UNIVERSITY 



<JH. v.] B?8 OP s ^THbDE RAYS 45 



favourable case, being always of the order 10 9 c.g.s., 

 while the electrochemical equivalent was of the order 

 10~ 7 c.g.s., or about T ^Vo that f hydrogen. 



Changing the kind of residual gas in the tube, and 

 changing the electrodes, made no difference to this 

 last value. The cathode rays were evidently inde- 

 pendent of the nature of the matter present : an 

 exceedingly momentous fact. If they were matter at 

 all, they appeared to be matter of some fundamental 

 kind, independent of the distinctions of ordinary 

 chemistry. Their velocity, however, depended on 

 the potential difference between the electrodes, in a 

 way that suggested that they were really projectiles 

 urged by the potential gradient acting along a given 

 length of path. They were propelled by the cathode 

 through an aperture in the anode, and the measure- 

 ment of their speed was made in the tube beyond the 

 anode, where they are travelling by their own 

 momentum. The distance apart of anode and 

 cathode did not, and on .the projectile hypothesis 

 ought not to, affect this speed ; for though the 

 potential gradient is steeper when anode and cathode 

 are put close together, the length of path during 

 which the particles are subject to it is diminished 

 by a compensating amount, -so that the velocity is 

 theoretically independent of the distance between the 

 electrodes, as long as the total difference of potential 

 is maintained ; it is the absolute difference of 

 potential that determines the speed. (This is a 

 familiar result of the conservation of energy, and is 

 illustrated by ordinary falling bodies.) But mani- 

 festly if the electrodes are too close together it may 

 be difficult to secure a high difference of potential 

 between anode and cathode, since they may spark 

 into each other outside the tube; and if there is 



