THE METALLOGRiAPHY OF METEORIC IBOIN 75 



structure oriented with reference to the surrounding oxide, somewhat 

 as dendrites in steel are oriented with reference to the enclosing 

 mould. 



That the iron droplets solidified first is indicated both by the 

 structure and by the relative fusibility of the two components. 

 Though purified ferrosic oxide melts at from 1,527° to 1,600°, natu- 

 ral magnetic oxide has been found to melt at from 1,210° to 1,260°, 

 owing to the probable presence of impurities, and even at lower 

 points. The presence of sulphide (see Chapter XV) might have 

 reduced the melting point to a still lower temperature. Thus the 

 rapidly growing dendrites would have rejected the still fluid oxide 

 imtO the iron had gathered into rounded masses around which the 

 oxide finally solidified. 



Cosmic reheating. — Quite different from the foregoing structures, 

 produced during a meteor's flight through the air, are those resulting 

 from reheating before it reached our atmosphere. Such reheating 

 is reflected in various alterations of structures previously established, 

 which are not confined to a superficial zone but may extend through- 

 out the mass. 



Numerous iron meteorites show such changes of structure, varying 

 from slight to extensive alteration. Examples of partial alteration 

 are the Mount Sterling, Social Circle, and Dungannon octahedrites, 

 which show a strong secondary granulation, with diffusion in dense 

 plessite areas but with the coarser taenite lamellae little affected. 

 More extensive change is shown in the unidentified iron illustrated 

 in plate 67, although even in this case the octahedral pattern is 

 clearly retained. The most complete alteration is found in some 

 of the irons designated as nickel-poor ataxites. 



Nickel-poor ataxites. — That low-nickel irons should occasionally 

 have a structure apparently similar to that of very high-nickel irons 

 was an anomaly not explainable by the older methods of research. 

 Now the microscope reveals that low-nickel and high-nickel ataxites 

 have very different structures and that there really is nothing in 

 common between the two groups except a superficial similarity in 

 appearance. 



Experiments in the heat treatment of meteoric irons have proved 

 that great changes could thus be produced in low-nickel irons, in- 

 cluding the obliteration of Neumann lines and of the octahedral 

 structure, secondary granulation, and the diffusion of taenite and 

 phosphide. 



Cosmic reheating, before the meteor arrived at the earth's atmos- 

 phere, has produced like effects, ranging from slight alteration in 

 some hexahedrites to their complete alteration into nickel-poor 

 ataxites. Chesterville (pi. 9) is a typical example of such an ataxite 

 produced by cosmic reheating of a hexahedral iron, the diffusion of 



