12 THE VOYAGE OF H.M.S. CHALLENGER. 



work of lines which intersect one another at an angle of about 124 30'. This cleavage is 

 well marked, and forms a very important element in determining the nature of the 

 substance. Other sections are elongated, and form long parallel lamellae ; but their 

 prismatic cleavage is not always distinctly marked, although with higher powers the 

 lamellae can be detected. Some of these sections are extinguished parallel and per- 

 pendicular to the lateral edges, others are extinguished at an angle of about 15. This 

 mineral is therefore clinorhombic, and belongs by its angle of cleavage and optic 

 properties to an amphibolic mineral. The fibrous structure and colour which the sections 

 sometimes assume, would afford some ground for classing it as uralite ; but I have 

 nowhere found even the slightest indication of its having been derived from augite. If 

 it be uralite the paramorphosis is complete in all the specimens submitted to me for 

 examination. Another circumstance which makes it difficult to form an opinion is that 

 the contours are vague and so ill-defined that it is impossible to discover forms resembling 

 pyroxene, or in any way recalling its presence. The sections are dichroitic, but very 

 feebly so, owing to lightness of colour. The presence of hornblende in an olivine 

 rock is not an isolated fact ; such an occurrence was noticed by Tschermak in the 

 eclogites accompanying the granulites of Lower Austria. He considered the amphibole 

 as smaragdite. He also found hornblende in the olivine rocks of Greenland : the 

 specimen from Simmetak, in the Museum of Vienna, is an aggregate of olivine, horn- 

 blende, bronzite, and biotite. If the occurrence of this amphibolic mineral in the olivine 

 rock of St. Paul were more constant, it might be compared to that of Varollo, in which 

 Stelzner found olivine, bronzite, and hornblende. 1 I am disposed to refer it to actino- 

 lite, of which it seems to possess the most characteristic properties. 2 In some of its 

 sections it is not so well characterised, having lost all trace of cleavage, and looking like a 

 chloritic decomposition product. Even then, however, the optic properties remain the 

 same in the altered as in the fresh and normal sections, and in the most thoroughly 

 decomposed, characteristic enclosures in the direction of the principal axis are always 

 found. 



I consider these enclosures to be colourless transparent microliths, in form generally 

 prismatic, their greater axis being always parallel to the vertical axis of the enclosing 

 mineral. They are frequently black and opaque, at least in some parts of their 

 length, and some of them look like hollow cylinders whose inner walls are covered with 

 black grains. It is very difficult to pronounce upon their nature. Were they merely 

 microscopical pores or gas enclosures with linear arrangement, I would compare them to 

 the enclosures noticed by Dathe 3 in the enstatites of the olivine rocks of Saxony, and by 



1 Stelzner, Zeitschrift d. d. geol., Gesellschaft, 1876, p. 623. 



2 Some preparations of the rocks of Greece, described by Becke under the name of Serpentin Olivin-Gesteine, have 

 shown me the same amphibolic mineral, which that author refers to actinolite, as I do myself. (Becke, Gesteine von 

 <}riechenland, Tschermak, Min. Mitth., 1878 and 1879.) 



3 Dathe, Neues Jahrbuch, &c., foe. cit., p. 235. 



