492 PROCEEDINGS OF THE AMERICAN ACADEMY 



In connection with this, it is interesting to note that, on polishing 

 and etching octahedral, dodecahedral, and cubic faces on crystals of 

 galena, lines were developed precisely similar in character and posi- 

 tion to the Neumann lines. Now it is to be remarked that on the 

 cube faces of the Hauptmannsdorf iron, which Reichenbach regards as 

 consisting of kamacite only, — ihat is, the purest form of iron, — we 

 have a prevailing cubic structure, as indicated by the so-called Neu- 

 mann lines. But in proportion as the iron in the jarocess of crystal- 

 lization has excluded foreign material, as shown by inclusions such 

 as graphite, sulphide of iron (troilite), phosphide of iron (Schreibersite) 

 and the like, the octahedral form prevails. 



It is further to be noticed in this connection, that on the sections of 

 kamacite plates in characteristic octahedral irons fine lines appear 

 called " SchrafRruDgs " by the Germans, which are evidently identical 

 with the Neumann lines, and these follow the direction of the pre- 

 vailing crystalline form. Hence, on this view, the coarser structure 

 which gives rise to the Widmanstattian figures with the well-marked 

 trias of Reichenbach is connected with the exclusion of incompatible 

 material in the process of crystallization, while the more uniform 

 structure of the so-called cubic meteorites, marked by the Neumann 

 lines, depends on the circumstance that the material was susceptible 

 of crystallization as a whole. It is by no means necessary that, to 

 fulfil this condition, the material should be pure iron. It may be an 

 alloy capable of crystallizing in mass, as is the case with many 

 crystalline alloys. 



We conclude, then, that the crystalline structure of the coarsest 

 octahedral irons is not more definite than that of the cubic irons ; but 

 in the first case the process of crystallization determines the elimina- 

 tion of the various materials which form the Widmanstattian plates, 

 while in the second case no such elimination takes place, and between 

 these extremes there is every gradation. The element of time may 

 also be an important, and perhaps the chief, condition in determining 

 the result, for we should naturally expect that, during a very slow 

 crystallization, foreign material would be more completely eliminated 

 than during a process wliich was comparatively rapid. That the 

 crystalline structure is equally definite in both cases, is shown by 

 the fact that cleavage forms may be developed with equal readiness 

 in either class of irons. 



Though there are no very definite data in regard to the crystal- 

 lization of nickeliferous alloys, yet it has been suggested that the 

 presence of nickel in the meteoric iron might determine its crystalliza- 



