696 Prof. E. P. Adams and Mr. A. K. Chapman 



do current in the galvanometer we must have 



M / C' + MC = m 1 I, 



M'C'-MC=m 2 I, 

 from which 



MC = l(>i-m 2 )l=rcl. 



To determine the circular current which is produced by the 

 magnetic field it is thus necessary only to take the change 

 in the inductance of the standard coils on reversing the 

 magnetic field. 



The strength of the magnetic field was measured with a 

 Grassot fluxmeter, using a coil of four turns of wire of a 

 diameter equal to the diameter of the disk employed. 



The mutual inductance M between the disk and coils was 

 determined experimentally. A brass disk of the same dimen- 

 sions as the disks employed, with a hole at its centre the 

 same size as the central wire, had a seetor of small angle 

 cut out of it, and to the edges two very heavy copper bars 

 were soldered which served to lead in the current. Under 

 these conditions the current in the disk is circular, the 

 current density being inversely proportional to the radius. 

 The coils were placed in the same position relative to this 

 disk as to the disks actually employed, and the whole placed 

 between the pole-pieces of the magnet so as to measure the 

 mutual inductance under the required conditions. The 

 mutual inductance was measured by comparison with a 

 known standard. As iron was one of the metals employed, 

 the mutual inductance between the coils and an iron disk, 

 similar to the brass disk, was also determined. 



The direction of the circular current produced in the 

 magnetic field was determined by comparison with the direc- 

 tion of deflexion of the galvanometer when a current was 

 sent through a single circular turn of wire which took the 

 place of the disk. If the current is carried wholly by free 

 negative electrons which collide with the metallic atoms in the 

 way assumed in the previous paper, then the circular current 

 should have the direction of the magnetizing current when 

 the radial current flows outward from the centre. Metals 

 for which this is the case will be said to have a negative 

 Corbino effect, in agreement with the convention adopted 

 for the Hall effect. 



The disks employed had a thickness of 009 cm. The 

 diameter of the central wire soldered into them was 

 9^ = 0*266 cm., and the internal diameter of the brass ring 

 into which the disks were soldered w T as r 2 = 5'0 cm. A 

 description of the disks employed, their mode of preparation, 



