142 Transactions of the Boyal Society of SoutJi Africa. 
made to remove the cause of the effect. The apparatus was steamed for 
two hours between Ti and T^, and fresh mercury put in the manometers 
and traps. On now allowing vapour under saturation pressure to stand 
in the apparatus, the effect was found to be enormously reduced. 
Initial pressure : 0*000 mm. 
Vapoue admitted (under Saturation (12 mm.) Pressure). 
Intervals of Time. 
Pressures as per Gauge. 
1 min. 
0-001 min. 
7 
0-004 
14 
0-005 
31 
0-0045 
51 
0-0045 
The seat of this minute remaining effect was found to be in the bulb of 
the McLeod gauge itself. This result is natural enough ; for, during the 
process of steaming the apparatus, the gauge bulb, which afforded no exit 
for the steam, could not have been subjected to so thorough a cleansing as 
the other parts of the apparatus. In the course of the next few days, this 
residual effect disappeared. 
The X-ray tube, after having been thoroughly steamed, was then re- 
fitted, the previous effect tested for, and found unrenewed. 
Preparatory to repeating experiments on the passage of the discharge 
through water vapour, the discharge was passed for some time through 
two or three thousandths of a millimetre of air, pumping being at the 
same time kept up ; the object being the driving out and removal of as 
much as possible of the gas which might be occluded in the walls of the 
tube and in the electrodes. 
After this had been satisfactorily done, it was found that on allowing 
water vapour under its saturation pressure to stand in the apparatus, the 
pressure as indicated by the gauge began to increase steadily. The dis- 
charge, at the time, was not being passed. The previous effect, then, was 
once more present, and must have been brought about solely by the last 
passage of the discharge. On repeating the experiment, the effect in- 
creased in magnitude ; and very shortly afterwards, on allowing vapour 
under its saturation pressure to stand in the apparatus, the pressure of gas 
formed after 15 hours, reached a value of the order of 1^ cm. At the 
same time, the cathode had become incrusted with a greyish growth. 
Enough gas had been formed to allow of its being pumped out and 
