ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
637 
the negative discharge there with its attendant molecular streams had had time to 
leave the surface of the glass. 
The above conclusions are strongly fortified by an experimental result which at first 
sight appears to be inconsistent with them. We were engaged in producing the 
above-described effects in a tube of only moderate exhaust. They were very well 
manifested, showing that even in such a class of exhausts the durational character of a 
negative discharge which is violent enough to give phosphorescence is strongly marked. 
But to our great surprise we found that when we were working with a positive air- 
spark, the tinfoil nearest to the negative terminal of the tube gave a small but distinct 
patch of phosphorescence. Now the effects at that piece of tinfoil were positive- 
special effects—that is to say, there was first a positive discharge, and then, after the 
discharge within the tube had come up, the negative leit behind by the former positive 
discharge passed off to meet it. This latter discharge, therefore, must have had all 
the essentials requisite to produce phosphorescence, and that which was produced was 
a true case of positive-special-phosphorescence. But we have already shown that the 
phosphorescence of the positive-special is very decided]}" weaker than that of positive 
relief, for the latter can extinguish it and cause phosphorescence on the very spot of 
the surface of the tube at which the positive-special takes place and from which its 
molecular streams are pouring. How, then, is it that we can produce the weaker phos¬ 
phorescence when we fail to produce the stronger ? The reason is that, in the case we 
are now considering, the requisite time is allowed for its production. It is true that 
the negative discharge at the further tinfoil is far less than the relief-discharge which 
originally occurred at the nearer tinfoil, but there is no revocation in the case of the 
former. It is allowed to pour off during the whole of the time that the tube remains 
charged with positive electricity—that is to say, until the negative electricity pouring 
in at the negative terminal has reduced the tube to an electrically inert state, or 
approximately so. This is clearly a duration of the right type, since it suffices for the 
production of phosphorescence both in the case of relief and in the case of the dis¬ 
charge from the negative terminal. Thus we see that the presence of the requisite 
time-element will enable a negative discharge to produce phosphorescence hi cases 
where its absence was a sufficient bar to the production of phosphorescence by a dis¬ 
charge of far greater initial violence.* 
These experiments reveal to us the necessity of considering two other time-quantities 
which, but for the information derived from the consideration of the phenomena of 
* We may add that all question as to the phosphorescence in this case being due to positive-special was 
set at rest by slightly raising from the tube the piece of tinfoil nearest the air-spark terminal. This 
eaused the phosphorescence to fade, because when the tinfoil was in that position the positive-special 
effect was produced by a much feebler inductive action than when the tinfoil was on the tube. As the 
tinfoil was raised higher the phosphorescence continued to fade for a short time, and then grew bright 
again, the system having got so far from the tube that it had ceased to produce positive-special effects and 
had become a relieving system. The effects upon the luminosity bore witness to this series of changes 
taking place. 
MDCCCLXXX. 4 N 
