12 Stokes, on the Nature of the Rant gen Rays. 



through the plate ? Do they get through the plate ? 

 I do not believe they do. Do they riddle the plate like 

 a bullet going through a thin piece of board ? I do 

 not think it. Suppose you have a trough containing a 

 solution of sulphate of copper, and at the ends of it 

 you have two copper plates ; if you send an electric 

 current through the trough, copper is eaten away at 

 the anode and deposited at the cathode. Now, suppose 

 you divide this trough into two by a plate of copper, 

 you still have copper eaten away at the original anode 

 and copper deposited at the original cathode. The 

 interposed plate really divides the cell into two, in each 

 of which electrolysis goes on, so that you have not 

 only copper eaten away at one end of the trough and 

 deposited at the other, but in your interposed plate 

 you have copper eaten away at one side and deposited 

 at the other. So it may be that the second surface of 

 the aluminium foil becomes, as it were, a new cathode, 

 and starts cathodic rays. This, perhaps, is not what 

 we should have anticipated beforehand. Still, there is 

 nothing unnatural in it, and nothing, it seems to me, 

 in consequence of which you would be obliged to reject 

 the theory which makes the cathodic rays to be streams 

 of molecules. There are one or two other difficulties 

 mentioned by Wiedemann, but I do not think they are 

 at all serious ; they are certainly not so serious as the 

 one I have just referred to. I will therefore pass on. 

 The possibility of deflecting the cathodic rays by elec- 

 trostatic and magnetic forces seems to be an insuperable 

 difficulty in the way of the theory which makes them 

 to be a process going on in the ether; but both of these 

 are perfectly in accordance with what was to be expected 

 on the supposition that they are streams of molecules, 

 provided you remember that these molecules are highly 

 charged with electricity. A moving charged body behaves 



