Quantitative predictions of the effects of acid deposition onterrestrial and aquatic systems require physically based, process‐oriented models of catchment soil water and streamwater chemistry. A desirable characteristic of such models is that they include terms to describe the important phenomena controlling a system's chemical response to acidic deposition, yet be restricted in complexity so that they can be implemented on diverse systems with a minimum of a priori data. We present an assessment of a conceptual model of soil water and streamwater chemistry based on soil cation exchange, dissolution of aluminum hydroxide, and solution of carbon dioxide, all processes that occur in catchment soils and that have rapid equilibration times. The model is constructed using an “average” or lumped representation of these spatially distributed catchment processes. The adequacy of the model is assessed by applying it to 3 years of soil water and streamwater chemistry data from White Oak Run, Virginia, a second‐order stream in the Shenandoah National Park. Soil properties predicted by the model are in good agreement with presently available measurements of those soil properties. The success of the model suggests that lumped representations of complex and spatially distributed chemical reactions in soils can efficiently describe the gross chemical behavior of whole catchments (e.g.,pH, alkalinity, and major ionic concentrations in surface waters). Further assessment of the adequacy of this conceptual approach will require more detailed empirical knowledge of the soil processes being modeled, particularly soil cation exchange and the variability of soil CO2partial pr