MR. A. E. OXLEY ON THE INFLUENCE OF MOLECULAR 
114 
mass of air which fills the volume m w occupied by the water.. In all the experiments 
described below the current was kept constant at 8'6 # amperes, to within 1 part in 
500. Hence the factor involving the magnetic field on the R.H.S. of (a) and (/3) is 
constant, and we get by division 
yM = x,M a + {x w m w -xa'ma) ^ .(y) 
The specific susceptibilities of water Xw and air Xa were taken as 7' 25xl0~ 7 and 
210 x 10 -7 (at 20° C.) respectively. 
(2) Experiments Down to the Temperature of Liquid Air. 
The apparatus designed for the measurement of susceptibility at low temperatures 
is shown in fig. 4. 
The phial P, supported from the torsion arm at A', was surrounded by a thin- 
walled glass tube G, the latter being sealed at the lower end and supported by the 
cork flange CC. Conduction of heat from or to the torsion arm was prevented by 
inserting a strip of wood W between the phial and the projection of the torsion arm. 
A copper cylinder N, which just fitted into the glass tube, served to keep uniformity 
* In the last two experiments the current was increased to 9 • 0 amperes. 
