118 THE TORTUGAS AND FLORIDA REEFS. 



by the currents, would explain a gradually increasing elevation of the surface till the 

 level was reached at which reef-building corals can flourish, and at which a reef would 

 naturally be formed. Darwin has noted the close resemblance between encircling 

 barrier reefs and atolls. It seems to me that the structure of the Marquesas 1 (Plate V.) 

 and of Alacran proves conclusively that not one point of difference exists between a 

 barrier reef and an atoll. Darwin has also called attention to the fact, that in shal- 

 low seas, such as the Persian Gulf and parts of the East Indian Archipelago, the reefs 

 lose their fringing character and appear as irregularly scattered patches, often covering 

 a considerable area; and observes that many reefs of the West Indies have been 

 formed in like manner upon large and level banks lying a little beneath the surface, 

 banks which he believes to have been formed by the accumulation of sediment. 

 Such patches of reef-building corals would seem, from their analogy with the Tor- 

 tugas, to be the beginning of more extensive reefs. 



Judging from my 



of the Tortuiras reefs, it would seem that corals do 



not thrive below a depth of from six to seven fathoms. It is, of course, impossible to 

 determine whether that is their bathymetrical limit, or whether they are killed from 

 the accumulation of ooze in the channels and adjacent slopes. We find them confined, 

 however, to the same shallow depths, along the whole of the main reef to the north- 

 ward (Agassiz). Captain Moresby also showed that at a depth of ten fathoms in the 

 Maldive and Chagos archipelagoes the masses of living coral are scattered at greater 

 distances separated by patches of smooth white sand, and that at a slightly lower 

 depth even these patches merge into a smooth, steep slope wholly bare of coral. All 

 the evidence accumulated by Dana, Ehrenberg, Quoy and Gaimard, tends to show 

 that the limit of reef-building corals is found at about twenty fathoms. On the 

 Yucatan, as on the Florida Bank, the conditions favorable for coral reef growth have 



1 The plan and section of the Marquesas Keys on Plate V. show the formation of the keys on a knoll rising from the general 



platform 



ling 



and the corals were killed from the accumulation of silt upon them. The filling up of a lagoon like that of the Marquesas must 

 be a slow process, for we find the water of the inner lagoon deeper than that of any other part of the reef immediately surrounding 

 the outer slope. We can imagine that when the outer ring of the reef surrounding the inside lagoon is once connected, or nearly 

 so, the enclosed calm area is so placed as to be subject to but few disturbing agencies, and is practically excluded from receiving 



sediment 



passages form 



current through the narrow 



will 



forms 



sive mud flats to the northward of the keys. This mud is, however, generally covered by a thin dark-colored layer of decom 



posed 



The Marquesas are covered by a thick growth of mangroves. 



