval is characterized by a broad, but irregular reversal in Mg number and Cr/V, analogous to stratigraphically equivalent horizons elsewhere in the Bushveld Complex. The rocks have distinctly higher concentrations in incompatible trace elements, particularly large ion lithophile elements (LILE) and light rare earth elements (LREE), than other Main zone rocks and thus appear to contain an elevated crustal component. The interval hosts several thick (meters to 10s of meters), but mostly relatively low grade and laterally discontinous PGE reefs that have been delineated over a strike length of ca 1 km. PGE contents locally reach ~16 ppm Pt + Pd + Au in rocks containing up to ~4 to 5 percent sulfides. The PGE tenors of most sulfides are >100 ppm, Pt/Pd ratios are above unity, and Pd/Ir is >200. Based on well-defined positive correlations between Pt and Pd with Ir we interpret the mineralization as being largely of primary magmatic origin.
Most previous models interpreted the Pyroxenite Marker to have formed from a major replenishment to the Bushveld magma chamber with relatively primitive magma. The new magma would have been emplaced on top of an ~4-km-thick cumulate package consisting of the ultramafic-mafic Lower and Critical zones and the lower to central portions of the gabronoritic Main zone. Here, we propose that the Upper Main zone including the Pyroxenite Marker did not result from a magma replenishment. Instead, it crystallized from the S-saturated residual magma of the upper Critical zone cumulates and originally directly overlaid the latter. The ~1-to 2-km-thick lower-central Main zone package formed from a crystal-charged magma emplaced at a relatively late stage in the solidification history of the Bushveld chamber, between the largely consolidated Upper Critical and Upper Main zones. We envisage that the emplacement of this large magma batch resulted in subsidence of the center of the layered suite, perhaps in response to the evacuation of a putative lower staging chamber from which the Bushveld Complex was fed. Semiconsolidated portions of the cumulate package sagged and slumped toward the center of the Complex, and the flowing crystal mushes underwent mechanical unmixing of light, relatively buoyant plagioclase from denser pyroxenes, oxides, and sulfides. This resulted in the formation of an interlayered sequence of relatively melanocratic sulfide-enriched with relatively leucocratic sulfide-poor layers. reef formation is thus partly the result of cumulate reorganization in response to the emplacement of the Main zone magma.