METEORIC AND ARTIFICIAL NICKEL-IRON ALLOYS. 
51 
which the permeability was unexpectedly low, possibly because of insufficient 
preliminary heating (see later). On raising the temperature to about 575° C., the 
permeability rose rapidly, and on allowing the temperature to fall to 465° C. again 
it was found that, although the permeability fell considerably, it was now 25 per 
cent, greater than the value at this temperature when the cooling was interrupted. 
Subsequently, when the temperature had fallen to that of the air the permeability 
was much larger than it would have been if the cooling had been uninterrupted 
(see 53). It was, however, less than the value that would have been obtained by a 
further reheating to 575° C., as previously described ( cj '. 37 and 38). 
The ring was now heated again above the critical temperature, and allowed to cool 
uninterruptedly when, as shown in the table, it showed the old permeability, at the 
ordinary temperature, corresponding with such treatment. (The first value, 54, was 
probably too high because the ring had not been allowed sufficient time to acquire a 
steady state.) The ring was then cooled to —10° C. without showing any appreciable 
variation of permeability. The object of this cooling was to find whether there 
was any sign of the comparatively sudden increase of permeability observed by 
Hopkinson in the case of a 25 per cent, nickel steel. 
Finally^ the ring was left undisturbed for some weeks, and the permeability was 
found to be the same at the end as at the beginning of this period. 
§ 6. The experiments just described showed that the permeability was subject to 
peculiar, but nevertheless definable, variation as the result of thermal treatment. 
Repetitions of a given process had led always to practically the same result, and 
hence, apart from the question of the interpretation of the data already obtained, it 
seemed worth while to study further, at various temperatures, those effects of 
interrupted heating and cooling which had been shown already, at some temperatures, 
to be accompanied by marked changes in the magnetic properties of the material. 
Thus, for example, it still remained to examine the effect of cooling from a tempera¬ 
ture between that corresponding to the maximum permeability and that at which the 
permeability became imperceptible. I was anxious also to test more fully whether 
the remarkable reversibility between the two steep branches of the /jlO curves, 
described by Guillaume in the case of artificial alloys, was shown by the meteorite. 
But at this point it became necessary, through want of time, to abandon the work or 
to enlist the services of another observer. Fortunately, Mr. J. Satterly, B.Sc., 
formerly a student and assistant at the Royal College of Science, volunteered to 
complete the examination of the meteoric iron on the lines proposed and also, if 
necessary, to test in a similar way a sample of artificial nickel iron, of nearly 
the same composition (see Section I., § 7, p. 26), which I had obtained for this 
purpose. 
After Mr. Satterly had made a few preliminary observations corroborating those 
described above, the insulation of the ring secondary for some reason became 
defective, and it was decided to take off all the coils and re-wind them. The original 
H 2 
