The Hypersolvus-Subsolvus Classification of Low-Ca Granites Fifty Years Later: Restrict to Anorogenic Suites!

The classification of granites and related rocks into those of hypersolvus type and those of subsolvus type constitutes one of the important applications of the seminal work of Tuttle and Bowen (T&B). In a low-Ca granitic magma, did crystallization occur at a solidus that was separated from the crest of the alkali feldspar solvus (hypersolvus case), or was the crest of the solvus intersected by the solidus, so that the melt crystallized two primary alkali feldspars (one sodic, the other potassic, to give the subsolvus case)? Tuttle and Bowen proposed that a subsolvus granite could also arise by thorough low-T recrystallization of a hypersolvus rock, although this has not proven to be commonplace. Post-T&B work on the haplogranite system has demonstrated the importance of P(H2O) of crystallization; an increase in P(H2O) lowers the solidus, so that it intersects the crest of the solvus at all pressures of 3.5 kbar or so in the case of H2O saturation. The hypersolvus–subsolvus classification applies only to A-type granites (first defined 10 years after T&B) and related anorogenic felsic rocks. The coexistence of hypersolvus and subsolvus granites in some A-type plutons can best be explained by the progressive buildup in Li, B, F and P in addition to P(H2O). The additive nature of these variables in lowering the solidus effectively causes the magma to switch from a hypersolvus to a subsolvus product. Subsolidus ordering and exsolution reactions then both contribute to the shrinkage of feldspar grains, in addition to thermal contraction. The pore fluid has the ability to deposit a late feldspar, typically albite, in the resulting intergranular spaces as a “swapped margin” overgrowth. This texture resembles the expulsion of the exsolved phase illustrated with sulfides by T&B, but the hydrothermal model is considered more realistic than a purely solid-state process.