PROCESSES THAT LINK FERTILE PROTOLITH, MELT-DEPLETED CRUST AND PERALUMINOUS GRANITE
In western-Maine, transpressive deformation within a crustal-scale shear zone system during the Acadian orogenic phase generated high-T/low-P metamorphism culminating in melting. Migmatites are stromatic in belts of apparent flattening strain but heterogeneous in lenses of apparent constrictional strain, and smaller-volume granites in migmatites vary from sheets to cylinders according to structural zone, all recording syntectonic melt flow through deforming crust. The coeval Phillips pluton was sourced primarily from rocks with with isotope geochemical compositions consistent with those of the host stratigraphic succession. Migmatites have melt-depleted compositions relative to metapelites, and migmatite leucosomes have compositions that are unlike frozen primary liquids derived by muscovite-dehydration melting. We interpret leucosomes to represent cumulate products of fractional-crystallization after variable loss of an evolved liquid. Smaller-volume granites are peraluminous with a range of compositions, but none of the sample represents likely primary melt. We interpret these to be composed of variable amounts of entrained residual-cumulate quartz-plagioclase-biotite and melt that remained after loss of most evolved liquid. Common leucogranites of the Phillips pluton have a range of silica similar to smaller-volume granites, but chemical compositions that suggest crystallization of evolved liquids derived by fractional-crystallization of primary muscovite-dehydration melts. They could represent liquids extracted after up to 20% crystallization of melt derived from a fertile protolith similar to metapelites of the host stratigraphic succession.
We propose a general model involving multiple processes to relate leucosomes and smaller-volume granites in depleted migmatites to peraluminous granite in middle-upper crustal plutons. Thus, melting, fractional-crystallization and separation of residual-cumulate material from evolved liquid, and melt ascent and emplacement enable intracrustal differentiation.