On the Magnetization of JS^ickel. 117 



I cannot but think that sound philosophy would be pro- 

 moted could these limitations of the doctrine of the conser- 

 vation of energy be made clear to those philosophers who see 

 in the doctrine only a special case of a general law of mind 

 and matter. 



XV. Combined Effects of Torsion and Longitudinal Stress on 

 the Magnetization of Nichel. By H. Nagaoka, Rigakushi, 

 of the Imperial University, Japan. With a Note by J. T. 

 BoTTOMLEY, F.R.S., and A. Tanakadate, Rigakushi ^. 

 [Plates II.- v.] 



THE effect of torsion in altering the induced magnetism 

 of iron has long engaged the attention of many phy- 

 sicists. Among the experimenters in this field of research 

 may be named Wertheim, Wiedemann^ Thomson, and Tom- 

 linson (whose latest work on such subjects I have not yet seen). 

 The experiments of Wiedemann were made by twisting and 

 untwisting the wire, which was placed horizontally, in a mag- 

 netizing solenoid, till the changes of magnetism became cyclic. 

 But it was Thomson who first investigated the effect of torsion 

 on the magnetism of iron, the wire being at the same time 

 subject to definite longitudinal stresses. In his experiments 

 the soft iron wire was placed vertically in the earth's field. 

 No one seems to have made similar experiments in different 

 magnetizing fields. Scanty though this kind of investigation 

 has been for iron, it is still more scanty for nickel. Indeed, 

 so far as I am aware, the effect of torsion on the magnetism 

 of nickel wire under various longitudinal stresses has not 

 hitherto been investigated. This accordingly was the problem 

 I resolved to attack ; and the results that have been obtained 

 will, I believe, be found to contain distinct novelties. 



It is well known from the results of various experimenters 

 that, by the application of longitudinal stress, the magnetism 

 of iron increases up to a certain critical load, while that of 

 nickel always diminishes. Thus we should naturally expect 

 that the effect of torsional stress on nickel will be opposite to 

 that on iron. In fact, this is exactly reproduced in one of 

 Wiedemann's experiments f, the curve obtained being just 

 the reverse of one given by Thomson |. But Wiedemann's 

 result was obtained by simply clamping the wire in a hori- 

 zontal position, so that the wire was subject to a weak longi- 



* Commimicated by Sir William Thomson. The paper of Mr. Naga- 

 oka was originally published in the Journal of the College of Science, 

 Imperial University, Japan, 1888. 



t Wiedemann's Annalen, Bd. xxvi. S. 376 (188(3). 



X Philosophical Transactions, 1879, p. 72. 



