G02 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
XXI.— On the relief-effects in tubes of high exhaustion with a positive air-spark. 
I. Relief-phosphorescence. 
If a tube of moderate exhaust be taken, and a continuous current be passed through 
it, no phosphorescence will, it is well-known, be visible. But if an air-spark of 
considerable length be interposed between the positive terminal of the machine or 
other source of electricity and the tube, the usual green phosphorescent light will be 
seen near the negative terminal. When this is the case if the finger be placed upon 
the tube, a bright green patch will ordinarily appear on the further side of the tube 
immediately opposite the finger. In order to produce this phenomenon, the finger 
may be placed at any part of the tube, except in the immediate neighbourhood of the 
negative terminal; but as a rule the green patch will be brighter the nearer the finger 
is to the positive terminal of the tube. If the same experiment be tried with a tube 
of high exhaustion it will be found still easier to produce this phosphorescent patch; 
for the air-spark necessary to produce it in a tube of high exhaust is much shorter 
than that which would be necessary in a tube of lower exhaustion. There is, however, 
the correlative disadvantage that it is not so easy to distinguish it, inasmuch as the 
whole of the inside of the tube of high exhaust may itself be phosphorescent from the 
action of the negative terminal. 
It is not difficult to interpret this phenomenon after the theory of the intermittent 
discharge is once comprehended. The inner surface of the glass beneath the finger 
acts as a negative terminal under the influence of the advancing positive electricity 
that has come from the positive terminal, and the relief-discharge from it is sufficiently 
violent to send off streams of molecules capable of causing phosphorescence on the 
opposite side of the tube. 
The importance of the p»henomenon is much greater than at first sight would appear. 
In the first place, it affords direct evidence that there is negative discharge from the 
inside of the tube close to the finger, for these molecular streams only occur as an 
accompaniment of negative discharge. This alone is a result of great value. The 
phenomena of positive relief which in the previous paper were attributed to negative 
discharge could only be identified with the phenomena characteristic of negative 
terminals by a long and intricate process, and even after this identification had been 
satisfactorily made it was often difficult to trace the resemblance between the two sets 
of phenomena so as fully to realise their identity. But here we have no such difficulty. 
The indication cannot be mistaken; and we are enabled to affirm just as certainly that 
there is discharge from the side of the tube, and that such discharge is of negative 
electricity, as if we could test the electricity actually coming from it. And again, the 
fact that the same indication of negative discharge is obtainable from all parts of the 
tube (except perhaps in the immediate vicinity of the negative terminal) is direct 
evidence that the discharge passes throughout the tube in the form of positive 
electricity, since the response is throughout in the form of negative discharge. And 
