118 



correspondence pi'inciple we can directly conclude that the Larmor theorem in general 

 does not hold for atoms containing several electrons, obtains further confirmation 

 from the observations of Paschkn and Back^), according to which in the magnetic field 

 new components arise in the complex structure of the series lines, which correspond 

 to transitions where the third quantum number changes by more than one unit. In 

 fact from this phenomenon we may conclude that the magnetic field exercises a direct 

 influence on the motion of the series electron relative to the inner system. 



As mentioned in the text, these circumstances seem to be connected with 

 the general character of the magnetic properties of the atoms of the elements in 

 their normal state. Although this problem is still unsolved, a clue to the inter- 

 pretation of these properties seems, as pointed out in Essay III, to be alforded by 

 the circumstance, that the occurrence of atomic paramagnetism is most intimately 

 connected with the presence in the atom of inner groups of electrons, which ai-e 

 in a state of further development and which therefore display a marked lack of 

 symmetry. 



') F. Paschen and E. Back, Physica, 1, (Zceman jubilee number) S. 261 {1921). 



At the conclusion of this work I wish to thank my collaborators for their valuable 

 help, especially Dr. H. A. Kramers, who has not only contributed materially to the 

 subject, but has also kindly assisted me in the redaction of the manuscript of all 

 three parts of this treatise. On this occasion I wish also to express my sincere thanks 

 to the direction of the Carlsbergfond for considerable grants in a number of years, 

 which have enabled me to acquire assistance in the elaborate calculations on spe- 

 cial problems underlying the work on the development of the theory. 



Universitetcis Institut for teoretisk Fysik. 



Copenhagen, September 1922. 



Færdig fra Trykkeriet d. 30. November 1922. 



