ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
639 
wholly or to a considerable extent, before the main discharge has in its passage along 
the tube arrived at the tinfoil; as there would be otherwise on such arrival a revoca¬ 
tion of the impulsive electrical action in the tinfoil, that is the cause of the very 
discharge that forms the hollow cone. But we are able to get quite clear and typical 
effects by this method, so that we are entitled to assume that the positive discharge 
has actually left the quasi-positive terminal beneath the tinfoil and proceeded along 
the tube with a velocity which we have already considered, forming the positive 
luminosity on its path. 
The phenomenon which shows in the most striking way the rapidity with which 
positive electricity leaves its source is the jwoduction of positive-special in a high 
tension tube. Attention has already been called to the wide sweep which the thin 
positive column takes to avoid the spot at which the positive-special is being produced. 
Now this avoidance is due to the fact that there is negative electricity there which has 
been left behind by the positive discharge caused by the positive impulses within the 
tinfoil. Thus the side of the tube beneath the tinfoil behaves precisely as a negative 
quasi-terminal, showing that the positive must have wholly passed away from the 
place at the time the advancing positive discharge arrives at the tinfoil, and has left 
the corresponding negative electricity free to produce its full effect. 
The evidence is strengthened if we consider the case of negative relief produced, by 
the use of a similar arrangement, at the piece of tinfoil that is nearer to the negative 
terminal. Here we are able to compare directly the two cases of the relieving tinfoil 
being on the tube, and at a distance from it. And we find that there is no difference 
in favour of the latter in the luminous phenomena of relief thereby produced. Hence 
we are fairly entitled to conclude that the positive discharge passes off in a time shorter 
than that required for positive electricity to advance along the tube, and that in so 
doing positive luminosity is formed by it as it goes.* 
So far we have nothing that would lead us to draw a distinction in time between the 
emission of electricity and the formation of the luminous phenomena which accompanies 
that emission. But when we take the correlative phenomena for negative discharge, i.e., 
the blank-space, a difficulty arises. It not only must, from its nature, be formed in 
the same time as the positive luminosity which it limits, but all the phenomena of 
revocation testify to the fact that it is so. But we have already shown that negative 
discharge does not take place with a rapidity at all comparable with that of positive 
discharge, so that we can no longer view the formation of the blank-space simply 
* It is doubtless owing to this extreme rapidity of discharge and propagation that there is so much 
luminosity and so little heat in the vacuum discharge. It has been experimentally shown that the tem¬ 
perature of striae is not greater than 100°. But all measurements of temperature are measurements of 
average effects, and such a result would be quite consistent with the hypothesis that they are in a state 
of intense heat for a very small fraction of the total period, and such a hypothesis would account for so 
large a proportion of the energy imparted to the gas passing off in the state of light inasmuch as the 
proportion of rays of high refrangibility to those of low refrangibility increases with the temperature. 
4 N 2 
