4 AKT. 'J. — Ü. NAGAOKA, AND S. K US AK ABE : EFFECT OF 



The ovoid, together with the element leaving its extremities 

 outside the eoil, was thickly coated with asbestos paper, so that 

 the temperature variation of the ovoid was very small even at 

 1000°C. The ovoid was placed, vertically in the middle of the 

 coil, as shown by the dotted lines in Fig. 1 and Fig. 2. The 

 vertical component of the terrestrial magnetic field was compensat- 

 ed by a coil wound outside the magnetizing coil. 



A magnetometer with a small bell-shaped magnet suspended 

 by a fine quartz fibre was placed in such a position that the 

 effect of a small vertical displacement of the ovoid was quite 

 negligible. This was effected by placing the magnetometer in 

 such a position that the line joining the centre of the ovoid with 

 the magnetometer magnet was inclined at an angle, whose tangent 

 = i with the horizontal line, in the vertical plane containing 

 the axis of the coil and the magnetometer. On the other side of 

 the magnetometer symmetrical with the magnetizing coil, was 

 placed another coil of the same strength, to compensate the effect 

 of the coil on the magnetometer. In order to ensure the smallest 

 possible displacement of the coils so as to get accurate compensa- 

 tion, they wore placed on a wooden bench with V-shaped grooves 

 lined with brass plates, on which the levelling screws of the coils 

 rested. These coils were fed by accumulators. The current was 

 measured by a Siemens and Halske amperemeter. 



The temperature of the ovoid was measured by a thermo- 

 electric junction with a D'Arsonval galvanometer in series, and 

 the constant of the pyrometer was tested by means of a mercury 

 thermometer to about 300° and by the melting point of sodium 

 chloride at 780 . 



The annexed figures show the arrangement in plan and in 

 elevation. 



