Glass‐bearing inclusions in Shergotty and Chassigny: Consistent samples of a primary trapped melt?

Glass‐bearing inclusions hosted by different mineral phases in SNC meteorites provide important information on the conditions that prevailed during formation of early phases and/or on the composition of the primary trapped liquids/melts of these rocks. Although extensive previous work has been reported on such inclusions, several questions are still unresolved. We performed a chemical and petrographic study of the constituents (glasses and mineral assemblage) of glassy and multiphase inclusions in Shergotty and Chassigny. We focused on obtaining accurate trace element contents of glasses and co‐existing minerals and discussing their highly variable REE contents. Our results reveal an unusual geochemistry of trace element contents that appear to be independent of their major element compositions. Chemical equilibrium between phases inside inclusions as well as between glasses and host minerals could not be established. The LREE contents of glasses in glass inclusions can vary by up to two orders of magnitude. The depletion in trace element abundances shown by glasses seem to be inconsistent with these phases being residual melts. The light lithophile element contents of glasses are highly variable with enrichment in incompatible elements (e.g., Be, Sr, Ba, and LREE) indicating some processes involving percolation of fluids. All of these features are incompatible with glass‐bearing inclusions in the host minerals acting as closed systems preserving unmodified primary liquids/melts. Glass‐bearing inclusions in Shergotty and Chassigny appear to have been altered (as was the rock itself) by different postformational processes (e.g., shock, metamorphism, metasomatic ?] fluids) that affected these meteorites with different degree of intensity. Our results indicate that these inclusions could not preserve a reliable sample of the primary trapped melt.