BODIES WHICH ARE FEEBLY MAGNETIC OR DIAMAGNETIC. 663 



Let there be two systems of identical coils, each composed of an 

 inducing coil and of a concentric induced coil.* The inducing 

 coils are traversed by the same current, and the two induced coils 

 are connected by a delicate galvanometer, so that their currents 

 have no action on the needle. By introducing the bismuth bar 

 in one of these coils, the inductive effects are diminished ; the 

 galvanometer measures then the difference of two induced cur- 

 rents, so that the coefficient k' may be determined. 



It must be observed that with a conducting core without mag- 

 netic action a diminution of the same kind would be observed, 

 owing to the induction currents developed in the mass ; but it can 

 be seen that a copper bar produces much greater diminution than 

 a bismuth bar. Nevertheless there is a source of error which 

 must be taken into account in the case of bodies which are very 

 feebly magnetic or diamagnetic, if they are not bad conductors ; 

 they tend to give too high a value for diamagnetic bodies. 



1225. This same method has been applied to perchloride of 

 iron. A vessel filled with the solution is taken as core. M. Silowf 

 uses a long tube placed vertically, in which is coiled the mag- 

 netising spiral; the induced coil, which is much shorter, is placed 

 equidistant from the two ends. M. Bergmannf places the liquid 

 between two concentric coils of the same lenght. The coefficient 

 of mutual induction M (771) should then be multiplied by i+^irk. 

 The same physicist also used annular coils. Silow has shown that 

 solution of ferric chloride obeys the same law as powerfully mag- 

 netic metals; the coefficient begins to increase rapidly with the 

 magnetising force, passes through a maximum, and then slowly 

 decreases. The maximum is with a magnetising force of 0*4 C.G.S. 

 units, and amounts to lyQ.io" 6 . 



It is generally assumed that, for feebly magnetic or diamagnetic 

 bodies, the value of k is a constant. It must, however, be re- 

 marked that intensity of magnetisation has been compared, not to 

 the intensity of the field itself, but to the intensity of the current, 

 which cannot be considered equivalent in the case of electromagnets. 

 We have seen that in any case this law does not hold for per- 

 chloride of iron in solution at any rate for small forces. For 

 greater forces the coefficient seems to be tolerably constant. 



* TOPLER. Pogg. Ann., Vol. CLIV., p. 60. 1875. 



t SILOW. Wied. Ann., XL, p. 324. 1880. 



BERGMANN. Beiblatter der Physik, Vol. in., p. 812. 1879. 



