38 BULUETIN 18 4, UNITED STATES NATIONAL MUSEfUM 



until from about 450° downward, the mass would be wholly in the 

 alpha phase. A meteorite of that composition would be a hexa- 

 hedrite (or a nickel-poor ataxite), consisting wholly of alpha kamacite. 



With a nickel content above 6 percent the structure would remain 

 heterogeneous at room temperature, the iron being an octahedrite 

 or (with nickel above about 13 percent) a nickel-rich ataxite. 



The supposed eutectoid. — Osmond and Cartaud assumed a eutec- 

 toid line BD at about 350°, corresponding with the combined A3 — Ai 

 line FG at 723° in the iron-carbon diagram, and they assumed also 

 that the line CE was a phase transformation line. Both assumptions 

 are now generally considered as erroneous, as there is believed to 

 be no iron-nickel eutectoid, and the line CE marks only the begin- 

 ning of ferromagnetism in cooling, and not a phase change. 



The views of Osmond and Cartaud, supported by Benedicks and 

 later by the Hansons and other authorities, were substantially em- 

 bodied in the diagram of Guertler (1910) for meteoric irons. Pfann 

 (1917), accepting it with slight modifications, fixed the eutectoid 

 ratio of nickel around 17 to 18 percent, at which composition a 

 meteoric iron actually is an ataxite with a paraeutectoid structure. 



In accordance with the Guertler diagram, he held, quite correctly, 

 that an iron with less than 6 percent nickel would be a hexahedrite 

 consisting wholly of kamacite, and from 7 to 13 percent it would 

 be an octahedrite, with the amount of plessite (the assumed eutectoid) 

 increasing. With more than 13 percent it would consist of plessite 

 with vestiges of kamacite; at 17 to 18 percent of eutectoid (para- 

 eutectoid) plessite; then taenite would increase and plessite diminish, 

 until at about 27 percent it would be pure taenite. Such, in fact, 

 are substantially the changes that are observed with rising nickel 

 content; examples of the various types will be found in the plates. 



The writers supporting that theory have differed somewhat as to 

 the eutectoid percentage. The Hansons give the composition range 

 for all structures between pure kamacite and pure taenite as 9 to 

 30 percent, as compared with Pfann's 7 to 27 percent. They place 

 the eutectoid temperature at about the same point as did Osmond 

 and Cartaud. 



Eutectoid theory untenable. — The earlier theory of the iron-nickel 

 system has been outlined at some length because of its historical 

 interest and the distinguished authorities that have supported it. 

 That support has continued almost, if not quite, to the present; for 

 the diagram presented as recently as 1927 in International Critical 

 Tables retains the eutectoid line, and an analogous line is seen in the 

 still more recent diagram of Bradley and Goldschmidt (1939) referred 

 to later. 



