Discharge of Electricity through Exhausted Tubes. 445 



The chance of the collision in question is proportional to the 

 relative velocity (v—pcoscj)) ; and in the stationary state the 

 whole gain of energy is zero. Hence 



JJp e~ h P 2 dp d(j){2qp cos <j> (v-p cos (j>) 2 + 2q 2 v 3 \ = 0. 



In the integration with respect to </> the odd powers of cos <fi 

 vanish. Hence 



so that 



as in (61). 



f* 00 



2qv\ pdpe- h P\qv 2 -p 2 ) = 0; 

 h = l/qv 2 , 



Terling Place, Witham, 

 August 19, 1891. 



LIY. On the Discharge of Electricity through Exhausted 

 Tubes without Electrodes. By J. J. Thomson, M.A., 

 F.R.S., Cavendish Professor of Experimental Physics, 

 Cambridge. 



[Continued from p. 336.] 



Phosphorescence produced by the Discharge. 



THE discharge without electrodes produces a very vivid 

 phosphorescence in the glass of the vessel in which the 

 discharge takes place ; the phosphorescence is green when 

 the bulb is made of German glass, blue when it is made of lead 

 glass. Not only does the bulb itself phosphoresce, but a piece 

 of ordinary glass tubing held outside the bulb and about a 

 foot from it phosphoresces brightly ; while uranium glass will 

 phosphoresce at a distance of several feet from the discharge. 

 Similar effects, but to a smaller extent, are produced by the 

 ordinary spark between the poles of an electrical machine. 



The vessel in which the discharge takes place may be 

 regarded as the secondary of an induction-coil, and the dis- 

 charge in it shows similar properties to those exhibited by 

 currents in a metallic secondary. Thus no discharge is pro- 

 duced unless there is a free way all round the tube ; the 

 discharge is stopped if the tube is fused up at any point. In 

 order that the discharge may take place, it is necessary that 

 the molecules of the gas shall be able to form a closed 

 chain without the interposition of any non-conducting sub- 

 stance ; indeed the discharge seems to be hindered by the 



