442 PROCEEDINGS OF SECTION E. 



one of the Tonga gi'oup. Against tliis fact it is well to remember that 

 the great volcanoes of the Taupo region showed no miusual activity^ 

 and that beyond a slight change in the activity of several hot springs 

 at Rotoriia and other places close to Tarawera there were no signs of 

 disturbance in the volcanic zone of New Zealand itself. Again it fre- 

 quently happens that Ruapehu is unaffected by the spasmodic action 

 of Ngauruhoe, and a similar independence is shown by Te Mari ; for 

 during its period of greatest violence both Ruapehu and Ngauruhoe 

 were quiescent. These facts suggest that too much importance should 

 not be attached to the time relation of the eruptions of 1886, and even 

 suggest that the occurrence of these within a few months of one 

 another was fortuitous. 



Outside of the Tonga-Kermadec ridge there is a volcano in a pro- 

 nounced condition of activity at Savaii, and apparently the actual 

 cone of present eruption only came into existence when the emption 

 commenced, and, notwithstanding the long and continuous eruption,, 

 only a small cone has been foi'med, though an immense amount of 

 material has been ejected from the volcano. Previous to 1902, when 

 there was an eruption from another cone on the same island, there had 

 not been an eruption in Savaii from any of the cones for, perhaps, 200 

 years. 



Further west the New Hel>rides are now the scene of much 

 volcanic action. The islands of Tanna, Ainbiyn, Lopevi, and Vanna 

 Lava all have volcanoes, some of which are amongst the most active 

 of the world's volcanoes. It appears that no very exact record has 

 hitherto been kept of the eruptions that have occurred in this group 

 of islands. In the Santa Cruz group Tinakolo is the best known for 

 its activity. In South Victoria Land it is well known that Mount 

 Erebus is frequently active. 



The Volcanic Rocks. 



Microscopic petrogi"aphy lias within recent years made exact 

 study of volcanic rocks a necessaiy portion of geological research. By 

 its use we have now become familiar with a great variety of rock 

 types, and have a considei-ajjle knowledge of their distribution. 



Within the South-west Pacific rhyolites are known from New 

 Zealand only. Here with the fragmental pumice material they cover 

 the greater part of the volcanic plateau of the North Island. 



The occurrence of trachytes is veiy limited, so far as the islands 

 is concerned, though very important, as shown by Jensen, on the 

 eastern margin of Australia. They occur at Auckland and Campbell 

 Islands, and at Dunedin, but are unknown elsewhere in New Zealand. 



Andesites have a far more general distribution. Within New 

 Zealand, Dunedin and Banks's Peninsula have andesites of somewhat 

 peculiar characters. In the North Island typical andesites are found 

 over a wide area. Tlie greater part of the older Tertiary rocks of the 

 Coromandel Peninsula are andesitic and similar t3rpes are widely dis- 

 tributed in the north of Auckland. Tlie rocks of the present active 

 volcanoes are exclusively andesitic. The Kermadecs. the rocks of 

 which were described by Thomas and Speight, are formed mainly of 

 andesitic rocks, though there is a variation towards a basaltic type. 



