8 BULLETIN 184, UNITED STATES NATIONAL MUSEUM 



comprising about 75 percent of the total. Hexahedrites and nickel- 

 poor ataxites (the latter very few) comprise about 15 percent and 

 nickel-rich ataxites about 10 percent. 



Interpreting structures. — Such terms as bands, lines, and needles 

 are inexact, as they merely describe the appearance of structural 

 details as viewed in section. 



The kamacite plates in an octahedrite produce an appearance of 

 bands, which vary in width according to whether the section is normal 

 to their planes or oblique. If the section happens to be made parallel 

 with the plane of a kamacite plate, it might be cut flatwise, in which 

 case it would appear as a triangular or quadrangular field. 



True plessite fields do not usually represent cross sections of large 

 interstitial masses of that constituent, but oftener sections of plate- 

 like bodies. Except in the fine and finest octahedrites the inter- 

 stices among the kamacite plates are similar to them in form — that 

 is, they also are plates bounded by octahedral planes. Thus, on a 

 polished surface the section of such a tabular mass of plessite may 

 appear (1) as a large field, if cut parallel with its plane; (2) as a wider 

 or narrower band, if cut transversely to its plane; or (3) as a triangle 

 or rectangle, if cut across a corner of the tabular mass. 



The foregoing observations apply to a Widmanstatten structure 

 in which the bands are closely packed. When they become sparse 

 and widely separated, as in fine and finest octahedrites, the interstices 

 are larger, and the plessite instead of being m thin tabular form ap- 

 pears in relatively large masses bounded by the octahedral planes. 

 In such cases the fields represent cross sections of such masses, and 

 they would be large regardless of the direction of the section. 



The larger fields of plessite in the finest octahedrites (e.g., Butler, 

 Carlton, Laurens Countv) (pis. 12, 13, 15) often enclose areas of 

 minute kamacite needles or spindles, which are not found in coarser 

 octahedrites in wliich the bands are less widely spaced. These are 

 not true needles but sections of small lamellae. The thicker, rounded 

 particles, usually termed spindles, are probably sections of small so- 

 called "swollen" plates. This may be inferred from the fact that sec- 

 tions often show many spindles of similar shape but few or no 

 rounded kamacite areas that would represent cross sections of a true 

 spindle form. 



Long needles of schreibersite (pi. 52) are probably always cross- 

 sections of very tenuous lamellae, and the same is true of the long 

 threads of taenite often found in plessite fields. Very short minute 

 lines of taenite or schreibersite might conceivably be true needles, 

 but even they are more lilcely to be sections of flakes of the precip- 

 itated component. 



Minute rounded dots of taenite (pi. 40, fig. 2) may be assumed to 



