156 THE ANTARCTIC. 



latitude 6i' to 63 \' S., they fall, on closer observation, into 

 three groups. The eastern one, separated by a striking 

 distance from the central group, clearly represents the 

 link with the South Orkneys, while the western sub- 

 division is not widely separated from the middle islands by 

 the much narrower Boyd Straits. In common with the 

 chain formed by the South Orkneys, and corresponding to 

 them, they have a distinct direction : first, in the eastern 

 group, from east to west ; and then, in the central group, 

 from east north-east to west south-west. At a trifling- 

 distance from these on the south side are three islands 

 with an essentially different structure, while the western 

 subdivision consists of two islands, of which the southern 

 one is a definite continuation of the central group, and 

 the northern one projects to the north. 



The eastern group consists of two larger islands, 

 Clarence Island and Elephant Island, so called from 

 the great number of sea elephants formerly found there. 

 Between the two lies Cornwallis Island, somewhat to the 

 north, and at the southern end of Elephant Island, though 

 at some distance, lie the four islets Narrow Island, Gibb 

 Island, Aspland Island and O'Brien Island. Clarence 

 Island, which is approximately circular in form, lies on 

 parallel 6i° 16' S. and meridian 54° io' W. Though of 

 no great size (30 miles are mentioned as its circumference 

 and about eleven miles as its greatest length) it attains a 

 considerable elevation. This is estimated at 4,300 feet, 

 rising abruptly from the sea and forming several steep 

 summits, of which the most considerable seems to be in 

 the west. The whole island is covered with snow, with 

 the exception of the highest points and the precipitous 

 descent on the coast, where the angle is too great to 

 afford a support to the snow, and glaciers descend on 

 all sides. Nothing is reported of any landing on the 

 island, and for this reason nothing is known of its 

 geological structure ; it is possible to conjecture that 



