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



are shown by sanidine. On the other hand, the abundance of liquid inclusions with 

 moving bubbles, and, in certain cases, the outlines and their relation to the direction of 

 cleavage, the smooth surface of the sections, and the optical characters, leave no doubt 

 as to the true determination. Some sections, in fact, show the black cross and that 

 the mineral is positive. Epidote is also well developed in the rock, this secondary 

 product appearing in the form of grains, often grouped or scattered uniformly 

 between the crystals of hornblende. The epidote is distinguished in this case by 

 the brilliancy of its polarisation colours, and a very feeble pleochroism, citron- 

 yellow or an almost colourless shade of green. This mineral is sometimes crystal- 

 lised in fibro-radial bundles. Titaniferous iron is also somewhat common, and is 

 decomposed into leucoxene. Crystals of grey titanite, probably derived from the 

 decomposition of ilmenite, may also be detected. The last-mentioned mineral has 

 sometimes left hexagonal hollows where leucoxene and epidote have subsequently 

 developed. The ground-mass is chiefly formed of quartzose grains, epidote, and 

 the remains of a few indistinct crystals of plagioclase. The microscopical structure 

 shows that this rock is a diorite, and this conclusion is confirmed by the examination 

 of the sections of hornblende. The completeness of these crystals as to their external 

 form, and their freshness, clearly show, it seems to us, that this mineral is prim- 

 ordial, and does not take its origin from the paramorphosis of augite into hornblende, 

 as frequently happens in altered diabases. The presence of quartz also indicates that 

 the rock may be related to the quartziferous diorites, but in order to establish this 

 determination one element — plagioclase felspar — seems to be wanting. Still, on taking 

 into consideration certain other specimens of similar rocks from the same place, which 

 show sections of plagioclase associated with quartz and epidote, it is easy to believe 

 that, in the rock under consideration, the plagioclase has undergone alteration into 

 epidote. It is necessary to mention the fact that for the classification of the rock as 

 diorite there is no other ground than the mineralogical composition and structure, all 

 stratigraphical data being wanting. 



Another rock presenting considerable analogy with the preceding is finer grained, 

 massive, greyish in colour, and breaks with a slightly conchoidal fracture. It also may 

 be classed as a diorite. The naked eye detects in the mass very small crystals of felspar, 

 and more rarely of hornblende. The microscope shows a ground-mass containing 

 rather large sections of hornblende, the crystallographic and optical characters of which 

 are like those of the same mineral in the rock just described, only it is more decomposed, 

 and the zonary structure does not appear. On the other hand, plagioclase, of which only 

 traces were perceptible in the former rock, is here much less altered, and it is possible 

 to determine the species. The lamellar sections of this plagioclase gave an average 

 extinction of about 6° on the trace of M. The symmetrical angles of extinction 

 on the two sides of the polysynthetic lamellse gave as an average 5°. Sections of 



