MHCO,) PHZS (MHS) PCO, K3 = These 'I constants " will depend upon thezconcentration and ~ .are brieRy reviewed, and the results are plotted in Fig. I after recalculation to a common basis (concentrations in g. mols./ litre, partial pressures in mm. Hg.). The numbers refer to literature cited at the end of the paper. Considerable discrepancies are noticeable, and the dependence of the constants upon temperature and concentration has not been systematically explored. The original paper of reference' is not available at the time of publication, so that results cannot be included in Fig. I. Values of the ionisation and solubility wnsfanfs.-The best published values give a complete set only at 25'. These are : k, = 3 -5 0 x 10-7@), k, = 5.73 x 10-11(0), k, = 1-15 x I O -~ (*I. Taking values of the Bunsen coefficients from the International Critical Tables, calculation gives at infinite dilution : Experimcntsl work Getter& details.-Samples of solution were taken in IOO ml.glass-stoppered bottles j the gas space above t;oe liquid surfpce was very small, and samples were water-cooled to room tem-peratureimrnediateIy after being taken. Analysis was commend at once, and completed the same day. The following order was observed :ff2S.' Treatment with excess of acidified iodine solution, back titration. CO,. A modified Schrotter flask method, preceded, if H,S was present, by addition of a C0,-free suspension of precipitated copper oxide in water.Cafion. Direct titration, with CO, elimination.Free CO, and H,S were estimated and subtracted. Bunsen coefficients were corrected for this purpose by the factor (1-0.15 M), where M is the cation normality. Thermometers and standard solutions were cross checked. Kl = 0.27, K, = 0'27, K3 = 1.0.have been determined ovei a range-of concentrations (0.2-2N) and of temperatures (0-60"). The results cover a wider field than those of previous investiaators, and are more consistent. The heats of reaction 1Kna) PLU; rmmpukd from the variations with temperature agree well with published heats of formation.