486 On the Change of Absorption produced by Fluorescence. 



exist in a dilute solution of fluorescine or eosine a small difference, 

 but a strong solution of either does not permit its fluorescent light 

 to penetrate, except a very small thickness of the liquid, whether 

 fluorescing or not. 



Sulphate of quinine is too transparent to the luminous rays, so 

 that even if a change did exist it could hardly be detected. The case 

 is otherwise with uranium glass, in which the effect is well marked. 



The compounds of uranium are remarkable for many characteristic 

 properties connected with fluorescence, which they seem to exhibit 

 more readily than most other bodies. The spectrum of the. fluor- 

 escent light emitted by uranium glass shows certain maxima and 

 minima, as was noticed by Stokes in 1852. 



The experiments which have been carried out with a view to 

 testing whether a change in the absorption is produced by fluores- 

 cence, have been almost exclusively upon uranium glass. Regarding 

 it as the most suitable substance to experiment upon, a careful 

 series of experiments have been made to determine whether the 

 phenomenon really existed or not. 



If we call cc and ft the fractions of the light from B transmitted by 

 A, according as the latter is fluorescing or not, for uranium glass 

 1 cm. thick, the values of a in a particular set of experiments, in 

 which eye determinations were taken, were as follows. Each 

 experiment consisted of twenty observations. 



Exp. I * = 0-54 



II = 0-46 



III = 0-51 



IY = 0-36 



And for ft — 



Exp. I ft = 0-70 



II = 0-84 



III = 0-72 



IY = 092 



The mean value of a. obtained from these being a = 0*47, ft = 

 0-79. 



The ratio — @— was also independently determined, and found as 

 1-hac 



the mean of eighty observations to be 0*507. 



The values of a and ft have also been determined photographically, 

 giving 



a <0'48 ft >0-75 



>0-43 <0-89. 



If we take the maximum value of a given photographically 0*48, the 



