Electric Discharge in Gases. 37 



and flat electrodes, and when the gas has the density for 

 which only a small potential is necessary for the discharge, 

 the image in the mirror often appeared continuous when 

 resistances were included, and, when they were removed, 

 resolved itself into a series of separate images. 



With discontinuous discharges, whether with or without 

 included resistances, in both cases the result obtained * by 

 Gr. Wiedemann and R. Riihlmann held good: — 



" If the positive electrode is connected with the machine, 

 and the negative put to earth, the number of discharges is 

 always smaller than with the reverse mode of connexion ; the 

 quantity of electricity necessary to bring about a discharge is 

 therefore greater in the former case than in the latter." 



It is a matter of indifference which of the two poles of the 

 machine is positive and which negative. A reversal of the 

 poles of the machine can always be produced by simply stop- 

 ping the machine and starting it again. 



The influence of the introduction of resistances may be 

 explained as follows :— 



To bring about a discharge from a metallic electrode, the 

 electricity must be accumulated upon it with a perfectly 

 definite density. If quantities of electricity are present upon 

 conductors connected with the electrode, these may, when once 

 the discharge has begun, and if they can reach the electrode 

 quickly enough, leave the electrode at a smaller potential than 

 is actually necessary for the commencement of the discharge. 

 Such accumulated electricity is, however, present in the larger 

 electrophone machines in a high degree, and in the smaller 

 machines in a smaller degree. To the conducting wires to the 

 poles and to the combs of the machine and the neighbouring- 

 parts of the glass plates there are, between any two discharges, 

 considerable quantities of electricity given off, which, as soon 

 as discharge from an electrode has been established, flow to it 

 and continue the discharge through the discharge- tube. If 

 the connexions are metallic, the flow is much more rapid ; the 

 time required to pass from the machine to the electrode is small 

 in comparison with the definite, although immeasurably small ? 

 interval of time occupied by the discharge, and the whole 

 quantity of electricity passes through at once. But at the 

 same time the density upon the whole conducting system in 

 connexion with the electrode sinks at once to zero ; and a 

 longer time must elapse until it is again raised by the machine 

 so high that a new discharge can take place. If we include 

 larger resistances, the electricity will only flow slowly to the 

 electrode when once the discharge has taken place. The 

 * Pogg. Ann. cxlv. pp. 235, 344 (1872). 



