Action of Selenium Cells. 213 



its iron plate is exposed to radiation. I did not succeed in 

 observing such a current. 



Prof. W. G. Adams was the first to show that selenium 

 on illumination develops an electromotive force. Still sup- 

 posing that in the photophone a direct correlation between 

 the energy of light and that of the electric current might 

 reveal itself, I endeavoured to observe and to measure the 

 electromotive force developed in selenium cells on their 

 exposure to light. 



I made selenium cells according to the method of Bell and 

 Tainter as modified by Bidwell. Two thin copper wires were 

 wrapped parallel side by side several times round a strip of 

 mica. Selenium was then melted, and crystallized between 

 them. In this way I obtained cells which were very sen- 

 sitive and yielded, when connected with the battery, a clear 

 sound in the telephone. According to my last measurement, 

 the resistance of one of those cells in the dark is 3000 ohms, 

 diffuse daylight reducing it to 2700. The measurements were 

 made with an astatic-mirror galvanometer, the resistance of 

 the coils of which was 6860 ohms. The scale was about one 

 metre distant from the mirror, each division one fortieth of an 

 inch. There were other galvanometers at my disposal, and a 

 battery of twenty Leclanche cells. The galvanometer proved 

 itself more sensitive than the telephone. 



The above-mentioned selenium cell yielded by intermittent 

 radiation of lime- or electric light clear and distinct sounds ; 

 illumination by the same sources of light deflected the image 

 of the slit strongly. 



But the same light which, when a current of an external 

 battery flows through the cell, makes the telephone to sound 

 or the spot of light to slide along the whole scale, is not able 

 to produce an observable electromotive force in the selenium 

 cell. This non-appearance of a current (the circuit being 

 formed only of selenium cell and galvanometer, or of selenium 

 cell and telephone) is, in my opinion, very noteworthy ; for 

 the problem which I wish here to solve is not to make photo- 

 phones without batteries, but to explain the efficiency of the 

 selenium cells by the qualities they actually possess. 



It is possible to prepare pieces of selenium which show an 

 electromotive force on being illuminated. I shall return to 

 this point further on. But cells and pieces of selenium not 

 possessing this property nevertheless show photophonic action. 

 Therefore the photophonic action cannot be explained by this 

 property, which is, or at least may be, wanting in the selenium 

 cells. 



I proceeded further, always endeavouring to find a correla- 



