The origin of calcite, Mg-calcite, and aragonite cements formed near the water table and in the intertidal zone of tropical and subtropical beach sediments has been the subject of extended debate. Following Field (1919), it is proposed here that much of this cement is precipitated as a consequence of loss of CO 2 from carbonate-saturated beach groundwaters. Mass-transport calculations support the further proposal that vertical fluid dispersion in the phreatic zone resulting from tidal oscillation of the water table is sufficient to induce degassing of CO 2 from a seaward-flowing groundwater. Loss of CO 2 is further enhanced by tidal pumping of the gas phase in the vadose zone across the sediment-atmosphere interface. As sediment porosity is lowered by precipitation of cement, the ability of the groundwater system to degas and form new cements is reduced. As long as the system remains thermodynamically open with respect to CO 2 and close to saturation with respect to calcite or aragonite, however, it remains an unlikely site for the precipitation of dolomite. The hypothesis that degassing alone is sufficient to cause supersaturation and cementation is supported by an experimental study of degassing of mixed beach and marine waters from St. Croix, U.S. Virgin Islands, in which 30 mu m-thick, low-Mg calcite crusts were formed which closely resemble natural water-table cements. The waters were spiked with Hg 2 Cl 2 to retard biological involvement. The maximum observed rate of precipitation of calcite in solutions in direct contact with the atmosphere was 10 (super -9) moles cm (super -2) sec (super -1) , which is a rate sufficient to indurate a beach sediment within a period of 12 hours. Thermodynamic calculations indicate that precipitation could not have been induced by mixing of marine and meteoric waters, as has been proposed by others, but that loss of CO 2 , a process independent of mixing, was necessary to cause supersaturation.