Radiation produced by slowly moving Cathode Rays. 375 



reasoning it must vary somewhat with the time) ; this branch 

 of the curve can only extend to S', where again 



Aire. _ 2r/X^ 

 /q 3 dx ' 



it will be succeeded by other similar branches. Thus 

 we should get the mean electric force in the positive column 

 represented by a fluctuating curve; this, as Graham and 

 Wilson have shown, is the case where the positive column is 

 striated. 



Negative Dark Space. — The energy communicated by the 

 corpuscles to the gas through which they move will be pro- 

 portional to the product of the number of collisions between 

 the corpuscles and the molecules and the kinetic energy 

 possessed by a corpuscle. Now we have seen that the number 

 of collisions is nn/\, while the energy possessed by the cor- 

 puscles is proportional to X\. Thus the energy given to the 

 molecules by the corpuscles is proportional to n . wX. Now, on 

 the view we have taken, n is very small close to the cathode, 

 while uX. is very small in the region about Q (fig. 8). Thus 



Fi<r. 8. 



this product would be small close to the cathode, and also 

 outside the negative glow; there would thus be some place 

 between the two where the product is a maximum. Thus, if 

 we take the luminosity of the glow as a measure of the energy 



