ON THE SENSITIVE STATE OF VACUUM DISCHARGES. 
627 
response from the other terminal when it has arrived in its immediate locality. The 
argument here is the same as that used in the former paper, and therefore need not 
be repeated, and it is only referred to in order to show that the evidence and the 
reasoning that sufficed to demonstrate it in the case of tubes of low exhaust are 
equally applicable and equally effectual here. And it is important to notice how 
persistently the characteristics of the positive luminosity in the case of the two air- 
sparks which we met with in tubes of moderate exhaust remain unchanged when the 
exhaust is pushed further. We know that with a positive air-spark the tendency is 
for the luminosity to shrink from the sides of the tube into a bright central column, 
smaller than the interior diameter of the tube, while the tendency of the positive 
luminosity with a negative air-spark is to spread out through the whole of the interior 
of the tube (or, at all events, to fill all the peripheral parts of it) and to become hazy 
in so doing. The pencil-like column of the positive intermittence and the diffused 
haze of the negative intermittence in high vacuum tubes represent the extreme forms 
of these peculiarities. 
We shall not, therefore, pursue the question of negative discharges any further at 
present. The general result that they correspond to the converse effect of the positive 
intermittence, except that the definition is less perfect, and that the feebler and 
secondary effects are difficult to obtain, will render the remarks in the previous 
sections applicable to the case of the negative intermittence. The special peculiarities 
of negative discharge will be dealt with in a subsequent section. 
XXVI.— General conclusions as to the electric discharge. 
I. The comparative magnitudes of the small time-quantities of the discharge. 
The problem of the physical nature of electricity is so closely bound up with the 
question of the distinction between positive and negative electricity, that the most 
hopeful way of approaching the greater problem is by solving the lesser. Now there 
is no class of electrical phenomena where the differences between the two kinds of 
electricity manifest themselves so strikingly as in the disruptive discharge, and hence 
this is the best field for studying the contrasts between the two kinds of electricity, 
with a view of ascertaining the source of this contrast. The subject is naturally a 
very wide one, and we do not purpose to deal with it generally in the present paper. 
Our object is at present simply to record a few conclusions to which we have come 
bearing upon the modus opercindi of the discharge. 
We have frequently had occasion to remark upon the extremely short duration of 
the phenomena of the electric discharge. This discharge is, however, not equally in¬ 
stantaneous in all its phenomena. In the present section we shall examine the 
various small time-quantities of the discharge, in order to get a clear idea of the 
relative shortness of the periods which they occupy, for the purpose of guidance in our 
future speculations as to their relationships one to another. 
