475 
on Discharge in a Vacuum Tube. 
after the discharge stopped, were 505, 504, 504, but these were 
always three or four volts lower while the discharge was passing. 
The same “ weakening” of the gas, noticed above, was found here, 
for a later reading gave 503 volts. It is hoped that this point 
may be investigated before long, but at present all that can be 
stated with certainty is the following — : 
(1) The “negative drop” was found to be the same, whether 
the radiation was falling on the kathode or not. 
(2) No change was detected in the magnitude of the current 
when the kathode was illuminated or when the light was shut off. 
The change — for probably a real change does occur— was at least 
small in comparison with one of one-half per cent. 
(3) A marked change in the potential required to start the 
discharge was produced by the action of the ultra violet light : 
the effect being a lowering of the required potential. In the 
case under examination the reduction was from 558 volts to 
510 volts. 
(4) By gradually lowering the potential difference after the 
discharge had been started and then all rays cut off, it was found 
possible to carry the potential to a point at least three or four 
volts lower than the minimum potential required to start the 
discharge by the action of the ultra violet light. 
(5) The effect is due to the action of the radiation at the 
kathode and not — at least not to any appreciable extent — to 
a volume ionization by the rays as they pass down the path of 
the discharge. 
This last point is shown by making the copper rod the kathode 
and the zinc electrode positive. A metal screen completely 
shielded the copper from the light. The following numbers were 
obtained : 
Copper negative , Zinc positive. 
291 cells will start the discharge in the dark ; 
290 cells will not start the discharge in the dark, nor when 
the rays are passing down the tube. 
Zinc negative , Copper positive. 
(a) In the dark, 
268 cells start the discharge ; 
267 cells do not start discharge. 
