E. Cohen — Meteoric Iron from Soudan. 257 



the troilite bad been crowded back through the softened mass 

 during the flight, or worked up in consequence of their low 

 specific gravity. It may also be especially emphasized that there 

 is an entire lack of any perceptible association with daubreelite, 

 graphite or schreibersite, as well as of nodules of larger dimen- 

 sions, such as usually prevail in iron rich in troilite. 



Schreibersite is unusually sparsely represented in these sec- 

 tions. In most of them it is entirely lacking, and only the 

 largest sections, of 118 sq cm , contained, directly under the 

 burned crust or near it, lamellae of which the largest was 3 cm 

 long and 2 mm thick. 



Observation of the unetched sections alone show that 

 JSPGoureyma belongs to the comparatively rare group of 

 coarsely granular irons. The fractures marking the limits of 

 the granules, however, as usual, are brought out more dis- 

 tinctly and completely by etching, which makes them broader 

 and deeper. Isolated fractures open to a width of f mm ; others 

 are so small as to be perceived only under the microscope. 

 Occasionally the fractures are found to be filled up with troi- 

 lite, the separation of which must have continued after the 

 formation of the fractures, although the fluid arrangement 

 justifies the conclusion that the main separation had occurred 

 earlier ; these granular components are bounded by irregular 

 edges, though the majority are in general fairly isometric, 

 with a thickness varying between \\ and 3 cm . Isolated ones 

 diminish to f cm . Many are elongated, especially the larger 

 ones, which then attain a length of 7 cm and a thickness of 2 cm . 

 In this respect the five cross sections are exactly alike. The 

 section made from a tongue-shaped projection (Plate IV, 

 fig. 2) appears, however, to be an individual ; evidently a 

 single granule which has been drawn out in the direction of 

 the longitudinal axis of the meteorite ; in which case, because of 

 the slenderness of the projection, contraction fractures (perhaps 

 parallel to the section) would certainly not be at all obvious. 



Etching develops glistening plates (Plate IV, figs. 1 and 2), 

 the arrangement of which is suggestive of Widmanstatten 

 figures. Under the microscope, however, one sees that there 

 are really no connected lamellse at all, but that tiny spangles 

 or granules, better reflecting than the remaining nickeliferous 

 iron, collect in indistinctly defined and small elongated clus- 

 ters, arranged in more or less connected rows which intersect 

 in groups. At any rate, there is nothing perceivable under 

 the microscope which could suggest the kamacite blades and 

 their taenit fringes found in normal octaheclrite. The remain- 

 ing nickeliferous iron, which perhaps is plessite, appears to 

 predominate in places and then again to be less prominent ; 

 always, however, forming a coherent body like the nickeliferous 

 iron of compact ataxite. 



