342 Mr. W. Hibbert on the Gladstone 



in the ratio of 5 to 1. It is surely improbable that molecules 

 of very varying atomic structure should have the same tem- 

 perature increase, but the same improbability does not attach 

 to " domain/' 



There is, however, one aspect of this question which may 

 prove to be still more important. It may be objected that the 

 empirical * expression here used for obtaining the domain is 

 too fanciful to be regarded with serious consideration. It 

 may further be pointed out that it is equivalent to 



O- 1)0-/3) x 7, 

 r 

 as we have seen, and that the first two factors give a constant. 

 Any relationships like those just alluded to must therefore 

 be dependent on a more or less common temperature function 

 for the refractive index. 



It is evident, apart from arithmetic error, that this must 

 be so. But such an objection, while having weight against 

 an expression obtained by the method we have pursued, at 

 once raises a very suggestive question. Does the evidence 

 really justify the idea that the refractive indices of various 

 substances change in the same ratio between the same tem- 

 perature limits ? In other words, is the ratio magnitude of 

 the change independent of the nature of the liquid ? The 

 following table gives the evidence available. The substances 

 quoted are those already dealt with, but the comparison deals 

 directly with the refractive indices, and is therefore free from 

 all objections which might be urged against the hypothesis of 

 this paper. 



To facilitate comparison, the substances are again divided 

 into groups, the order being slightly different from that pre- 

 viously given. The refractive indices are given in column 3 ; 



* In the June number of the Chem. Soc. Journal appears an abstract 

 of a paper by W. F. Edwards (American Chem. Journ.), who obtains the 

 expression now proposed, as a new formula for specific refraction, thus : — 

 If V and V l are the velocities of light in space and the substance re- 

 spectively, IX = — . Let A= V- V„ then A = ^— V. If the difference 



Vi h* 



A is proportional to the amount of substance, and assuming as unit one 

 gramme of substance in 1 cub. centim., and that X is the retardation of 

 velocity for unit substance, then 



/x a . 



This is the same expression as that I propose for domain, multiplied by 

 V* As will be seen from my tables, and as Edwards shows, the value of 

 X is not a constant. It cannot therefore serve as a measure for specific 

 refraction unless under stated conditions of temperature. He promises to 

 return to this, but will, I belieye, only find the solution by substituting 

 "volume of domain " for volume of substance. As will be seen, I obtain 

 his expression by a transformation of the Gladstone expression. 



