192 THE OCEAN WORLD. 



In order to give an idea of the general form of these atolls, 

 although they are rarely so regular, the reader is referred to Plate IV., 

 which represents one of these islands of the Pomotouan Archipelago, 

 in the Indian Ocean. It represents the island of Clermont-Tonnerre, 

 figured by Captain Wilkes in the American Exploring Expedition. 

 The exterior girdle of rocks here surrounds a basin nearly circular. 

 Such is the general form — the typical form, so to speak — of the coral 

 islands, of which this is a fair representation. 



The animals which form these immense accumulations belong to 

 diverse groups, and nowhere have the results of observations made 

 upon these atolls been more minutely described than in Mr. Darwin's 

 remarks on the grand Cocos Island situated to the south of Sumatra, 

 in the Indian Ocean. 



No writer, it seems to us, has reasoned on these atolls more com- 

 prehensively than the author of the "Origin of Species." "The 

 earlier voyagers," he says, " fancied that the coral-building animals 

 instinctively built up their great corals to afford themselves protection 

 in the inner parts ; but so far is this from the truth, that those massive 

 kinds, to whose growth on the exposed outer shores the very exist- 

 ence of the reef depends, cannot live within the lagoon, where other 

 delicately-branching kinds flourish. Moreover, in this view, many 

 species of distinct genera and families are supposed to combine for 

 one end ; and of such a combination not a single instance can be 

 found in the whole of Nature. The theory that has been most 

 generally received is, that atolls are based on submarine craters, but 

 when the form and size of some of them are considered, this idea 

 loses its plausible character. Thus, the Suadiva atoll is forty-four 

 geographical miles in diameter in one line by thirty-four in another ; 

 Rimsky is fifty-four by twenty miles across ; Bow atoll is thirty miles 

 long, and, on an average, six miles broad. This theory, moreover, 

 is totally inapplicable to the Northern Maldivian atolls in the Indian 

 Ocean, one of which is eighty-eight miles in length, and between ten 

 and twenty in breadth." 



The various theories which had been propounded failing to 

 explain the existence of the coral islands, Mr. Darwin was led to 

 reconsider the whole subject. Numerous soundings taken all round 

 the Cocos atoll showed that at ten fathoms the prepared tallow in the 

 hollow of the sounding rod came up perfectly clean, and marked 

 with the impression of living polyps. As the depth increased, these , 

 impressions became less numerous, but adhering particles of sand 

 succeeded, until it was evident that the bottom consisted of smooth 

 sand. From these observations, it was obvious to him that the 



