Brown — Mechanical Stress and Magnetisation of Nickel. 33 



wire when it is subjected to the combined action of circular and longitudinal 

 magnetisation. From tlie fact that the magnetic field in which the maximum 

 twist takes place increases as the load on the wire increases — that is, as the 

 load increases the peak of the curve is displaced towards the right as shown 

 in fig. 1 — one would suppose that a wire placed under these conditions of 

 load and magnetic field would offer more resistance to any change in its 

 internal configuration than when placed in any other field of different 

 strength. 



In order to test this, a series of experiments were made on tlie rate of 

 subsidence of torsional vibrations in the nickel wire when it was surrounded 

 by longitudinal magnetic fields of different strengths. 



The results of some experiments of this kind on nickel and other wires 

 have been published by Gray and Wood,' in which nickel wires one metre 

 long were employed, the initial amplitude of oscillation being 90°. These 

 authors point out that this latter condition (the magnitude of the amplitude 

 of oscillation) plays a very important part in the final results, which is 

 confirmed in a way by the results obtained by the present writer, who used 

 an initial amplitude of about 7° only in the results given in this and the 

 following section. The above-named authors, however, omit to state the value 

 of the longitudinal load that was on their wires when under test. 



In the experiments described below the length of the nickel wire used 

 was 226 cms., of diameter 0'168 cm. and simple rigidity 708 x 10'^ grammes 

 per square centimetre; and by means of the long solenoid already mentioned 

 the wire was in a uniform magnetic field throughout its whole length. The 

 amplitudes of the torsional oscillations of the wire were read off by means of 

 the light spot on the millimetre scale, which was placed at a distance of 116'5 

 cms. from the mirror on the lower end of the wire; and the largest amplitude 

 used was about 7° or 300 millimetres on each side of the zero on the scale. 

 This initial amplitude was chosen because the maximum twists obtained in the 

 expeviments on Magnetism and 2'orsion fell between this value and the smallest 

 atnplitude taken in the experiments now under discussion. 



The method of experiment was as follows : — With a given load on the end 

 of the wire and a certain longitudinal magnetic field round it, it was made to 

 oscillate round its own axis ; and when the initial amplitude of oscillation was 

 correct — that is, wlieu the light spot was on the division marked 300 on the 

 scale — the observer began to count the oscillations and read off the amplitude 

 at every fifth oscillation up to the twentieth, and then every tenth oscillation, 

 until 70 oscillations in all had taken place. The load on the wire being kept 



1 Pioc. Eoy. Soc, vol. Ixx., 1902, and vol. Ixxiii., 1904. 



