162 



THE ROYAL SOCIETY OF CANADA 



Light of any desired wave length could be introduced into the 

 cell and the rate of drift of the needle noted on a scale suitably mounted 

 in front of the electrometer. The wave'length of the light introduced 

 into the cell was altered from 7260 A.U. to 3140 A.U., by successive 

 stages, and the rate of drift taken for each particular stage. This was 

 repeated for each set of readings taken. The intensity curve for the 

 variation of the deflection with the wave length for the different sets 

 of readings varied but slightly, a typical curve being that shown in 

 Fig. 3. In this curve we see that the maximum drift was obtained at 



7000 



6000 



SOOO 



fooo 



9000 



Figure 3 



a wave length of 5160 A. After a set of readings as indicated were 

 taken the rubber was placed directly in front of the slit of the spectro- 

 graph. The thin sheet of rubber was held between brass plates about 

 8 cm. by 5 cm., in the centre of which had been cut a small rectangular 

 opening about 1 cm. long and -3 cm. wide. The brass plate cut out 

 all the extraneous light and only that passing through the rectangular 

 opening where the rubber was exposed, was permitted to fall on the 

 slit of the instrument. Readings for the rate of drift of the needle 

 were again noted. In this way two sets of readings were obtained 

 for any particular wave length, the first due to the effect of the light 

 direct, the second due to the effect of the light after passing through 



