1909.] THEORY OF ATOLL FORMATION. 677 



claimed that the bathymetrical limit of the reef-building corals is 

 intimately associated, if not coincident, with the limiting line of 

 sedimentation, and that it is therefore a variable plane depending 

 on the local conditions of the sea. The reasons for this coincidence 

 are to be found in the study of the living corals themselves ; 

 and I have come to the conclusion that the presence of matter 

 suspended in the water is the most ^^otent factor in determining 

 the unsuitability of an environment for coral life. Where sedi- 

 ment is at all times liable to fall upon the living zooids, reef-corals 

 will not flourish : we would therefore not look for their luxuriant 

 presence below the limiting line of sedimentation. In the wave- 

 stirred area above this line, however, they can and do flourish. 

 We therefore arrive at the pi-esumption that sediment can build 

 banks up to this hypothetical line, and reef-corals can build banks 

 from this line up to the surface of the sea. There is therefore no 

 reason why coral colonies should not settle upon the bank and 

 start the development of a reef. As a matter of fact several other 

 forms of life that possess calcareous skeletons outrun the reef- 

 corals in bathymetrical range, and it is likely that they (calcareous 

 algje, deep water corals, &c.) first populate the summit of the 

 bank. 



The process now becomes less a matter for hypothesis and more 

 one for actual observation, for the growth tendencies of reefs and 

 of colonies may be more easily studied. It i-s claimed that the 

 tendency is for such reefs to become " basin-shaped reefs," and 

 to develop as flat banks, with edges raised from their general 

 surface and abundantly covered with coral colonies. The chief 

 factor in this process is again the action of sedimentation. The 

 surface waters still di'op their burden of suspended matter over 

 the reef, and it is deposited upon the uneven surface of the coral 

 colonies, for, though it could no longer come to rest upon the 

 open sedimentation bank, it more easily finds a lodgment upon 

 the broken coral surface of the reef. At the edges of the reef the 

 sediment becomes more easily washed ofi" by wave action, and the 

 corals of the circumference of the reef flourish most. 



To obtain a concrete picture of the process it is only necessary 

 to turn to the colonies to be found any day in quiet pools in which 

 sediment is accumulating. A colony of Porites grows as a 

 spherical mass. In time it develops to such a size that its rounded 

 upper surface becomes sufliciently flat to afibi-d a lodgment for 

 sediment. Then the activity of its central zooids wanes, and, by 

 the upgrowth of the peripheral ones, the flattening increases. At 

 length the central area dies — the zooids choked by sediment, — and 

 a raised ring of active living zooids surrounds a central depressed 

 area — an atoll in miniature. 



That this process is not due to the colony reaching tide-level 

 (Darwin, Semper) is proved by the abundant finding of such 

 colonies developed many feet below the level to which the tide 

 ever falls. 



The process that may be seen any day in the myriad colonies 



