Experiments on Positive Rays. 239 



of kinetic energy very large compared with that acquired by 

 the fall of the atomic charge through the potential difference 

 in the dark space. From the appearance of the curve a 

 triple charge would hardly be sufficient to account for the 

 close approach to the origin. Can it be that some of the 

 mercury ions linger on from one discharge of the induction- 

 coil until the next, and thus get exposed to the intense field 

 which acts on the gas in the early stages before the normal 

 discharge is established ? 



Besides the ionization due to the collisions with the positive 

 and cathode particles, there must be some due to the " Entlad- 

 ungstrahlen " which are emitted from the negative glow as 

 well as from the Rontgen rays which are produced when the 

 cathode rays strike against the walls of the tube. At these 

 low pressures there would be little absorption of such rays by 

 the gas in the tube, so that the ionization due to this source 

 would be uniform throughout the dark space. We are at 

 present without the data which would enable us to estimate 

 what proportion of the ionization is due to these rays and what 

 to ionization by collision. The Rontgen rays, when they strike 

 against the cathode, cause it to emit negative corpuscles. 

 Now the experiments of Mr. Whiddington and others show 

 tnat the velocity of the corpuscles ejected when Rontgen rays 

 fall on a metal plate is approximately the same as the velocity 

 of the cathode rays which produce them. If we apply this to 

 the case of the discharge-tube, it indicates the existence of 

 two streams of cathode particles, one due to the impact of the 

 positive particles which emerge with comparatively small 

 velocities (the experiments of Fiichtbauer and myself indi- 

 cate that this velocity is comparable with that acquired by 

 the fall of a. cathode particle through a potential difference 

 of between 20 and 30 volts) ; and another stream due to the 

 incidence of the Rontgen rays which emerge with a velocity 

 comparable with that due to the potential difference between 

 the anode and cathode in the discharge-tube, which amounts 

 to many thousands of volts. The comparative homogeneity 

 of the stream of cathode particles shows that the produc- 

 tion of cathode particles by these two methods cannot be 

 comparable in importance. Indeed, it would seem that any 

 preponderating production of these particles by the second 

 method would lead to instability in the discharge, for the 

 ejected particle would start with a velocity comparable with 

 that due to the whole electric field ; it would in its course 

 through the dark space increase this velocity ; the Rontgen 

 ray it produced at the end of its course would eject from the 

 cathode cathode rays with a higher initial velocity than their 



