Commonly cited models for controls on variations in carbonate mineralogy are only partly successful in accounting for natural mineralogical trends. This limited success may result in part from a failure to recognize or address the ramifications of assumptions inherent to such models. In general, changes from calcite to aragonite or vice versa, and variations in MgCO3 incorporation in calcite are interpreted as resulting from: (1) kinetic rather than thermodynamic effects, (2) the effects of fluid chemistry rather than hydrologic, biological or substrate effects, (3) the effects of single parameters of fluid chemistry rather than the interactive or aggregate effects of several components, and/or (4) inorganic rather than organic parameters of fluid chemistry. Recent laboratory data suggest the necessity of a new approach to modeling controls on mineralogy. These data indicate that no single fluid parameter is likely to be a sole or dominant control on carbonate mineralogy in natural environments. In some cases, thermodynamic stability differences may be as important as kinetic factors in determining mineralogy. Furthermore, the effects of substrate, hydrologic and biological factors, and of organic compounds cannot be ruled out for most natural situations although, at present, reliable data are lacking to quantify the influence of these factors on mineralogy. Future models of controls on carbonate mineralogy need to consider the aggregate and interdependent effects of all of these factors.