1896 - 97 .] Influence of Rant gen Rays on Electric Conduction. 405 
same sheet as the surrounding aluminium tube, was supported at 
one end by a small piece of paraffin so placed as to be out of 
reach of the action of the Rontgen lamp. The rays from * the 
lamp were allowed to pass from a lead cylinder surrounding it by 
a small hole about *3 of a square cm. in area. They fell on the 
aluminium sheath transparent to them, and rendered the air 
between it and the insulated aluminium within conductive. 
We tried various differences of potential, ranging from a few 
volts to 2200 volts. In one series of experiments we charged the 
insulated metal to — 97*5 volts, and then disconnected the battery 
electrodes. The lamp was then set a-going, and the electrometer 
deviation taken each half -minute for a minute and a half with one 
pair of quadrants insulated. The rays were then stopped, the quad- 
rants metallically connected, and metallic zero again found. Then 
the reading during another period of one and a half minutes, with 
the rays acting, was observed, and so on until no deviation from the 
metallic zero of the electrometer was found with one pair of quad- 
rants insulated, and the rays falling on the aluminium outside 
coating of the Leyden A. The sensibly complete discharge thus 
observed took place in about a quarter of an hour. We found that 
the rate of deviation from the metallic zero was the same as the 
difference of potential fell from - 97*5 volts to about - 4 volts. 
With differences of potential of - 930, - 1750, and - 2000 volts the 
rate of deviation was not appreciably greater than with ± 20 volts. 
This confirms and extends, through a very wide range of voltage, 
the interesting and important discovery announced by J. J. Thom- 
son and McClelland, in their paper in the Cambridge Philosophical 
Society Proceedings of March 1896, to the effect that the conduc- 
tion of electricity through air under influence of the Rontgen rays 
is almost independent of the electric pressure when it exceeds a 
few volts per centimetre. 
§ 15. In the experiments on paraffin, the outside coating of the 
Leyden A consisted of an aluminium cylinder 27 cms. long, 4 cms. 
diameter, connected to sheaths. A metal bar about 1*73 cms. in 
diameter, and 30 cms. long, supported centrally on paraffin filling 
the whole space between it and the aluminium sheath, constituted 
the inside coating. With this arrangement we made experiments 
with differences of potential of ± 94, ±119, ± 238, - 2000, + 2500, 
