18 THE 189G ANNUAL ADDRESS. 



Leiiard constructed au apparatus commencing with a Crookes 

 tube, in which there was very higli, though not too high, 

 exhaustion, with a cathode which was either flat or cup- 

 shaped at one end. and opposite to that, in the part where the 

 cathodic rays woukl strike the glass if it were there, instead 

 of glass it was closed by a thin plate of aluminium foil, so 

 thin that it Avould not support the atmospheric pressure 

 although it was impervious to air. But as a continuation of 

 that tube he had another tube, which was also capalde of 

 exhaustion. The two tabes had glass tubes leading from 

 them to the same air-pump. There was communication with 

 the air-pump and communication between the two tubes, and 

 you could exhaust them togetlier, and the pressure would be 

 so far reduced that the aluminium plate was strong enough to 

 sustain the reduced pressure. They were both exhausted 

 together until a suitable exhaustion was produced for the 

 production of the cathodic rays in the first tube, and then the 

 connection between the two tubes was intercepted, and the 

 exhaustion of the second tube, Avliich was kept connected 

 with the air-pump, was continued for several days, until, as 

 near as he could get it, there was nothing at all, in the way 

 of gas, left in it. Wliat was the result? In the first tube 

 the cathodic rays were produced by the electric discharge. 

 They fell on the aluminium foil at the end, and then there 

 was a continuation of cathodic rays in the highly exhausted 

 tube — the vacuum tube I will call it — and these went on as if 

 they had been rays of light. They were deflected by the 

 magnet just like the original cathodic rays. 



Now at first sight that looks very much as if you had to 

 deal with actual rays, which passed through the aluminium 

 foil, just as rays of light Avould pass through a plate of glass. 

 But I think the real explanation of it is altogether different. 

 I belie^^e it to be of this nature. First I will use rather a 

 gross illustration, in order that you may the better apprehend 

 the nature of the other explanation that I am about to bring* 

 before you. Suppose that I have a row of ivory balls in 

 contact, such as billiard balls, and that another similar ball 

 strikes the first of these. The result is that the last of the 

 balls is sent off, and the striking ball and the intermediate 

 balls remain approximately at rest. Now it is conceivable 

 that something analogous to that may take place as regards 

 these so-called cathodic rays, supposing they are not rays at 

 all, but streams of molecules. It is conceivable that the 

 molecules proceeding from the cathode or negative electrode 



