336 
PROFESSOR W. G-. ADAMS AND MR. R. E. DAY ON THE 
end of the selenium. On screening off the light the galvanometer-needle came back at 
once to zero. This was repeated several times with the same result. 
The light was then brought to bear upon the centre of the plate, and then upon the 
end remote from the marked electrode. In neither case could any current be detected. 
Hence this plate appears to have one sensitive end. 
No. 10 was then examined in a similar manner, but without our being able to detect 
any perceptible current. 
No. 11, when under examination in this way, gave the following results: — 
The light was brought to bear upon the marked end, and gave a current of 6 to the 
left of zero, i. e. in a direction from the selenium to the platinum at the illuminated 
end. Now, if platinum stands above selenium in the thermoelectric scale, the current 
due to heating of the junction in this case would have been from platinum to selenium, 
or in the opposite direction to the observed one. 
When the light was brought to bear on the centre we obtained a deflection of 25 
divisions to the left of zero, and then a current in the opposite direction seemed to 
commence, which gradually sent the deflection down to zero and away to the right. 
On screening off the light there was always a throw of the needle to the right. 
The poles were then reversed in the mercury-cups of the commutator, so as to enable 
us to bring the light to bear upon the other side of this piece of selenium. 
The light was first directed on the end remote from the mark, and gave a current of 
15 to. the right, i. e. in a direction from the platinum to the selenium. 
No. 7 was then examined in the same way. When the lime-light was directed on to 
one end of it we obtained a steady deflection of 7 to the right of zero, and when brought 
to bear upon the other end a steady deflection of 4 to the left. On screening off the 
light, the deflection in each case came back at once to zero. 
Hence in the case of No. 7 the currents produced by light falling on the two ends are 
in opposite directions through the selenium, and pass from platinum to selenium at the 
illuminated end. 
No. 25 comes next. This piece had been only recently annealed, and had never pre- 
viously had a current through it. 
The lime-light was brought to bear upon the marked end, and produced a current of 
6 from the selenium toAvards the platinum. The electrodes were then reversed, and 
the light was directed on to the end remote from the mark, and produced a current of 
11 from the selenium toAvards the platinum at the illuminated end. In both cases we 
found that on screening off the light the deflection came back at once to zero. 
In another experiment when the light was rather more brilliant, the currents produced 
by directing on to the two ends were 25 to the left and 30 to the right respectively. 
The direction of the current in each case Avas from the selenium to platinum. 
Another piece, No. 23, also a neAV one, and through which no current had ever been 
sent from a battery, Avas next examined, with the following results : — 
