MEASUREMENT OF PERMEABILITY 273 



total field will be H x + H 2 and the total induction B x + B 2 . To 

 find the induction per sq. cm. in the iron, we must divide the 

 total induction by the cross-section S. The current may be 

 increased step by step to any desired amount and then taken 

 round a cycle in any desired way, the values of H and B being 

 determinate at every point. 



Fig. 209 represents a test carried out by Ewing (loc. cit. pp. 

 70-72) on a wrought-iron ring, mean radius 5 cm. and cross-section 

 0-0483 sq. cm. The magnetising coil had 474 turns and the 

 secondary had 167 turns. H was gradually raised from to 9'14, 

 then taken back to and again raised to 9'] 4. 



The induction per sq. cm. in the iron rose to 12,440 at the 

 maximum. The maximum value of /* = B/H was 1740 when the 

 value of H was 495.* 



Magnetisation in very weak fields. Baur,f experi- 

 menting by the ballistic method on a ring of soft iron with values 

 of H rising to 0'384, found that I and B could be represented by 

 the parabolic formulae 



I = 14'5H+HOH 2 

 B = 183H+1382H 2 



whence K = 14'5+110H 



and M = 183+1382H. 



These results suggest that for extremely small values of H, K and 

 IUL are constant. This point was investigated by Lord Rayleigh J 

 and the suggestion was confirmed. He used the magnetometer 

 method with a second compensating coil through which the 

 magnetising current passed, adjusted in such a position that it 

 neutralised the effect of the iron wire tested (Swedish iron un- 

 annealed) on the magnetometer needle when the intensity of the 

 magnetic field within the solenoid did not exceed 0'04. When 

 the intensity of the field was still further reduced the compensation 

 remained perfect and the reduction was carried down to an intensity 

 about one two-thousandth of the initial value. It might therefore 

 be concluded that, between H = and H = 0'04, K and JUL are constant 

 and that within this range there is no retentivity. The value of 

 IUL was about 100. Above 0'04 the constancy no longer held. 



Lord Rayleigh also found that if a specimen was already 

 magnetised by a moderate force, and there was then made a small 

 change in the magnetising force opposite in direction to that 



* For other arrangements in which the ballistic method is used, especially 

 adapted for technical requirements, the reader should consult Ewing on Magnetic 

 Induction in Iron and other Afetals, chap. xii. 



t Wied. Ann., xi. p. 399 (1880). 



j Scientific Papers, vol. ii. p. 579. 



