186 . ■ . . KNOTT AND TANAKADATE 



are now mounted in position (see Plate VI, Fig. 1), everything is 

 ready for making a determination of the magnetic meridian. The 

 stage and all its belongings are clamped to the base of the theodolite 

 and rendered quite free of the Y's ; and, consequently, the coil can 

 be turned round independent of everything else. The magnetometer 

 stage is adjusted until some convenient division on the scale, as re- 

 flected from the magnet mirror, is brought to coincidence with the 

 cross-wire of the observing telescope. 



The coil is now put in circuit with a small cell, which is shown 

 in Plate VI, Fig. 1, hanging from the centre of the theodolite. This 

 is done by putting the terminal double-plug (see small cut on page 

 184) into one of the holes of the resistance box, whose terminals (see 

 Plate VI figure 1) are' joined to the poles of the battery. At the 

 first trial the direction of the current should be such as to make the 

 magnetic field due to the current in the coil have the same (general) 

 direction as that of the earth. This is readily judged of by the 

 quickened movement of the magnet. The reflected image of the scale 

 will in general be seen to move. With the current always on, let 

 the azimuth of the coil be shifted until the originally observed reading 

 of the scale is brought back again to the cross-wires. Since the 

 magnetometer and telescope have been absolutely fixed in position 

 during the whole operation, this gives to the first approximation the 

 direction of the declination. The current in now reversed by simply 

 turning the double-plug half-way round in its hole. In general, the 

 result of this will be that the image of the zero scale reading will 

 slowly move to one or other side of the cross-wire. The resistance 

 in circuit is then adjusted until the current is such as to cause the 

 time of oscillation of the magnet to be some three or four times as 

 long as that under the earth's force alone. The original division of 

 the scale is again brought back to the cross wire by careful adjust- 



