24 Mr. H. Tomlinson on the Sudden Loss of the 



M. JSTouel and the author that, at any rate in the case of 

 annealed iron, there is no sudden change in the thermal ex- 

 pansibility at this critical temperature. 



When the temperature, after it has been raised to the 

 critical point, is again lowered, the centres w l5 m 2 , m B , will 

 creep closer together again and the magnetic properties will 

 reappear as suddenly* as they were lost. 



According to the theory, it would seem that if we could make 

 the magnetic molecules assume a sufficiently closed magnetic 

 circuit, iron 1 and nickel might be converted into diamagnetics. 

 Indeed we can understand that diamagnetics may be such in 

 consequence either of a paucity of magnetic molecules in 

 each molecule of the substance, or from having their magnetic 

 molecules forming nearly closed circuits. 



Again, if the temperature be raised high enough it would 

 seem possible to partly restore the lost magnetic susceptibility; 

 for then, by the further expansion of m^ m 2 , and m B , the N 

 and S poles would begin to be separated. If, therefore, such 

 metals as bismuth are diamagnetic at ordinary temperatures, 

 partly because their magnetic molecules form closed circuits, 

 rise of temperature should make them more magnetic or less 

 diamagnetic. This seems to be the case, for the experiments 

 of Pliicker and Matteucci led them to conclude that the 

 susceptibility of diamagnetics diminishes with increase of 

 temperature ; in the case of bismuth the decrease between 

 ordinary temperatures and its melting-point is said to be 

 about one sixth or more f. 



The following experiments were made with the apparatus 

 which has already been described J. The wire to be examined 

 was suspended vertically in the axis of a glass tube, being 

 clamped at its upper extremity into a brass block resting on 

 a wooden support fitting on to the top of the tube, and pro- 

 vided with a terminal screw for making connexion with one 

 pole of a battery of thirty Grove's cells. The current from 

 this battery was employed to heat the wire and passed in or 

 out of the latter through the intermediation of a mercury- 

 cup, into which dipped a stout copper wire attached to a brass 

 block clamped to the lower extremity of the wire. 



Experiment I. 

 A piece of very soft and well-annealed iron wire, 30 cm. 

 long and 1 mm. in diameter, was suspended in the glass 



* The suddenness of both loss and gain of magnetic susceptibility may 

 also be conditioned by internal friction, 

 t Chrystal, Enc. Brit., Art. Magnetism. 

 % Phil. Mag. vol. xxiv. p. 258, 



