194 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
presence cannot be satisfactorily ascertained, so that the column on the 
positive side of the ring appears hollow, as it is on the negative side. 
(2.) As the ring contracts from a large size, so that we get at first only the inductive 
effects from a comparatively distant conductor, the first appearances are 
those we have described for a small air-spark, and the others come in turn as 
the ring is decreased in size. - 
(3.) As the ring is moved away from the air-spark terminal the stria-effect is turned 
into the double hollow-cone-effect, and to reproduce the former effect the 
ring has to be widened. 
Next, taking the air-spark in the negative and observing the relief-effects, they 
are found to be as follows :— 
(1.) An increase of the air-spark from its initial value, zero, produces exactly the 
same series of effects as that just described under (1), with minor differences 
in the definition of the various phases. The bright and sharply-defined 
termination of the positive column of pointed or annular shape according as 
the air-spark is smaller or larger is often very beautifully shown. 
(2.) Increasing the ring was equivalent to decreasing the air-spark. 
(3.) Moving away from the positive made a pointed termination become annular. 
But this phenomenon is generally very difficult to observe. 
This identity of effect in the two cases is the strongest confirmation possible of the 
correctness of the conclusions previously arrived at. We know that the special effect 
when the positive is the air-spark terminal is caused by pulses of positive electricity 
passing to the tinfoil co-periodic with the elevations of potential at the positive 
terminal produced by the passage of charges of electricity across the air-spark interval. 
And we find that such pulses of electricity in the tinfoil produce the hollow-cone 
discharges in the tube, and these form the base of a subsequent positive column while 
the original positive column stops short within the hollow cone and its discharge 
apparently finds satisfaction in the negative electricity set free there. Applying the 
knowledge thus obtained to explain the exactly similar effects which we obtain by 
connecting the tinfoil to earth and making the negative the air-spark terminal, we see 
that we must ascribe the pulses of positive electricity which give these effects to the 
inductive action of the negative charges which rush through the tube from the negative 
terminal at each passage of electricity across the air-spark interval. These pulses set 
free positive electricity within the tube, which rushes from the glass in its normal 
form of the hollow-cone discharge to satisfy the negative discharge which by its 
inductive action originated it, and the negative electricity on the inner side of the 
tube, set free again by reason of this satisfaction of the original discharge, passing 
from the interior of the hollow cone, continues the discharge through the tube. It will 
be observed, therefore, that the interior of this luminous hollow cone is taken to act 
the part of a negative terminal in both cases. 
One difference between the two cases must be noticed here. The positive pulses in 
