METEORIC AND ARTIFICIAL NICKEL-IRON ALLOYS. 
95 
and if we assume that, in round numbers, £' = 40, £ = 27 and h = 6'5, then 
x = 39 approx., 
or, roughly, 40 per cent, of the eutectic will consist of kamacite. 
Having, for the sake of directness, stated this view at the outset, the arguments 
in its favour may now be given. 
§ 4. A preliminary experiment was performed soon after winding of the ring 
to find approximately the maximum permeability of the material. It gave the 
following numbers :— 
Temperature. 
H. 
/x. 
o 
658 (approx.) 
0-20 
0-31 
7- 91 
8- 23 
Previously, the ring had been heated only once to a high temperature, viz. to 
a dull red, during a preliminary testing of the heating apparatus. The permeabilities 
are given in the same arbitrary units as those of the figure for the first winding 
(see Section III., § 6, p. 39). 
After the above measurements the ring was heated twice to a temperature above 
800° C. Some months later the permeability was again measured, at temperatures 
comparable with the above, with the following results :—- 
Temperature. 
H. 
p.. 
o 
622 
0-43 
4-72 
661 
0-445 
5-17 
678 
0-44 
4-38 
Making every allowance for possible error of experiment and for the isolated 
character of the experiments first cited, it would seem to be proved that heating 
above 800° C., in the early stages of the heat treatment, is followed by a considerable 
lowering of the maximum permeability. 
This conclusion is supported by the numbers given in the previous tables (pp. 42 to 45). 
The maximum permeability seems to have become gradually smaller as the experi¬ 
ments proceeded. Thus, whereas the maximum permeability first quoted above 
corresponds approximately to p, = 1000 C.G.S., the highest value observed towards 
the end of the experiments during the first winding did not exceed 700 C.G.S. 
(H = 0 - 48). The data obtained during the second winding show that at the end of 
the experiments the maximum permeability of the ring probably did not exceed 
600 C.G.S. (H = 0-43). 
Although part of the decrease may have been due to oxidation of the ring, the 
