Chemical differentiation of magma on Earth occurs through physical separation of liquids and crystals. The mechanisms of this separation still remain elusive due to the lack of information on solidification fronts in plutonic magmatic systems. Here, we present records of fossilized solidification fronts from massive magnetitites of the Bushveld Complex in South Africa, obtained by two-dimensional geochemical mapping on field outcrops. The chemical zoning patterns of solidification fronts indicate that nucleation and crystallization occur directly at the chamber floor and result in near-perfect fractionation due to convective removal of a compositional boundary layer from in situ growing crystals. Our data precludes the existence of thick crystal mushes during the formation of massive magnetitites, thus providing no support for the recent paradigm that envisages only crystal-rich and liquid-poor mushy reservoirs in the Earth’s crust.

An understanding of magma chamber dynamics relies on answering three important yet highly controversial questions: where, why, and how magma chambers crystallize and differentiate. Here we report on a new natural phenomenon—the undercut-embayed chamber floor in the Bushveld Complex—which allows us to address these questions. The undercut-embayed floor is produced by magmatic karstification (i.e. erosion by dissolution) of the underlying cumulates by replenishing magmas that form basal flows on the chamber floor. This results in a few metres thick three-dimensional framework of spatially interconnected erosional remnants that separate the floor cumulates from the overlying resident melt. The basal flow in this environment is effectively cooled through the floor, inducing heterogeneous nucleation and in situ growth against much of its three-dimensional framework. The solidification front thus propagates in multiple directions from the surfaces of erosional remnants. Fractional crystallization may occur within this environment by convective removal of a compositional boundary layer from in situ growing crystals and is remarkably efficient even in very confined spaces. We propose that the way magma crystallizes and differentiates in the undercut-embayed chamber floor is likely common for the evolution of many basaltic magma chambers.

the formation of some earth's monomineralic igneous rocks appears to be prohibited by constraints imposed by liquidus phase-equilibria on evolution of mantle-derived magmas. Yet, these rocks exist as stratiform layers in many mafic-ultramafic intrusions. One conspicuous example is monomineralic anorthosites in the Bushveld complex that occur as stratiform layers up to hundreds of kilometres in length. Such monomineralic anorthosites appear to require parental melts saturated in plagioclase only but where and how to produce these melts remains a contentious issue. Here we argue that they are likely sourced from deep-seated magma reservoirs. in response to pressure reduction, these ascending melts become first superheated and then saturated in plagioclase after stalling and cooling in shallow-level chambers. Adcumulus growth of plagioclase from such melts at the chamber floor results in the formation of monomineralic anorthosites. We propose that stratiform layers of monomineralic anorthosites in layered intrusions are products of the chamber replenishment by melts whose saturation in plagioclase as a single liquidus phase is triggered by their transcrustal ascent towards the earth's surface.

Layered mafic intrusions commonly contain non-cotectic, foliated igneous rocks that are traditionally attributed to processes involving settling, transport, and redeposition of crystals. Here we examine the chemistry of magnetitite layers of the Bushveld Complex using a portable XRF spectrometer on drill core and dissolution ICP-MS analysis on pure magnetite separates. While magnetitites contain foliated plagioclase grains in non-cotectic proportions, the magnetite is characterized by a regular upwards-depletion of Cr which is best explained by in situ crystallization. We suggest that plagioclase nucleation in thin residual compositional boundary layers atop a solidification front causes in situ growth of plagioclase in proportions much lower (<10%) than those expected from cotectic crystallization (±85%). Crystallization in such a boundary layer also favours lateral growth of the plagioclase, producing the foliation. We suggest that some non-cotectic, foliated rocks that are commonly interpreted to arise from gravity-induced sedimentary processes may instead be produced by in situ crystallization.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.