Brown — 3Iechanical Stress and Magnetisation of NicJccl. 47 



centimetre ; and, as we know from page 34, above, witli this load tlie 

 greatest damping of the torsional oscillations will take place when the wire is 

 subjected to a longitndiiial magnetic field of 25 o.g.s. nnits. Each wire was 

 therefore tested for the subsidence of torsional oscillations: (1) when there 

 was no magnetic field present ; and (2) when a magnetic field of 25 units 

 was round the wire ; two sets of observations on the rate of subsidence were 

 therefore made on each of the six wires, slatting in each case from an 

 initial amplitude of 300 divisions on llie scale (the distance of which from 

 the mirror on the end of the wire was 116'5 cms.), or from an angular 

 distance from tho zero of 7j°, the amplitudes being read off the scale at every 

 fifth or tenth oscillation, until 70 complete vibrations had been made. If 

 the results so obtained be plotted with the values of the cross-sectional areas 

 of the wires as abscisssB, and as ordinates the corresponding values of the 

 amplitudes attained at tlie 70th vibration, the six points will be found to lie 

 in a straight line in the case of no magnetic field round the wire, and in a 

 lower and practically parallel straight line in the case wliere the wire is 

 surrounded by a magnetic field of 25 units. These curves, not here repro- 

 duced, as well as the table below, show that when the cross-sectional area of 

 the wire is ineieaaed five times, tho decrccme in the amplitude of torsional 

 oscillation, after seventy vibrations have been made from the same initial 

 amplitude, is 10 per cent., with no field round the wire, and 11'4 per cent, 

 with a magnetic field of 25 units round it. The No. 16 wire was tested first 

 in the hard state wlieu its rigidity was 770 x 10'' grammes per square 

 centimetre, and, secondlj^ in the soft state when its rigidity was 708 ;< 10^ 

 grammes per square centimetre, this latter state being obtained by hanging 

 the wire vertically under its own weight only, and raising it three times in 

 succession to the tempeniture represented by a bright cherry-red by means 

 of a broad liunsen burner. 



Tlie following table gives some of the results obtained with tlie smallest 

 and the largest wires, both in the hard state, and the results for the No. 16 



SCIENT. PKOC. K.D.S., VOL. XUI., NO. III. 



