The electrical properties of granite appear to be dominantly controlled by the amount of free water in the granite and by temperature. Minor contributions to the electrical properties are provided by hydrostatic and lithostatic pressure, structurally bound water, oxygen fugacity, and other parameters. The effect of sulfur fugacity may be important but is experimentally unconfirmed. In addition to changing the magnitude of electrical properties, the amount and chemistry of water in granite significantly changes the temperature dependence of the electrical properties. With increasing temperature, changes in water content retain large, but lessened, effects on electrical properties. Near room temperature, a monolayer of water will decrease the electrical resistivity by an order of magnitude. Several weight-percent water may decrease the electrical resistivity by as much as 9 orders of magnitude and decrease the thermal activation energy by a factor of 5. At elevated temperatures just below granitic melting, a few weight-percent water may still decrease the resistivity by as much as 3 orders of magnitude and the activation energy by a factor of 2. Above the melting temperature (650¿ to 1100¿C depending upon water pressure), a few weight-percent water will decrease the resistivity by less than an order of magnitude and will barely change the activation energy. Remarkably, the few weight-percent water must be present as free water. Experiments with hydrated hornblende schist (with structural water) indicate an electrical resistivity very similar to that for dry granite. The implications of these results, together with the findings of deep magnetic sounding and magnetotelluric surveys, suggest much more free water than is commonly associated with the lower crust and possibly into the upper mantle.