Stabilization of metastable aragonite and Mg-calcites is an inevitable process and constitutes a large and important part of carbonate diagenesis. Stabilization can occur in only a limited number of ways. "Solid state" stabilization takes place by inversion or exsolution for aragonite and Mg-calcite, respectively. Stabilization can take place by total dissolution and reprecipitation, or in the case of the Mg-calcites, by incongruent dissolution yielding a replacement product of calcite plus a solution enriched in magnesium. Lastly, stabilization can take place by replacement by either dolomite or a non-carbonate phase. The processes of stabilization which skeletal reactants undergo have been investigated by hydrothermal techniques, and the products of the reactions studied in thin sections. Aragonite inversion produces coarse-grained equigranular mosaics, and the rate of inversion of skeletal material seems to be only temperature-dependent near 300 degrees C. Non-skeletal aragonite is less reactive. Mg-calcite exsolution is very slow compared to aragonite inversion, and the phase which exsolves is a very calcium-rich "disordered" protodolomite. The kinetics of dolomitization have been studied semiquantitatively near 300 degrees C in aqueous solution. Dolomitization is speeded by: increased instability of the reactant, increased calcium plus magnesium concentration of the dolomitizing solution, increased magnesium/calcium ratio of the dolomitizing solution, d. increased solution/solid ratio, and increased temperature. Protodolomite, rather than dolomite, forms as the system becomes more dilute in magnesium. Analytical "steady state" solubilities of skeletal reactants in distilled water and in sea water are presented. Long term experiments suggest that Mg-calcites undergo an incongruent dissolution to produce calcite plus a solution enriched in magnesium. The different stabilities of the various kinds of skeletal reactants permit different mechanisms of diagenesis to operate, and allow a variety of products to form. Stabilization seems, in many cases to be an early process in the history of carbonate rocks, and therefore preferential reactions of metastable phases indicate an early age for the reactions. We are not yet sufficiently astute to interpret all the various products.