In a contemporary study of graphite deposits occurring in granulite facies, with those of Sri Lanka, India and Dillon being the prime examples, all three show similar stable carbon isotope values exhibited by both vein graphite and flake graphite dispersed in the host rock, typically a gneiss or pegmatite.

It is proposed that the graphite was deposited from a
CO₂-rich fluid, the possible sources of which may include magmatic, mantle-derived fluids, metamorphic decarbonation reactions, and devolatilization reactions of organic matter. Derivation from a mantle source is favored based on the carbon isotopic signature.

One question that has not been answered satisfactorily in the literature is the mechanism by which open cracks or vugs occur under the great pressure and stress present at the time of carbon precipation as graphite. It is clear that there were open fissures, and this resulted in unencumbered crystal growth from the walls-inward, giving rise to vein comb structures and rosettes.

Flake graphite was also deposited as disseminations in the host rock over a broad area. The flake graphite is most concentrated near the occurrence of veins.

It is theorized that the open fractures occurred as a result of differences in thermal expansion of the marble and granite gneiss. As the strata cooled the marble contracted to a greater extent than the surrounding gneiss, and this shrinkage resulted in the formation of open fractures which were held open by the high-pressure, high-temperature fluids, whether the fluids were hydrous silica melts or gasses emanating from the melts.