The Bushveld Complex Did Not Form from Externally Derived Crystal Slurries, Whether Crystal-poor or Crystal-rich

The Bushveld Complex is a classic example of an open-system magma chamber that grew incrementally through numerous replenishment events, with parental magmas sourced from deep crustal to mantle reservoirs. A central question is whether these replenishing magmas were crystal-free melts (normal to superheated) or crystal-bearing slurries (crystal-poor to crystal-rich). Here we address this question by integrating three independent lines of evidence: (1) depth and pressure of the staging chamber, (2) density and rheology of orthopyroxene-laden slurries, and (3) chemistry of cumulus orthopyroxene. First, regional geology indicates that the Bushveld Complex was emplaced into the ~23 km thick Transvaal Basin, with the floor of the intrusion lying at a depth of ~10–12 km (~3.0–3.5 kbar). The lack of intrusive remnants of staging chambers anywhere in the basin succession indicates that the staging chambers feeding the Bushveld Complex must have been located beneath the basin. Recent seismic and gravity imaging identifies the remnants of this staging chamber at depths of ~40–45 km (~10 kbar), thereby defining the most likely pressure conditions under which any parental melts or crystal slurries must have originated. Second, crystal slurries originating in this lower-crustal staging chamber could reach the shallow Bushveld chamber only if they were sufficiently buoyant to rise through the Transvaal Basin sediments (average density ~2700 kg m −3). Density considerations indicate that only crystal-free to crystal-poor melts satisfy this requirement, whereas the proposed crystal-rich slurries (from 22 mass% orthopyroxene to 62 vol % chromite) have densities of ~2790–3810 kg m −3 and are therefore too dense to ascend through the basin. Where orthopyroxenite layers are interpreted as sills, inferred crystal contents in the parental slurries (49–98 vol %) overlap and substantially exceed the rheological lock-up range, precluding their mobility and further challenging their emplacement. Third, if crystal slurries transported orthopyroxene crystals from a ~10 kbar staging chamber, those crystals should preserve the chemical signature of high-pressure crystallisation. Experimental crystallisation of Bushveld-type magmas at ~10 kbar produces orthopyroxene enriched in Al2O3 (~3.5–7.0 wt %) and Cr2O3 (~1.7–2.2 wt %), providing a quantitative benchmark for high-pressure orthopyroxene. In contrast, cumulus orthopyroxene throughout the entire Bushveld Complex is consistently low in Al2O3 (<1.5 wt %) and Cr2O3 (<0.5 wt %). This discrepancy cannot be explained by post-cumulus re-equilibration, as all plausible mechanisms capable of lowering Al2O3 and Cr2O3 in high-pressure orthopyroxene are inconsistent with the petrographic and chemical characteristics of Bushveld orthopyroxene. This chemical constraint also excludes crystal-poor slurries carrying deep-crystallised orthopyroxene, even when their low crystal contents permit buoyant ascent. Finally, crystal slurry models require abundant phenocrysts and/or antecrysts, yet convincing field, textural, and chemical evidence for this externally derived population of crystals has not been presented from Bushveld cumulates. This proposed crystal cargo therefore remains a postulate rather than a demonstrated feature of the Bushveld rocks. Taken together, the above three constraints, reinforced by the lack of evidence for the postulated phenocrysts in the rocks, rule out models in which the Bushveld Complex was assembled from externally derived crystal slurries. Instead, they indicate that the Bushveld Complex was built by repeated injections of crystal-free melts, with nearly all crystals forming directly within the shallow Bushveld magma chamber under low-pressure conditions.