Variation in the Pressure of Saturated Vapours. 89 



for large amplitude of temperature, others do not. Among 

 the former are, sulphur (390° to 570°), sulphide of carbon 

 (-20° to 130°), water (-30° to 100°) benzene (40° to 170°), 

 ether (0° to 110°) and others, while the latter include carbon 

 tetrachloride, ethene bromide, methyl alcohol, and others. 

 Yalta, January 1891. 



Note on Prof. Kraevitch's Paper. By Prof. Geo. Fras. 



Fitz Gerald. 



It may be worth while pointing out in connexion with this 

 paper that the author seems to attribute too much importance 

 to the fact that his two methods of calculating the specific and 

 latent heats agree. His second method consists in calculating 

 the latent heat of vaporization r from the thermodynamic 

 formula 



r=k(v-wJS .^ 



and then using this value in his own formula 

 ,.=r„-( C - Cl )(T-T ). 



His first method consists in substituting r out of the second 

 of these equations in the first and then integrating through 

 the small range of temperature T to T . In both cases, for 

 reasons he gives he assumes that the vapour obeys Boyle's 

 and Dalton's laws, and that c — Cj is constant throughout the 

 ranges of temperature employed. This he does explicitly in 

 his first method, but these assumptions are equally necessary 

 for his second method, for the first of these equations, as 

 used by him, is modified by an assumption of the validity of 

 Boyle's and Dalton's laws, and the second is explicitly derived 

 from a differential equation by assuming c — c 1 constant within 

 the limits of integration, and these limits are mostly the same 

 as those he subsequently employs to obtain the equation he uses 

 in his first method. Hence the two methods are not indepen- 

 dent, and the fact that they do not lead to exactly equal 

 numerical results is largely due to the difficulty he takes so 



much trouble to surmount of calculating ~ at the tempera- 



ture required. There are in addition several arithmetical 

 mistakes which would require laborious calculations to correct, 

 as some of them may be mistakes in the elaborate calculations 

 which are not given fully in the paper. 



