532 W. M. DAVIS SUBSIDENCE OF REEF-ENCIRCLED ISLANDS 



lowered Glacial ocean; but the former depth of 50 or more fathoms, at 

 the time when the elevated fringing reef was forming, can not be safely 

 explained without subsidence. 



On the other hand, even with the aid of subsidence, it may seem diflS- 

 cult to explain the vast Seychelles bank as a drowned and partly rebuilt 

 almost-atoll, for its dimensions are far greater than those of any known 

 sealevel atoll. The difficulty here is, however, more apparent than real, 

 for while the time demanded for the production of a bank by edgewise 

 abrasion increases more rapidly than the increase of its diameter, the 

 aggradation of an atoll lagoon by locally formed sediments, such as are 

 provided by floating foraminifera and bottom nuUipores, is independent 

 of the diameter, inasmuch as the agency here grows with the area. Hence 

 if this bank be a drowned almost-atoll, it may have as small a propor- 

 tion of coral-reef limestone to lagoon limestone as Vaughan has found 

 in the reef and lagoon area of southern Florida. The upshot of all this 

 is that in so far as the Seychelles bank may speak for its neighbors, the 

 submarine banks of the Indian Ocean have suffered changes of level in 

 late geological time; and this is more consistent with Darwin's theory 

 of coral reefs than with any other. 



RECEXT ORIGIN OF GREAT OCEAN DEPTHS 



In view of these various considerations one is led to think that, as 

 Suess has suggested, the deep mediterranean seas, of which the China 

 Sea is a typical example, resemble lofty mountains in resulting from 

 comparatively recent deformation. Certain members of the Philippines 

 as well as the bottom of the neighboring China Sea basin may therefore 

 be regarded as having, on the whole, subsided so much, so rapidly, and 

 so recently that reef upgrowth could not keep pace with their submer- 

 gence. They are therefore characterized rather by imperfectly rimmed 

 submarine banks and by narrow fringing reefs of a new generation than 

 by barrier reefs and atolls. I believe the same statement may be made 

 regarding various other parts of the Australasian archipelago, not that 

 subsidence alone has taken place in this great region, for elevated reefs 

 occur on various islands up to altitudes of 1,000 or 2,000 feet; but that 

 diverse and rapid downward movements of the sea borders and basins, 

 often associated with corresponding upward movements in the larger 

 islands, to which Molengraaff (1916) and Abendanon (1917) have lately 

 given much emphasis, result in the present prevalence of fringing reefs 

 here, in association with submerged barrier reefs and atolls, while sea- 

 level atolls and barrier reefs prevail in the quieter regions of the open 

 Pacific, where submerged atolls are almost unknown. It is certainly 



