186 
Mr Secirle, An experiment with a pair 
clearance between the spheres and the drawing board is 3 or 4 
millimetres. The set screw is an important feature since it 
enables the needle to be changed or re-sharpened when its point 
has deteriorated with use. No satisfactory results are likely to be 
obtained with a damaged needle-point. 
In attaching the U -piece to the magnet care is taken that the 
axis of the needle intersects the axis of the magnet at a point 
midway between the centres of the spheres. Any small want of 
balance, due to the magnetic “ dip ” or to other causes, is corrected 
by a small rider. 
The positions of the centres of the spheres are marked down 
on the drawing board by aid of a simple optical device. Two 
straight lines aob, cocl are drawn by a diamond upon a piece of 
c 
mirror glass, as shown in Fig. 3, and a circle, slightly greater in 
diameter than the spheres of the suspended Robison magnet, is 
described about o as centre. Two of these mirrors are provided, 
one for each end of the suspended magnet. 
A bell jar or other transparent screen is placed over the 
magnetometer to shield it from draughts. The bell jar rests on 
three blocks 2 or 3 millimetres high and a strip of thin card passing 
through the gap is used to move the mirrors into position. The} 7 
are in adjustment when, on looking vertically down, the observer 
sees the sphere symmetrically within the circle. The bell jar is 
then lifted and marks are made on the drawing board at a, b, c 
and d. On joining these marks by straight lines the projection 
of the centre of the sphere on the board is obtained. 
§ 2. The magnetometer may be used in an experiment to find 
the pole strength of a second Robison magnet. This magnet, 
