Magnetic Properties of Iron and Nickel, 25 



tube, so that it could freely swing torsionally. The glass 

 tube was placed inside a magnetizing solenoid, which was 

 in turn coaxial with a secondary solenoid connected with 

 a Thomson's reflecting-galvanometer. The magnetizing sole- 

 noid was actuated by 2 Grove's cells, which were kept in 

 action throughout the experiment. As soon as the cur- 

 rent from the 30 Grove's cells had heated the wire to a 

 dull red, there was a slight deflection of the Thomson's 

 galvanometer, indicating that the iron had suddenly lost part 

 of its magnetic susceptibility ; then, after a pause of three or 

 four seconds, there was a much greater deflection of the 

 galvanometer in the same direction. Again, when the wire 

 was allowed to cool, there was first a slight deflection of the 

 galvanometer, indicating partial restoration of magnetic sus- 

 ceptibility followed immediately by a much larger deflection*. 

 It seems likely that the interval between one deflection and 

 the next, when the w T ire is being heated or is cooling, is the 

 first of the two periods during which, according to Pionchon, 

 heat is becoming latent. There is a similar pause, though 

 not quite so marked, in the sudden jerk which occurs in an 

 iron wire which has been permanently strained when the 

 critical temperature C has been reached f, and when, according 

 to Pionchon, heat a second time becomes latent. 



Experiment II. 



In this experiment an attempt was made to abolish the 

 deflections of the Thomson's galvanometer, due to the sudden 

 loss or restoration of the magnetic susceptibility, by torsionally 

 oscillating the wire during the whole period of heating or 

 cooling. But no amount of torsional oscillation availed any- 

 thing. 



Experiment III. 



The experiments of Faraday and others, quoted previously, 

 show that there is a temperature at which permanent mag- 

 netization in steel begins to be rather suddenly lost, and that 

 this temperature is considerably lower than the temperature 

 at which the temporary magnetic susceptibility begins to 

 rapidly diminish. The author has also found that when a 

 permanently twisted wire is heated it begins at a certain 

 temperature to rapidly \ lose its permanent strain. It seemed, 



* This pause between one deflection and another has already been 

 noticed bv Mr. Newall (Phil. Mag. vol. xxiv. p. 435). 



t Phil. Mag. voL xxiv., L887. 



\ This sudden loss of permanent strain most not he confounded with 

 the sudden jerk which takes place at the critical temperature C. 



