176 



Professor Sir J. J. Thomson [J 



[April 10, 



The magnet had been readjusted and furnished with new pole-pieces. 

 If d was the magnetic deflexion, it was found that when 1 ampere 

 was the current through the coils \ 



e 

 •mv 



d 



4-5 X 10* 



With this tube the following values of e/m and v were obtained for 

 the positive rays proceeding from the cathode in the opposite direc- 

 tion to the Canalstrahlen. The value of Y was 400 volts. 



The deflexions of the Canalstrahlen in the same tube were also 

 observed : this could easily be done by making the tube A the cathode. 

 It was found that the structure of the Canalstrahlen was more com- 

 pHcated than that of the retrograde positive rays, the former showing 

 two types of rays, characterised by values of e/m in the ratio of 

 2 to 1 : the latter showed only one type of ray, this type coinciding, 

 however, both as to value of e/m and the velocity with the type of 

 Canalstrahlen rays having the maximum value of e/m. Some of 

 the retrograde rays, as well as some of the Canalstrahlen, are deflected 

 in a direction which shows that they carry a negative charge, and 

 that their mass is comparable with that of the positively charged 

 particles. I have determined the value of e/m and the velocity of 

 these negative constituents of the retrograde rays, and find that the 



