Inductive Capacity and Latent Heat of Vaporization. 115 



the complete series are given from n = to n = 5, the first 

 members being formic acid, acetic acid, and water respectively, 

 and the last the amyl compounds *. For the acids the specific 

 inductive capacity is not known, but I have calculated it from 



the mean ratio ^ of each particular series. If these calculated 



values of K should in future be compared with those found 

 by direct observation, it must be borne in mind that the values 

 found by Favre and Silbermann for the latent heat of these 

 acids are probably not quite accurate. For the specific in- 

 ductive capacity of water, E. B. Rosa's f value, viz. 75*7, is 

 adopted, Tereschin's figure, 83*8, being notably higher ; 

 whilst the results obtained by the same observers for ethylic 

 alcohol are almost identical, viz. 25*7 and 25*8 respectively. 



It will be seen from Tables I., II., and IV. that the devia- 

 tions of the individual values of the quotient ^ from the 



mean are no greater than one would expect, considering that 

 the two constants were obtained by different observers and 

 with entirely different samples of liquid. In the case of the 

 acetates and alcohols, which form two almost complete series, 



the constancy of the quotient = is particularly satisfactory ; 



and when the observed values are plotted out, taking the 

 latent heats as abscissae and the inductive capacities as 

 ordinates (vide the diagram), the points of intersection are 

 nearly equally distributed on both sides of a straight line, 

 thus unmistakably showing the linear dependency of the two 

 physical constants. Returning to Table III., which embodies 

 the results for five different ethyl esters, it will be noticed 



that the ratio t? is not quite so constant, but rather seems to 



increase with the molecular weight ; however, if the first 

 member of this group, viz. ethyl formate, is disregarded, the 

 deviations from the mean are by no means very great. 



In order to show the legitimacy of employing calculated 

 values of latent heat, where no direct determinations are 

 available, e.g. for the benzoic-acid esters, I have also calculated 

 them for those bodies where experimental values actually do 



* Schiff's determinations of the latent heat refer to the isoamyl esters, 

 hut it is doubtful if this is also the case with Tereschin's specific inductive 

 capacities. The normal amyl formate is not mentioned in Beilstein's 

 Handbook. 



t Phil. Mag. [5] xxxi. p. 200 (1891). 



12 



