ACTION OF LIGHT ON SELENIUM. 331 
the commutator and joined by a binding-screw, so as to allow the polarization-current 
gradually to subside. 
On a subsequent occasion, when the current from 20 Leclanche cells had been sent 
through this same plate in the same direction for two successive intervals of 5 minutes, 
the consequent polarization-currents were —27 and — 51 respectively. 
A plate of selenium was next used which in outward appearance exactly resembles 
the last one, but which has evidently not been so well annealed, as its resistance is about 
13 - 1 megohms. The current from the 20 Leclanche cells was sent through it for a short 
and also for a considerable interval of time, but no signs of a polarization-current could 
in either case be detected by the galvanometer. 
The tubes numbered 3, 4, 8, and 12 were then tested in the same manner, with currents 
both of short and long duration from the battery of 20 Leclanche cells. They all showed 
polarization-currents, but only of feeble intensity, the deflection of the galvanometer- 
needle varying from 2 to 7 divisions on the scale. Though small, these currents were, 
however, quite distinct. 
That all these currents were real polarization-currents was also evident from the fact 
that, on lifting any tube out of the commutator and replacing it with its electrodes 
reversed in the mercury-cups, the deflection was always on the opposite side of zero to 
what it had been previously. 
The next piece used was that in tube No. 15, whose resistance is comparatively small, 
being about ’0247 of a megohm. The positive pole of the above battery having been 
connected with the marked end of the tube, the current was sent through it for five 
minutes, and gave rise to a polarization-current of — 25. The same current was then 
sent through it in the same direction as before for 15 minutes, but at the end of that 
time the resulting polarization-current was only —11. 
The next day the same tube was tried again, with the following results : — 
Positive current from 20 cells for 
5 minutes in at the marked end gave polarization-current of . 
10 
5 
10 
10 
55 
55 
55 
55 
unmarked end 
55 
55 
55 
55 
. -12-5 
— 13-5 
. -10 
. -15 
. -19 
Tube No. 14 was then used, and the current from the above battery was sent through 
three times for 10 minutes at intervals of 15 minutes. The following observations 
were made : — 
Strength of current. Polarization-current. 
250 xlO 3 -30 
235 xlO 3 -25 
236 xlO 3 -26 
3 a 2 
