218 Sir J. J. Thomson : Further 



this it' the relative velocity of th^ two is greater than a 

 certain limiting velocity. This velocity is determined by the 

 condition that it is the velocity with which a corpuscle must 

 be moving when it has sufficient kinetic energy just to carry 

 it away from the attraction of the positively electrified 

 particle. The work required to separate a corpuscle from a 

 positively charged particle may be expected to be of the 

 same order as that required to ionize the particle when 

 neutral. Thus, if V is the velocity of the corpuscle when it 

 can just ionize the particle ; then, if the velocity of the 

 particle were appreciably in excess of V, a positively charged 

 particle would not get neutralized, while if the velocity of 

 the neutral particle were appreciably below T V it would not 

 get ionized. Hence the velocity of the secondaries which 

 arise in the way we are considering must be very approxi- 

 mately equal to V, a velocity which depends only oa the 

 nature of the particle, and not on the electromotive force 

 applied to the discharge-tube. This accounts for the fact 

 that although the primary line may be due to particles moving 

 with a wide range of velocities, the particles in the secondary 

 are all moving with the same velocity. 



Energy required to Ionize a Gas. — By measuring, bv the 

 method described in former papers, the velocity of the 

 particles forming the secondary rays corresponding to any 

 atom, we can determine V the smallest velocity with which a 

 corpuscle must move to ionize that atom. I find this velocity 

 for the hydrogen atom to be about 2 x 10 s cm. /sec, a 

 Velocity which would be acquired by the fall of a corpuscle 

 through a potential difference of 11 volts. Hence we may 

 take 11 volts as the measure of the energy required to ionize 

 an atom of hydrogen. 



To give to the hydrogen atom the velocity 2 X 10 8 cm. /sec. 

 requires a potential difference of 11 x 1*78 x 10 3 (taking 

 <?/m=l-78xl0 3 ), or about 20.000 volts: to give the same 

 velocity to an atom of oxygen would require 16 x 20,000 or 

 320,000 volts, a potential difference much greater than that 

 which which we apply to our tubes. Hence we can see why 

 the secondaries of the kind we are considering are only found 

 with the lighter elements such as hydrogen and helium. 



In addition to the secondaries which are due to the 

 ionization of neutral atoms, there may be atoms which start 

 with a positive charge and attract a corpuscle in their path 

 through the electric and magnetic fields. When these fields 

 are coterminous and uniform, these secondaries will, if the 

 Velocity is constant, be straight lines passing through the 



