206 THE ANTARCTIC. 



It is difficult to gain any more detailed notion of 

 the connection among themselves of isolated portions 

 of Victoria Land, without entering upon over-fantastic 

 speculations. A lofty mountain chain is seen rising steep 

 and abrupt from the sea, extending from north-west to 

 south-east and then on the whole to the south-west, till 

 growing lower, they turn south-east and east in high 

 latitudes with a final curve south. The summits probably 

 consist principally of volcanic matter, like the outlying 

 islands, and the mineral specimens obtained support 

 this conjecture. At the same time it would be absurd 

 to assign a purely volcanic structure to the whole of 

 the extensive country, and non-eruptive minerals brought 

 up from the ocean bed moreover militate against such a 

 theory. The single block of true granite brought up 

 by the dredging net near Coulman Island is sufficient 

 to prove that in the region of Victoria Land also old 

 crystalline rocks must be present, that an early geological 

 formation has to be reckoned with, in which the great 

 outlines of the mountain structures are merely accompanied 

 on the coasts by volcanoes of far later origin. It is worthy 

 of notice that from the only volcano in Victoria Land 

 known to be active, a chain of volcanic islands runs in 

 nearly a direct line northwards — Beaufort Island, Franklin 

 Island, Coulman Island and Possession Island. Hans 

 Reiter, in the work already frequently referred to, gives 

 it as his opinion that the coast of Victoria Land is formed 

 by a contorted range, the continuation of the great Western 

 Pacific curve, of which the last recognisable ranges are 

 those of Stewart Island to the south of New Zealand. 

 He further assumes that the Parry Mountains turn east, 

 and that the contorted island range seen by Ross at the 

 east end of the ice barrier is continued across Graham's 

 Land, thus completing the circle of the Pacific range. 

 Even though it be necessary to guard againt rash or 

 premature conclusions, and though the hypothesis by 



