114 THE TORTUGAS AND FLORIDA REEFS. 



have also the still older phenomenon of such islands as Barbados, where the ter- 

 races formed by the raised coral reefs mark the successive elevation of the volcanic 

 cone ; or we may have still another combination, like that of Guadeloupe, where a 

 high volcanic peak forms the main island, an elevated plateau forming the Basse Terra 

 with a growing coral reef to the windward of the latter. 



The fact that these great submarine banks of modern limestone lie in the very 

 track of the great oceanic currents sufficiently shows that these currents hold the 

 immense quantity of carbonate of lime needed in the growth of the bank. Its 

 amount has, besides, been actually measured by Murra3^ He has shown that, if the 

 pelagic fauna and flora extend, as the experiments carried on by the " Challenger " 

 and the " Blake " seem conclusively to prove, to a depth of one hundred fathoms, we 

 should have sixteen tons of carbonate of lime for every square mile one hundred 

 fathoms deep. But the greater the depth at which these plateaux begin to form, 

 the less rapid must be their formation. The fact that the deeper part of the ocean, 

 below three thousand fathoms, does not contain any of the larger shells of pelagic 

 type, can be readily explained on the supposition that, being thinnest, like Pteropod 

 shells, they pi'esent a large surface to the action of the carbonic acid, which is most 

 abundant in deep water. Attacked as soon as they reach the bottom by this action, 

 these shells of thinnest surface are reduced to a bicarbonate, and are carried off in 

 solution. They do not, therefore, appear at these greater depths, and are indeed 

 rarely to be found below two thousand fathoms. The thicker-sheUed Foraminifera 

 extend to a greater depth, not because they are of different chemical composition, but 

 because their greater amount of substance yields less easily to the action of the acid. 

 At shallower depths the solvent action of carbonic acid must be far less efficient, on 

 account of the rapid accumulation of dead siUceous and calcareous shells of Forami- 

 nifera, Sponges, Hydroids, Corals, Halcyonoids, Mollusks, Polyzoa, Echinoderms, etc., 

 which must long have lived upon the bank before they had by their accumulation 

 brought it to a level at which coral reefs could begin to grow. 



The bathymetrical sections (Plate VII.) of the peninsula of Florida to the east- 

 ward into the trough of the Gulf Stream are very different from those taken on the 

 western side of the peninsula. Proceeding northward from Cape Florida, you pass 

 out of the action of the current from the Straits of Bernini, where the velocity of 

 the Gulf Stream is the greatest. As soon as you reach a latitude at wdiich the 

 trade winds do not blow, you come gradually upon the usual comparatively gentle 

 slope off shore, showing less traces of disturbance either from currents or from the 

 action of the prevailing winds. Judging from what I have seen of the east coast 



