Longitudinal Stress on the Magnetization of Nickel. 129 



The general results of these experiments may be thus 

 summarized. 



In all magnetic fields with moderate loading, the effect of 

 twisting nickel wire is to increase the magnetization. But 

 the increase depends on the strength of the field as well as the 

 longitudinal stress applied. If the field is weak and the lon- 

 gitudinal stress sufficiently great, the magnetization increases 

 in one direction of twist and decreases in the other. Even- 

 tually, for a particular stress which is approximately propor- 

 tional to the field, the wire begins to show opposite polarity ; 

 and the cyclic curve of magnetization passes gradually from 

 a two-looped to a single-looped form. For stronger fields 

 similar effects exist. But in fields higher than a critical 

 value the increase and decrease of magnetization take place 

 for reversed directions of twist, and at the same time the 

 course of the curve becomes reversed. 



In a recent paper (" Magnetische Untersuchungen/'' Wiede- 

 mann's Annalen, March 1886) Professor Wiedemann has 

 described certain experiments on the combined effects of mag- 

 netization and twist in iron and nickel. He does not seem, 

 however, to have investigated the effect of longitudinal stress 

 in conjunction with these. His chief aim seems to have been 

 to adduce facts in support of his theory of frictionally rotated 

 molecules. Some of the results above described may be ex- 

 pressed in terms of his theory. Thus, when the external 

 magnetizing force is great, the magnetic molecules will be 

 held in position more strongly, and consequently the change 

 of magnetization due to twisting will be diminished. This 

 agrees with experiment. But, again, we saw that by suffi- 

 ciently loading the wire we could bring the apparent magneti- 

 zation down to zero, and eventually reverse its sign by mere 

 twisting. Now if this be due to the frictional rotation of 

 molecules, the molecules must, notwithstanding the directive 

 force of 30 units, be rotated through more than a right angle 

 from their first position, while the amount of mechanical twist 

 amounts only to '079 radian per centim. in each direction. 

 Admit it to be so ; what effect then may we expect increased 

 loading to produce on the rotation of the molecules? The 

 magnetic molecules in strong fields are acted on only by a 

 greater directive force, and consequently they must tend to 

 remain more in the direction of the magnetizing force ; but 

 why they should assume nearly the position of magnetic 

 neutrality when they are subjected to sufficient longitudinal 

 stress is a question which every supporter of the theory of 

 frictionally rotated molecules is bound to answe."^. 



Phil. Mag. S. 5. Vol. 27. No. 165. Feb. 1889. K 



