of the Selentum Cell. 35 
plates into a layer about 1 mm. in thickness, and converting 
this into the metallic modification, excellent reflecting 
surfaces may be obtained. By employing the katoptric 
method of obtaining the angle of maximum polarization, it 
ought to be poss sible to determine values of n and & (refrac- 
tion and extinction coefficients) through a considerable range, 
and thus to determine whether the anticipated relationship 
exists or not. 
Mode of Action of the Selenium Cell. 
Jt is the purpose of this part of the work to gain some 
slight insight into the molecular mechanism called into play 
when a cell changes its resistance upon exposure to light. 
The problem was aitacked by investigating whether the 
flowing of an electric current is essential to the development 
of sensibility in a selenium cell. 
The following experiment was performed to decide whether 
light in falling upon a cell produced some conducting com- 
pound, and whether the current eventually decomvosed this 
compound, bringing the resistance of the cell back to its 
original value. <A selenium cell, in circuit with two dry 
batteries, galvanometer, and key, was exposed to the light of 
an incandescent lamp for 15 seconds, which gave rise toa 
large deflexion of the galvanometer. When the light was 
extinguished, the galvanometer-needle returned to its original 
position, requiring 65 seconds to do so. For the second 
part of the experiment the circuit was broken, so that no 
current was flowing through the cell, and again an exposure 
of 15 seconds was given. When the circuit was closed 
65 seconds later, it was found that the resistance of the cell 
was the same as in the beginning. 
If a stable, conducting compound had been formed in the 
second case, its presence would have been detected upon 
closing the circuit by giving rise to a galvanometer deflexion 
greater than the original one which corresponded to the 
resistance of the cell in darkness. It is possible to account 
for the above negative result by assuming either that nothing 
occurs in a cell unless an electric current is flowing, or that 
an unstable condition is set up which breaks down of its own 
accord in darkness. An experiment, based upon the following 
consideration, was undertaken to clear up this point. 
When a cell is taken from light into darkness its conduc- 
tivity falls off rapidly at first, and more slowly afterwards. 
If changes in resistance do actually occur while no current is 
flowing, they ought to be observable by exposing a cell on 
open circuit to light for perhaps 30 seconds, and then closing 
D2 

