MAGNETISATION OF STEEL AND CAST IRON. 639 



According to M. Berson, the temporary magnetisation of nickel 

 increases from to 250 or 260, then rapidly diminishes, especially 

 from 280, and disappears towards 340. 



The magnetic moment of a nickel rod, magnetised in the cold, 

 becomes gradually weaker as the temperature rises, and is null 

 towards 330, as for steel ; but it has this curious feature, that 

 if we magnetise it at a high temperature of 200 to 290, the 

 residual magnetisation increases first during cooling, then slightly 

 diminishes, always remaining at the ordinary temperature higher 

 than what it was at the temperature of magnetisation. 



For cobalt, the temporary and residual magnetisations increase 

 steadily with the temperature between 25 and 325. Contrary to 

 what is the case for nickel, the residual magnetic moment of cobalt 

 always diminishes, when we get away from the temperature of mag- 

 netisation, in one direction or the other. 



These results agree with the older experiments, in which Faraday* 

 had observed that, at the temperature at which olive oil begins to de- 

 compose, the magnetic properties of nickel have scarcely diminished, 

 while, on the contrary, those of cobalt have increased. 



1209. STEEL AND CAST-IRON. The results for steel are less 

 concordant, the complexity of the phenomena being greater. When 

 the steel is magnetised for the first time, the composition of the 

 metal, the degree and the duration of the temper, the temperature, 

 and the duration of the annealing, the shape and dimensions of the 

 bar, etc., must be taken into account. We may add also, that in bars 

 which are somewhat large, the material is necessarily heterogeneous, 

 as the effects of tempering are not equally felt in the various layers. 

 Lastly, after several successive magnetisations, in the same or in 

 opposite directions, the residual magnetism may have a very irregular 

 distribution, and may even be formed of several successive layers, 

 with magnetisations in opposite directions. 



We owe to MM. Barus and Strouhalf numerous experiments on 

 steel bars, always of small diameter, so as to obtain a more homo- 

 geneous material. 



The different kinds of iron, soft iron, steel, cast-iron, differ greatly 

 in conducting power ; the specific resistance at 0, which is equal to 

 about 10 microhms for soft iron, rises to almost 100 for hard cast- 

 iron, and varies from 47 to 15 for steel, according to the degree of 



* FARADAY. Experimental Researches, xxx., 3424. 1855. 



t BARUS and STROUHAL. Bulletin of the United States Geological Survey, 

 No. 14, p. i. 1885. 



