218 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
that there is a change of electric tension in the tinfoil at every pulse, and it cannot 
possibly tell us at what point in the period of the pulse the change occurs. To 
ascertain this we must use some tests which in themselves partake of the periodicity 
of the current itself. 
The simplest experiment is that of taking two pieces of tinfoil, one near the air- 
spark end and one near the opposite end of the tube, and connecting them with a 
thin wire. 
In an ordinary sensitive discharge, if we thus place a piece of tinfoil somewhat 
near to the air-spark terminal and connect it with a larger piece of tinfoil at a 
considerable distance along the tube, we shall find that we have complete relief-effect 
at the first-mentioned spot. And if we examine the effect at the larger piece of 
tinfoil we shall find indications of the non-relief-effect. If now we reverse the 
arrangement and place the larger piece near the air-spark terminal and the smaller 
piece at a considerable distance from it, we shall find that while we have feeble 
relief-effects at the larger piece we have strongly marked non-relief-effects at 
the other.* 
Nor must it be thought that it is only the less definite types of relief and special 
effects that can be thus produced. If the pieces of tinfoil be made in the form of 
rings surrounding the tube, all the effects previously described can be produced in 
this manner. If, for instance, with a positive air-spark a broad ring of tinfoil near 
the positive end of the tube be connected to a narrower ring near the opposite 
terminal, we shall, with a proper adjustment of the air-spark, obtain the segmentation 
of the luminous column and the formation of the hollow cone, just as definitely as 
if we had joined the narrow ring to the positive terminal of the tube. A similar 
effect can be produced with a negative air-spark, provided the narrower ring is the 
one that is nearer to the air-spark terminal. 
Now these effects can scarcely be reconciled with any theory which makes the 
discharge simultaneous throughout the tube. Suppose that it is so, then if we take 
A to be the smaller piece of tinfoil (which we shall suppose is the nearer to the 
air-spark terminal) and B to be the larger piece, it is obvious that the sudden change 
* We have seen that the negative special effect is identical with the positive relief-effect; and it might, 
therefore, be thought that, after all, the effects seen are nothing but the relief-effects belonging to a 
discharge from the terminal nearest to the tinfoil, somewhat as was the case when a coil is used. But it 
is easy to show that this is not the right view to take of the matter; for if, instead of connecting the 
tinfoil that is further from the air-spark with that which is nearer, we connect it with earth, so as to 
give to that part the full relief it desires, we find it to be an effect of an opposite character, showing that 
the former effect was truly a special effect. 
Another conclusion of some importance may be drawn from these experiments. We see the relief-effect 
in A causes B to be so raised in potential as to give marked non-relief-effects. Hence the magnitude of 
the electric displacements which accompany relief-effects, though of course not so great as those 
accompanying non-relief-effects, which are the same as those of the discharges themselves, are yet 
comparable with- tliem, and are quite capable of causing inductive discharges in the tube. 
