181 



fying glass to find and remove every foreign particle, and 

 had washed the whole apparatus out with methyl alcohol itself. 

 These good conditions lasted only a short time, and unfortu- 

 nately a second cleansing process was not equally effective. 



I must point out that the results for benzene and acety- 

 lene are now close together. In the preliminary paper I be- 

 lieved them to differ considerably, and used them as an illus- 

 tration of the want of direst connection between the energy 

 spent and the ionisation produced. It will be seen later that 

 this effect is now clearly shown, but I was unfortunate in 

 using a comparison of benzene with acetylene as an illustra- 

 tion. 



§ 3. 



Though our knowledge of the process of ionisation 

 by the a particle is as yet only small and imperfect, it does 

 not seem out of place to draw together what facts we do 

 know, and to endeavour to connect them by some thread of 

 argument, which may be useful for a time. 



In the first place there is the fact that the ionisation 

 produced by the a particle increases as its velocity diminish- 

 es. Now, Rutherford has recently shown (Phil. Mag., Aug., 

 1906) that the particle spends energy at a uniform rate along 

 its path. It follows, therefore, that the ionisation produced 

 is not proportional to the energy spent. In my preliminary 

 paper I have already given a reason for supposing that the 

 eneigy spent and the ionisation produced are not directly 

 connected, viz., that the former is related to the atomic weight 

 by a simple law and the latter is not. 



As a temporar}^ hypothesis let us suppose that there is 

 an intervening link; that the a particle produces a primary 

 effect A, which in tiirn produces a secondary effect B. The 

 latter consists of ionisation, the former may or may not do 

 so. It is in the production of the primary effect that the 

 energy of the particle is spent. 



Since the energy spent is related to the atomic weight 

 by a simple law, since it is independent of velocity, and since 

 there is a critical speed at which all ionisation ceases, which 

 speed is the same for all atoms, it appears clear that A is 

 a sub-atomic effect. It consists in the performance of some 

 act which always involves the expenditure of the same 

 amount of energy ; and the stopping power of an atom is 

 proportional to the number of times that the act is performed 

 within it. The effect might consist, for example, in some 

 operation upon a common constituent of all atoms, such as an 

 a particle. The critical speed might be that at which the 

 moving a particle failed to penetrate, or, more generally, act 

 upon the a particle of the atom. 



