Prof. E. Bouty's Studies on Magnetism. 201 



nent magnetism produced by a force F comprised between C and 

 2C is proportional to F — C. 



Fie. 6. 



The original hypothesis of the coercive force explains, there- 

 fore, pretty accurately the only phenomena well known at the 

 time when it was invented ; it cannot enable us to foresee new 

 phenomena ; and the progress of the science of magnetism has 

 left little of it in existence but the name. 



The phenomena of temporary magnetism and of the maximum 

 of magnetization, as well as those of permanent magnetism, have 

 been explained collectively by Wiedemann, but again by means 

 of a comparison. He compares magnetic phenomena to those 

 which depend on the elasticity of solid bodies, specially those 

 accompanying torsion or flexion. This has the advantage of ex- 

 pressing an incontestable physical relation, since torsion modi- 

 fies the magnetic condition of a bar, and such modifications may 

 in their turn cause a variation in the torsion to which the bar 

 has been submitted beforehand. 



Wiedemann assumes that in the unit of volume of iron or steel 

 a finite number of magnetic elements (molecular currents) exist, 

 but that in the natural state their axes are distributed equally 

 in all directions. External magnetic forces tend to rotate these 

 elements about their centre of gravity ; but these oppose a re- 

 sistance to displacement, analogous to elastic reaction, in such a 

 manner that, according to the intensity of the acting forces, the 

 elements retain permanently a more or less considerable portion 

 of their temporary displacement. On submitting these hypo- 

 theses to calculation we obtain, for the representation of the 

 temporary and the permanent magnetisms, curves which, though 

 satisfying the same essential conditions, present only vague and 

 general relations with the curves furnished by experiment. 



For the very ingenious developments given by this author to 

 his theory the original work must be consulted*. We will 



* Galvanismus, vol. ii. 



