Geology.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 685 



since the surface of the island has been subject to the action of rain-water, as is 

 proved by the entire absence of clay and soil beneath the peat. This conclusion is 

 supported by the presence of only comparatively small alluvial fans at the heads of 

 the coves and harbours. 



The slipping of peat may in some measure have evened the slopes of the valley- 

 sides, but such slips cannot be regarded as agents that have modified the form of 

 the rock walls, for no stones or rock matter is visible in the shore exposures of peat, 

 except such small rounded stones as may well have been disgorged by sea-lions as 

 they crawled over the land. 



On the other hand, the semicircular heads of the valleys, their flat floors, their 

 straight or gently curved courses, are a.ll features which are usual in country that 

 has been glaciated. The soundings of Perseverance Harbour suggest that it has 

 been overdeepened where the ice which filled it received the large additions from 

 Mount Honey and Mount Lyall, and therefore shows the same features as the fiords 

 and lakes of south-west Otago. The valley east of Mount Honey becomes a hanging 

 valley. It was occupied by ice tributary to a larger stream, which flowed from the 

 east of Mount Dumas, and received ice from Mount Eboule, Mount Filhol, and from 

 the south of Mount Honey, and had therefore a high surface, with which the surface 

 of the small tributary was confluent. The floor of the tributary, in consequence 

 of the small quantity of ice which occupied it, was less deeply eroded than that of 

 the main valley, and when the ice melted it remained perched on the side of the 

 main valley. 



The terraces on the north-west side of the North-east Valley are regarded as 

 lateral moraines, an opinion which was strongly supported by an inspection of the 

 banks of the small streams which pass through them. Here and there beneath the 

 peat some rocky matter was exposed. It was found to be fragmentary, and, though 

 no striated boulders were seen, the form of the deposit and the shape of the rock- 

 fragments supported the idea of glacial origin. 



Filhol' s explanation of the surface features of the island was largely based upon 

 the effects of volcanic action. The hills were regarded as volcanic cones. It will, 

 however, be shown subsequently that the close resemblance between the lavas on 

 different hills, or in some cases their actual identity, may be held sufficient to prove 

 that they must be regarded as parts of the same lavas which have become separated 

 by erosive action. 



Geological Structure. 



There is a greater variety of rocks in Campbell Island than would be expected 

 from a mere consideration of its dimensions ; but this was clearly shown by Filhol. 

 Both sedimentary and volcanic rocks are represented, and both occur in a good num- 

 ber of varieties. Their mutual relations allowed of very definite conclusions being 

 arrived at as to the geological history of the island. 



Mount Menhir is formed of a gabbro which extends to the sea on the western 

 side of North-west Bay, and perhaps reaches to Dent Island. Near the coast its 

 upper portion is much decomposed. The surface of the gabbro appears to be ex- 

 tremely uneven. This is suggested by the fact that it forms the whole of Mount 

 Menhir, reaching to an elevation of about 600 ft. above sea-level, whereas half a 

 mile away it reaches the sea-level on the west and north, but is not seen in the cliffs 



