262 THE ROYAL SOCIETY OF CANADA 



A set of about 12 solutions with different known concentrations 

 was carefully prepared and measurements of their fluorescent in- 

 tensity taken within a day or two to avoid any complications due 

 to decay of the solutions. The usual form of curve was obtained 

 (Table I, Fig. 3). For very dilute solutions the intensity was pro- 

 portional to the concentration but reached a maximum for higher 

 concentration. 



An experiment was made to detect the effect of the presence 

 of the decayed solution on the fluorescent power of the unchanged 

 molecules. Solutions were diluted the same amount, one by the 

 addition of distilled water, the others by the addition of decayed 

 solution of varying strengths. 



As the fluorescence was equal in all cases, the fluorescent power 

 was taken to be unchanged by the presence of transformed molecules 

 and the intensity of fluorescence of any solution was assumed to be an 

 indication of the number of unchanged molecules present. Values 

 of the concentration during the exposure were obtained from Fig. 3, 

 and a graph plotted (Table 2, III, Fig. 5). It is seen that the number 

 of molecules transformed was very large at first, and decreased as 

 the exposure continued and the solution became weakened by the 

 ultra violet light. 



By drawing tangents to this curve a calculation was made of the 

 percentage of the molecules present transformed per sec. One would 

 suppose that after the solution became very dilute the light would 

 penetrate the entire solution and the percentage transformed per 

 sec. would be constant, but Fig. 6 shows the rate to decrease very 

 rapidly at first, then steadily. Perrin has found that for very con- 

 centrated solutions the rate of change is slow, increasing as the con- 

 centration is lessened by the decay of the fluorescent particles. 



These results would lead one to believe that the rate would 

 increase to a maximum, then decrease as the concentration becomes 

 dilute. 



Further examination of Fig. 6 shows that on recommencing the 

 exposure after a rest of several hours the rate of decay was much 

 more rapid than before the interval, although the fluorescent intensity 

 had weakened in the meantime. The rate of change is then not 

 merely a function of the concentration but depends also on the dura- 

 tion of exposure. A slight extension of Perrin's theory would be 

 necessary to account for this. 



It was found that the fluorescence of aesculin could also be 

 destroyed by bubbling ozone through the solution. The reaction 

 proceeded at the same rate whether ultraviolet radiation was present 



