536 THE CONTINENTAL ORIGIN OF FIJI, II., 



but is slightly decomposed along the cracks, with production of 

 dark brownish-yellow serpentine. The colour of this alteration 

 product indicates a rather ferriferous variety of olivine. The 

 fairl}' good cleavage parallel to (001) indicates the same thing; 

 the other cleavages are not apparent, but rough cracks are 

 numerous. Refractive index and double refraction are normal. 

 Unlike the other phenocrysts, the olivine shows little or no trace 

 of solution by the magma, the breaking up of the crystals being- 

 due to mechanical causes. In convergent polarised light a section 

 perpendicular to an optic axis gives a distinct^ curved brush, 

 showing that the optic axial angle is smaller than usual. The 

 dispersion is very distinct, p < v. Interpositions are rare, and 

 include occasional grains of magnetite and apatite. Unindivi- 

 dualised inclusions were not observed. 



Magnetite is fairly abundant, but calls for no special comment. 



Apatite is a notable constituent in rough prisms which pene- 

 trate the magnetite. 



Porphyritic Basalt, approaching Limburgite (Nadari- 

 vatu). Plate xxxvi., fig. 9. 



Macroscopic characters. — In hand specimen the rock is dark 

 bluish-grey in colour. It is very markedly diflerent at first glance 

 from the andesites hitherto described, in that porphyritic felspar 

 is not developed. The rock is nevertheless very strongly porphy- 

 ritic, the phenocrysts being black augite and yellow-green olivine. 

 The former is strongly idiomorphic, the forms being ideally 

 perfect, bounded by \ 010 [>, -j 100 |^, \ 110 ^ and -{ 111 1^. 

 The olivine, too, is beautifully crystallised ; it occurs both as 

 isolated crystals and as clusters of these. The two minerals are 

 about equally abundant; if anything, the olivine is predominant. 

 Magnetite is plentiful. 



The rock is slightly vesicular, the spaces being filled with white 

 zeolitic products. Specific gravity 2-90. 



Microscojnc characters. — In thin section it is obvious that this 

 rock is much the most basic encountered during the whole 

 expedition, Macroscopically no porphyritic felspars are visible, 



