Longitudinal Stress on the Magnetization of Nickel, 127 



+ j/('— 3) becomes infinite. In other words, when the 

 strength of the field is such that, under sufficient loading, the 

 reversal of polarity vanishes away, this seems to be the signal 

 for the new set of conditions to appear. Up to this critical 

 strength of field, it is the left-hand loop in the typical sym- 

 metrical curve that gradually diminishes under loading. But 

 for strengths of fields higher than this critical value, it is the 

 right-hand loop that disappears when the load is great enough. 

 This reversal of effects also seems to be accompanied by a 

 lingering of the magnetization near the zero (see figures 18 

 and 19). 



It is not necessary to discuss in detail other combinations 

 of field and stress which were experimented upon. There are 

 certain minor differences depending on strength of the field ; 

 but the principal features for fields higher than 8 are the 

 same. The essential characteristics can be gathered from 

 figures, brief explanations of which I shall content myself 

 with giving. 



Figs. 22-23 (PL IV.).— These illustrate the changes of 

 magnetization in field 11*9. 



Figs. 24-27 (PI. IV.).— These were obtained in field 13-85. 

 The curious transition-curve, fig. 30, Pl.V., is specially worthy 

 of note. 



Figs. 28-31 (PI. v.). — These show the gradual changes of 

 magnetization for field 15*78. 



Figs. 32-34 were obtained in field 23*50. 



Figs. 35-40 were obtained in field 33'54. 



This last was the highest strength of the field at which it 

 was possible to notice the changes of magnetization. For 

 the stronger the field the higher is the load necessary to bring 

 out the curious changes. The critical load for fields higher 

 than 33'54 is greater than the tenacity of the wire. 



There is one other point that calls for remark. The range of 

 the change of magnetization under feeble stresses begins gra- 

 dually to diminish after a certain strength has been reached, so 

 that for a field of 20 or 30, the change of magnetization by 

 twisting becomes almost inappreciable. This can be accounted 

 for by the fact that nickel wire, whether in the normal or 

 twisted condition, behaves practically the same as regards mag- 

 netization in higher fields. Indeed, nickel is more easily satu- 

 rated than iron, so that when Jp = 20 or 30, it is already far 

 beyond the saturation-point. Consequently the differences in 

 the susceptibilities of nickel in the normal and twisted con- 

 ditions, to which this alteration of magnetic intensities must 

 be ascribed, become less and less marked as the strength of 

 the field is increased. Many of the features of magnetization- 

 curves are quite simply accounted for by this consideration. 



