4 Messrs. J. H. & G. Gladstone on the Refraction and 



So far as the well-known halogens chlorine, bromine, and 

 iodine are concerned, these figures are principally interesting 

 as compared with the values already determined for them in 

 other organic compounds. From their combinations with the 

 compound radicals (C n H 2 «+i), the numbers given in the first 

 division of the following table have been deduced ; but from 

 some other organic compounds, such as chloroform and its 

 congeners, define compounds, &c, somewhat higher numbers 

 have been obtained, as in the second division of the table*. 

 The third division gives the figures ol tained from the benzene 

 compounds. In each case merely the molecular refraction 

 for the line A is given, with the molecular dispersion, that is, 

 the difference between R H a-nd R A . 





Compound 

 radicals. 



define 

 compounds. 



Benzene substi- 

 tution compounds. 



K 



^H — ^A. 



*A- 



R -Ea. 



R A" 



^h—Ra- 



Chlorine 



Bromine 



Iodine 



9-9 

 15-3 

 24-5 



0-45 

 103 

 2-60 



10-0 

 15-2 

 25-4 



0-50 

 1-22 

 365 



100 

 15-2 

 252 



0-70 

 1-41 

 3-43 





It is evident that the values of these ordinary halogens in 

 the benzene substitution-products are analogous to their values 

 in the olefine compounds rather than the others- This is more 

 significant in the dispersion than in the refraction. 



But a greater interest attaches to the optical properties of 

 fluorine. No organic compound containing this element had 

 hitherto been examined, while it was known that the refrac- 

 tion of fluorine in its mineral compounds is extremely small. 

 A glance at the sixth Table will show that in the benzene 

 compounds also the effect of fluorine upon the light transmitted 

 is very little. But there is another peculiarity. Whereas 

 the values of chlorine, bromine, and iodine increase from A 

 to H, those of fluorine regularly decrease ; so that, in fact, 

 while the refraction-equivalent for A is only 0*63, the disper- 

 sion for H— A is actually a negative quantity, — 0*28. 



Lest this anomalous result might be due to our having 



* This distinction is only partially recognized in the paper on " Dis- 

 persion Equivalents" in Proc. Roy. Soc. xlii. (1887). 



