INDUCED MAGNETISATION. 627 



the case of a cylinder of great length, for which C = or C = - 27r, 

 according as its axis is parallel or perpendicular to the field. Lastly, 

 we have C = 0, for a ring surrounded by an annular coil. 



The forms which present most advantages in practice are then : 

 a sphere, an elongated ellipsoid of revolution, which may be com- 

 pared with an ellipsoid of the same cross section and the same 

 length, or which may better be considered infinitely long ; and 

 lastly, for a circular ring. 



Experiment shows that the coefficient /,. for a body of a given 

 kind iron, for instance depends not only on its actual condition, 

 its purity, temperature, temper, but also on the successive stages 

 through which it has passed that is to say, on its previous magnetic 

 history. 



Any modification in the magnetic condition of a body, even if 

 apparently transient, should be considered as having produced a 

 permanent alteration in its constitution, and this alteration can only 

 disappear if the body is raised to a red heat. On the other hand, 

 such an operation modifies the chemical or physical condition, so 

 that nothing is more difficult than to obtain specimens of the same 

 substance which may be considered as being perfectly identical. 



The uniform fields which are ordinarily used are the internal 

 field of a coil, which is long in comparison with the dimensions of 

 the body ; the field of a system of frames, conveniently arranged ; 

 the field produced by a system of magnets ; or simply the terrestrial 

 field. 



The preceding remark shows, further, that the manner in which 

 the body is brought into the field is not a matter of indifference. If 

 the body is placed in a coil, and the current is suddenly made or 

 broken, we may have effects of induction which, at a given moment, 

 give a value quite different from its permanent value, especially when 

 the coil contains a magnetic core ; in a field which is thus variable 

 the body undergoes a sort of shock, which modifies the final 

 condition. 



Thus, when a bar of soft iron is magnetised in a coil, and is 

 slowly removed, its residual magnetism may be almost twice that 

 which would result from the sudden suppression of the current. 



The method of induced discharges can thus only be employed 

 with certain precautions. If we introduce the body into the coil 

 when the current is already made, it passes through a variable field 

 before attaining the uniform part. The best plan would be to in- 

 troduce the body in the coil, and restrict ourselves to making and 

 breaking the current in a slow and progressive manner. 



SS 2 



