i>8t 



II. \ At I. voie. \ 



I rosolvod to iittacl; : tmd the results (Iml h;i\o 1)ccii (ilihiiiifd will, f 

 believe, he. foiiml to contain distinet novelties. 



It is Well-kill iwn from tlie I'esults of \ iirioiis experiinenfers that Tiv 

 the application of longitudinal stvf.ss, tlie magnetisni of iron inereases 

 up to a certain critical load, wliile that of nickel always diniinishes. 

 Thus we should naturally expect that the etfect of torsional stress on 

 nickel will he opposite to that on iron. According to Thomson's 

 experiments, the effect of torsion on the niagnetistn of iron is to 

 increase it at first, but wlien the twist exceeds a certain angle, it tends 

 to diminish, while on untwisting it inci'eases and attains its former 

 value, and similar things take place when the wire is twisted in tlie 

 opposite direction. So with nickel, it seemed quite likely that there 

 will be decrease of magnetism till the twist reaches a certain value, 

 and lievond this, the magnetism of nickel will increase, which on 

 untw^isting ngahi decreases to its former value. In fact, this is 

 exactly repi-oduced in one of "Wiedemann's experiments,* the curve 

 obtained being just the reverse of one given by Thomson. f l>ut 

 Wiedemann's result was obtained by simply clamjnng the wire in a 

 horizontal position, so that the wire was subject to a weak longitu- 

 dinal stress only. On this account, the combined effect of jtull and 

 torsion on the magnetism of nickel was still a matter tobe determined. 



\n the following experiments, 1 have examined these points, and 

 liave found that the longitudinal stress produced a singular eft'ect. For 

 weak stresses, the changes of magnetism came out as was to ])e 

 expected, but when the load exceeds a certain limit, this is no longer 

 the case. The changes of magnetism becomt' gradually altered, and 

 beyond a critic;d value of the longitudinal stress, one <-nd of the 

 nickel wire acquires the two opposite kinds of magnetism during the 



•Wiodemann's Annalon, Bd. 20, S. 37G, ISSiV 

 fPhilosophiciil 'I'ransaotiiinB. 1S79, p. 72. 



