174 Professor Sir J. J. Thomson [April 10, 



due to the positive rays coming out in front of the cathode is, under 

 the action of electric and magnetic deflexion of the rays, drawn out 

 into a continuous band, even when the pressure is such that the 

 phosphorescence due to the Canalstrahlen shows well-developed 

 patches. In the case of the Canalstrahlen, there are some rays whose 

 deflexion shows that they are negatively charged and have a mass 

 comparable with that of the positive rays. We find, too, in the case 

 of the rays travelling in the opposite direction to the Canalstrahlen, 

 that a considerable number of them are negatively charged particles, 

 indeed I think the proportion of the negative to the positive is greater 

 in this case than in that of the Canalstrahlen. I have observed cases 

 in which the phosphorescence due to the negatively charged particles 

 was little, if at all, less than that due to the positively charged ones. 



A large number of determinations were made, by the method 

 described in my first paper,* of the velocity and values of ejm for 

 these rays ; in consequence of the spot of phosphorescence being 

 drawn out into a band, the values of elm ranged continuously from 

 to a maximum value. 



Two tubes were used for this purpose, in one of them (fig. 1) 

 the electrode B was connected with a stopper ground into the tube, 

 one side of the electrode was aluminium, the other calcium ; by 

 turning the stopper either side could be made to face the tube A, 

 down which the rays passed. L M are parallel plates, which can be 

 connected with a battery of storage-cells ; when this is done, rays 

 passing between the plates are acted on by a strong electric field. 

 S is the screen at the end of the tube ; H, K the pole-pieces of the 

 electro-magnets. C is a wire fused into the tube, to serve as an 

 electrode, thus allowing A to be insulated or connected with one or 

 other of the electrodes at will. The dimensions of the parts of the 

 tube which affect the deflexions of the phosphorescence areas follows : — 



Distance between the plates L and M = 0'45 cm. 

 Length of these plates . . . .=:-}• 8 cm. 

 Distance between the screen and the 



end of the plates =4*0 cm. 



If V is the potential in volts between the plates, e the charge, 

 m the mass, and v the velocity of the rays, D the deflexion due to 

 the electrostatic field, we can easily prove that 



e ^ D 

 mv^ 5 X 10^^' 



While if d is the deflexion due to the magnetic field when a current 

 of 1 ampere is passing through the coils, it was found by the method 

 described t that 



e _ d 

 m/v ~ 2-8 X 10*' 



* Phil. Mag. xiii. p. 561, 1907. t Loc. cit. 



