INFLUENCE OF TEMPERATURE ON MAGNETISATION. 637 



of magnetisation /, if the ratio of the diameter of the rod to its 

 length is so small that the magnetisation may be considered uni- 

 form. 



Experiment shows at first that in all circumstances f>f. These 

 coefficients, while very different for short cylinders, become gradually 

 nearer, and approach unity as the rod is longer. In the second 

 place, the field in which the maximum of / or of /' is produced 

 is stronger the shorter is the rod. 



With iron wires of the same kind, when the length is from 

 40 to 600 times the diameter, the maximum value of the co- 

 efficients/or /' varied from 25 to 190, or from 40 to 220, while 

 the corresponding fields diminished from 20 C.G.S. units to 3 units. 

 The same wire used as rings gave for the coefficient k a maximum 

 of about 200 with a field of 3 units. The agreement of these 

 results may be considered as sufficient if we' observe how difficult 

 it is to obtain perfect identity of the specimens on which we work. 



We may then investigate the magnetic properties of iron when 

 formed into cylindrical rods, provided the length is at least 500 

 times the diameter. We have this advantage, that we can know 

 at any moment the magnetic condition of the metal, and that it 

 is easy to demagnetise it so as to subject it to fresh tests. 



1207. NICKEL AND COBALT. With nickel and cobalt, Rowland 

 observed the same phenomena as for iron except as regards the 

 intensity. The curves of permeability as a function of the in- 

 tensity of magnetisation have the same aspect (Fig. 247). At the 

 ordinary temperature, the maximum of magnetisation is about 500 

 for nickel and 800 for cobalt. The maximum permeability is about 

 10 times as great for soft cobalt as for tempered cobalt. 



Berson,* finally, has found that rolled nickel is more coercitive 

 than fused nickel : the permanent magnetisation is then superior 

 to the temporary magnetisation. 



1208. INFLUENCE OF TEMPERATURE. The experiments of Pro- 

 fessor Rowland, confirmed by those of Trowbridge and McRae,f 

 have shown that the permeability given is virtually independent 

 of the temperature between and 280. Berson verified this pro- 

 perty as far as 330, For higher temperatures the permeability 

 should diminish, for the metal is no longer magnetic at a cherry- 

 red heat. 



* BERSON. Ann. de Chim. et de Phys. [6], Vol. vm., p. 432. 1886. 



t J. TROWBRIDGE and McRAE. Proceedings of the American Academy of Arts 

 and Sciences, May ij, 1885. 



