Electrical Resistance of Nickel in Magnetic fields. 653 



The electromagnet produced fields up to 15,000 C.G.S. units 

 with pole-faces 8 mm. in diameter and '98 cm. apart. The 

 semi vertical angle of the conical pole-pieces was approxi- 

 mately 39°. 



A sensitive Broca galvanometer the combined resistance 

 of whose coils was 90 ohms was used, and worked very 

 satisfactorily. 



In taking readings, the balancing point on the bridge was 

 first obtained with the specimen in zero field. The specimen 

 was then slid in between the pole-pieces. The magnet current 

 was turned on and the new balancing point obtained with a 

 second rider. The difference of readings gave the step on 

 the bridge. Demagnetization was effected by means of a 

 reversing key operated automatically by the movement of one 

 of the plates of a liquid-resistance box, which slowly diminished 

 the current at the same time. Good demagnetization was 

 obtained by this method. The specimen was then slid from 

 between the pole-pieces of the magnet and balance again 

 obtained with the specimen in zero field, the cycle of opera- 

 tions being repeated. 



Field strengths were measured by the ballistic method, 

 using an exploring coil 5 mm. in diameter. 



In determining the accurate orientation of the specimen 

 with regard to the external field the method of procedure 

 was as follows. The pointer in connexion with the vertical 

 scale was kept at the zero (or the corrected zero) . The speci- 

 men could then be rotated in the horizontal plane containing 

 the common axis of the pole-pieces by turning the screw 

 connected with the pointer for the horizontal scale. This 

 pointer was placed at say the division marked 110°, for 

 which position the specimen may or may not be accurately 

 at an angle of 110° with the longitudinal field. A series of 

 observations was taken, the results being plotted in a curve. 

 The pointer for the horizontal scale was then rotated through 

 a few degrees towards the transverse position and another 

 curve taken. This process was repeated until a series of 

 curves for short intervals on each side of the transverse 

 position had been obtained. Assuming the curves sym- 

 metrical for equal intervals on each side of the transverse 

 position (a result justified by the fact that reversal of the 

 field has no effect on the magnitude of the change of resist- 

 ance), the comparison of similar curves furnishes a guide for 

 further experiments at intermediate intervals. In this 

 manner we can finally arrive at a series symmetrical for 

 equal intervals on each side of the transverse position, and 

 thus determine the true transverse position very exactlv. 



Phil. Mag. S. 6. Vol.27. No. 160. April 1914. 2 X 



