CORAL FORMATIONS. 623 



The barrier sometimes becomes wooded for long distances, like the 

 reef of an atoll ; but the wooded portion, when there is any at all, is 

 usually confined to a few islets. 



The barrier and fringing reefs are formed precisely like the atoll 

 reefs ; and special explanations are needless. 



The absence of reefs from parts of coasts of islands, within coral-reef seas, is due to 

 several causes: (1) to the depth of water, for corals fail if the depth exceeds one hun- 

 dred feet; (2) to fresh-water streams, especially if bringing in detritus, which destroys; 

 the living corals ; as such fresh waters flow over the surface of the salt, they do not pre- 

 vent the corals from growing below, unless impure with detritus; (3) tidal and other 

 currents, which keep passages open, by means of the detritus they often bear along 

 their course. These are the principal causes that prevent the harbors from becoming 

 filled with corals and thereby destroyed. 



The growth of the different parts of a reef, or its prolongation in one direction or an- 

 other, depends much on the tidal and other currents that sweep through the channel or 

 by the side of the island. As in the case of silt along other sea-shores, the coral detri- 

 tus made by the waves is distributed by these currents: and hence the increase of a 

 reef is not dependent solely on the number of growing corals over its surface, or their 

 kinds. 



Breadth of Reefs. — The reefs adjoining lands have sometimes great 

 width. On the north side of the Feejees, the reef-grounds are five to 

 fifteen miles in width. In New Caledonia, they extend one hundred 

 and fifty miles north of the island, and fifty south, making a total 

 length of four hundred miles. Along northeastern Australia, they 

 stretch on, although with many interruptions, for one thousand miles, 

 and often at a distance, as just stated, of fifty or sixty miles from the 

 coast, with a depth between of fifty or sixty fathoms. But the reefs, 

 as they appear at the surface, even over the widest reef-grounds, are 

 in patches, seldom over a mile or two broad. The patches of a single 

 reef-ground are, however, connected by the coral basement beneath 

 them, which is struck, in sounding, at a depth usually of ten to forty 

 or fifty feet. 



The transition in the inner channels, from a bottom of coral detritus 

 to one of common mud or earth, derived from the hills of the encir- 

 cled island, is often very abrupt. Streams from the land bring in this 

 mud, and distribute it according to their courses through the channels. 



Tldckness of Reefs. — The thickness of a coral formation is often 

 very great. From soundings within a short distance of coral islands, 

 it is certain that this thickness is in some cases thousands of feet. 

 Within three-quarters of a mile of Clermont Tonnerre, in a sounding 

 made by Hudson, the lead struck and brought up an instant at two 

 thousand feet, and then fell off and ran out to three thousand six hun- 

 dred feet, without finding bottom ; and seven miles from the same 

 island, no bottom was found at six thousand feet. 



The barrier reefs remote from an island must stand in deep water. 



