METEORIC AND ARTIFICIAL NICKEL-IRON ALLOYS. 
93 
first place, that the iron consists largely of an alloy containing about 6 per cent, of 
nickel. This agrees with the result derived from chemical analysis, that the kamacite 
of octahedral iron contains between 6 and 7 per cent, of nickel. As already shown 
(see, e.g ., the magnified photograph of the core, Plate 1, fig. 4), the present meteorite 
consists mainly of kamacite. 
§ 2. The thermomagnetic behaviour of the meteorite at temperatures below 500° C. 
shows, however, that its structure is complex, that it does not consist wholly of the 
Gf- per cent, alloy, kamacite. This is seen in the curves for the first winding (fig. 9), and 
again very clearly in fig. 14, second winding. The latter shows, in greater detail than 
the former, that there is an irreversible alloy present in which magnetism disappears 
at about 480° C., and reappears near the temperature of the air. It is seen that this 
alloy behaves, on interrupted heating, in the way characteristic of irreversible alloys. 
Inspection of the available experimental data shows that the alloy having the 
thermomagnetic properties just stated is the richest in nickel of the irreversible 
alloys, and contains about 27 per cent. Ni. 
The results thus point to the conclusion that the thin bands of nickel-rich alloy 
which occur in the meteorite contain between 25 and 30 per cent, of nickel. The 
geometrical distribution of the taenite in the ring has already been described 
(Section I., § 5, p. 25) and its loss of magnetism would clearly produce gaps in the 
magnetic circuit and consequent diminution of permeability. 
The amount of nickel-rich alloy in the material is proved to be small not only 
by inspection of the etched surface, but also by the chemical analyses (Section I., 
§§ 2, 3, pp. 23, 24) of the meteorite as a whole. If the meteorite contains m per cent, of 
Ni, and the kamacite and taenite k and t per cent, of Ni respectively, then the number 
of grammes of taenite in 100 grammes of the meteorite is 
x = 100 ( m—k)/(t—k ). 
This assumes, of course, that the distribution of the kamacite and taenite is 
uniform ( cf. the analyses by Foote, p. 23, and Wraight, p. 24). 
If we assume that in the present case, in round numbers, m = 7‘3, k = 6'5, and 
t = 27, then x — 4 approximately, i.e., only about 4 per cent, of the meteorite can 
consist of taenite. 
Inspection of the curves of fig. 14 (see p. 56) shows that when the nickel-rich 
constituent is magnetic the permeability of the ring is over 60 per cent, greater than 
when it is not magnetic, but 4 per cent, of the nickel-rich constituent is quite 
sufficient to account for this difference. Thus, if a ring contains x per cent, of an 
alloy of permeability g 2 arranged as a single transverse gap in the rest of the material 
(which is of permeability g x ), the equivalent permeability of the whole ring will be 
g = g x g 2 
