Tlie Effect of the Electric DiscJiarge on Water Vaj^our. 141 
a large surface. The anticathode, of platinum, was not used as an 
electrode. 
Experiments toith Water Vapour — It was found impossible to deter- 
mine what pressure of vapour would cause a cessation of the emission of 
rays ; for immediately on passing the discharge through the vapour the 
pressure, as indicated by the McLeod gauge, increased rapidly, showing 
that gas was being formed at a rapid rate, while the fluorescence of the 
X-ray tube ceased almost immediately. 
These increases in pressure were not consistent, owing, no doubt, to 
variations in temperature, variations in the physical conditions in the 
X-ray tube, &c. In every case, however, these increases tended to 
a limit. 
If, after gas had been formed in this way, the trap T2 was opened, so 
that connection between the X-ray tube and the P2O5 bulb was brought 
about, the pressure as indicated by the gauge was found to fall at a 
diminishing rate as the passage of the discharge continued. 
From these preliminary experiments it would appear that the effect 
of the discharge on small quantities of water vapour is to cause partial 
dissociation. A definite fraction of the vapour would become dissociated 
into the electrolytic gases H2 and O2. If the vapour, in the meantime, 
remained unabsorbed, this dissociation would continue till the pressure of 
electrolytic gas attained to a steady value. If, however, the vapour, which 
would still exist undissociated, were now absorbed, the pressure conditions 
would no longer be in equilibrium. Some of the electrolytic gas would 
recombine to form H2O ; but as fast as this recombination took place, 
the water would be absorbed by the P2O5, a steady state of pressure would 
no longer be possible, and the pressure indicated by the McLeod gauge 
would fall off with time. 
Beduction Effect. — After the above experiments with water vapour had 
been conducted, it was found that immediately on introducing any water 
vapour, gas began to be formed, quite independently of the passage of 
any discharge. The pressure as indicated by the McLeod gauge 
increased steadily to a maximum and permanent value. This effect 
seems to have been due directly to some change in the internal con- 
dition of the X-ray tube brought about by the passage of the previous 
discharges. 
The effect grew rapidly with time. Thus, five days after the discharge 
had been passed for the first time, on allowing vapour under saturation 
pressure (12 mm.) to stand in the apparatus (initially evacuated to 
0-000 mm.), it was found that after 18 hours, the pressure of gas formed 
had attained to the order of 1 cm. 
The seat of the effect was found to be in the X-ray tube; for, on 
removing the latter, the effect was greatly reduced. An attempt was then 
