ACTION -OF LIGHT ON SELENIUM. 
319 
With 1 candle the deflection was 
„ 4 candles „ „ 
» 8 J5 
q 
55 ° 55 95 55 
90 divisions. 
180 
250 
270 
380 
55 
55 
55 
These experiments clearly show that the change in the resistance of this bar of 
selenium varies directly as the square root of the illuminating power. 
Diminution of Resistance with Increase of Battery-power. 
At an early stage of these experiments it was found that, with the same piece of 
selenium and at the same temperature, the resistance diminished as the battery-power 
was increased. As an illustration, we may take the following observations made with 
Mr. W. Smith’s plate of selenium : — 
Number of Leclancbe cells. Resistance of the plate. 
5 ... . 
. 2-7 
megohms. 
35 ... . 
2*2 
55 
5 . . . . 
. 2-7 
55 
30 ... . 
. 2-3 
55 
And after some hours we had with 
30 ... . 
. 2-4 
55 
and almost immediately afterwards with 
5 ... . 
. 2-9 
55 
It was also found that the electrical resistance of this plate was different for currents 
going through it in opposite directions : that is to say, that when the selenium plate 
was in the dark, and the positive direction of the current was from the electrode A to 
the electrode B, and its resistance was balanced by the Wheatstone-bridge arrangement 
in the usual manner, on reversing the current so that the positive direction was B to A 
through the selenium, the numerical value of the balancing resistance was always found 
to be different from that previously obtained. If the electrical conductivity of selenium 
followed the ordinary law of metallic conduction, this difference would not exist ; and 
as it seemed probable that a careful investigation of these phenomena might afford some 
clue to the causes of the peculiar behaviour of selenium under certain conditions, we 
have of late been devoting especial attention to this branch of the inquiry. 
Description and Mode of Preparation of new Specimens. 
In the experiments hitherto described the original plate of selenium belonging to 
Mr. Willoughby Smith had been exclusively employed; but as it now seemed desirable 
to make the inquiry more general, and not to limit it to the examination of the beha- 
viour of one specimen only, several pieces of selenium were fitted with electrodes, and 
