Brown — Mechanical Stress and Magnetisation of Nicltel. 31 



twist was read off with the circuit open ; the circuit was then closed, and the 

 current reversed, and 8 or 10 steps again taken up to maximum current, 

 the twist or deflection being read ofi at each step, and so on until a complete 

 cycle was obtained. These values were then plotted on millimetre paper, with 

 the values of tiie current through the wire as abscissae and the corresponding 

 values of the twist or deflections on the scale as ordinates, the curve was 

 drawn and its area determined. The curve or loop so obtained was very 

 symmetrical about both the horizontal and vertical axes, but was longer and 

 narrower than the curve obtained in a similar manner with an iron wire.' 



20 



40 60 



Fig. 1. — Lonaitudinal Magnetic Field. 



100 



Cycles were taken in this way for seven loads in all ; and in each ease the 

 longitudinal magnetic field round the wire was that which gave the maximum 

 twist with the given load; for example, when the load on the wire was 

 1-5 X 10^ grammes per square centimetre, the magnetic field round it was 

 20 c.g.s. units; and when the load was 4x 10° grammes per square centimetre, 

 the magnetic field round the wire was 44 units ; and so on. 



I Scient. Proc. Roy. Dub. Soc, vol. xii., No. 17, p. ISb. 



