482 
MESSES. W. DE LA RUE AND H. W. MULLER ON THE 
the discharge, however, took place with more facility across the outside of the tube, 
at ordinary atmospheric pressure, or between the terminals of the coil which were 
distant 0’5625, inch (1*43 centim.). 
The tube was sealed off and laid by for a few days, when it was found that 5100 
cells would not pass, but that 8700 did so intermittently, and it required 10,160 to 
produce a continuous discharge. Subsequently 1200 cells passed, 20 would not do so 
even immediately after the discharge of the former number; 800 cells, 700, and then 
600 passed; but after a while the last number was insufficient to produce a discharge. 
This tube shared the fate of the first tube and ultimately cracked. It is very possible 
that the strong discharge volatilized a portion of the terminals, which were of platinum, 
and that this condensed, or that they absorbed the residual gas so completely as to 
produce a vacuum too perfect to admit of a discharge taking place, and that, 
ultimately, sufficient of the occluded gas was again given off to render the discharge 
again possible. 
Occlusion of gas by terminals. 
The power of terminals to occlude gas and then under an electric discharge to give 
it off again is well exemplified by tube 48, the terminals of which are both of palladium. 
The tube is shown in fig, 7, one terminal is in the form of a helix, the other straight; 
Fig. 7. 
it is 7 inches (17'8 centims.) long, and 1^ inches (3’8 centims.) in diameter, and con¬ 
tains residual hydrogen. When the tube was new the terminals were bright, but by 
continued use in consequence of occluding the gas and giving it off again they have 
become mat or frosted, and porous to a certain depth. We usually cause the spiral to 
be negative; after the discharge of a battery of 1200 cells has passed for a few seconds 
the tube which was at first free from deposit, as in fig. 7, becomes coated more especially 
near the negative, with a mirror-like metallic deposit, as in fig. 8. After the tube has 
remained at rest for a few days, seven in one instance, this metallic mirror disappears 
entirely, being absorbed by the terminals; this shows not only that the terminals give 
off occluded gas and reocclude it, but that there is formed in this particular case a 
volatile hydrogen-palladium compound. Moreover, the vacuum appears to alter con¬ 
siderably during the discharge, for a stratification which is produced in the first 
instance ceases in a short time, but the tube regains its original condition by standing 
for a few days, and stratification is again produced. 
The phenomena described have been produced very many times and shown repeatedly 
to friends who have visited our laboratory since March 13th, 1875. 
