644 CONSTANTS OF MAGNETISATION. 



behaves like iron, but the reverse is the case for nickel : mag- 

 netisation shortens the bar. According to Professor Barrett,* the 

 same magnetising force, in analogous conditions, produces an 



elongation of ; for iron, of for cobalt, and a 



260,000 425,000 



shortening of for nickel. 



130,000 



In like manner Sir W. Thomsonf found that, in general, trac- 

 tion increases the magnetisation of an iron bar and diminishes 

 that of a nickel bar. On increasing gradually the intensity of the 

 field in which the bar is placed, we find that the increase of mo- 

 ment due to traction goes on increasing to a maximum, then 

 decreases and becomes null for a certain value of the field ; be- 

 yond this critical value traction diminishes magnetisation. There 

 appears also for nickel to be a critical value of the field, for which 

 the effect of traction on magnetisation changes the sign ; but this 

 limit is higher than for iron, and has not been attained. 



The deformations which accompany magnetisation explain the 

 sounds produced by iron and steel when these metals are submitted 

 to a rapid series of magnetisations and of demagnetisations. The 

 sounds due to this cause may possibly play an important part in 

 the telephone. 



A bar of soft iron becomes heated when submitted to a series 

 of successive magnetisations and demagnetisations, or to magneti- 

 sations in opposite directions. Joule J even used this phenomenon 

 for one of his determinations of the mechanical equivalent of heat. 

 But it is impossible to separate the heat developed by induced 

 currents in the mass of iron from that which might correspond 

 to the work of magnetisation or of demagnetisation. An analo- 

 gous question, and which seems quite as difficult to solve, is the 

 retardation which a bar of soft iron presents when placed in a field 

 which periodically varies, and owing to which the maximum of 

 magnetisation is at a greater or less interval after the period of 

 the maximum of the field. It is difficult to say whether this 



* BARRETT. Natiire, Vol. xxvi. 1882. 



t Sir W. THOMSON. Proceedings of the Roy. Soc., Vol. xxni., pp. 445, 473, 

 1875 ; Vol. xxxii., p. 442, 1878. Phil. Trans. Roy. Soc. Loud. [2], Vol. CLXVI., 

 p, 693. 1877. 



J JOULE. Phil. Mag. [3], Vol. XXIIL, pp. 260, 347, 405. 1843. Scientific 

 Papers, Vol. I., p. 120. 



