CONTEMPORARY ADVANCES IN PHYSICS, XXIX 321 



(it is perhaps significant that all of these have even atomic numbers 

 and odd mass-numbers).^^ As for its numerical magnitude, I recall 

 that the insufficiencies of the theory render the recorded values — of 

 which there are some twenty-five — subject to much doubt. Such as 

 they are, it is a striking fact that none of them is more than a couple of 

 times as great as that of the proton, and some are a good deal smaller. 

 This injures still more the scheme of constructing nucleus-models out of 

 electrons and protons, for the magnetic moment of a single electron 

 not compensated by some other should of itself make the nuclear 

 moment enormously larger. Yet if we adopt the proton and the 

 neutron as the sole constituents of nuclei for the sake of banishing 

 this difficulty, it will return to haunt us so soon as we attempt to 

 reduce the number of elementary particles by conceiving either the 

 neutron as a proton united with a negative electron, or the proton as a 

 neutron with which a positive electron is combined. 



Acknowledgments 



I am much indebted to Professor I. I. Rabi and Dr. R. F. Bacher 

 for having read over the manuscript of this article and having made 

 many helpful comments, and to Professors G. Beck, S. Goudsmit and 

 A. E. Ruark for correspondence. 



15 Two isotopes of Cd M = HI and 113), two of Sn (117 and 119), one of Hg (201), 

 one of Kr (83), one of Xe (129). 



