540 THE CONTINENTAL ORIGIN OF FIJI, II., 



regarded as primary, like the analcite of the S3^dne3' basalts.* 

 The evidence, however, is not strong enough for any such 

 hypothesis to be advanced in tliis case. 



EXPLANATION OF PLATES XXXV. -XXXVI. 



Plate XXXV. 



Fig. 1. — Granite — Narokorokoyawa — showing effects of strain in the bending 



of the cleavage lines in biotite ( x 20 diameters). 

 Fig. 2.~Quartz-Diorite — Nadranikula ( x 20 diameters). 

 Fig. 3. — Quartzite — Togicedra— showing injection of the rock by quartz 



veins; crossed nicols ( x 20 diameters). 

 Fig. 4. — Augite Andesite — Namulowai — showing characteristic group of 



augite crystals and general features of rock ( x 20 diameters). 

 Fig. 5. — Glass-cavities in felspar, augite andesite — Namulowai. The 



cavities contain relatively large bubbles which remain fixed. The 



larger inclusions are irregular, while the smaller ones have the form 



of negative crystals ( x 400 diameters, about). 



Plate xxxvi. 



Fig. 6. — Hornblendic Andesite — Buki Levu — general character of the rock 

 ( X 20 diameters). 



Fig. 7. — Hornblende Andesite — Korobasabasaga ( x 20 diameters). 



Fig. 8. — Olivine-bearing Andesite — Tama ni Ivi — to show the mode of 

 occurrence of the olivine. The particular portion of the rock 

 photographed contains no augite ( x 20 diameters). 



Fig. 9. — Porphyritic Basalt — Nadarivatu. This photograph shows well the 

 characteristic development of the olivine, and fairly well the 

 tendency to Spheroidal Cracking ( x 20 diameters). 



*G. W. Card, Bee. Geol. Surv. N.S.W., Vol. vii., p. 93, tt seq. 



