184 Professor Sir J. J, Thomson [^pril 10, 



mass M, and at a given distance from it, the condition that the ion 

 should be able to get away from the molecule is that 



m + M 



should be greater than a certain quantity depending on the force 

 between the molecules and the ion and apsidal distance between 

 them. Now if the force were independent of the mass of the mole- 

 cule, we see that it would require a greater value of ?7zY- to separate 

 the ion from the molecule, when M is small compared with m, than 

 when it is large. Thus if the kinetic energy of the ions were gradu- 

 ally to diminish say by collisions with the molecules, then if there 

 were molecules of different masses in the gas through which the 

 Canalstrahlen are moving, combination would occur first with the 

 molecules of smallest mass, while the heaviest molecules would be 

 the last to combine. The lightest molecules would thus have the 

 first pick of the ions, which would therefore tend to be absorbed by 

 the lightest gases. The force between an ion and a molecule is 

 proportional to the volume of the molecule ; and if the volume of a 

 molecule were to increase as rapidly as its mass, the preceding 

 considerations would not be valid. We have every reason, however, 

 to believe that the changes in the volume of the molecules are not 

 comparable with the changes in the masses : that, for example, the 

 volume of a molecule of oxygen, instead of being sixteen times that 

 of a molecule of hydrogen, is hardly more than twice, so that the 

 increase in the forces exerted by the heavier molecules is not sufficient 

 to counteract the influence of the increased mass. 



The Nature of the Positive Ions in different Gases when 

 the Ionization has settled into a steady state. 



The Canalstrahlen are formed in very intense electric fields, and 

 the kinetic energy which they possess tends to prevent them combin- 

 ing with the molecules and corpuscles around them ; they are thus 

 under quite different conditions from the ordinary ions produced in 

 a region where the electric force is small or absent, for these have 

 time before being removed from the field to enter into combination 

 with the molecules, the system of molecules and ions getting into a 

 steady state if the source of ionization is constant. The difference 

 between this case and that of the Canalstrahlen may be compared with 

 the difference between the state of a mixture of different chemical 

 substances after they have entered into combination and settled into 

 a state of equilibrium and when they were first mixed. I thought 

 that it would be interesting to determine the values of e/m for the 

 ordinaryLions simultaneously with the determination of e/m for the 

 Canalstrahlen in the same discharore-tube. The method used to 



