216 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
In corroboration of the statement that these tongue-shaped luminous columns are 
parts of two distinct incomplete discharges, we may add that the magnet shows that 
they represent discharges going in opposite directions, the positive electricity in each 
proceeding from the base of each column towards the apes. 
Similar phenomena, save in respect of the shape of the luminous columns, are seen 
when the two terminals are joined to the negative terminal of the machine; but we 
shall not dwell here on the resemblances or differences between the two phenomena. 
The two kinds of unipolar discharge, however, are at once distinguishable from one 
another, and this may be made use of as a final proof of the correctness of the 
conclusions arrived at as to the nature of the electric displacements which cause the 
relief-effects. If a piece of tinfoil be placed on a tube through winch a discharge 
passes, and this tinfoil be electrically connected with one terminal of a tube of suitable 
resistance, it will be found that when an air-spark is introduced into the discharge a 
unipolar discharge will appear in the tube whose terminal is connected with the tinfoil 
(Plate 20, fig. 24). If the air-spark be in the positive circuit, this will be found to be 
a positive unipolar discharge ; if the air-spark be in the negative circuit it will be a 
negative unipolar discharge. This test is made still more certain and useful by taking 
the wire from the tinfoil to both terminals of the auxiliary tube. The result will then 
be a double unipolar effect, the nature of which is much more readily recognisable. 
This gives a very convenient test of the sign of the disruptive discharge at any part of 
the tube, and has been frequently employed with great advantage in cases where the 
complexity of the circumstances has made it difficult to ascertain it directly. 
The following experiment is instructive in relation to unipolar discharges. If, with 
the arrangement described on page 214, we place upon a tube containing a positive 
unipolar discharge a ring of tinfoil not very far from the further end of the tube, and 
connect it with the positive terminal, the discharge is at once completed, and continues 
to pass throughout the tube (Plate 19, fig. 23). On examining it we find that from 
the interior of the tube within the ring of tinfoil proceeds the usual hollow conical 
discharge characteristic of the non-relief state with the air-spark in the positive, and 
this stretches in the usual way towards the negative terminal of the tube, and the 
luminosity that starts from the positive column comes up to and stops at the limit 
of the dark space within this cone in the usual manner. Thus we have by our tinfoil 
ring cut the tube into two, each of which shows its own separate discharge; and the 
tension which was insufficient to cause a discharge through the whole tube is equal 
to the task of supporting the two shorter discharges. And, just as in the correspond¬ 
ing case in the ordinary sensitive discharge, the pulsations in the tinfoil cause positive 
induction discharges which are strong enough to reach the negative terminal, and 
the relief of the tension in the tinfoil winch ensues on the occurrence of the dis¬ 
charge into the tube from the positive terminal sets free negative electricity from 
beneath the tinfoil which satisfies the discharge from the positive terminal. Thus 
the interior of the tube within the ring of tinfoil acts alternately as a positive and 
