REPORT ON THE PETROLOGY OF OCEANIC ISLANDS. 65 



lava, seems to show that igneous action has had something to do with the development 

 of the inclusions. The form of the grains into which the quartz sections are divided 

 shows clearly that the parts belong to one large individual. The cause just invoked to 

 account for what may perhaps be termed the scorification of the felspar probably 

 produced this cracking of the quartz. Sections of micropegmatite are less common than 

 in the granitite previously described, but they sometimes appear. Besides the rare 

 scales of biotite, there are some small crystals of hornblende ; these are almost colour- 

 less, but some sections with the characteristic cleavage show incontestably that they 

 are amphibolic. 



Other specimens of older rocks from Green Mountain must be classed with diabase, 

 although, as we shall see, the micro-structure is not altogether identical with that of 

 rocks of this type. These fragments are very much altered, and easily crumble down. 

 The naked eye distinguishes felspar, biotite, and a granitoid structure. The microscope 

 shows that the rock is formed of an aggregate of plagioclastic lamella?, augite, and 

 biotite, with hornblende as an accidental constituent. The triclinic felspar shows 

 extinctions which lead one to believe it to be labradorite. The auaite shows itself in 

 excessively broken-up sections, formed of an accumulation of irregular granules. The 

 grains of augite do not appear to result from fractures along the lines of cleavage ; the 

 mass rather resembles a crushed crystal. Fibrous hornblende appears between the 

 grains, and shows itself most clearly at the extremities of the pyroxenic sections, where 

 it may be seen to pass into black mica. The augite is greenish in colour, and more like 

 that of diorite than of diabase. The lamellae of biotite are often twinned, the limit of 

 the twin being parallel to the lamellae ; the composition plane is probably the pinacoid 

 OP. Grains of augite sometimes appear associated with biotite ; in this case it is not 

 uncommon for the former to be oriented with the vertical axis parallel to the lamellae 

 of this mica. Hornblende, which is rather rare in the preparations, is only distinguish- 

 able from biotite in ordinary Ught by its structure, and by a decided prismatic cleavage. 

 Sometimes, when this mineral borders augite, it is fibrous. Finally, we note the 

 transparent prismatic crystals of a mineral which appears grey from the number of 

 inclusions it contains. It would be classed as cordierite if its colours with polarised 

 light were a little more vivid ; perhaps it is an altered felspar. 



At Eed Hill, as at Green Mountain, fragments of old rocks are found which have 

 been brought up by recent eruptions. The specimens from Red Hill may be classed 

 with the gabbros, and microscopic examination shows them to be olivine gabbros. The 

 rock has to the naked eye a granitoid texture ; in colour it is reddish, being impregnated 

 with limonite. Triclinic felspar is distributed through the mass in the form of grains, 

 and is intimately associated with a pyroxenic mineral. The elements of this rock 

 measure about 5 millimetres in diameter. 



(FHYS. CHEJI. CHALL. EXP. — PART VII. — 1889.) 9 



