Structures in Ohsldian, Ferlite, and Leucite. By F. Butleij. Ill 



sections similar to those which an octahedron would yield (Fig. 8). 

 They are often feebly translucent towards their margins. This 

 probably results from peroxidation of the protoxide of iron which 

 the mineral, if magnetite, would contain. I believe that these 

 crystals are magnetite altered at their margins into limonite 

 or possibly in some cases hematite. Lasaulx has noted the occur- 

 rence, in the altered lavas of Auvergne, of microscopic pseudomorphs 

 of hydrated oxide of iron after magnetite, in distinct octahedra of 

 a brownish-red colour and translucent.* I merely record this point 

 because, from the slight marginal translucency of these crystals, 

 some doubt might be experienced as to whether or not they 

 really represent magnetite. The hand specimen, from which the 

 section was cut, attracts the magnetic needle, and seems, at certain 

 spots, to repel it also to a very slight extent. Fig. 12 shows 

 similar, but much smaller crystals, which are plentifully dissemi- 

 nated in the crystal of leucite, presently to be described, but they 

 are so minute, as a rule, that unless magnified five or six hundred 

 diameters they appear to be mere amorphous specks. At various 

 points along the spherulitic bands in the obsidian, minute tube-like 

 bodies may be seen under the microscope, with a quite low magni- 

 fying power (such as 20 diameters). The tube-like bodies appear 

 to emanate from sometimes one, sometimes the other, of the inner 

 layers of the spherulitic bands, pass through the outer zone, and 

 protrude, often for a considerable distance, into the glassy, colour- 

 less obsidian. Sometimes they are seen to terminate in a rounded 

 end, like the end of a test-tube (Figs. 4 and 5), but they are 

 often cut off by the planes of section, which in some cases gives 

 them the appearance of ending abruptly. After a careful examina- 

 tion of these forms, I have come to the conclusion that they are 

 not tubes, but solid rods of glass. An inspection of the form shown 

 in Fig. (5 may serve to render this point more apparent. Here, 

 one of these rods is seen to pass through the cortical zone of 

 a spherulitic band, and to completely surround an outlying, isolated 

 spherule. Now there is no reason to doubt that this clear, glassy 

 ring, which appears as a mere girdle in the section, is in reality a 

 slice through a vitreous envelope which completely encased the 

 spherule. Had this rod been a tube, and the envelope of the 

 spherule merely a bulbous continuation of that tube, it is manifest 

 that, whether vacuous or filled with a gas or a liquid, the enveloped 

 spherule would infallibly have stripped out during the process of 

 grinding the section ; and even assuming that a vacuum existed 

 between the cup and the piece of the spherule, and that by atmo- 

 spheric pressure or by some accident it had remained in situ, there 

 is little doubt but that some small amount of emery mud would have 

 * ' Neues Jahrb. f. Min.,' 1870, p. 695. 



