SOME CONTEMPORARY ADVANCES IN PHYSICS— X 135 



tions Init drawn forthright from viewing atoms in equihl)rium in their 

 Normal States. 



When from the diverging pencils one proceeds to determine the 

 orientations of the atoms and their magnetic moments, one is con- 

 fused by a possibility made clear in the foregoing pages, but unsus- 

 pected at the time when the first of these experiments were performed. 

 I illustrate with the case of silver, the atoms of which flock into two 

 diverging pencils with a quite vacant space between. At first it was 

 naturally supposed that one pencil consists of atoms oriented parallel, 

 the other of atoms oriented anti-parallel to the field. The deflections 

 of the two pencils are such, that if this assumption is true then the 

 numerical value of the magnetic moment of the silver atom agrees 

 within the error of experiment with the value of eh/4:Trfxc — agrees, 

 therefore, with the notions that the angular momentum of the silver 

 atom in its normal state is h/2Tr and that the magnetic moment stands 

 in the right and proper ratio e/2fxc to the angular momentum. The 

 data were supposed to prove these notions. They also agree, how- 

 ever, with the suppositions that one pencil consists of atoms inclined 

 at 60^ to the field and the other of atoms inclined at 120^; in which 

 case the magnetic moment of the silver atom would be 2eh/4:7rfjLC, 

 suggesting that the ratio of magnetic moment to angular momentum 

 has twice the right and proper value. This inextricable tangling of 

 the efifect of orientation with the effect of magnetic moment makes it 

 impracticable to deduce quite so much from the data as was at first 

 thought possible; but plenty still remains. It is found that copper and 

 gold behave like silver, as w^as to be expected from their positions 

 in the Periodic Table. It is found that lead atoms, and (most sur- 

 prising of all!) iron atoms are not deflected at all; so that either the 

 magnetic moments of their parts balance one another completely, or 

 else they all orient themselves crosswise to the field. Nickel, on the 

 other hand, behaves as though its atoms had each a magnetic moment 

 surpassing 2eh/4:iriJ.c, while thallium responds as though that of its 

 atoms were much less than eh/AivtJiC. Finally — lest the results seem 

 too gratifying — it is found that bismuth atoms are deflected in a 

 manner quite unforeseeable. 



There is not time nor space to speak of the other method for de- 

 termining the magnetic moments of atoms, by measuring the sus- 

 ceptibilities of great quantities of them in gases or solutions; but the 

 measurements so made are also very helpful in determining the 

 magnetic moments of various atoms and ions — various groupings, 

 that is to say, of electrons around nuclei.^- All such data are of im- 



12 For the status of such measurements in 1923, the first of this series of articles 

 may be consulted. (This Journal, September, 1923.) 



