THE METALLOGRAPHY OF METEORIC IRON 43 



This fact has a bearing on the modification of meteoric structures 

 by reheating. With an octahedrite containing 7 percent nickel, for 

 example, the gamma-alpha transformation begins at about 700° and 

 is completed at about 500° (fig. 5). But in reheatmg (fig. 6) the 

 alpha-gamma transformation does not begin at the line AC, or even 

 at the line AB. The temperature must reach the line AC, about 

 700°, before any change in the established structure would begin to 

 take place. Then in recooling the iron would again have the same 

 long sojourn in the gamma-alpha range, during which its former 

 structure might be reestablished. 



A meteoric iron-nickel diagram. — The diagram shown in figure 5 

 relates to artificial iron-nickel alloys, embodying approximately the 

 results of experimentation. If a diagram could be accurately drawn 

 for meteoric iron it probably would show considerable variation in 

 the position of the transformation points, although the nature of 

 the transformations would be similar to those in artificial iron-nickel 

 alloys. Unfortunately the construction of such a diagram is impos- 

 sible because of our limited knowledge of conditions of production 

 of meteoric irons and the thermal changes to which they have been 

 subjected. 



In the artificial iron-nickel diagram the transformation points 

 are determined positively from the behavior of alloys of definite 

 composition under controlled conditions; but in meteoric iron only 

 the final cooled product can be studied, and the transformation points 

 can merely be inferred from the observations of various investigators 

 upon irons of varying compositions. 



One sample, for instance, may contain a substantial proportion 

 of phosphorus, which depresses the gamma-alpha transformation, 

 while another may be phosphorus-free. The rate of cooling also 

 markedly affects the position of the transformation, which is depressed 

 as the rapidity of cooling increases, and it is impossible to allow for 

 that factor. Meteoritic structures generally were produced by ex- 

 tremely slow cooling, which would make the transformation take 

 place at higher temperatures than in artificial iron-carbon alloys. 



It is believed that the diagram as presented approximates in some 

 degree the equilibrium conditions of iron and nickel in meteorites. 

 We may faii'ly assume, however, that in meteoric irons both lines 

 are less depressed than in artificial alloys, so the corresponding struc- 

 tural changes would take place with higher percentages of nickel 

 than in artificial alloys; or conversely, that with the same percentage 

 of nickel they would take place at higher temperatures because of 

 much slower cooling than is possible to attain artificially. Cooling 

 of secular slowness might produce a more perfect transformation at 

 lower temperatures than could be accomplished in a laboratory. 



It may well be that, for meteoric nickel-iron, the line AC in figure 



