26 FLORIDA REEFS. 



water mark, but lies several feet below that. The able engineer now build- 

 in o- a lighthouse on that very spot has provided against a similar misfortune 

 by uncovering the reef as the first step in his work. Mr. Dana has already 

 pointed out another fact of the utmost importance, with reference to the 

 construction of lighthouses. Even the lower and more solid mass of the 

 reef is not continuous throughout, but has cavities filled with sand, debris, 

 loose material of all kinds. Any point selected for the erection of a light- 

 house should, therefore, be carefully surveyed, and the solidity of the reef 

 should be tested by boring small holes throughout its thickness. If not 

 weakened by such cavities, it will afford a safe foundation for beacons, sig- 

 nals, or lighthouses. The reef-building corals are, indeed, not only the 

 largest, but also the most compact and hardest, and their order of succession 

 in their growth, with the closer, heavier kinds below and the lighter ones 

 above, give durability and strength to their masonry. The foundation of the 

 reef is built by the massive heads of Astrgea and Porites, in the middle range 

 Maeandrinas and Milleporas come in, and only above these do we find the 

 Madreporas, the stoutest and strongest of which form the uppermost growth 

 upon the reef of Florida. As if to give additional firmness to the structure, 

 the dying corals, as they reach the surface, become incrusted with Nullipores 

 of different kinds. One species forms a close coating of the hardest limestone, 

 not less than half an inch thick, over the dead corals. In the empty spaces 

 grow other more branching kinds, which gradually fill the small cavities and 

 act as a natural mortar cementing the whole. Upon this basis the broken, 

 floating materials are thrown up, forming that loose cap of sand and debris 

 which rests on the top of the reef and is liable to be carried away by storms. 

 The reef in its easternmost part north of Fowey Rocks, and in the 

 direction of Bearcut and Virginia Bay, converges toward Cape Florida. 

 Here it is covered by silicious drift sand, as are also the kej^s north of Cape 

 Florida. The character of these formations, evidently the work of corals 

 like other parts of the reef, favors the supposition that the longitudinal narrow 

 islands stretching in almost unbroken continuity along the eastern coast of 

 the peninsula are, like the Everglades upon the main-land, a prolongation 

 of the reef now covered by drift sand. 



ShijJ-Channel. 

 The broad channel extending the whole range of the reef, between 

 the main keys and the outer reef, is rather uniform, having the same 



