ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
647 
the free positive electricity of the discharge and the locus of demand for negative 
electricity than we could have predicated independently of this experimental evidence. 
It will be noticed that this experiment showed that the positive luminosity, plus an 
enveloping shell of the breadth of a blank-space, was the limit of the effective electric 
field. The rest of the tube was not used in the discharge. 
A tube was taken of tolerably good exhaust enclosing a spiral of wire. There was 
a positive column in the tube sufficiently nearly filling up the whole section of the 
tube to make it pass through the coils of the spiral, so that the spiral was, as it were, 
bathed in positive luminosity. So long as the tube was not touched, the wire did not 
act in any way upon the positive luminosity, but if the finger were brought in contact 
with the side of the tube at a point where it was touched by the spiral, a blank-space 
appeared surrounding the wire of the spiral whenever it had been touched by the 
positive luminosity. This shows that throughout the luminosity there was a demand 
for negative electricity, which caused the formation of a blank-space around the wire 
and negative discharge therefrom so soon as the relief afforded to the wire from the 
finger permitted it to part with its negative electricity. 
The drawback to this experiment was that it was not possible to test whether there 
was negative discharge equally from all parts of the wire, or whether it was only 
from those parts which were in the midst of the positive luminosity. The only test of 
the existence of negative discharge in such a tube being the appearance of a blank-space, 
it is obvious that no conclusive evidence could be obtained of its existence where 
there was no luminosity out of which the blank-space could be cut and by which it 
could be bounded. But this want is supplied to some degree when we come to 
tubes of higher vacua, for the negative discharge is there accompanied by molecular 
streams ; and by projecting relief-phosphorescence upon the conductor within the 
cylinder, and observing the bulging out of the shadow, we can detect negative dis¬ 
charge even when no luminosity is visible. These experiments have been referred to 
in Section XXIII., and they show that the thin positive column in high vacua indicates 
the local presence of a very intense demand for negative electricity greater than that 
which is experienced generally throughout the tube, since the result of the positive 
column passing across the end of a wire within the tube is to cause a very considerable 
bulging out of the shadow even when no relief is given to the wire from without the 
tube, and therefore when the wire is subjected throughout its whole length to the 
ordinary electric tension produced by the presence of the discharge in the tube.'" 
* Since the above was written these conclusions have received striking confirmation from the following 
experiment. A long tube of high exhaust was taken in which there was a loose wire of considerable 
length, and straight in its general direction, but bent in one or two places into sharp angles. With a 
positive air-spark the tube gave the long, thin, pencil-like column of positive luminosity of which we have 
spoken. When this did not come in contact with the wire no special appearances presented themselves, 
but when by the approach of the hand or otherwise it was made to move up to and come in contact with 
one of the angles in the wire, which rested on the inner surface of the tube, a bright patch of phosphor¬ 
escence of an oval contour appeared, commencing at the angle and extending from it, evidencing incon- 
4 O 2 
