Experiments on Positive Raj/s. 22'^ 



between parallel plates maintained at different potentials and 

 measuring the current sent across the plates, I have found 

 that these heavy particles can produce a very considerable 

 amount of ionization. The question arises, how is this 

 ionization produced ? The negatively electrified corpuscle 

 must, in order to escape from the attraction by which it is 

 bound to the atom, possess a certain amount of kinetic 

 energy; this, if the mass is as small as that of a corpuscle, 

 requires it to have a very high velocity. One method by 

 which a comparatively slow positive ray might start off a 

 corpuscle at a high speed is considered on p. 235 when we 

 discuss the difference between ionization by cathode and 

 positive rays. Again, if the corpuscle were firmly attached 

 to a very much greater mass, this system could possess the 

 requisite amount of energy with a very much smaller 

 velocity, a velocity no greater than that possessed by the 

 heavier positive rays, and which these rays could com- 

 municate to the system by colliding against it. This method 

 of ionization might occur in a compound molecule if the 

 negative charge clung, say, to an atom of oxygen, while it 

 would not occur if the atoms were uncombined, when there 

 would be no body o£ a mass comparable with that of an 

 atom for the corpuscle to cling to. There are some phe- 

 nomena which strongly suggest some process of this kind. 

 For example, when the liquid alloy of sodium and potassium 

 is bombarded by positive rays, it is only the specks of oxide 

 on the surface which give out the sodium lines, the clean 

 parts of the surface are quite dark. On the view just given 

 this arises from the atom of oxygen being knocked off the 

 atom of sodium with which it is combined and carrying off 

 a negative charge, leaving the atom of sodium positively 

 electrified, and in a condition to give out its spectrum : with 

 the pure metal this process does not occur. There are 

 some phenomena connected with the luminosity of flames 

 which are also in accordance with this idea. 



The great sensitiveness of willemite to cathode rays and 

 to fast positive ones, and its inertness under the slower 

 positive rays, would on this view be due to its luminosity 

 being mainly produced by ionization of the type due to 

 particles which have a velocity of the order 2 x 10 8 cm./sec. : 

 velocities of this magnitude are possessed by the atoms of 

 hydrogen and helium and the molecule of hydrogen, but not 

 by the other atoms in an ordinary discharge-tube. 



The fact that on the photographs the intensity of the lines 

 due to particles of different kinds is not proportional to the 



