CATHODE RAYS. 163 



from the cathode rays acts itself like a cathode, though kept perma- 

 nently to earth. In the case of the thick brass the effect seems much 

 more likely to be due to a sudden change in the potential of the outer 

 cylinder at the places where the rays strike rather than to the penetra- 

 tion of any kinds of waves or rays. If the discharge in the tube was 

 perfectly continuous the potential of the outer cylinder would be con- 

 stant, and since it is connected to earth by a wire through which no 

 considerable current flows, the potential must be approximately that of 

 the earth. The discharge there can not be continuous; the negative 

 charge must come in gusts against the ends of the cylinder, coming so 

 suddenly that the electricity has no time to distribute itself over the 

 cylinder so as to shield off the inside from the electrostatic action of 

 the cathode rays; this force penetrates the cylinder and produces a 

 discharge of electricity from the far side of the brass. 



Another effect which I believe is due to the negative electrification 

 carried by the rays is the following: In a very highly exhausted tube 

 provided with a metal plug I have sometimes observed, after the coil 

 has been turned off, bright patches on the glass. These are deflected 

 by a magnet and seem to be caused by the plug getting such a large 

 negative charge that the negative electricity continues to stream from 

 it after the coil is stopped. 



An objection sometimes urged against the view that these cathode 

 rays consist of charged particles is that they are not deflected by an 

 electrostatic force. If, for example, we make, as Hertz did, the rays 

 pass between plates connected with a battery, so that an electrostatic 

 force acts between these plates, the cathode ray is able to traverse this 

 space without being deflected one way or the other. We must remem- 

 ber, however, that the cathode rays when they pass through a gas make 

 it a conductor, so that the gas, acting like a conductor, screens off the 

 electric force from the charged particle, and when the plates are 

 immersed in the gas, and a definite potential difference established be- 

 tween the plates, the conductivity of the gas close to the cathode rays 

 is probably enormously greater than the average conductivity of the 

 gas between the plates, and the potential gradient on the cathode rays 

 is therefore very small compared with the average potential gradient. 

 We can, however, produce electrostatic results if we put the conductors 

 which are to deflect the rays in the dark space next the cathode. I 

 have here a tube in which, inside the dark space next the cathode, two 

 conductors are inserted; the cathode rays start from the cathode, and 

 have to pass between these conductors ; if, now, I connect one of these 

 conductors to earth there is a decided deflection of the cathode rays, 

 while if I connect the other electrode to earth there is a deflection in 

 the opposite direction. I ascribe this deflection to the gas in the dark 

 space either not being a conductor at all, or if a conductor, a poor 

 one compared to the gas in the main body of the tube. 



Goldstein has shown that if a tube is furnished with two cathodes, 



