1882.] Effects of Retentiveness in Magnetisation of Iron, fyc. 39 



III. " On Effects of Retentiveness in the Magnetisation of Iron 

 and Steel. (Preliminary Notice.)" By J. A. EwiNG, B.Sc, 

 F.R.S.E., Professor of Mechanical Engineering in the 

 University of Tokio. Communicated by Professor Sir 

 William Thomson, F.R.S. Received May 6, 1882. 



The term Hysteresis was introduced in a paper,* recently comniuni- 

 r cated to the Royal Society, to designate a peculiar action which, was 

 observed in the inquiry then recorded, and which had also presented 

 itself in an earlier investigation — of the effects of stress on thermo- 

 electric quality. f It was found that when a stretched iron wire was 

 gradually loaded and unloaded the changes of thermoelectric quality 

 lagged behind the changes of stress, so that curves exhibiting the 

 relation of stress to thermoelectric quality during the putting on and 

 taking off of the load were far from coincident, but inclosed between 

 them a wide area. J 



In prosecuting those experiments it occurred to me that there is 

 much room for investigation of hysteresis § in the changes of mag- 

 netisation of iron and other substances produced by (1) change of the 

 magnetic field ; (2) change of stress ; (3) change of temperature. In 

 (2) and (3) two cases are to be considered : — First, when the substance 

 is exposed to a constant magnetising force ; second, when the magneti- 

 sation which is changed is wholly residual. 



From the known character of residual magnetism we may at once 

 infer that when magnetisation along any axis is changed so con- 

 siderably that its sign is reversed there must be hysteresis, but it is 

 not clear that any such phenomenon need appear when the action is 

 •confined to one sign. In fact, Maxwell's extension of Weber's theory 

 of induced magnetism |] assumes that residual magnetism resembles 

 the "permanent set" of a strained solid, and implies that any sub- 

 sequent application of a magnetising force in the same direction with 

 and not exceeding that by which the residual magnetism has been 

 produced, will give changes of a quasi-elastic character not exhibiting 

 the action which I have called hysteresis. 



By the direct magnetometric method, and also by the ballistic 



* " On the Production of Transient Electric Currents in Iron and Steel Con- 

 ductors by Twisting them when Magnetised, or by Magnetising them when 

 Twisted." " Proc. Roy. Soc," vol. 33, p. 21. 



f "Effects of Stress on the Thermoelectric Quality of Metals. Part I." "Proc. 

 Roy. Soc," vol. 32, p. 399. 



% Since the paper cited was laid before the Royal Society, I have learnt that M. 

 Emil Cohn has anticipated me in the discovery of this peculiar feature of the effects 

 ■of 6tress on thermoelectric quality. (" Pogg. Ann.," N.F., VI, 305.) 



§ Or effects of retentiveness — Note by Sir William Thomson, May 5, 1882. 



|| " Treatise on Electricity and Magnetism," II, chapter vi. 



