ON ELECTRICAL DISCHARGES THROUGH RAREFIED GASES. 
167 
produced by artificial means, and where the periodicity is therefore capable of being 
varied at will. Moreover, the usefulness of an investigation into the phenomena of 
discontinuous discharges is not wholly dependent on the theory which regards all 
vacuum discharges as discontinuous; for, as the mathematician is often obliged to 
reason from the discontinuous to the continuous—from the polygon to the curve—so 
the physicist may ultimately be best able to arrive at the causes of the phenomena 
accompanying the ordinary vacuum discharges (whether these be absolutely continuous 
or only relatively so) by a study of the phenomena which rapidly intermittent discharges 
present. And the importance of this method is greatly increased when we consider 
that no one of the usual phenomena of ordinary vacuum discharges is necessarily 
absent from or incompatible with the discontinuous discharge; and that if the theory 
of the discontinuous character of the ordinary vacuum discharge be correct, this method 
oilers us the prospect of constructing a continuous chain of phenomena, extending from 
the single flash of a Leyden-jar discharge to the steadiest striated column which can 
be produced by a battery-current. 
II —The sensitive state is due to a periodic intermittence in the discharge of consider¬ 
able rapidity and regularity , the quantity of electricity in each individual 
discharge being sufficiently small to permit the discharge to be instantaneous* 
In order to establish this proposition we shall first describe all the modes known to 
us of producing sensitive discharges, and shall show that in all cases the circumstances 
of the discharge give great probability to the idea that it is discontinuous or inter¬ 
mittent in the manner above described. We shall then give direct experimental 
evidence that in all these cases such intermittence actually exists in the resulting 
discharge. And lastly we shall examine the cases in which we know that inter¬ 
mittence actually exists in non-sensitive discharges, and shall show that such inter¬ 
mittence has not the requisite characteristics stipulated for above, but is of a wholly 
different type from that which we have described as the cause of sensitiveness in 
vacuum discharges. 
The simplest method of producing the sensitive state is by conducting through the 
exhausted tube a current from a Holtz machine or some other constant source of 
* In speaking of the sensitive state as being due to an intermittence in the discharge, the authors of 
this paper are treating of discharges lasting for a finite time. It was, naturally, in this form that sensitive¬ 
ness was first known to them; but in the course of the inquiry they have discovered, as will hereafter 
appear, that each individual discharge is sensitive, and that the sensitiveness of the intermittent discharge 
is due solely to the sensitiveness of its individual members. The effect of the intermittence is only to 
change through rapid iteration the instantaneous phenomena presented by each discharge into an appear¬ 
ance which to the eye is steady and continuous. 
