BODIES WHICH ARE FEEBLY MAGNETIC OR DIAMAGNETIC. 66 1 



If m is the magnetic moment of the sphere, and C the coefficient 

 of the wire, the equation of equilibrium is 



If, for instance, the field is obtained by means of Faraday's 

 electromagnet, provided with armatures symmetrical in reference to 

 a common axis, the needle will be arranged so that the sphere can 

 move along this line ; the original position, moreover, being outside 

 the centre. The centre, in short, would represent a position of 

 unstable equilibrium for diamagnetic bodies. 



It is even advantageous to place on the electromagnets armatures 

 which are symmetrical in respect of the centre, but dissymmetrical in 

 reference to the common axis, and to use a long needle, the middle 

 of which is in the centre of symmetry.* As the direction of equili- 

 brium of the needle makes an angle with the axis, the couple 

 due to the action of the field on an element of volume du is 



ty 



-k<$> du; the resultant couple gives, as condition of equilibrium, 

 ou 



C A - k \ d> 



Vxxi ~~ rC I **p 



The integral by which we should multiply the coefficient k will 

 be constant for needles of the same form placed in the same field, 

 provided the position of equilibrium is not modified. 



M. Becquerel also used the ordinary balance, and this method 

 is especially convenient for the case of gases. A glass globe, placed 

 below the armature of an electromagnet, is suspended to the beam 

 of a balance. The weight necessary to restore equilibrium, when 

 the electromagnets are excited, gives a measure of the action of the 

 field on the globe, and the gas which it contains. The globe being 

 at first empty, and then filled with a gas at known pressure, the 

 difference of the two weighings gives the action of the gas. 

 Becquerel found that air is magnetic, and that this magnetism is 

 due to the oxygen alone. 



1223. As Faraday has shown, f experiments made in air only 

 give apparent values of the coefficient of magnetisation that is 



* E. BECQUEREL. Ann. de Chim. et de Phys. [3], Vol. LXXXII., p. 68. 1851. 

 t FARADAY. Experimental Researches, 2362. 1845. 



