442 



Electrodynamic Qualities of Metals. [May 23. 



stress.* This influence illustrated by experiments made a few days 

 ago for the author, by Mr. Macfarlane, in Glasgow, and Mr. Bottomley, 

 in the Physical Laboratory of King's College, London, by kind per- 

 mission of Professor Adams, which, show in two very different ways 

 that twisting a brass tube through which a current of electricity is 

 maintained gives to the electric stream lines a spirality of opposite 

 name to that which the twist gives to longitudinal filaments of the 

 substance, and so proves that in aeolotropically stressed brass the 

 electric conductivity is greatest and least in the directions of greatest 

 and least pressure. The some law probably holds for iron. Wiede- 

 mann's result is that the end of the iron wire by which the current 

 enters, becomes a true north or a true south pole, according as the 

 twist is that of a right handed or of a left handed screw. This is the 

 same direction of effect as would result from the aeolotropy of the 

 magnetic susceptibility produced by the stress if the tangential magne- 

 tizing force in the outer part of the wire is less than the critical value, 

 for which the effect of the stress is isotropic ; but it is opposite to the 

 effect due to the aeolotropy of the electric conductivity. Yet the 

 author in repeating Wiedemann's experiments has found his result — 

 the same in direction, and greatest in amount — with the strongest 

 currents he has hitherto applied — currents strong enough to heat the 

 wire seriously (but not yet measured or estimated in absolute measure) . 

 The reconciliation of the Wiedemann result with the conflicting influ- 

 ence of conductive aeolotropy, and with the influence of aeolotropy 

 of magnetic susceptibility, which also is conflicting when the magne- 

 tizing force is great enough, is a difficulty which calls for investiga- 

 tion. 



The paper includes a series of experiments on the effects of twist 

 on magnetization of iron wire under longitudinal magnetizing force 

 (the Glasgow vertical force alone in this first series). It confirms 

 results of previous experimenters, Matteucci, Wertheim, and Edmond 

 Becquerel, according to which twist in either direction diminishes the 

 magnetization ; and extends them to wires under different amounts of 

 longitudinal pull. When the pull was great — approaching the limit of 

 elasticity of the wire, the twist, even when well within the limits of 

 elasticity, had -much less effect in diminishing the magnetism than 

 when the pull was small. The results are recorded in curves which 

 show a very remarkable lagging of effect, or residue of influence of 

 previous conditions. 



The paper concludes with a description of experiments, showing in 

 bars of nickel and cobalt effects of longitudinal pull opposite to those 

 found by Yillari for iron with magnetizing force below the critical 



* See § 161 of Part V of this paper ("Electrodynamic Qualities of Metals," 

 "Trans. Roy. Soc," vol. cxlvi, Feb., 1856. Also Tomlinson, "Proc. Roy. Soc," 

 vol. xxvi, p. 401, 1877. 



