METEORIC AND ARTIFICIAL NICKEL-IRON ALLOYS. 
I 09 
the permeability change (under the same field H = '43 C.G.S.) during heating to 
100° G. of a ring, not previously heated, cut from the core already described. 
In this case there is no initial decrease, but an increase of about 5 per cent, 
between 15° C. and 100° C.—as would be expected if the material contained only 
6 per cent, and 40 per cent, alloys. 
It may be mentioned further that the micro-photographs of taenite given in 
fig. (5 cl), Plate 2, closely resemble those of well-known examples of “sorbitic,” or 
finely-segregated, pearlite (cf Osmond, ‘ Microscopic Analysis of Metals,’ pp. 88 and 
89, 1904). The two fragments of taenite (seen in fig. 5 d) appear to belong to 
differently orientated plates ; but such structural details as are shown in the original 
negative are difficult to reproduce, and no quantitative value is claimed for the figure. 
Better photographs could probably be obtained after a more prolonged search than 
was made while the apparatus was at my disposal ; but even the best microphoto¬ 
graphs of fine-grained structures frequently leave room for doubt with respect to 
their significance. On the other hand, if two constituents of a material have 
thermomagnetic properties which are sufficiently distinct, their coexistence will be 
demonstrable without doubt by the magnetic method, however fine grained the 
microstructure may be. 
§ 12. In conclusion, it seems reasonable to hope that the theory advocated will 
prove to be an adequate basis of correlation of the properties of all alloys of nickel and 
iron, and that the thermo-magnetic method is capable of further development as a 
means of observing continuously and of interpreting changes in the internal structure 
of magnetic alloys. 
7B88ENTSD 
S APR. 1908 
