﻿Mr. Norman Campbell on Delta Rays. 57 



always fulfilled, — E and T, respectively the currents repre- 

 senting the whole of the emergent and the whole of the incident 

 radiations, are given by 



E = (M-)-Zn m 



I = (M + )-Z'J [L) 



On the other hand, if the simple theory of the last paper 

 were correct, and, in the absence of an electric field, all the 

 rays from one electrode were received by the other, it was 

 shown that 



E.-Z-(M + )-|. 

 I = Z-(M-)J K) 



It is to be noted that, on these suppositions, if the emergent 

 and incident radiations are equal, we should have 



Z' = Z. 



It was noted in the last paper that the estimates of E and 

 I given by (1) and (2) do not, in general, agree ; and the 

 question arises which of the two is the more reliable. In the 

 last paper, complete confidence was placed in the simple 

 theory and the estimates given by (2) were accepted. This 

 choice was made because, in order to interpret the results so 

 as to give information as to the quality of the rays, it was 

 necessary to reckon both branches of the curve from Z as 

 zero ; it was natural to reckon from the same point when 

 estimates of quantity were to be made. But the arguments 

 of the foregoing paragraphs have shown that the theory of 

 the last paper is absolutely unreliable and that, accordingly, 

 no trust can be placed in estimates of E or I based on (2). 

 At least in cases when the surface of the two electrodes are 

 different, Z is not the current which flows when all the rays 

 from one surface strike the other ; for, if the rays can escape 

 more easily from one surface than the other, the electrode 

 from which the rays can more easily escape will lose more 

 electrons than it receives, even if the number of rays generated 

 is the same for both electrodes. 



This point is strikingly illustrated by the curves of fig. 2. 

 It will be observed that Z is very much smaller than Z', but 

 it does not necessarily follow that the quantity of the rajs 

 generated at the two electrodes is different. For we imagine 

 that the rays leave a metal surface much more easily than a 

 surface of soot ; and, when there is no difference of potential 

 between the electrodes, A will be receiving from B more 

 electrons than it gives up to B. Accordingly, as is found, 

 the current received by A will be less algebraically than it. 



