32 
MR. S. W. J. SMITH ON THE THERMOMAGNETIC ANALYSIS OF 
will correspond truly with the change of induction in the inductor secondary and will 
be free from the effect of hysteresis in the secondary of the ring. This follows 
because the total induction in the ring secondary will be practically the same when 
the transient current has ceased as it was before the inductor primary circuit was 
completed for the second time. The induction in the ring will have changed in the 
way represented by the dotted curve b 2 b± and by the full curve bj ) 2 , i.e. its value at 
the end will be practically the same as at the beginning, viz., O b 2 . 
This argument was verified by direct observation. The throw obtained by the 
method just described was indistinguishable from that obtained on replacing the ring- 
secondary by an equivalent resistance not containing iron. 
§ 5. The possible effects of external fields, constant or variable, existing during the 
measurements remain to be considered. 
(a) The majority of the permeability measurements described were made while 
using magnetising fields comparable with that of the Earth, and it might seem that 
the observations could be influenced appreciably by the presence of the latter field. 
Assuming the ring to he placed with its axis of figure in the direction of the 
Earth’s field, it is of course seen at once that the effect of the external axial 
magnetising force upon the magnetic state of the ring is greatly diminished, owing to 
the shortness of the ring in the direction of its axis, by the demagnetising force 
which results from the magnetisation of the ring by the field. But it might be 
expected that if this axial field were such that it produced an intensity H a within 
the ring, appreciable in magnitude compared with the intensity H of the magnetising 
field due to the ring primary, its presence would materially affect the amount of 
induction observed on establishing the field H. 
(b) Another source of possible error, more likely to be serious in the measurements 
at the higher temperatures, was the field due to the current in the coil by which the 
ring was heated. Thus, to take a particular case, the current in the ring primary for 
permeability measurements was about 0‘16 ampere; whilst in the bifilarly wound 
heating coil surrounding the ring a current of more than 2 amperes was required in 
order to keep the temperature of the ring at 700° C. It might seem therefore that if 
the bifilarity were not perfect, the permeability measurements, with a field corre¬ 
sponding to OH6 ampere, taken at ordinary temperatures, would not be exactly 
comparable with those taken at high temperatures under the influence of the same 
primary field. 
It was impossible of course to make the bifilarity perfect in practice. The following- 
numbers, obtained by means of a ring wound as described later, serve to show the 
degree of approximation to bifilarity attained. A current of 0'23 ampere in the ring 
primary gave an inductive throw of 25 - 0 centims. of the galvanometer in the 
secondary circuit. A current of 0"66 ampere in the heating coil gave a throw of 
0‘48 centim. Hence the ratio of the fields produced by the same current in the 
primary and in the heatmg coil respectively would be approximately 150 : 1. Thus, 
