( 362 ) 



n' = 1 + E, 



which contains the theory of the longitudinal Zeeman effect. 



Among the terms arising from the natural activity that of* most 

 interest is =fc A, which is generally much larger than &. This term 

 which has opposite signs for the -|- an( ^ tne — rotating waves, but 

 which reaches its maximum al the same place as E, expresses a 

 dissymmetry of the Zeeman doublet somewhat similar to that obtained 

 in the experiments of Onnes and Becquerel. In order that this 

 dissymmetry should have an appreciable magnitude it is only neces- 

 sary thai iff is commensurable with sh. Experience has shown that 

 at the absorption bands of the crystals under consideration, s/, is 

 very small and so no difficulty is in the way. 



Notwithstanding there is lack of agreement between the above 

 formula and observation in two quite different directions. In the 

 firs! case the observers did not notice any natural activity in the 

 crystals they used, and from the symmetry of these crystals such 

 was not to be expected, excepl in the case, that at low temperatures 

 the constitution of the molecules change-. On the other hand the 

 observed behaviour by a reversed magnetic field is not in agreement 

 with the above formula. 



Both objections disappear if in the formulae (3) we substitute for 

 the terms dhA', dhB', dhC' on the right hand side the terms 

 O, dftPA', óhPB' and in the formulae (5) for the right hand terms 

 resp. A, B -f PSdhVk'/sb r + P2 dh\h ' • Such a series of terms 

 corresponds exactly to the symmetry of the magnetic field and leads 

 to the same formula (6) for. n as that given above; only we have 

 /Vf/t in place of 6h. By this step both the above mentioned difficulties 

 are removed; however the substitution leads from the sound basis 

 of experience into the region of hypothesis. Observation show- 

 that oh increases as the temperature diminishes. 



II. In some crystals of rhombic symmetry it has appeared that in 

 each of the three chief spectra certain absorption lines correspond 

 to nearly the same wave length. If we observe in the direction of an 

 axis of symmetry each of these lines is seen broken up into a doublet 

 when the magnetic field is excited. In the present special case of 

 equal wave length of the lines the distance of the doublet compo- 

 nents is the same in some cases, in other cases differs. If we let 

 the axes A', Y, Z, be parallel to the axes of the crystal and call 

 R the direction of magnetic force, to the direction of propagation of 

 the ray, <> the direction of vibration, then the following table gives 

 the result of observation: 



