127 
1921-22.] On Models of Ferromagnetic Induction. 
In order that this should happen, the transient deflecting force must do 
an amount of work sufficient to cause W to break away from its original 
position of stable equilibrium. If W break away, the encounter has the 
result that a deflnite amount of energy is communicated to the atom of which 
W is a part, and this energy is the source of the consequent radiation. 
Suppose, on the other hand, that the work done by the transient de- 
flecting force is insufficient to cause W to break away. In that event no 
energy is communicated in the encounter as a whole. During the first part 
of the encounter, work will be done in reversibly deflecting W, within the 
limited range in which it remains stable ; but since, by hypothesis, that 
range is not exceeded, the work done in deflecting W is recovered without 
loss as it returns to its initial or zero position. The encounter is elastic : 
there is no dissipation of energy. Oscillations are not set up. The atom 
does not emit radiation in consequence of the encounter. 
Return now to the case of an inelastic encounter in which W breaks 
away from a position of stable equilibrium and settles down, Avhen its 
oscillations have subsided, again in a position of stable equilibrium. If 
we assume the energy to be equal in both positions, it follows that the 
work done in causing W to break away is entirely spent in producing 
radiation. That work represents energy which was received in the 
encounter. Suppose the energy to have come from bombardment by an 
isolated electron. Then the least kinetic energy which such an electron 
must possess, in order to make the encounter inelastic and consequently 
to set up vibrations, is equal to the definite amount of work that is 
required to make W break away from its position of stability. If the 
electron possesses just that amount of energy (and encounters the atom 
in the most effective manner), it will give up all its energy in the 
encounter. If it possesses more, it will still give up only that amount of 
energy. If it possesses less, it will give up no energy. 
Thus the model serves to illustrate the mechanical possibility of an 
encounter which involves no scattering of energy except under quantum 
conditions. It receives energy in quanta and gives out corresponding 
quanta of radiation. 
This action of the model, as Professor Whittaker points out, has a direct 
bearing on the experiments of Franck and Hertz, McLennan, and others, 
on the emission of radiation by metallic vapours and other gases under the 
stimulus of a bombardment of electrons.* In these experiments the bom- 
* Franck and Hertz, Verh. d. deutsch. Phys. Gesells., 1914, p. 457 and p. 512 ; M‘Lennan 
(and pupils), Proc. Roy. Soc., vol. 91, A, p. 485 ; 92, A, p. 305 and p. 574 ; Phil. Mag.., Dec. 
1918. See also Davis and Gonclier, Phys. Rev., Ang. 1917 and Jan. 1919 ; Gonclier, Proc. 
