286 PAPERS FROM THE DEPARTMENT OF MARINE BIOLOGY. 



for a comprehensive discussion of them, but they seem to represent a distinc- 

 tive class. 



{b) Lagoon deposits, some of which are essentially like the fine-grained 

 muds forming on the west side of Andros Island. Sizing indicates three 

 grades of these deposits; the finest corresponds to the Andros Island muds; 

 the next grade is represented by the surface deposits in Marquesas lagoon 

 (clay range from 14.3 per cent to 19.9 per cent; silt range from 9.7 per cent to 

 33.3 per cent); the specimens from Tortugas Lagoon average less clay than 

 those from the Marquesas, but there is overlapping of the grades. The 

 MgCOa percentage of the Tortugas material (3.86) is somewhat higher than 

 in the other specimens, a fact probably to be accounted for by the abundance 

 of foraminferal shells and alcyonarian spicules. 



{c) The deposit in 60 fathoms, south of Sand Key light, is intermediate 

 in character between a Globigerina ooze and a shoal-water deposit. It rela- 

 tively is coarse grained, has a high (5.03 per cent) MgCOs content, and con- 

 tains both bottom-living and pelagic foraminifera. 



Specimens of sands from behind the reefs have been collected, but will 

 not now be described. The classes of deposits are closely similar to those 

 recognized in the Bahamas. 



(2) As compared with other agents, corals are subordinate as extractors 

 of CaCOs from the sea-water; but a more accurate evaluation of the work of 

 the different agents must wait until the completion of the study of the com- 

 position of the samples from the standpoint of the source of its ingredients. 



(3) The remarks on the precipitation and possible solution of CaCOa 

 made on pages 265-268 apply to Florida as well as to the Bahamas. 



(4) The Si02 content of the Miami oolite (8.19 per cent) is strikingly 

 different from that of the Bahaman oolites and bottom deposit. The silica 

 is clear sand, around which oolite grains have often formed. The oolite on 

 Boca Grande Key, however, contains only 0.03 per cent Si02, showing that 

 in Pleistocene time, during the formation of the latter oolite, terrigenous 

 material did not reach so far westward. The beach sand at Sands Key 

 contains 1.15 per cent insoluble matter; the Marquesas Lagoon sample 

 (No. 98) 1. 1 8 per cent Si02. Except in Biscayne Bay and the sounds just 

 southward, next the shore, there is almost no sand in the bottom deposits 

 along the east coast of southern Florida and there is none in the other parts 

 of the key and reef area.^ The amount of (Al, Fe)203 ranges from 0.21 to 

 0.42 per cent in the Pleistocene oolites; it is 0.37 per cent in the Marquesas 

 sample, No. 98 — somewhat more than in similar deposits in the Bahamas. 

 The waters of the Florida reef tract afford a superb example of almost pure 

 limestone forming near a land-mass which is of low relief and across whose 

 surface no large streams flow. Ca3P208 is absent or is represented only by 

 traces. CaS04 ranges from traces to 0.50 per cent. The relations of the 

 latter ingredients are as in the Bahamas. 



'See Carnegie Inst. Wash. Pub. 133, pp. 1 14-129, 1910. 



