BETWEEN LONGITUDINAL AND ClKCULAll MAGNETIZATIONS, oil 



in high fiehls is not characteristic for nickel ; with a magneti- 

 cally softer nickel, it would recede towards a weak field as in 

 the case of iron. 



Another principal cause of the discrepancy is the effect of 

 residual magnetism. When the longitudinal field is kept con- 

 stant and the transverse field is gradually increased, the residual 

 magnetism due to the gradual change of direction of the result- 

 ant field must have some effect on the successive resultant intensity 

 of magnetization. This effect we shall now proceede to consider. 



It was found by experiments that magnetization of iron 

 or nickel in a given direction was very little affected by the exist- 

 ence of residual magnetism in the same direction at the beginning 

 of magnetization, provided the magnetizing field is not less 

 than that which caused the residual magnetism. Let us now 

 consider any two consecutive stages of applying transverse fields, 

 and let Ii and I. be the corresponding intensities of mag- 

 netization. We shall then investigate the contribution of Ii as 

 a residual maonetism on the formation of I,. We resolve the 

 residual magnetism due to Ii into two components, one along 

 and the other perpendicular to the intensity of magnetization 

 1/ which gives L when acted on by the residual magnetism. Of 

 these two, the parallel component very little affects the 

 intensity of magnetization, as we have seen from experiments. 

 The only effective part is the component perpendicular to 

 L', the action of which tends to change its direction. Hence 

 we may conclude that in whatever manner the residual 

 magnetization may act, it is only to prevent the shifting of the 

 direction of magnetization towards tliat of the resultant 

 field. The intensity of magnetization in the longitudinal or trans- 

 verse direction of our tubes must therefore depend on the 



