METEORIC AND ARTIFICIAL NICKEL-IRON ALLOYS. 
71 
of producing such figures artificially has not yet been discovered), be regarded as a 
sign of “ meteoric ” origin, yet the absence of such figures does not prove that the 
iron is not meteoric. There is a considerable number of specimens of iron which, 
judged by their mode of occurrence and general appearance alone, would be accepted 
as of similar origin to those classed as meteoric, but of which the origin, on account 
of the absence of the Widmanstatten figures, is regarded as doubtful. In each of 
these doubtful cases the percentage of nickel is either appreciably lower than 6 per 
cent, or appreciably higher than 11 per cent., and the composition is in consequence 
described sometimes as “ anomalous.’ But it is clear that this point of view is 
conventional. 
T1 le essential fact is that no nickel-iron alloy, meteoric or terrestrial, is known 
which gives Widmanstatten figures when the percentage of nickel, in the portion of 
the material showing the figures, is less than about 6 or more than about 15. 
Neither of these limits has been fixed precisely, because the structure of an alloy 
which gives the figures is heterogeneous and the amount of the meteorite used for the 
analysis is generally small (cf Cohen, I., p. 86). 
§ 8. Numerous analyses are available of meteoric irons, poor in nickel, which are 
practically homogeneous (Cohen, I., p. 97). From these it is seen that the percentage 
of nickel may be as high as 7 per cent, without occurrence of Widmanstatten figures. 
Moreover, the crystalline characteristics and other physical properties of “ hexahedral” 
iron (the commonest of the homogeneous types) are practically identical with those of 
the kamacite of octahedral iron (Cohen, I., p. 85), and, finally, although such iron 
may contain only 4 per cent, of nickel or less (cf. Cohen and Brezina, ‘ Akad. 
Sitzungsb. Wien,’ vol. 113, 1904; Moissan, ‘ Comptes Rendus,’ vol. 139, p. 773, 
1904), those specimens richest in nickel contain practically the same percentage of 
nickel as the kamacite of octahedral iron. Hence it seems clear that the percentage 
of nickel can reach some value between 6 and 7 without occurrence of heterogeneous 
crystallisation in the material fr'ftm which the meteorite is formed; and that when the 
amount exceeds this value, the whole of the material cannot crystallise as kamacite. 
Another alloy richer in nickel (taenite) then makes its appearance, and the production 
of Widmanstatten figures by etching becomes possible. 
At first, when the percentage of nickel only slightly exceeds 7, the structure is 
relatively coarse. The layers of less nickeliferous alloy (kamacite) are broad and the 
thin lamellae of the alloy richer in nickel (taenite) which border the layers of kamacite 
are sometimes absent. But as the percentage of nickel rises, the kamacite layers 
become narrower and the layers of taenite more abundant. The kamacite is gradually 
displaced from its position of predominance and the figures eventually become too fine 
to be visible except when highly magnified (Cohen, I., pp. 100, 101) (cf. Section VII., 
§ 15, p. 103). 
§ 9. Data are not available by which to decide the question whether there is a 
definite percentage of nickel at which the Widmanstatten figures disappear. Modern 
