216 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



reefs. Plates 27 and 28 are reproductions of two of Saville-Kent's 

 photographic illustrations of the Great Barrier Eeef of Australia. 

 These three i)lates illustrate true coral reefs, which in my opinion 

 should be defined as follows: Coral reefg are ridges or mounds of 

 limestone, the upper surfaces of Avhich lie, or lay at the time of 

 their formation, near the level of the sea, and are predominantly 

 composed of calcium carbonate secreted by organisms, of which the 

 most important are corals.^ 



The composition of Avhat I consider true coral reefs is very com- 

 plex. The main framework of the reef is formed by coral heads and 

 stout coral branches, while the interspaces are filled by small corals, 

 and the skeletons of other organisms, some of which in the course of 

 time are more or less broken up by the waves. In many cases it is 

 difficult to decide whether or no to apply the designation "coral 

 reef " to richly coralliferous deposits that are obviously bedded. 

 However, it seems to me that it should be applied wherever corals of 

 reef facies seem sufficiently abundant to have formed appreciable 

 rugosities on the sea bottom, although the deposits are bedded. Reefs 

 predominantly composed of the remains of calcareous algae should be 

 designated " nuUipore " or " Lithothamnion reefs." But, where the 

 proportion of these organisms to corals is so nearly the same that 

 only exact computation will decide between the two, such a reef may 

 be designated " coral." 



SOME KIXDS OF LIMESTONE THAT HAVE BEEN CONFUSED WITH CORAL- 

 REEF ROCK. 



To many it may seem superfluous in a definition of coral reefs to 

 say that the remains of corals should be an important constituent 

 of the rock ; but the term " coral rock " or " coral-reef rock " has 

 been repeatedly applied to limestone Avith the making of wdiich 

 corals have had either nothing, or practically nothing, to do. An 

 excellent instance of such a popular, and until recently scientific, 

 misconception is supplied by the Bahama Islands. 



According to Alexander Agassiz the Bahamas are composed of 

 wind-blown coral-sand. The sand composing the ridges in the Ba- 

 hamas, at least those I have seen on New Providence and Andros 

 Islands, has certainly been wdnd-blown. Plate 30 illustrates an 

 exposure along East Street in Nassau, and plate 29, figure B, rep- 

 resents the face of a small cliff at the south end of Morgan Bluff, 

 Andros Island, both in the Bahamas; while figure A of plate 29 

 is from a photogi'aph of a section of a sand dune at Cape Henry, 

 Virginia. These illustrations show the essential similaritv of the 



^ Vaughan, T. W., Physical conditions under which. Paleozoic coral reefs were formed : 

 Bull. Geol. Soc. America, vol. 22, p. 238, 1911. 



