1882.] Retentiveness in the Magnetisation of Iron and Steel. 41 



which Thomson has shown must occur on account of the fact that 

 the susceptibility to magnetic induction is a function of the tempera- 

 ture. 



I have endeavoured to account for the static hysteresis by sup- 

 posing that the rotation of Weber's magnetic molecules is opposed by 

 a frictional couple of constant moment, not necessarily the same for 

 all the molecules in a given piece. It seems not unlikely that residual 

 magnetism itself maybe due to this frictional sticking of the molecules 

 rather than to the quasi-plasticity suggested by Maxwell. The ex- 

 amination of this theory, as well as the description of the experiments, 

 some of whose results have been briefly mentioned in this notice, will 

 form the subject of a more detailed communication. 



Another portion of the work has consisted in looking for hysteresis 

 in the changes of the longitudinal magnetism of iron wires, produced 

 by pulling and relaxing pull, the wires being under the influence of 

 the vertical component of the earth's magnetic force, which in Tokio 

 is about 0'34 C.G.S. unit. Sir W. Thomson* has investigated very 

 extensively the general effects of stress on the magnetisation of iron 

 and other metals, in magnetic fields of various strengths, but without 

 special reference to this point. Only in the case of torsion (alter- 

 nately to opposite sides) is mention made of any action of the kind 

 which I have termed hysteresis. His researches were for the most 

 part conducted by the ballistic method, f by which the currents in- 

 duced in a solenoid surrounding the wire were observed when a single 



* " Electrodynamic Qualities of Metals," " Phil. Trans.," 1856, 1876, 1879. 



f [Note by Sir William Thomson of May 3, 1882.— This is not quite so. The 

 experiments described in §§ 214 — 244 of my " Electrodynamic Qualities of Metals," 

 Part YII ("Phil. Ti*ans." for 1879, p. 55), were performed by the magnetometrie 

 method. My earlier experiments described in §§ 178 — 213 (" Phil. Trans." for 

 1876, p. 693, and for 1879) were performed by the ballistic method. 



The following is taken from a preliminary statement (§ 178) : — Early in the 

 year 1874, I made arrangements to experiment on the magnetisation of iron and 

 steel wires in two different ways — one by observing the deflections of a suspended 

 magnetic needle produced by the magnetisation to be tested, the other by observing 

 the throw of a galvanometer needle, due to the momentary current, induced by each 

 sudden change of magnetism. The second method, which for brevity I shall call 

 the ballistic method, was invented by Weber, and has been used with excellent 

 effect by Thalen, Poland, and others. It has great advantages in respect of conve- 

 nience, and the care with which accurate results may be obtained by it ; but it is not 

 adapted to show slow changes of magnetism, and is therefore not fit for certain im- 

 portant parts of the investigation. On this account I am continuing arrangements 

 for carrying out the first method, although hitherto I have obtained no good results 

 by it." 



The first method was accordingly followed in all the latter part of my experi- 

 ments on this subject ; not only those described in §§ 214 — 244 referred to above, 

 but also in further investigations which I have continued up to the present time, . 

 and of which I hope to offer results to the Poyal Society before long.] 



