THE METALLOGRlAiPHY OF METEORIC IROflS^ 81 



suiting structures in meteoric irons, although its effects are not 

 determinable. 



By further depressing the gamma-alpha transformation (already 

 depressed by the presence of nickel) it favors the preservation of 

 unstable structures (i. e., transformation structures that normally 

 disappear in slow cooling) by extending downward the range in which 

 such structures remain unchanged. The presence of phosphorus 

 therefore would tend to promote the de\elopmcnt of paraeutectoid 

 structures in nickel-rich ataxites, as will presently be explained. 



Genesis of phosphide inclusions. — VogePs conclusion that droplike 

 masses of schreibersite originate from the melt, while rhabdites and 

 needles separate in the solid (gamma) state, seems to be fully sus- 

 tained by the ternary iron-nickel-phosphorus diagram, by the results 

 of heating experiments, and by the habits of the various types of 

 inclusions. The segregations of needles (lamellae) in the gamma 

 phase would naturally be interrupted if a permanent octahedral 

 structure v/ere formed; therefore we find long needles confined to 

 hexahedrites and nickel-poor ataxites. In such irons the gamma- 

 alpha range was narrow and at a high temperature, and the result- 

 ing octahedral structure was incipient and transitory, causing only 

 a temporary barrier to the formation of such lamellae. This seems 

 evident from the preservation of lamellae of exceeding fineness and 

 regularity, as in Walker County (pi. 52). 



The orientation of phosphide needles usually seems unrelated to 

 the Neumann lines in hexahedrites or the Widmanstatten pattern in 

 octahedrites. They are, in fact, necessarily unrelated; for the planes 

 on which precipitation occurs are determined by the crystalline habit 

 of the precipitant, not by that of the matrix, and in an artificial 

 iron-phosphide solution the phosphide separates on planes that are 

 neither cubic nor octahedral. Therefore the orientation of a phos- 

 phide needle (lamella) may or may not coincide with some plane of 

 an octahedrite, or with any of the twinning planes developed in the 

 alpha phase which are evidenced by Neumann lines in hexahedrites. 



Brezina (1894) from a study of the schreibersite lamellae in certain 

 octahedrites concluded that they follow dodecahedral planes. That 

 conclusion, however, seems clearly erroneous (p. 64). 



Phosphorus and plessite.— Phosphorus has some influence on the 

 formation and the structure of plessite, though its role is variable 

 and in most cases not important. 



Inasmuch as the kamacite and taenite in octahedrites are rela- 

 tively phosphorus-poor, the phosphide content of the original melt 

 is correspondingly concentrated in the plessite fields. The effect 

 of such enrichment, however, is chiefly cumulative, tendmg more or 



