The products of rock alteration and their properties must be known so that the capacity of the host rock to retain radionuclides can be predicted. The mineral and major element compositions of altered natural rock systems are known, but it is difficult to predict the sequence of host rock alteration from this information because the conditions, especially the composition of the solution phase, generally have not been determined. Knowledge of mineral behavior during rock alteration is extended by thermodynamic information, but this approach is limited by incomplete thermodynamic data for many minerals and complications resulting from kinetic and compositional factors. Additional research on naturally altered rock systems is needed to show that the results of rock alteration processes can be predicted from the thermodynamic properties of the system’s components. These studies must include complete mineral, chemical, and textural analyses of the solid phases, and solution composition and element speciation of the solution phase. The experimental difficulties of obtaining accurate thermodynamic data for complex silicates can be overcome by careful thermodynamic studies of high purity end member minerals combined with schemes that estimate thermodynamic data for members of solid solution series.