( 364 ) 



Uliz Pt n„ 



ia/ IX (p^ + pJ/Nt 

 \ol/Y Pt fN„ 



J?,Y ,,Y \ C H T tP*+PM N * 



R//} ' W//X U* P JN 9i 



\o; x P JN tl 

 „,,z\ Gl ! x " »» 



R/IZ . w X 

 tol V 



wl/ Z 



ollY /<, N x% 



o z (pt -/':> \; 



oilZ (p M +p 9 ) AT, 



Supposing now the absorption lines to have the same positions in 

 the three chief spectra, then the real parts of Pi,p a ,p t and also of 

 Pi> p t > Pt are equal; supposing- the intensities of' the absorption lines 

 to be equal, then the imaginary parts are the same. If* on the other 

 hand g x ,g„g 3 and f Xi f ti f t retain different values from each other, 

 then also N x , N„ N t and N~ 3t , $ 81 , iV la have different values from 

 each other. In this case the system of distances between the doublet 

 components obtained from theory agrees exactly with that observed. 

 If the three absorptions are slightly different from each other, then 

 the theoretical distances show small deviations from the mentioned 

 law, which however will scarcely be within the range of perception. 



What is interesting in the observations is that if we assume 

 g 1 =z g i = g S) and therefore XT 33 = JV §1 = N~ Xi then theory is not in 

 agreement with experience. 



The parameters gu give the direct influence of the external magnetic 

 field on the vibrating electron. This influence appears then to be 

 different when the field acts along the X-, the Y-, the Z-axis of the 

 crystal. This result seems to me to verify the view I have deduced 

 from other considerations, that the magnetic field inside the molecule, 

 where an electron is in motion, can be very different from that in 

 outside space. Reasoning in this way it is quite natural to imagine 



