382 PEor. J. w. JTJDD ox the teettaey axd 



separation of a black or dark brown, usually opaque, substance, 

 probably magnetite and other iron-oxides, along certain planes 

 within the crystals. But in the olivines the separated material 

 assumes ver}' peculiar and highly characteristic forms. In 1871 

 Prof. Zirkel, in describing the olivines of the gabbros of Mull, pointed 

 out that they contained great numbers of blackish or brownish 

 needles arranged in curious combinations, of which he gave 

 a drawing*. Prof. Zirkel insisted on the fact that similar ap- 

 ])earances are never found in the olivines of basalt, but that they 

 occur in the olivines of many gabbros like those of Yolpersdorf, the 

 Yalteline, and other localities. He also pointed out the fact that 

 these dark inclusions sometimes become so numerous in the olivine 

 as to render the mineral black and opaque, so that it may be easily 

 mistaken for magnetite, except in very thin sections which have 

 been carefully prepared to illustrate the structure of the mineral. 

 Dana has also pointed out this change of olivine into a black opaque 

 substance resembling magnetite, often exhibiting a fissile structure 

 similar to that of micaf, and Wadsworth has described the same 

 phenomenon. 



The study of a series of specimens which have originally existed 

 at different depths from the surface, in the Western Isles of Scotland, 

 enables us to fully understand and explain this interesting change. 

 This is rendered more easy by the fact that the results are, in this 

 case, not complicated by serpentinization, a totally different kind of 

 change due to qnite other causes. 



The first appearance of the change in question takes place along 

 those irregular fissures that so frequently traverse olivine-crystals. 

 Along these incipient or completed fissures, when they are examined 

 by the aid of high powers, small stellar-groups of black or deep- 

 brown filaments are seen making their appearance mingled with 

 reticulations of cavities containing liquid or solid substances, like 

 those formed in the felspars and pyroxenes. The stellar groups 

 have much the appearance of dendritic markings (see PL XII. 

 figs. 1, 2), and when seen foreshortened, or partially within the range 

 of view of the higher powers of the microscope, present the cha- 

 racters represented by Zirkel t. Sometimes these star-like bodies 

 become crowded together so as to make the surfaces of the cracks 

 and sometimes also the outer portions of the crystal black and opaque 

 (see PL XII. figs. 3, -1). 



Precisely similar appearances then make themselves visible along 

 certain planes within the crystal, which- are certainly parallel to the 

 optic axis, but the more exact crystaUographic relations of which I 

 have as yet been unable to determine. With these stellate groups 

 of fibres"^ flat brown plates, like those appearing in the pyroxenes, 

 sometimes appear in considerable numbers. Examples may be 

 found of the gradual conversion of the stellate enclosures into 

 tabular ones, by the fiUing-in of the intervals between the rays 



* Zeitschr. d. d. geol. Gesell. Bd. xxiii. (1871) pp. 59, 60. 



t System of Mineralogy, 5th ed. (1875) p. 258. 



X Zeitschr, d. deutsch. geol. Gesell. vol. xxiii. (1871) p. 59, Taf. iv. fig. 11. 



