90 Journal of the Mitchell Society. [Nov. 



late E,. Density at 0° C is theoretical, the molecular weight 

 used being 60.06. 



Acetic Acid. See Table 22. Data from Jour. Chem. Soc. 

 49, 1886, p. 790, and in 1891. Vol. 59. p. 903, preference being 

 given in all cases to the later measurements. For calculat- 

 ing the latent heat, we have, 



A = antilog (1.9257964 — 0.00149898/) 



-f antilog (1.1390566 — 0.0130424O, 

 where t = t° C — 120. 



APPLICATION OF THE THEORY. 



For details of the calculations involved the appended tables, 

 2 to 22, must be consulted. A summary of the results 

 obtained for the constant by the use of the equation 



L — E 



• y 57= = constant, is given below in Table 1. Certain 



■p d — v D 



mean values of this constant somewhat arbitrarily chosen are 

 given at the top of each column. All results not in agree- 

 ment with this mean value of the constant by as much as two 

 per cent, are in italics. 



Since from 0° C to within 10° of the critical temperature 

 eleven of the substances show altogether, out of two hundred 

 and seventeen observations, only two divergences greater 

 than two per cent, from the average values, it is clearly evi- 

 dent that the relation given by the equation cannot possibly 

 be accidental. But is the relation in these and in the other 

 cases as accurate as the excellent experimental data will permit ? 

 Only by such an agreement could the theory be finally estab- 

 lished. To answer this question we must examine the man- 

 ner and extent to which errors in the observations will affect 

 the constant. 



A cursory examination of the equation ., — J = con- 

 stant would make it appear that as the critical temperature is 



