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



electroscope and charged up positively to about 40 volts ; the disk 

 now acquired a positive charge, when the cylinder was charged to 

 - 40 volts the disk got a negative charge. This shows that the gas 

 between the cylinder and disk was ionized, though the magnetic 

 field prevented any negative corpuscles from entering this region. 



Though we have given reasons for thinking that the Rontgen 

 rays are not the cause of the ionization in the side tube when this 

 is exposed to strong magnetic fields, soft Rontgen rays are produced 

 by the impact of the primary cathode rays against the molecules of 

 the gas in the tube. This was proved by covering the end of the 

 side tube (Fig. 8) with thin aluminium foil and placing in the 

 side tube behind the foil an insulated metal plate connected with a 

 charged electroscope. The escape of electricity from this plate 

 could not be ascribed to ionized gas making its way from the main 

 tube into the side one, for the only channel of communication was 

 through a long stretch of glass tubing from the main tube to the 

 pump, and then through another long tube from the pump to the 

 side tube ; since the opening between the main tube and the side 

 tube was closed, it was necessary to exhaust them separately. When 

 the primary and secondary cathode rays were well developed and the 

 main tube filled with a bright glow, the charge from the electroscope 

 rapidly leaked away whether it were positive or negative. The gas 

 in the side tube is thus ionized by rays which have passed through 

 the thin aluminium foil. The leak was, however, completely stopped 

 when, by means of a strong magnetic field, the primary and secondary 

 cathode rays were rolled up into a small bundle at the mouth of the 

 tube, from which they emerge just above the aluminium foil. In 

 this case the length of the path of the rays after coming through 

 the tube was only 2 or 8 mm., and there was hardly any luminosity 

 in the tube. The aluminium foil prevents the ionization in the side 

 tube in this case, for if the foil is removed the gas, as we have 

 already stated, is ionized. 



The preceding experiments are in harmony with the view that 

 neutral doublets are one of the stages in the process of ionization ; 

 they must, however, be regarded as only preliminary. More extended 

 experiments are necessary before we can be certain that the effects 

 are not due to some very easily absorbed kind of radiation or to the 

 diffusion of very slowly moving ions. 



AVe have hitherto considered the case when the primary ionization 

 was due to cathode rays, but there are reasons for thinking that 

 similar doublets are produced when the ionization is produced by 

 positive rays. Thus Fiichtbauer* found that the velocity of the 

 secondary cathode rays from metals was the same whether they were 

 produced by cathode rays or Canalstrahlen. It is sometimes argued 

 that the much greater difficulty experienced in saturating a gas 



* Loc. cit. 



