144 
Tramactions of the Royal Society of South Africa. 
During the latter part of the experiment the cathode was at a red 
heat, and intense disintegration was taking place. The tube, in the 
neighbourhood of the electrodes, more especially the cathode, was now 
very black. 
The above experiments were repeated, with the tube in the same con- 
dition, on three occasions when the temperature conditions were much the 
same. Curve 3 shows how the pressure as indicated by the McLeod 
gauge varied with time on each occasion. The pressure increased rapidly 
during the first few minutes to a maximum value, after which it fell at a 
diminishing rate. Gurve 4 shows a case where the P2O5 bulb was put into 
connection with the remainder of the apparatus immediately after this 
maximum value of the pressure had been attained. It serves to compare 
the action of the PgO^ with that of the agency responsible in these last 
experiments for the falling ott" of the pressure with time. 
This agency must be the platinum-black on the walls of the tube ; 
for the falling off of pressure with time was absent when the tube was 
clean, and increased simultaneously with the formation of the deposit. 
That the seat of the effect was actually in the discharge tube was verified 
experimentally. 
Experiments with the Electrodeless Discharfje. — In order to remove the 
effect apparently due to the platinum-black, a tube with no electrodes was 
substituted for the last discharge tube. The tube, 15 cm. in 
^ ^\ length, was made of 1-5 cm. -bore soda-glass tubing. The ends 
/ j were knob-shaped and covered with tinfoil caps to which were 
\ ^ connected the wires from the induction coil. 
The gases from the walls of the tube were driven out and 
removed by pumping at intervals while the discharge was 
passed through air under a pressure of a few thousandths of 
a mm. Hg. During this process a belt of brownish-black 
coloration appeared near the cathode extremity of the tube. 
The discharge was passed through vapour under 0-9 mm. 
pressure on three successive occasions, the temperature con- 
Uditions being very similar (about 14° C). Curve 5 'shows how 
I the pressure as registered by the McLeod gauge varied with 
the time on each occasion. During the first few minutes there 
was a rapid increase to 0-08 mm., after which there was a 
falling off at a diminishing rate. The results are very similar to those 
obtained when corresponding experiments were conducted with the 
previous tube. The coloration on the walls of the tube remained 
throughout the experiments. 
The discharge was passed through vapour under 1*7 mm. pressure, 
the temperature being 14-5° C. iVs before, the pressure increased 
to a maximum, and then decreased at a diminishing rate ; but the 
