864 Mr. G-. P. Thomson on the Application of 



rays bad a constant velocity due to the anode fall of 

 potential, which is probably not the case, and the magnetic 

 deflexion, on the measurement of which their value of e/m 

 depended, was only 1*4 mm. in all. It seems probable that 

 their result is to be attributed to experimental error. 



In view of the comparative ease with which even electro- 

 negative elements can acquire a double charge in ordinary 

 positive ray analysis, it is rather surprising that no trace 

 of double charges was found for any of the metals of 

 the alkaline earths which undoubtedly have two valency 

 electrons. This casts doubt on the view that the mechanism 

 of the anode rays is to be regarded as a kind of electrolysis, 

 and suggests that it is more analogous to the spluttering 

 of the metal of a cathode. It is known * that the impact 

 of high-speed cathode particles can reduce many metallic 

 halides to the metal, so that we should expect a layer of 

 (probably molten) metal on the surface of the anode. This 

 surface is subject to a severe bombardment both from the 

 cathode rays and also from the negatively charged halogen 

 atoms, which can acquire a considerable energy by falling 

 through the unusually large anode fall of potential. Either 

 or both of these agents might be expected to liberate 

 positively charged metallic atoms from the surface of the 

 anode, which then will acquire a considerable energy from 

 the anode fall of potential. In positive rays doubly charged 

 atoms appear to be formed most easily as the result of the 

 dissociation of a compound, and one would hardly expect a 

 sufficient proportion to be visible on the photographs so far 

 taken if the process is of the nature suggested above. 



No trace of negatively charged rays was observed. This 

 might have b<-en expected from the electro-positive nature 

 of the metals and the absence of free electrons due to the 

 presence of halogens in the tube. 



In spite of the fairly high pressures used, there were very 

 few "secondary''' lines visible on the plates, though in a few 

 cases faint lines were seen going to the lithium parabolas 

 when these were so strong as to be solarized. 



Non-metallic parabolas. 



In addition to the metallic parabolas, a few due to non- 

 metallic elements and compounds were sometimes faintly 

 seen. They may be regarded as ordinary positive rays 

 derived from the gas between the anode and cathode. 

 In accordance with this view, the least deviated parts of 

 their parabolas generally had a greater deflexion than those of 



* Elster & Geitel, Wied. Ann. vol. lix. p. 487 ; Phys. Zeit. vol. xi. 

 pp. 257 and 1082. 



