ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
£QO 
Duo 
obtain the most splendid virtual shadows. Now the discharge from beneath the 
tinfoil will certainly not last longer than that from the effective negative terminal of 
the tube, and yet we find that the negative discharge from beneath the tinfoil and its 
accompanying molecular streams are still in full vigour when the molecular streams 
from the negative terminal arrive at that part of the tube where they are situated, for 
they interfere with one another. We need scarcely repeat that there is every evi¬ 
dence that this interference continues for a substantial portion of the period during 
which these streams respectively last. 
These experiments suffice to show that the duration of the negative discharge is 
not less than the time occupied by the passage of molecular streams along the tube, 
but is comparable with it. We shall now proceed to show that the time occupied by 
the passage of either hind of electricity along the tube is incomparably shorter than that 
occupied by the emission of these molecular streams , or (ivhich is the same thing) than 
that occupied by the negative discharge. 
The truth of this proposition, so far as the positive discharge is concerned, is shown 
by the following experiment:—If a piece of tinfoil be placed near the positive end of 
a tube through which a discharge with strong positive intermittence is passing, and 
be connected by a wire along the tube with a similar piece of tinfoil near the negative 
terminal, we shall obtain, as we have already said, positive relief-effects at the former 
piece of tinfoil and positive special at the latter. The relief-effects will not be in¬ 
creased by raising the latter piece of tinfoil from the tube and placing it as far from 
the tube as possible, keeping it at the same distance from the other piece of tinfoil. 
But a very remarkable exception presents itself. If the tube be of sufficient exhaust 
or the air-spark sufficiently long to cause the relieving system, formed by the tinfoil 
and the wire, to give rise to relief-phosphorescence when at a distance from the tube, 
it will be found that this disappears when the system is again lowered down so as to 
rest along the tube. The relief-effects, so far as luminosity is concerned, will be un¬ 
changed, or, if anything, increased. But the relief-phosphorescence which should have 
accompanied them will be found to have disappeared. 
Extraordinary as this phenomenon appears at first sight, it will become perfectly 
intelligible if we accept the hypothesis that the electricity, on its discharge into the 
tube, spreads along it with a rapidity which enables it to pass to the other end in a 
very small fraction of the time required for a negative discharge, or its attendant 
system of molecular streams, to pass off from their source. The positive electricity, on 
bursting into the tube and arriving at the place where the first piece of tinfoil is 
situated, produces an impulsive electric tension upon it, which makes it summon from 
the more distant piece a supply of negative electricity. This summons passes along 
the wire in so short a time that, as we know, it arrives at the further piece of tinfoil 
before the discharge has arrived at the part of the tube in its immediate neighbour- 
or shorter than that to the terminal. Moreover it has been demonstrated in the former part of this 
section. 
