1891.] Luminous Discharge of Electricity through a (jfas. 91 



to produce a discharge from the positive electrode ; a much more im- 

 portant consideration in this case is the relative time required by the 

 two electricities to leave their respective electrodes. If the time 

 taken by the positive electricity to leave the anode is very much less 

 than that taken by the negative to leave the cathode, and especially 

 if this second time is greater than the time taken by the luminosity to 

 pass over a considerable length of the tube, there could be no difficulty 

 in understanding how the luminosity of the positive column, which 

 in these experiments practically fills the tube, should have its origin 

 at the anode. 



Now, Spottiswoode and Moulton, in their very remarkable paper 

 on the " Sensitive State of the Electric Discharge " (' Phil. Trans.,' 

 1879, p. 165), investigated the relative magnitudes of the times 

 occupied by the various processes which go to make up the electric 

 discharge, and by means of the phenomena which are observed in the 

 revocation of what are called by them relief effects, show (1) that the 

 time taken by the negative electricity to leave the cathode is so much 

 longer than the time taken by the positive electricity to leave tlie 

 anode, that the two times may be considered to belong to different 

 orders of small quantities; and (2) that the time taken by ihe 

 negative electricity to leave the cathode is greater than the time 

 taken by the luminosity to travel over the length of the tnbe (in their 

 case the tube was not very long) ; remembering these facts, the result 

 which we have obtained by the use of the revolving mirror need 

 occasion us no surprise. 



These experiments lead us to regard the discharge as the sweeping 

 down of the positive electricity from the anode with an enormous 

 velocity (about half that of light in our experiments), accompanied 

 by what is comparatively a very slow discharge from the cathode. 



The fact that the positive electricity leaves the anode more quickly 

 than the negative does the cathode, explains a very prominent feature 

 of the electric discharge : the accumulation of positive electricity in 

 the neighbourhood of the cathode. The positive electricity arrives 

 at the region surrounding the cathode before the discharge from this 

 terminal is completed; thus there will, during the greater part of the 

 discharge from the cathode, be an excess of free positive electricity 

 in the neighbourhood of the electrode, and, if the discharges succeed 

 each other with sufficient rapidity, the positive electricity will 

 accumulate until the effect of its attraction is sufficiently great to 

 cause the negative electricity to leave the cathode as fast as the posi- 

 tive electricity arrives. 



The explanation of the exceedingly rapid rate of propagation of 

 the positive column is of primary importance in any theory of the 

 mechanism of the electric discharge. The theory which seems to me 

 the most probable is that the passage of electricity (or, from another 



YOL. XL1X. H 



