262 



Professor Fleming 



[June 5, 



We Lave in this way examined the behaviour of magnets made 

 of a very large number of steels — chromium-steels, aluminium-steels, 

 tungsten-steels, silicon-steels and nickel-steels, in various states of 

 temper, hard and soft. We find that in some cases there is no initial 

 decrease of magnetism at all, and that the steady state begins at 

 once. Broadly, however, the results amount to this : — A steel magnet 

 when plunged into liquid air generally loses some fraction of its 

 magnetisation, but that after a few such immersions it arrives at 

 a fixed condition in which the effect of cooling it is in most cases to 

 produce an increase of magnetic moment, but in a few exceptional 

 cases to produce a decrease of magnetic moment. In the case of the 

 nickel-steels we have found very curious changes of magnetic 



Soft Carbon Steel. 



Fig. 15. 



19 % Nichel Steel. 



Diagram showing changes of magnetic moment of a magnet when alter- 

 nately cooled in liquid air and warmed up again to -|-5° C. The length of the 

 firm lines represents the value of the magnetic moment when cooled, and that of 

 tlie dotted when warm. 



moment as the magnet is heated up from — 186° C. to -f 300°. There 

 is a maximum magnetic moment at about 40° C. (see Fig. 16) in the 

 case of the 19 per cent, nickel-steel. 



In the technical use of magnets for instrumental purposes they 

 have to go through a process called ageing to get rid of the sub- 

 permanent magnetism. One of the best ways of ageing a magnet is 

 to plunge it several times into liquid air. 



We have given a large amount of attention to a study of the 

 changes taking place in the magnetic qualities of soft or annealed, 

 and also in hard iron when cooled to very low temperatures. 



In the first place, we have examined the change in the permea- 

 bility of iron at the temperature of liquid air. If a ring of iron 

 is wound over with a coil of wire and subjected to gradually in- 



