‘Postcumulus Processes Recorded in Whole Rock Geochemistry: A Case Study from the Mirabela Layered Intrusion, Brazil’ by S.J. Barnes and M. Williams: Reply to Comment by A.E. Boudreau

Abstract In our 2024 paper, we investigated a database of more than 60 000 whole-rock analyses of a pyroxenite–harzburgite–gabbronorite interval across the entire width on the Mirabela Layered Intrusion. We concluded that during the solidification of these cumulates they behaved as closed systems with respect to intercumulus components, where sufficiently large samples (3- to 10-m drill core composites) were considered. Boudreau (2025) reinterprets individual drill holes sampled at 1-m intervals to reach the opposite conclusion from ours. He claims that detailed geochemical trends in the plagioclase-excluded element P relative to plagioclase-compatible components Na and Sr indicate that residual trapped liquid was preferentially removed from the cumulate pile during compaction in the centre of the intrusion. We show that this conclusion is incompatible with trends in element ratios across the intrusion, particularly in the ratios of Al and Na to Ti. Concentrations of Ti show no evidence of depletion relative to Na and Al and correlate strongly with plagioclase-compatible elements throughout. Our original conclusion remains robust on this evidence. However, we find that Ti and P are decoupled to some extent and therefore agree that some selective mobility of P probably has occurred, although part of this apparent decoupling may be due to nugget effects in low-level P analyses. Detailed microstructural studies are needed to supplement a purely geochemical approach.