THE METALLOGRAPHY OF METEORIC IROiN" 59 



nounced, who concluded that the origin of the structure is substan- 

 tially as outlined earlier in this chapter, and by other investigations. 



The results of X-ray analysis also support the view that the 

 structure is formed at a relatively high temperature during very 

 slow cooling (Mehl and Derge, 1937; Derge, 1939). 



Another reason why the Vogel theory is untenable is the fact that 

 it requires a face-centered cubic phase to form upon the octahedral 

 planes of a body-centered cubic phase. In such a case the face- 

 centered phase would form in plates along the dodecahedral planes 

 of the body-centered phase. 



The Vogel theory also is inconsistent with the fact that although 

 troilite inclusions solidified at 970°, nevertheless they obviously are 

 older than the surrounding octahedral structure, which abuts against 

 them uncomformably. 



These and other considerations, including the known octahedral 

 crystalline habit of the gamma phase, seem to confirm the earlier 

 and generally accepted view as to the origin of the Widmanstatten 

 structure. 



A low-temperature equilibrium. — The investigations of Owen and 

 Williams (1938) and of Bradley (1939) with X-rays and heat treat- 

 ment indicate the existence of two phases in equilibrium between 

 350° and 580°. Such a two-phase equilibrium, previously not be- 

 lieved to exist in artificial nickel-iron, was produced by annealing 

 the alloy at 580° followed by very extended cooling, in which a 

 gamma phase developed with about 30 percent nickel. The struct- 

 ure produced between 350° and 380° was very minutely crystalline. 



This led these investigators to the conclusion that the coarse 

 octahedral structure in meteoric irons could have been produced 

 either by secular cooling within the 350°-580° range or by repeated 

 heatings and successive coolings within that range, which would 

 have the same effect as a single protracted cooling. This latter 

 condition of production, they point out, might be met if a meteor 

 traveled in an elliptical orbit that brought it periodically near the 

 sun. 



Some of the finer structures found in nickel-rich ataxites might 

 conceivably have been produced in such a manner, but the author 

 cannot accept the conclusion that the coarser Widmanstatten struc- 

 tures arose in so low a temperature range. It does not seem con- 

 sistent with the existence of a very coarse octahedral structure in 

 irons with only 6 or 7 percent of nickel, which naturally would have 

 arisen in a relatively high temperature range in which diffusion was 

 active. It is also a pertinent fact that the coarse precipitation of 

 phosphide on the crvstallographic planes in Horse Creek (pis. 60, 61) 

 show an Fe-FcsP eutectic structure that could not have formed at 



