We propose a new model to describe precipitation-induced permeability reduction during chemical diagenesis of sandstones. The model has two parts: (1) geochemical model for determining the expected mass of precipitates and (2) physical model relating how a given mass of precipitates affects permeability. The model is used to study the experiments reported by Tenthorey et al. , where reactions involving dissolution of quartz and feldspar and precipitation of authigenic minerals were observed to cause drastic permeability reduction in feldspathic sandstone. We find that the geochemical evolution of the system dictates permeability evolution, and thus permeability dependence on temperature and stress is related to geochemical responses to these variables. The model predictions for the experimental system of Tenthorey et al. provide a very good fit to the experimental permeability curves over all times, stresses, and temperatures within the experimental range. Following the experiments and model, we suggest that precipitation of secondary mineral phases is an extremely efficient method for permeability reduction but one that depends critically on the system's approach to equilibrium. The reaction products of chemical diagenesis observed in rocks are the consequence of chemical reactions that occur in natural systems as they approach equilibrium, and we predict extensive and rapid permeability reduction as a result of such reactions. ¿ 1998 American Geophysical Union