[ 533 ] 
XIV. Magnetization of Liquids. 
By John S. Townsend, M.A. (Diehl.), Cavendish Laboratory, Cambridge. 
Communicated by Professor J. J. Thomson, F.R.S. 
Received June 11,—Read June 18, 1896. 
The magnetization of liquids is a subject which has been examined by several 
methods, and although the relative values of the coefficient of magnetization for 
different bodies agree tolerably accurately, still there are large discrepancies in the 
absolute values given by different observers. An account of nearly all these methods 
is given in Wiedemann’s ‘ Electricitat,’ vol. 3. 
The difficulty of finding the magnetizing force is no doubt a serious objection to 
the accurate determination of the coefficient k in absolute measure by magnetometer 
methods, which, on the other hand, are very convenient for finding the relative 
values of k for different bodies when magnetized by large forces. 
The following null method of arranging an induction balance to determine k in 
absolute measure gives very accurate results, and is very sensitive even when the 
forces are as small as 1 C.G.S. unit. 
The apparatus which was used consisted of a primary and a secondary circuit. A 
quantity of electricity 2MJ/R 3 is induced in the secondary when the current in the 
primary is changed from — i to + i, M being the mutual induction of the two 
circuits, and R 3 the resistance of the secondary. A commutator is used to commute 
the connection of the primary with the battery, and at the same time to alter the 
connections of the secondary with the galvanometer in such a way as to send the 
induced currents in the same direction through the galvanometer, so that it will give 
ey AX i 
a deflection corresponding to a current V— p, p being the number of times per second 
the direction of i is changed. 
By this means a very small value of M can be detected. 
The mutual induction M consisted of three parts :— 
1 . The induction (a) of a long solenoid (wound on a glass tube) on a secondary 
wound round its centre. 
2 . The mutual induction (/3) of two large circular coils. 
3. An adjustable induction (y) which could be altered by small known amounts. 
M = a -f fi + y. 
21.10.96 
