Inductive Capacity and Latent Heat of Vaporization. 123 



depends solely upon the units chosen. For instance, when 



the latent heat is based, as usual, on the specific heat of water, 



the gramme weight, and the Centigrade scale, and when 



the inductive capacity is compared with that of air taken as 



C 

 unity, the ratio C'= — becomes e.g. for the acetic-acid series 



In place of the molecular weights, Trouton originally used 

 the vapour-densities D based on H 2 as unity ; the value of 

 his constant consequently lies between 10 and 13. It will be 

 noticed that these figures are not very different to the con- 

 stant <y obtained for the various ester series. In those cases 

 where the two happen to be identical, the relation between 

 specific inductive capacity, vapour-density, and absolute 

 boiling-temperature, can be expressed by the extremely simple 

 formula KD = T. 



De Heen has shown * that the absolute boiling- temperature 

 T at 760 millim. pressure and the coefficient of expansion a Q 

 at 0° are inversely proportional, or « T = constant ; it there- 

 fore follows that « KM = constant, i. e. the coefficient of expan- 

 sion at 0° is inversely proportional to the molecular inductive 

 capacity. The product a KM is perfectly analogous to the 

 first term ct p M in de Heen's equation for intramolecular 

 work, which represents that part of the heat spent for physical 

 dissociation as contradistinguished from chemical dissociation 

 forming the second term, and to which electrolytical decom- 

 position would here correspond. 



Tereschin has already tried if there is any direct connexion 

 between the specific inductive capacit} 7 of the various bodies 

 which he examined and their molecular weight, and found 



that the two formulae K = a + h ™ and K = Wl + A^), 



where a, h, and A are constants, <r the specific gravity, and 

 n the refractive index, fairly represent his observations. 

 These formulae are more complicated than the one which I 



T 



have proposed, viz. K = C / ^, without affording a notably 



better agreement with the experimental results, as will be 

 seen on comparing Tereschin's calculated values given in the 

 Annalen with those in the following Table VII. : — 



* Vide abstract of Schiff's article in the Beiblatter, 1. c. p. 691. 



