258 VOYAGE TO THE 



chap. All the points or angles of these islands descend 

 vj^w into the sea with less abruptness than the sides, and, 

 Fe 2 b - I think, with more regularity. The wedge-shaped 

 space that the meeting of the two sides would form 

 in the lagoon is filled up by the ledges there being 

 broader; in such places, as well as in the narrow 

 parts of the lake, the coralline are in greater num- 

 bers, though, generally speaking, all the lagoons are 

 more or less incumbered with them. They appear 

 to arise to the surface in the form of a truncated 

 cone, and then, their progress being arrested, they 

 work laterally, so that if several of them were near 

 each other they would unite and form a shelf simi- 

 lar to that which has been described round the mar- 

 gins of some of the lagoons. 



The depth of these lagoons is various: in those 

 which we entered it was from twenty to thirty-eight 

 fathoms, but in others, to which we had no access, by 

 the light-blue colour of the water it appeared to be 

 very small. It is, however, tolerably certain that 

 the coral forms the bases of them, and consequently, 

 unless depositions of sand or other substances, ob- 

 noxious to the coral insects, take place, their depth 

 must depend upon their age. 



Very little offered itself to our notice, by which 

 we could judge of the rapidity of the growth of the 

 coral, as the islands which we examined had ne- 

 ver been described with the accuracy necessary for 

 this purpose ; and there were, consequently, no 

 means of comparing the state in which they were 

 found by us, with that which was presented to our 

 predecessors ; but from the report of the natives, the 

 coral bordering the volcanic islands does not increase 

 very fast, as we never heard of any channels being 



