AFRO-INDIAN REGIONS. 



249 



In coasting around in a boat with Captain Wilkes, the cliff seemed 

 to continue vertical from a vast depth in the ocean ; but approach- 

 ing the surf as near as consistent with prudence, we were enabled to 

 distinguish living coral through the clear water, and to obtain sound- 

 ings; evidently on fallen portions of the cliff. That the mid-ocean 

 coral-islands are margined by vast submarine precipices, had been 

 already proved by attempts at sounding from the Peacock : "About a 

 mile from one of the coral-islands we first visited, Captain Hudson let 

 down eleven hundred fathoms of line, without reaching the bottom." 

 On second trial at a different spot, " after striking the bottom at three 

 hundred fathoms, the lead slipped off, and descended as many fathoms 

 more without striking; but the line brought up an entangled fragment 

 of coral ; giving evidence, that the strike was on the edge of a vertical 

 or overhanging precipice." 



As the boat kept close under Metia, the face of the cliff presented 

 marks left by portions that had fallen off; many of them very exten- 

 sive, but the fallen masses however huge, had disappeared beyond the 

 reach of the lead in the depths of the ocean. Other portions of the 

 cliff were evidently on the point of falling. In the absence of all means 

 of access, the island seemed a fortress that might dety the world : but 

 continuing around, we at length arrived at a slight strip of beach at the 

 base of the cliff, affording a landing-place. Another accessible spot 

 was found in a different part of the island; and Mr. Agate, landing 

 there from one of our boats, climbed a steep ascent consisting of a 

 succession of ledges, and succeeded in reaching the summit. 



The natives on the beach above-mentioned, received us in a friendly 

 manner; and we were conducted by a pass, part of the way rendered 

 practicable by art, to the top of the island. From the brink, there 

 was a very gradual rise for the first hundred yards, and then a per- 

 ceptible descent ; and at the end of about a mile, the trees being more 

 thinly scattered, we gained a somewhat extended view over what 

 seemed a slight central depression; perhaps only a general inclination 

 towards the opposite margin, which was concealed by trees. For it 

 was soon perceived, that the top of the island was not level ; the end on 

 which we landed being thirty or forty feet higher than the farther one. 



The island consisted exclusively of coral ; the component species 

 identical with those of the low coral-islands, all made up in like 

 manner principally of conglomerated " Astreas." Metia is therefore 

 a low coral-island, that by some local geological cause, has been pushed 



63 



■ 



