The origin of garnet in the anorthosite-charnockite suite of the Adirondacks

Detailed analysis of textural and chemical criteria in rocks of the anorthosite-charnockite suite of the Adirondack Highlands suggests that development of garnet in silica-saturated rocks of the suite occurs according to the reaction: {Mathematical expression}, where ?? is a function of the distribution of Fe and Mg between the several coexisting ferromagnesian phases. Depending upon the relative amounts of Fe and Mg present, quartz may be either a reactant or a product. Using an aluminum-fixed reference frame, this reaction can be restated in terms of a set of balanced partial reactions describing the processes occurring in spatially separated domains within the rock. The fact that garnet invariably replaces plagioclase as opposed to the other reactant phases indicates that the aluminum-fixed model is valid as a first approximation. This reaction is univariant and produces unzoned garnet. It differs from a similar equation proposed by de Waard (1965) for the origin of garnet in Adirondack metabasic rocks, i.e. 6 Orthopyroxene+2 Anorthite = Clinopyroxene+Garnet+2 Quartz, the principle difference being that iron oxides (ilmenite and/or magnetite) are essential reactant phases in the present reactions. The product assemblage (garnet+clinopyroxene+plagioclase ?? orthopyroxene ?? quartz) is characteristic of the clinopyroxene-almandine subfacies of the granulite facies. ?? 1977 Springer-Verlag.