212 MESSRS. W. SPOTTIS WOODE AND J. FLETCHER MOULTON 
electrically with the more distant terminal, the portion of the tube between this 
terminal and the tinfoil became totally dark, and the whole discharge was confined to 
the portion between the tinfoil and the nearer terminal. And this discharge presented 
most remarkable features. The hollow luminous cone of the positive non-relief-effect 
stretched from the interior of the tube within the tinfoil towards the nearer terminal, 
and also the blue luminosity within the tinfoil which we associate with the positive 
relief-effect made its appearance. On placing a similar ring of tinfoil in the same 
position relatively to the opposite terminal, and connecting it with the terminal most 
distant from it, an exactly similar discharge took place at that end of the tube, and the 
middle portion of the tube remained dark. 
The interpretation of these very striking phenomena follows readily from the theory 
we have enunciated. Indeed, the action in this case is so plain that it is easy to arrive 
at the interpretation of the phenomena without any reference to the theory of sensitive 
discharges ; and the importance of these experiments for our purposes is chiefly due to 
the support they give to the explanations we have put forward of the hollow lu min ous 
discharge in the positive non-relief state, and the other characteristic phenomena 
of sensitive discharges. A positive pulse coming from the secondary circuit rushes at 
once to the terminal and to the tinfoil ring. At the same instant a negative impulse 
rushes to the terminal nearest to the tinfoil ring and leaps into the tube. The impulse 
to the ring has already produced an inductive discharge into the tube of positive 
electricity, and this meets and satisfies the negative pulse from the terminal, leaving 
negative electricity predominating near the tinfoil, and this holds for a brief instant 
the positive charge in the tinfoil. That brief instant suffices for the next pulse from 
the coil (which will necessarily be of the opposite character) to come up and satisfy the 
positive charge, and thus set free the negative within the tube, which in its turn 
satisfies the positive pulse that now comes from the terminal, and then the cycle 
recommences. 
As it has reference to the phenomena of sensitive discharges described in the earlier 
part of this paper, it is worth remarking that if we do not properly adjust the breadth 
of the tinfoil, or if the tube be not of a suitable character, we fail to get a complete 
absence of discharge in the intermediate portion of the tube, but we find that super¬ 
posed upon the discharges as we have described them there is a somewhat feeble 
discharge going through the whole tube. This is precisely what we so often obtain in 
the case of the relief and non-relief-effects in tubes; the phenomena are there with all 
their characteristics, but blurred or masked by the superposition, as it were, of a further 
discharge which does not show the usual phenomena. It is reasonable to suppose that 
the cause is the same, viz. : that the external actions are not of the suitable strength, 
or the internal conditions are not such as to permit the effect upon the discharge to 
exist in all its completeness. 
We see, then, that observation of the relief-effects enables us to determine the 
ter min al from which the discharge proceeds, and the distance it goes without pro- 
