Abstract
The paper is divided into three parts, with an appendix. In the first, the general properties of the temperature-entropy diagram for water and steam are examined. By a simple device resulting from the use of Clapeyron's equation, it is shown that it is possible to give a very complete representation of the thermodynamic properties of saturated liquids by areas on this diagram. The second part of the paper constitutes a critical review of Callendar's hypothesis that saturated water contains in solution its own volume of saturated steam, and the pictorial representation of the properties of saturated water is here found to be very useful. In an appendix to the paper the use of Callendar's hypothesis as a working proposition is shown to give elegant and rapid evaluation of the relations required in the calorimetry of saturated water under ideal experimental conditions. This section of the paper is placed in an appendix as the ability to use the hypothesis under these conditions does not constitute a proof of its validity. It is, in fact, shown in the second part of the paper that, on the latest experimental data, the hypothesis fails to give an exact prediction of the thermal properties of saturated water. The third part of the paper shows that the failure of Callendar's hypothesis to fit the experimental data can be accounted for by supposing that the quantity of dissolved steam is related to, but less than, that postulated by Callendar. This leads to an interesting picture which bears some resemblance to certain aspects of the hole theory of liquids, to which Callendar's hypothesis has been compared. Both, however, are based on hypotheses, the validity of which can be tested only by the measure of agreement between theory and experiment.

This publication has 6 references indexed in Scilit: