36 BUULETEN" 18 4, UNITED STATE'S NIATIONAL MUSEnCJM 



at a temperature approaching that at which normal pearlite is formed 

 a similar black component is produced, appearing at moderate mag- 

 nification to have an obscure granular or amorphous structure. At 

 higher magnification its structure is more or less lamellar. This 

 was formerly designated as sorbite, but in reality it is a refined 

 pearlite. 



The terms troostite and sorbite are falling into disuse, both being 

 referred to as pearlite. The terms pearlitic and sorbitic, however, 

 will be used in this work as convenient designations for structures 

 having the appearance either of lamellae or of a black unresolved 

 aggregate. The latter structures, composed of particles of gamma 

 and alpha nickel-iron, the author believes may sometimes have been 

 mistaken for the true iron-carbon troostite or sorbite. 



Pearlite in meteorites. — True pearlite has not been found in mete- 

 oric irons, although some contain enough carbon to make a local 

 development of pearlite theoretically possible. A perfect pearlitic 

 pattern of high- and low-nickel lamellae, however, was observed by 

 Pfann (1917) in San Cristobal, a very high nickel iron, and also by 

 the author in Otumpa (H), Seelasgen (Ogg), Leeds (Om), Youndegin 

 (Og), and Walker County (H), examples of which are illustrated 

 in the plates. 



Such structures, however, are not strictly eutectoids, made up, 

 like pearlite, of components of definite compositions, but merely 

 two-phase gamma-alpha structures in which the composition of at 

 least one phase is variable. 



VIII. THE IRON-NICKEL SYSTEM 



When nickel is substituted for carbon the resulting diagram is 

 quite different from that of the iron-caxbon system, yet with inter- 

 esting and important analogies. 



Differences in transformation. — The phase changes and the general 

 processes of transformation are similar; but instead of a separation 

 of ferrite and cementite there is a separation of a body-centered 

 alpha phase (kamacite) from a face-centered gamma phase (taenite). 

 And just as a ferrite-cementite eutectoid is produced with 0.80 per- 

 cent of carbon, so with high (but indefinite) percentages of nickel 

 we find structures somewhat suggesting in appearance a taenite- 

 kamacite eutectoid; in the absence of any accepted designation for 

 such structures they will be referred to as paraeutectoid. 



It is clearly to be noted that the heterogeneous structure of meteoric 

 irons is not due to the segregation of iron and nickel, but to the seg- 

 regation of the nickel-richer and nickel-poorer solid solutions of iron 

 and nickel — taenite and kamacite — just as the structures of steel 

 are produced by the segregation of cementite and iron, not of carbon 

 and iron. 



