550 Messrs. K. Honda and S. Shimizu 



on 



by a quartz fibre in a thick copper case. A magnetizing 

 coil (length = 40 cm., 47ni = 394 , 4) and a compensating coil 

 of nearly the same dimensions were placed respectively due 

 magnetic east and west o£ the magnetometer. The magneto- 

 meter was placed in such a position that the specimen exerted 

 the maximum effect upon it. The vertical component o£ the 

 earth's field was compensated for. The deflexion of the 

 magnetometer was measured by means of a scale and 

 telescope. 



In measuring magnetization, the following precautions 

 were taken. The verticality of the two coils was tested by 

 means of a level. The line of the magnetometer, the com- 

 pensating coil and the magnetizing coil, was then tested by 

 a compass needle. The compensation of the earth field and 

 then that of the magnetizing coil were next effected ; lastly, 

 the scale and telescope were pLiced in correct positions. 



The precautions above enumerated were especially necessary, 

 as the magnetization and the magnetic hysteresis in strong- 

 fields were to be studied with the ovoids placed in vertical 

 positions. Though these precautions were taken with the 

 utmost care, the magnetizations by opposite currents of equal 

 strength were not exactly equal in absolute amounts, so that 

 a small asymmetry of the hysteresis curve was also observed 

 in the strong fields.. This difference was, in the most un- 

 favourable case, not greater than 1 per cent, of the total 

 magnetization for a field of 700 c.G.s. This probably arose 

 from a slight deviation of the coils from the vertical line. 

 If the lines of force at the centre of the magnetometer be 

 not vertical, but be in the meridian plane, the field due to 

 the coils may slightly affect the horizontal component of the 

 earth's field, without producing any deflexion of the magneto- 

 meter. If the horizontal component be increased by a current 

 in one direction, a current in the opposite direction will 

 diminish it. In the first experiment, therefore, two mag- 

 netizing curves for opposite currents were taken. These 

 two curves almost coincided with each other below a field of 

 200 C.G.S., but slightly deviated above that field. Since the 

 disturbing force is proportional to the magnetizing current, 

 the correction for the intensity of magnetization can be 

 derived from a pair of the opposite magnetizations by equal 

 and opposite currents ; hence in the second and third ex- 

 periments, the magnetizations by equal and opposite currents 

 of the maximum strength were taken, and the correction 

 was found, and applied to magnetization by currents of one 

 direction. 



