298 K. HONDA : ON THE MUTUAL INFLUENCE 



ourves with respect to the axis of d T shows that the change 

 of longitudinal magnetization is independent of the direction of 

 the transverse field, as first observed by Wiedemann/' 



The experiments of several previous investigators on this 

 subject are somewhat discrepant. As was already mentioned, 

 G. Wiedemann noticed only decrease of magnetization ; 

 Villari observed an increase of magnetization by passing a current 

 through an iron bar placed in a constant longitudinal field of 

 moderate strength, and a decrease in a strong. He-' also found 

 a decrease of magnetization in strong circular field, and the dis- 

 cordance between the two was ascribed to the eftect of heating 

 by the current. On the other hand, Knott, Siemens, Schnitze 

 and Pitcher did not observe any increase of magnetization. 



All these results are however not altogether discordant, and 

 can be easily reconciled from the results of the present experi- 

 ment. 



I shall next describe the experiment on the eftect of longi- 

 tudinal field on circular magnetization. 



The magnetizing coil was so placed that it gave no eftect 

 on the galvanometer. Putting the tube in the magnetizing coil 

 and demagnetizing it by reversals, the eftect of longitudinal field 

 on the circular magnetization, when there is no circularly 

 magnetizing field, was first examined by the ballistic method. 

 The effect was very small and can be easily corrected. Keeping 

 the circular magnetizing field constant, the tube was longitudinally 

 magnetized with the following result. 



1) loc. cit. 



2) loc. cit. 



