198 Professor Sir J. J. Thomson [April 10, 



was produced from the hot Wehnelt cathode C, these passed through 

 a hole in the anode A, the anode was earthed, the primary rays 

 passed over the top of the side tul}e, T ; across the top of the tul3e 

 were stretched two parallel pieces of wire gauze about a millimetre 

 apart : the upper gauze was earthed, the other could be charged 

 negatively by connecting it with the negative terminal of a battery 

 of small "^storage-cells, the positive terminal of which was earthed. 

 When the lower gauze was earthed as well as the upper, the tube was 

 filled with the glow due to the secondary rays. When the lower 

 gauze was charged to a negative potential of about 40 volts, this 

 glow become exceedingly faint ; but that the gas below the gauze 

 was ionized was shown by the fact that when the negative potential 

 of the lower gauze was increased to about 200 volts, a potential quite 

 insufficient to produce luminosity in an unionized gas, the tube 

 again became full of luminous glow. Thus something capable of 

 ionizing the gas was able to traverse the strong electric field. There 

 are two sources of ionization which have to be eliminated before we 

 can assign this ionization to the existence of neutral systems travers- 

 ing the electric field — the ultra-violet light coming from the 

 luminous discharge in the main tube, and soft Rontgen rays pro- 

 duced by the slowly moving primary cathode rays. To test whether 

 it was due to ultra-violet, a thin plate of quartz was placed over the 

 top of the upper gauze : with this arrangement no luminosity could 

 be detected in the side tube under the conditions as to potential and 

 so on which gave bright luminosity in the tube when the quartz was 

 absent. Hence I conclude that the luminosity was not due to ultra- 

 violet fight. To test whether it was due to soft Eontgen rays, 

 taking the quartz away, I got a bright luminosity in the side tube 

 with the primary rays passing horizontally down the tube, then by 

 means of a magnet I bent the primary rays so that they struck the 

 glass of the tube just above the side tube, the path of the rays being 

 represented by the dotted line of the figure. This made the rays 

 themselves further from the side tube, but brought the places where 

 they struck the glass, the sources of the Rontgen rays, much nearer 

 to that tube ; so that if the ionization in the side tube were due to 

 Rontgen rays it should be increased by the introduction of the 

 magnet, while if it were due to the neutral doublets it would be 

 diminished. As a matter of fact the luminosity in the side tube 

 almost disappeared when the rays were deflected in this way, showing 

 that it was not due to Rontgen rays, while the effect is what we 

 should expect if the ionization were due to uncharged systems. 



In the preceding experiments there is the possibility that the 

 ionization might arise in some such way as the following. The 

 secondary cathode rays would have to penetrate some way between 

 the two pieces of gauze before they were stopped, and if they collided 

 against the molecules of the gas they might ionize it : the positive 

 ions so produced would, under the action of the electric field between 



