306 Mr. A. K. Chapman on the 



It is the purpose of the present paper to report on the 

 variations o£ the Corhino effect with the magnetic field in a 

 number o£ metals and alloys and, at the same time, to 

 compare these variations with measurements of the Hall 

 effect in a disk exactly similar to the one used for the 

 Corbino effect. 



The experimental method used was the same as that 

 described in the former paper. In the earlier experiments 

 some difficulty was experienced with the rotating commutator. 

 It was in series with the galvanometer, and any inequality 

 in the pressure of the brushes or any undue pressure of the 

 brushes always resulted in undesirable electromotive forces 

 which made the galvanometer very unsteady. By lubricating 

 the brushes with a mixture of graphite and oil, and by taking 

 care that each brush had a smooth, broad, but not too heavy 

 contact, a great deal of this trouble was overcome. 



The specific resistances of the metals under investigation 

 were determined by measuring the potential difference 

 between the central wire and the circumference of the disk 

 when a known current was flowing radially through it. To 

 take account of any non-uniformity, four points on tho 

 circumference, ninety degrees apart, were chosen, the 

 potential difference between each point and the centre 

 measured, and the mean taken. In all cases the disks were 

 of the order of Q'9 mm. thick, so that such measurements by 

 means of a Wolff potentiometer are accurate only for metals 

 of considerable resistance. The values for copper and silver 

 were simply taken from a table. The disks were all immersed 

 in a well-stirred oil-bath during this process. 



Circular plates with four integral electrodes, ninety degrees 

 apart, were used for the Hall effect. The electrodes by 

 which the primary current was led into the plates were 

 heavy ; the others were much smaller and were connected 

 to the galvanometer circuit. These plates were placed in a 

 hard rubber mounting- and introduced into the magnet in 

 the same position as that occupied by the disks used for the 

 Corbino effect. They were made circular in order to include,, 

 as nearly as possible, the same flux. In every instance 

 where curves for both effects are given, the two disks were 

 either cut from the same sheet of metal or cast from the 

 same mass of molten metal, as the case may be. 



Copper. — It was thought advisabLe to repeat the measure- 

 ments on copper under more favourable experimental 

 conditions. The disk, cut from a sheet of commercial copper,, 

 was 0*9 mm. thick and 5"0 cm. in diameter. The radii, 

 r 2 and r l5 were 5'0 cm. and 0*266 cm., as they were for all 



