REPORT ON THE PETROLOGY OF OCEANIC ISLANDS. 103 



The fragment of included granite is a rolled pebble, large grained and very 

 micaceous. Grains of plagioclase, orthoclase, quartz, and mica are to be seen in it. The 

 felspathic sections are altered into micaceous matter. From the smallness of the angle 

 of extinction of this plagioclase it may be classed as oligoclase. Alteration has, one 

 might say, effaced the original characteristics of the mica which is transformed into a 

 greenish matter filled with secondary products. It also happens that fibro-radiated 

 chloritic plates have taken the place of the micaceous mineral. Colourless sections 

 polarising with blue tints are also observed ; these are lengthened and coated with 

 mica, and are perhaps cordierite. The quartz has the characters of that mineral in 

 granitic rocks. 



Another clastic rock from the same locality presents the appearance of a fine- 

 grained felspathic sandstone, penetrated by oxide of iron, and breaking with a plane 

 fracture. Microscopically it is an aggregation of grains of felspar and quartz with 

 heterogeneous particles of rock. Some of the last named are mica schist, formed of 

 grains of quartz ranged in lines with lamellae of muscovite between. Other fragments 

 are of a vitreous nature, the glass being altered, having been originally vesicular. 

 In this base there are numerous plagioclase microliths ; no bisilicates are to be seen. 

 These splinters may, all things considered, be referred to porphyrites ; sometimes a 

 glance is obtained of micaceous lamellse. Finally, there are found amongst this debris 

 of ancient rocks some grains which seem to be splinters of the paste of a red porphyry. 

 The broken felspars are principally jdagioclase ; some of the sections being very finely 

 striated, and giving small extinctions, are probably oligoclase ; others have few hemi- 

 tropic strige, and by this character may be taken as albite ; finally, there are others 

 presenting considerable resemblances to microcline. The titanite occurs as an inclusion 

 in a grain of felspar, the latter being perhaps albite. This idea is suggested on taking 

 account of the frequent association of both minerals in the more or less schistose 

 ancient rocks. Orthoclase only plays a subordinate part, sections of felspar being, in 

 fact, rarely seen without hemitropic lamella?. Titanite is, on the contrary, somewhat 

 common, and it tends to show that the original rock, the disaggregation of which 

 furnished the constituents of that we are considering, contained probably hornblende. 

 The quartz is in irregular fragments, which occasionally, though not often, show 

 undulating polarisation. Their crystalline outlines, which are discovered in certain 

 cases in the form of the sections, or in the arrangement of the inclusions, seem to 

 indicate that this mineral is more likely derived from a porphyritic rock than from a 

 granite. Amongst the minerals formed in situ, and developed in the interstices, we may 

 mention certain small greenish scales resembling chlorite. 



Some schistose rocks from Port Sussex are of an earthy grey-blue colour, with a 

 homogeneous ground-mass with darker blackish bands, recalling the appearance of an 



