Although many hydrous magmas may have phenocrysts that define silica activity (aSiO2) (e.g., quartz or olivine + orthopyroxene), it is only possible to determine γSiO2 and XSiO2 if the concentration of water in the magma can be established when the phenocrysts equilibrated. Thus experiments on hydrous magmas with known water concentrations are essential if RTlnγSiO2 is to be quantified. Using the relationship between RTlnγSiO2 and the composition of the magma, including water, the activities of silica in magmas can be set equal to either those defined by a mantle source, or to a phenocryst assemblage such as quartz. For intermediate magmas (52–63 wt% SiO2) from the Mexican volcanic belt, it can be shown that hydrous magmas with less than ~56 wt% SiO2 and more than 6.5 wt% MgO, could have equilibrated with a mantle source, as represented by olivine + orthopyroxene in lherzolitic nodules carried to the surface by an andesitic lava. These limits are imposed by water saturation of magmas at 10 kb, which is a poorly known quantity. So indeed is the partial molar entropy of water that controls the thermal response to the exsolution of a fluid phase on magma ascent.
In magmas bearing quartz phenocrysts, quartz can be used as a geobarometer, and the indicated pressures (±1.2 kb) for three samples of the Bishop Tuff agree with the water saturation curves at the temperatures of the Fe-Ti oxides, which indicate a pre-eruption magma column of ~1.5 to 3.3 kms. For Katmai rhyolitic ejecta, there are inconsistencies between the measured water concentrations of the glass inclusions, the pressures derived from the water saturation curves at the Fe-Ti oxide temperatures, experimental phase equilibria, and the pressures derived from the proposed quartz barometer.