Melting of plagioclase in the system Ab − An − $${\text{H}}_{\text{2}} {\text{O}}$$ and Qz − Ab − An − $${\text{H}}_{\text{2}} {\text{O}}$$ at P H 2 O =5 kbars, an equilibrium problem

Melting experiments were performed in the systems at P H 2 O =5 kbars. The compositions of the plagioclases coexisting with melt or with melt and quartz were determined. The data obtained in the plagioclase-water systems show that the reaction rate decreases exponentially with decreasing temperature. Plagioclase compositions do not change any further for run times ≧ 1 h at 1000° C and are considered to have reached equilibrium. At 900° C run times of approximately 200 h are necessary to approach equilibrium. At 800° C the reaction is too slow to attain constant compositions within reasonable times. The solidus of the system Qz-Ab-An-H2O was reversed using mixtures of 95% plagioclase and 5% quartz. This solidus gives the equilibrium compositions of plagioclase for any combination plagioclase-quartz-melt. A series of comparative experiments was also made with An20 and An40, mixed with quartz in the ratio plagioclase/quartz: 1/1, because theoretically the equilibrium solidus composition of plagioclase should be independent of the plagioclase/quartz ratio. However, at various run times and temperatures, the plagioclase compositions thus obtained approached the equilibrium solidus compositions only very slowly. It is extrapolated that at 730° C about 100,000 years are needed to reach equilibrium composition. The experimental data show that it is not possible to reach equilibrium in the investigated tonalite- and anorthosite-system around and below 800° C in run times available in the laboratory. It is also concluded that many investigations performed in the granite system Qz-Or-Ab-An-H2O did not reach equilibrium.