232 Sir J. J. Thomson : Furth 



er 



it is true that the cathode rays produced in a discharge-tube 

 are not all moving with absolutely the same velocity, yet at 

 the pressures at which these experiments on positive rays 

 were made (of the order of 1/300 of a mm.), the variation in 

 the velocities are not large, by far the larger portion of the 

 particles having velocities which do not vary in value by 

 more than a few per cent. So that we may conclude that 

 the larger part of the positive ions which are produced by the 

 cathode rays are produced beyond the dark space, and drift 

 into this space under the action of the feeble electric forces 

 which are found in the negative glow and the positive column. 



We have data which confirm this idea. Glasson has 

 shown (Phil. Mag. October 1911) that cathode rays with a 

 velocity of 4*7 x 10 9 cm. /sec. moving through air at a 

 pressure of 1 mm., produce 1*5 pairs of ions per cm. of 

 path. These are comparatively slow cathode particles ; in 

 our experiments the velocity of the cathode particles when 

 they left the dark spare was considerably greater than this ; 

 and as the ionization due to cathode particles varies inversely 

 as the square of the velocity, the ionization at the same 

 pressure would be less than that found by Glasson. In our 

 experiments, the pressure of the gas through which the 

 particles passed was only about 3/1000 mm. Hence, if Ave left 

 out of account the diminution in the ionization due to the 

 increase in velocity, the cathode rays would only produce 

 about 1*5 pairs of ions in a path of 333 cm. This distance 

 is much greater than the width of the dark space. This 

 confirms the conclusion that in the dark space itself the 

 number of positive ions produced by the cathode particles, 

 and therefore the number of positive rays arising in this 

 way, would only be a small fraction of the number of 

 cathode particles starting from the cathode. 



Let us now consider the ionization due to the positive 

 pnrticles themselves. Direct measurements show that these, 

 when they have passed through the cathode, produce ions 

 at tlie rate of a pair of ions for a. centimetre or so of their 

 path through air at a pressure of 3/1000 mm. 



Thus for the same number of rays the ionization due to 

 positive ones would be greater than that due to cathodic 

 ones : it is probable, however, that the cathode particles 

 are much more numerous than the positive rays. For we 

 know that when a positive ray strikes against a metal 

 plate such as the cathode of a discharge-tube, the plate 

 emits negative particles. Now the energy of a positive ray 

 when it hits the cathode is very much greater than the 

 energy required to liberate a negative particle: so that as far 



