ATOLL AND REEF FORMATION 217 



As tlie theory of Subsidence ushers in the modern epoch 

 of coral-reef theories, it will be best to explain, before these 

 views are discussed, what problems the question offers, and 

 what details must be explained by a theory that is to be a 

 useful one. 



In order to save wearying repetition, these problems will be 

 briefly taken in order, and the merits and the shortcomings of 

 the various theories will be measured for each individual 

 problem. In every case the consideration of the subject falls 

 naturally into three divisions : the views held before Darwin's 

 day, Darwin's work itself, and the theories put forward in 

 opposition to Darwin's representing three distinct epochs in 

 the evolution of our ideas of coral-island formation. 



The First Point that requires an explanation is the 

 nature and origin of the submarine bank on which the atoll 

 rests : — i.e., on what structure did the corals first start to 

 build the islands ? The question of the nature of the " basis " 

 assumes great importance from the fact that the depth of water 

 in which the reef-forming corals can grow is strictly limited. 

 The basis of every atoll must therefore be an elevation, rising 

 so near to the surface that reef-building corals may flourish 

 on it. 



This fact consequently presupposes the existence of a very 

 large number of submarine mountains, dotted all over the 

 Indian and Pacific Oceans, and all Avith their peaks approxi- 

 mately of the same height. 



Now no terrestrial formation such as this is known to exist ; 

 and it was this fact that led Darwin— rather than assume 

 such an apparently impossible condition of submarine con- 

 figuration — to bring forward his well-known theory of subsi- 

 dence. But it must always be borne in mind that he 

 admitted that, were such banks at hand, they could furnish 

 the basis for atolls. 



To Semper he wrote (Oct. 2, 1879) : " I always foresaw that 

 a bank at the proper depth beneath the surface would give 

 rise to a reef which could not be distinguished from an atoll 

 formed during subsidence " ; and again to Alexander Agassiz 



