30 
MR. S. W. J. SMITH ON THE THERMOMAGNETIC ANALYSIS OF 
§ 4. There is a farther source of inaccuracy in the ordinary method of measuring 
permeability, of which the effects are usually small, but can on occasion become quite 
noticeable. It may be of interest to describe a method of procedure by which such 
effects can be eliminated almost entirely. 
In the usual way of measuring permeability by the ballistic method the galvano¬ 
meter circuit includes, in addition to the ring secondary, one of the two coils of an 
inductor not containing iron. The convenience of this arrangement arises from the 
fact that by its use measurements of resistance are avoided which otherwise would be 
necessary owing to the fact that the total flow of electricity in a circuit, in which 
a given change of induction is produced, depends upon its resistance. The method is 
especially convenient when, as in experiments of the kind described in this paper, the 
resistance of the secondary circuit has a succession of different values. But, unless 
certain precautions are taken, values of the permeability determined in the way just 
indicated are frequently liable to quite appreciable error owing to the effect of 
hysteresis in the ring secondary. 
The quantity of electricity indicated by the galvanometer throw, when a current 
is established in the primary either of the inductor or of the ring, is a measure of the 
total change of induction in the secondary circuit. When a current is established in 
the primary coil of the ring the total change of induction in the secondary of the 
inductor is zero and the galvanometer throw therefore indicates the total change of 
induction in the ring secondary. But when a current is established in the primary of 
the inductor, the resulting total change of induction in the secondary circuit, 
indicated by the galvanometer throw, does not strictly represent the change of 
induction in the secondary of the inductor alone, because, owing to hysteresis, the 
total change of induction in the ring secondary is not zero. The induced current in 
the secondary, rising from zero to a maximum value and then again falling to zero, 
causes the iron in the ring to be subject to a magnetic force varying in the same way. 
In consequence the iron is left magnetised and the total quantity of electricity 
which circulates in the secondary circuit is less than would have been the case if there 
had been no hysteresis by an amount depending on the particular sample of iron and 
upon the magnitude of the changing field to which it is subjected in the way 
described. 
The following numbers will serve to show the relative importance of the effect in 
question in the case of the experiments herein described. The meteoric iron ring 
was demagnetised by the method of continued reversal and diminution and a current 
was then established in the primary of the inductor. This caused a galvanometer 
throw of 6’3 5 centims. The current was then broken and a deflection of 6'25 centims. 
was obtained in the reverse direction. The current was then again made and broken, 
and the respective throws were 6'3 centims. and 6 - 25 centims. The whole process 
was then repeated and the observed throws were 6'35, 6H5, 6 - 25 and 6'25 centims. 
The ring secondary was next cut out of the circuit and replaced by an exactly equal 
