The Merensky Reef in retrospect: a century of challenges, controversies, and conceptual shifts

Abstract The Merensky Reef (MR) of the Bushveld Complex is a thin package of broadly orthopyroxenitic rocks that contains sulphide and chromite that are extremely enriched in platinum-group elements (PGE). Since its discovery in 1924, it has become renowned for hosting the lion’s share of the world’s PGE reserves. Over time, ideas about the origin of the MR have significantly evolved and can be broadly grouped into three time periods. From 1924 to 1980, abundant information was accumulated on the geological setting and structure of the MR. The period was dominated by three unrefined approaches for the origin of the MR: (a) a new magma pulse emplaced into the chamber, (b) a sill intruded into solidified rocks and (c) trapping of hydrothermal fluids below an impermeable barrier. From the 1980s to 2010, a lively debate erupted between the proponents of two distinct schools of thought - ‘uppers’ and ‘downers’. The ‘downers’ model posits that mixing of new magma with resident melt leads to the separation of droplets of immiscible sulphide liquid which equilibrate with the convecting silicate melt, scavenging significant amounts of PGE due to their high sulphide/silicate melt partition coefficients. In this model, the MR forms by gravity settling of sulphide droplets and crystals of silicate minerals onto the chamber floor. In contrast, the ‘uppers’ model suggests that PGE-rich interstitial melt or fluid percolates upward from the underlying cumulates, resorbing primary PGE-bearing sulphides and stripping the lower cumulates of PGE, thus providing a source for PGE mineralisation higher in the sequence. In this model, the MR forms by precipitation of sulphides at some depth within the cumulate pile. Finally, from about 2010 onward, three contrasting concepts began to compete: (a) a crystal slurry, (b) a sill-like intrusion, and (c) an in situ growth. The first model proposes that the MR formed by accumulation of sulphides and chromite at the base of a crystal-rich slurry flowing along the chamber floor. The second model suggests that the MR was produced from a late-stage sill that intruded into the pre-existing cumulate pile. Lastly, the third model argues that the MR formed from a pulse of crystal-free magma that entered the chamber as a basal flow followed by in situ crystallisation directly on the chamber floor. We hope that the ongoing debate between proponents of these three distinct hypotheses, along with earlier models, may culminate in a resolution of the enduring enigma of the Merensky Reef in the Bushveld Complex.