180 Intelligence and Miscellaneous Articles. 



prisms and an achromatic lens ; the latter was separated from the 

 slit by twice its focal distance. This arangement gave me a spec- 

 trum sufficiently pure for the principal lines to be seen in it. The 

 spectrum was expanded on the side of a box in which was a 

 movable slit, which could be shifted to the different parts of 

 the spectrum, and the breadth of which could be modified at plea- 

 sure. Through this slit I let certain rays of the spectrum, which 

 I previously caused to traverse a flint-glass prism, enter the box, 

 which contains a vessel filled with fluorescent liquid. After this, by 

 means of a reflecting prism I direct these perfectly homogeneous 

 rays upon the fluorescent liquid. Between the surface of the liquid 

 and the slit in the side an achromatic lens is placed, which produces 

 the coloured image of the slit upon the surface of the fluicl. With 

 a second reflecting prism I direct the light which comes from the 

 fluorescent liquid into the slit of the collimator of a Brunner's 

 spectrometer. In the field of vision of the telescope of the spectro- 

 meter I get two coloured images — one produced by the fluorescent 

 light, the other that which is directly reflected at the surface of the 

 fluid. I then measure the minimum deviation of these two images. 

 I give the values I have obtained for fluoresceine : — 



Incident light. Fluorescent light. 



Width of the 



pencil. 



o 



2 



51 







46 







46 







48 







48 



Mean deviation. 



50 38 

 49 59 

 49 60 

 ■ 48 18 

 47 56 



Width of the 

 pencil. 



1 25 

 25 



53 



1 60 

 20 



n deviation. 



48 43 



48 18 



48 12 



48 60 



47 48 



These experiments show that the fluorescent has a lower refran- 

 gibility than the incident light. I have repeated the same experi- 

 ments with eosine, naphthaline-red, and chlorophyll, and obtained 

 the same results. 



In my researches I have taken the fluids at different degrees of 

 concentration, and in layers of different thicknesses ; and the 

 result has always been the same. Each fluid has in the spectrum 

 certain rays which excite in it the strongest fluorescence ; with 

 other rays the fluorescence will be feebler ; and it will disappear 

 if rays still less refrangible be operated on. All the rays in the 

 spectrum which are more refrangible than the fluorescent ones 

 excite fluorescence in these fluids. It is upon naphthaline-red that 

 I have obtained the greatest change of refrangibility of light, in 

 which case the incident rays, whose index of refraction for flint- 

 glass is T63917, were changed into rays having the refraction- 

 index 1-61521. 



After these investigations I think I may conclude that the law 

 of change of refrangibility of light is perfectly correct in the 

 general form in which it was given by Stokes. — Comptes Rendus de 

 VAcademie des Sciences, June 9, 1879. 



