126 Scientific Proceedings, Royal Dublin Society, 



by means of an exploring coil, earth inductor, and ballistic galvanometer, 

 and by plotting the total current in the wires as abscissas and the 

 corresponding values of the field as ordinates, the points were found to lie 

 on a straight line passing through the origin. Thus, if a line be drawn from 

 the origin to a point corresponding to current = 360 amperes (72 amperes in 

 each wire) and magnetic field = 1000 c.g.s. units, all the intermediate points 

 will lie on that line. 



For a detailed account of the arrangements and method of experiment I 

 would refer to my previous paper. 1 I may just say that, as before, the 

 distance from the mirror on the vibrator to the millimetre scale was 

 167 cms. ; and though the wire in some cases in the present experiments 

 was only one third the length of the wire formerly used, the maximum 

 amplitude of oscillation was the same, namely, from zero to the mark 300 

 on the scale, giving a twist of the lower end of the wire equal to about 5° 10'. 



On account of the strong magnetic fields used, the whole length of the 

 wire, 225 cms., could not be employed, because the wires were attracted to the 

 face of the slot ; so after some preliminary tests it was arranged for high 

 fields to use throughout the experiments a length of 75 cms of the wire, with 

 an oscillating load on the end of the wire equivalent to 2 x 10 5 grammes per 

 sq. cm., and 2670 grammes on the scale-pan end of the torsionless silk string 

 previously explained. 2 



Section I. 

 Nickel Wires. 



The wire first tested was a No. 16 soft nickel wire of simple rigidity, 

 about 708 x 10 6 grammes per sq. cm. Observations on the subsidence of 

 torsional oscillations were taken for many values of transverse magnetic 

 fields up to a maximum value of 800 c.g.s. units, both direct and alternating 

 of frequency 50 per second ; and, as before, the shape of the subsidence curve 

 was obtained in any case by plotting the number of vibrations as abscissas, 

 and as ordinates the corresponding values of the amplitude of oscillations. 



The more important of the values obtained are given in Table I, where the 

 numbers in columns 2 and 3 under the mark D.U., were obtained when the 

 nickel wire was subjected to no field or to a direct transverse magnetic field, 

 and those under the mark A.C., when it was under the influence of alternating 

 transverse magnetic fields of frequency 50 per second. 3 The value opposite 



1 Loc. cit. 2 Loc. cit. 



3 The alternating transverse magnetic fields are reckoned in root-mean- square values, 



