Recent advances in infrared speckle interferometry reported by Ridgway, et al. (1986) have made possible the determination of the temperatures at the inner radius of certain dusty outflows. When combined with recent data on the thermal annealing and hydrous alteration rates of amorphous magnesium silicate grains, this information allows one to predict that grains heated to high temperatures around stars such as NML Cygnus will be more crystalline than will cooler grains around stars like IRC +10420. In 1985, Jura and Morris (1985) showed that water vapor can condense on previously nucleated refractory grains in some stellar outflows. Stochastic heating events might provide sufficient energy to produce hydrated silicates from orginally amorphous grains provided that the loss of water from such materials does not occur too rapidly. Observable consequences of both types of grain processing are discussed.