1887.] 



Dispersion Equivalents. 



403 



published ; but while I am free to confess that there are many difu- 

 •culties in the investigation of dispersion which have not been felt in 

 dealing with refraction, I hold that the following conclusions are 

 fully warranted by the data : — 



1st. That dispersion, like refraction, is primarily a question of the 

 atomic constitution of the body; the general rule being that the 

 dispersion equivalent of a compound is the sum of the dispersion 

 equivalents of its constituents. 



2nd. That the dispersion of a compound, like its refraction, is 

 modified by profound differences of constitution ; such as changes of 

 atomicity. 



3rd. That the dispersion frequently reveals differences of constitu- 

 tion at present unrecognised by chemists, and not expressed by our 

 formulae. 



In this paper my object will be to point out the uniformity that 

 does exist, leaving apparent exceptions for future consideration. 



Before entering upon an attempt to determine the dispersion 

 equivalents of the different elementary substances, it may be well to 

 consider the difficulty which occurred at the threshold of the enquiry, 

 and another which appears to have deterred Briihl from prosecuting 

 his enquiries in the direction of dispersion. 



The original experiments of Mr. Dale and myself led to the belief 



that the " specific dispersion, ^ H ~^ A , slightly diminishes with in- 



iCrease of temperature "; but more accurate experiments made on the 

 same specimens of bisulphide of carbon, benzene, brombenzene, and 

 mint terpene, at the temperature of the observing room in the height 

 of summer and depth of winter, have made me less confident of this 

 .conclusion. The variations are certainly within the limits of experi- 

 mental error. The observations of Wiillner both upon bisulphide of 

 carbon and water, those of Bailie and v. d. Willigen upon water, as 

 well as those of Pisati and Paterno on benzene and cymene, show 

 that there is little, if any, appreciable difference in the specific dis- 

 persion at different temperatures. The general tendency of the 

 observations on the seventy substances which have been examined 

 more or less carefully, appears to be that the small difference of 

 specific refraction that exists at different temperatures is a little 

 greater in the case of H than in that of A. 



Briihl gives three cases of isomeric or quasi-isomeric bodies. He 

 measures the specific dispersion by the B of Cauchy's formula divided 

 "by the density. He shows that cinnamic alcohol, C 9 H 10 O, and 

 cinnamic aldehyde, C 9 H 8 0, both of which he conceives to contain four 

 pair of doubly-linked carbon-atoms, have a widely different specific 

 dispersion; that allyl paracresolate and anethol, C 12 H 12 0, having 

 four pair of doubly-linked carbon-atoms, are also quite different iu 



