236 
MESSRS. R. T. GLAZEBROOK AND J. M. DODDS ON THE 
the primary circuit, then the total induced current produced by reversing the primary 
. 2AD 
is 
Let ft be the first throw of the galvanometer needle produced by this reversal, T the 
time of a complete vibration, X the coefficient of damping, G the galvanometer constant, 
and H the horizontal intensity of the earth’s magnetism. 
Moreover let r be the coefficient of torsion of the suspending fibre. 
Then, 
2M% H( 1 +t) 
R 1 
■ (3 
sni 2 
(1) 
Again, let a current %' be passed through the same galvanometer directly afterwards, 
and let 6 be the permanent deflection produced. 
Then 
Hence 
But 
Therefore 
., H(l + r) n 
i ——— tan v 
v_T 
( 2 ) 
lfl+5 
tan 6 
T/3 
sm ~ 
i S + V 
i~ Y 
- 7 - 2ttM S + Y tan 6 
R= 
T 1 + ■ 
A 
V 
• 0 
sm — 
9 
(3) 
(4) 
T is, of course, the observed time of oscillation. 
The correction for the finite amplitude of the swing is too small to produce any 
error in the result. No correction for damping or torsion is required. 
In the experiments the deflections on the scale were measured, not the angles of 
deflection. We require, then, to reduce the scale readings to angular measure. 
Let and q be the scale values of the throw and deflection, 2 the distance of the 
point of the scale vertically below the axis of the telescope from the point which 
appears to coincide with the cross wire when the needle is at rest, z being measured 
in the same direction as and q, let a be the distance of the scale from the mirror. 
Thus in the figure (fig. 4) let M be the centre of the mirror, O the point on the 
