186 MAGNETISM 



is measured by the excess and the permeability by the multiplica- 

 tion of lines of force. We shall see later that it is convenient to 

 adopt such units of permeability /j. and of susceptibility K that 



yU=l+4 TTK. 



We see then that the permeability curve is like the suscepti- 

 bility curve, but we must multiply the ordinates of the latter by 

 4 TT and raise the whole curve by 1 to give the former. 



"Permeability" is the more fundamental idea, for it takes 

 account of the whole magnetic action going on. " Susceptibility " 

 is somewhat artificial, but it is convenient to use the term, as K 

 is more directly measured in certain methods of experiment. 



Magnetisation and temperature. If we heat a magnet 

 to a bright red heat and then cool it, we find that the magnetisa- 

 tion has entirely disappeared. If we place a bar of iron or steel at 

 a bright red heat within a magnetising coil it shows no suscepti- 

 bility. But as it cools to a dull red usually a little below 800 (', 

 its susceptibility is almost suddenly regained. If an iron ball 

 heated to a bright red is hung by a spiral spring over a strong 

 pole it is not attracted, but as it cools a point is reach rd when it 

 is suddenly pulled down to the pole. 



Hopkinson* found that with a certain specimen of iron the 

 permeability increased with rise of temperature up to 775 C., when 

 it was many times the value at ordinary temperatures. A slight 

 further rise to 786 sufficed to render it non-magnetic. He called 

 the temperature at which the magnetic permeability ceases to be 

 more than that of air the critical temperature. The critical tempera- 

 ture varies, with different specimens of iron and steel, from 690 to 

 870 C. It may be taken usually as not very far from 750 C. 

 This sudden change in property is related to other phenomena. 

 Gore j" observed that if an iron wire was heated to redness and 

 then allowed to cool, at a certain point a slight elongation occurred. 

 Barrett { observed that if a piece of iron, or, better, hard steel, is 

 heated to bright redness and then allowed to cool, at a certain 

 point it suddenly glows more brightly. This can easily be 

 observed in a dark room. The phenomenon was termed by 

 Barrett " recalescence," as implying a sudden rise of temperature. 

 Hopkinson measured this rise. He found that a certain piece of 

 hard steel, after cooling to 680, suddenly rose to 712C. and then 

 began to cool again, and he showed that the magnetic condition was 

 assumed at the same time. No doubt the lengthening observed by 

 Gore occurs at the same point. There is then some sudden change 

 in the molecular condition of the iron when it passes from the 

 magnetic to the non-magnetic condition, or rather when its sus- 

 ceptibility disappears and its permeability falls to that of air. 



Permanent magnets and temperature change. If 

 a permanent magnet has already been subjected to M-\eral vnria- 



* Phil. Trails. (1889), A, p. 443. f Proc. /?..*., xvii. (1869), p. 260. 



J Phil. Mag., Jan. 1874. Proc. R.S., xlv. (1889), p. 4 



