Abstract

The concept of compositional bimodality in carbonatites has become widely accepted and has been used to impose restrictions on the composition of carbonatite magmas. We agree that mineralogical bimodality exists in carbonatites (most are either calcitic or dolomitic/ankeritic), but we argue that there is no compositional bimodality. The idea of bimodality is based on the interpretation of a variety of element distribution diagrams which were compiled only from chemical analyses in which SiO 2 is < 10 wt.%. All others were rejected. Even with such a restricted data set the case for compositional bimodality is extremely weak, but the inclusion of analyses with higher SiO 2 content destroys it completely. Yet these more siliceous compositions must be included, for many carbonatites contain substantial amounts of Fe–Mg silicates which are an essential part of the magmatic mineralogy of the rocks. They account for much of the Mg in carbonatites that are otherwise calcitic. Many such carbonatites contain well in excess of 10 wt.% SiO 2. Supporters of the bimodality concept argue that liquids having compositions between calcite and dolomite can precipitate neither calcite nor dolomite because the minimum on the solid solution loops in the system calcite–dolomite permits only a carbonate of intermediate composition. Therefore, it is argued, liquids of such intermediate composition cannot be parental to calcitic and dolomitic carbonatites; their parent magmas must be calcitic and dolomitic. This deduction is incorrect. It is well established that dolomitic liquids have calcite as the liquidus phase over substantial temperature intervals, and that this is followed by dolomite precipitation. Mixed calcite–dolomite carbonatites are explicable in this way. Therefore, dolomitic liquids can be parental to calcitic carbonatites. However, dolomitic carbonatites cannot crystallize from a calcitic liquid. We suggest that intermediate composition carbonatite magmas are probably common. Bimodality in carbonatites is solely mineralogical, not compositional.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call