636 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
piece of tinfoil. This would be necessarily later than the revocation which would take 
place when the discharge arrived at the further piece of tinfoil, i.e., before it arrived 
at the negative terminal, and d fortiori before the molecular streams that issued from 
the negative terminal in response to it had returned to the first piece of tinfoil so as to 
suffer interference there. 
We have used the above experiments for the purpose of supporting our proposition, 
both on account of their great interest and the remarkable way in which they illus¬ 
trate the whole theory of the discharge, and also because of their analogy with an 
experiment by which we shall prove for the passage of the negative discharge along 
the tube the same proposition that we have proved above for the positive discharge. 
But so far as the positive is concerned, the phenomenon of virtual shadows suffices of 
itself to prove the proposition; for the response from the negative terminal occurs 
later than the relief-response at any point by a period equal to that taken by the 
positive discharge in passing to the negative end of the tube. The molecular streams 
that pass up the tube from the negative terminal find, on their arrival at the place 
where the relief is taking place, that the molecular streams there are still continuing, 
for they indicate this by being deflected by them, and thus forming the molecular 
shadow. Hence the time required by the discharge to pass to the negative end of the 
tube must have been less than the time during which these relief-molecular streams 
were being emitted. Simple as this proof is, however, it is inferior to the experimental 
proofs that we have previously given, inasmuch as it does not point so clearly to the 
great contrast in magnitude between the two small time-quantities under consideration 
as does the complete extinction of the relief-phosphorescence in the experiment first 
described. 
So much then for the velocity of positive electricity along the tube. It remains to 
demonstrate a similar proposition as regards negative electricity. This was done in 
the following way :—A piece of tinfoil of some considerable size was laid on a high 
vacuum tube near to the negative end, the discharge being one of strong negative 
intermittence. A smaller piece of tinfoil was placed near the farther end, and they 
were connected by a wire as before. We know that this arrangement must give us 
negative special effects at the farther piece of tinfoil— i.e., negative discharges there. 
These would, if strong enough, be accompanied under ordinary circumstances by mole¬ 
cular streams, which would cause phosphorescence on the opposite side of the tube. 
But no such phosphorescence appeared. In order to test whether this was due to the 
weakness of the negative impulses from the first piece of tinfoil, the second was placed 
upon a similar tube in an independent circuit (as in the standard-tube arrangement), 
the distance between the two pieces being retained unchanged. It gave most brilliant 
phosphorescence, showing that the negative impulses were abundantly strong enough 
to have caused phosphorescence had they not been prevented from doing so. And it 
is clear from our previous remarks that the cause which prevented their doing so was 
the arrival of the negative discharge in the tube at the farther piece of tinfoil before 
