188 
MESSRS. W. SPOTTISWOODE AND J. FLETCHER MOULTON 
VI.— On the nature of the electrical actions by which the relief and non-relief effects 
are respectively caused or accompanied. 
It is clear, as we have seen, that tinfoil placed upon the tube and connected to 
earth produces its effect on the luminous discharge by permitting the electric forces 
caused by or arising from the discharge within the tube to produce electric displace¬ 
ment within it and the system connected with it. And as it is immaterial how the 
actual displacement is produced, we may, without affecting the result, suppose that 
there is a suitable charge of -electricity sent into the tinfoil from some source without 
the tube. This charge will, of course, be the same in quantity, but opposite in sign, to 
that which passes out of the tinfoil to the earth. We may thus discard all thoughts 
of relief or non-relief, and consider both states as being due to charges sent into the 
tinfoil from external sources at the proper times. And in the case of the relief-effect 
it is evident that these charges will be co-periodic with the electrical changes in the 
tube since they are caused by them. 
In considering this subject we are at once struck with the fact that the non-relief- 
effect has been, up to this stage of the inquiry, produced only by connecting the 
tinfoil electrically with the air-spark terminal. It is evident that the peculiarity 
possessed by this source of electricity does not arise from its high or low potential, for 
we have seen that the potential does not affect the power of the conducting system to 
influence the luminous discharge ; nor, as we have also seen, would mere fluctuations 
of potential in the wire produce any other than the relief-effect, or interfere with or 
destroy it, unless they were of like period with the pulsations in the tube. Hence the 
special effect in question must be due to the fact that connexion with the air-spark 
terminal interferes in a special manner with the electrical state of the tinfoil, and the 
peculiarity must consist in the fact that the period of the charges (and consequent 
variations of potential) is such that they keep time with the pulsations in the tube. 
Now the displacement of electricity, which would otherwise give the standard relief- 
effect, may be interfered with either by exaltation, or by reversal, or by diminution. 
The last alternative is excluded by the experimental fact that these special or non¬ 
relief-effects do not resemble any of the series of relief-effects which we have described, 
and which range from zero to the complete to-earth effect. 
To decide between the remaining alternatives we may adopt a similar course. Like 
the relief-effect, the special or non-relief-effect may be obtained in other than the 
complete form we have already described. In that form a piece of tinfoil placed upon 
the tube is connected with the air-spark terminal. If, instead of resting on the tube, 
the tinfoil (still connected with the air-spark terminal) be fastened to an insulated rod 
and made to approach the tube, a series of special effects are produced which pass from 
zero to the complete non-relief-effect. Now if this complete non-relief-effect w T ere 
merely an intensification of the relief-effects we should expect to obtain the ordinary 
relief-effects, or some close approximation to them, at some part of this chain of special 
