22 On Production of Transient Elective Currents. [Nov. 17, 



The transient currents produced by twisting a magnetised wire had 

 been noticed by Matteucci as early as 1858 (Wiedemann's " Galva- 

 nismus," ii, § 484), bnt the signs of the effects as observed by him 

 were opposite to those observed by the author, and he did not observe 

 any current when a twisted wire was magnetised. These discrepancies, 

 as well as the evident connexion between the effects under considera- 

 tion and the remarkable experiments recently described by Professor 

 Hughes (" Proc. Roy. Soc," vol. 31, p. 525), led the author to examine 

 the subject at length. The results are described fully in the paper. 



The statement of the numerical effects, which are at first sight very 

 involved, is much facilitated by the conception that the combined 

 effect of torsion and longitudinal magnetisation is to produce a state, 

 provisionally called "polarisation," in the wire, which persists almost 

 unchanged when the magnetising force is removed. This polarisation 

 is measured by the transient currents which accompany its produc- 

 tion. The normal polarisation for a given angle of twist and a given 

 magnetising force is measured by half the ballistic deflection due to 

 the transient current which appeared in the wire when the magnetising 

 force was reversed at that angle of twist. In this way, curves con- 

 necting normal polarisation with angle of twist (within the elastic 

 limit) are given for iron and steel. When the wire, after being 

 normally polarised at +6°, is twisted over to — 0°, the polarisation 

 does not change to the full normal value for — U , but to something less, 

 and this difference becomes still more apparent after several twistings 

 from one side to the other. By dividing the full twist across into 

 several steps, cyclical curves have been obtained, showing the relation 

 of polarisation to torsion when the same magnetising force is kept up 

 without interruption or reversal. These curves exhibit, in a striking- 

 manner, a persistence of previous state, such as might be caused by 

 molecular friction. The curves for the back and forth twists are 

 irreversible, and include a wide area between them. The change of 

 polarisation lags behind the change of torsion. To this action, which 

 is like that formerly described by the author as a characteristic of the 

 curves connecting thermoelectric quality with longitudinal pull, in a 

 paper on the " Effects of Stress on the Thermoelectric Quality of 

 Metals," Part I, the author now gives the name Hysteresis (vareprjat^, 

 from ixjTepeoj, to be behind). The modifying influence of "hysteresis " 

 in the whole action is minutely described, and it is shown that the 

 effects of hysteresis may be wiped out by subjecting the wire to 

 mechanical vibration, and still more effectually by changing its mag- 

 netisation, which may be done either by breaking and re-making the 

 magnetising current in the solenoid, or by reversing it. 



A curve is given showing the relative effects of different values of 

 the magnetising force up to 24 e.g. s. units, from which it appears that 

 a maximum transient current is produced by reversal of the mag- 



