14 



Photographic 

 Plate 



Re 



consequently, so to speak, a large penumbra. Thus the 

 trace upon the plane of the paper of all the a rays is such 

 as is represented in Fig. 3, the deflections being all exag- 

 gerated so as to be capable of depiction. 



Now suppose an aluminium 

 plate is placed, as in M. Bec- 

 quereFs experiment, over the slit, 

 so that the a particles have to pass 

 through it on the way to the photo- 

 graphic plate. M. Becquerel sup- 

 poses that there ought therefore to 

 be an increased displacement of the 

 photographic image. But this is 

 not so. The path of any one a 

 particle is slightly deflected, but 

 the whole trace is not appreciably 

 disturbed. The aluminium dimin- 

 ishes the range of every a particle 

 by the same amount, but the only 

 result is to cut off all the rays which 

 would have gone past a certain 

 point, say Q, and to cause them to 

 take the places of those rays which Fig. 3 



fell short of Q; these latter being 



further shortened. This does not in the least affect the posi- 

 tion of the outer edge of the trace upon the photographic 

 plate; and though there must be a slight movement of the 

 inner edge, so that the trace is somewhat narrower, the 

 change is so small that it could not possibly be detected, as 

 a glance at the photograph will show. Magnetic dispersion 

 of the a rays does exist : it has been directly shown by 

 Rutherford,* and, as I think, indirectly by M. Becquerel's 

 own experiments, in the peculiarities of the curvature of his 

 photographic traces. But it could not be shown in the man- 

 ner of the experiment which M. Becquerel now describes. 

 That would be analogous to the search for evidence of the 

 motion of the stars in the line of sight in the displacement 

 of the visible spectrum as a whole, whereas the measurement 

 to be made is of the displacement of some Fraunhofer line 

 in the spectrum, i.e., of one set of waves which can be isolated 

 for consideration. It is here that Rutherford's experiment 

 is differentiated from that of M. Becquerel. The former em- 

 ployed as a source of rays a wire coated with a thin layer of 

 RaC emitting o. particles of uniform velocity, which is 

 analogous to confining one's attention in the star problem to 



1905. 



Also quite recently by Mackenzie, "Phil. Mag." Noyember 



