522 HISTORY OF SCIENCE. 



the connections are made as before, no luminous track is visible be- 

 tween the positive and negative poles, and whether the positive pole 

 be at P or P' the appearances are precisely the same. But from N a 

 very faint bluish light, a b c ;/, passes straight across the bulb in a 

 direction perpendicular to the surface a '. This beam widens a little 

 as it recedes from N, and if the bulb be made of glass tinged with 

 oxide of uranium, the place where the beam impinges upon the interior 

 of the bulb is marked by a circle of phosphorescent light. The posi- 

 tion of this circle and the direction of the faint blue illumination are 

 quite independent of the position of the positive pole ; but they are 

 changed on the approach of a magnet. The action of the latter is, 

 however, markedly different from that which it exerts on the current 

 indicated by the luminous track in inferior vacuums. Also whereas 

 in the latter the track follows all the sinuosities of a bent tube, the 

 faint blue light streaming from ihe negative pole in the very highly 

 exhausted tube passes in straight lines only, and refuses to turn a 

 corner. 



This question then presents itself : Is the faint blue light due to a 



FIG. 242. 



current of electricity, or is it a stream of negatively electrified material 

 particles shot off from the negative pole ? The solution devised by 

 Mr. Crookes, of this problem, is as conclusive as it is elegant. The 

 experiment was arranged as shown at Fig. 242. A is the exhausted 

 tube, with a platinum wire sealed in at/ to form the positive pole, while 

 at the other pole there are two negative poles n and n' exactly similar 

 to each other. At a short distance from these is a metallic screen 

 perforated by two small apertures a a'. When n or n' is the negative 

 pole, the faint blue light streams through the aperture parallel to the 

 axis of the tube. But the moment both n and ' are made negative poles, 

 the beams take the divergent directions a c and a <?. Now, if these two 

 streams conveyed currents of electricity, the currents being parallel, 

 would, according to Ampere's law (Chap. XX.), attract each other. On 

 the other hand, the observed divergence is precisely what ought to be 

 seen in the case of a stream of negatively electrified particles. 



When the negative pole is concave instead of plane, the stream 

 radiating from it converges to a focus. If, at this focus, a strip of 



