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



provide strong testimony against certain essential features of the Glacial- 

 control theory, which is today the only serious competitor of the theory 

 of subsidence. All other modern theories, of which I have elsewhere given 

 a somewhat detailed review (1914), are discredited by their neglect of 

 the submergence that is demanded by embayments and unconformities. 

 The principles involved in the Glacial-control theory were briefly stated 

 in an earlier paragraph, but will now be considered more deliberately. 



According to the Glacial-control theory, which assumes reef -encircled 

 islands to be stationary as a rule, there might have been narrow pre- 

 glacial fringing reefs on the Philippine Islands where the shores pitch 

 down steeply into deep water; and if the corals on the exterior slopes of 

 such reefs were killed by the cooled waters of the lowered Glacial ocean, 

 the reefs should have been cut away by the waves. Now if the abrasion 

 then operative endured long enough to cut do^vn the large hypothetical, 

 still-standing volcanic island in the center of the China Sea, at present 

 represented by the Macclesfield bank, which measures 95 by 35 nautical 

 miles, as is also assumed by the Glacial-control theory, then a compara- 

 tively broad platform backed by strong cliffs should have been abraded on 

 all the exposed island slopes of the Philippines ; and when the ocean rose 

 to its normal level again in postglacial time, a barrier reef should have 

 grown up on the outer margin of the abraded platform, and the tops of 

 the cliffs should still be visible above sealevel. 



On the other hand and as a matter of fact, barrier reefs and high shore 

 cliffs with submerged bases are hardly known in the Philippines. Sub- 

 marine platforms of various breadths up to 30 miles, and at various 

 depths from 20 to 60 fathoms, are truly enough found bordering some of 

 the islands, but the island shores are nevertheless characterized by unclift 

 promontories, sometimes without reefs, as along the mid-west coast of 

 Palawan, figure 10, but usually margined with unconformable fringing 

 reefs from 500 to 5,000 feet in breadth, and indented by drowned valley 

 embayments more or less filled with deltas. On a number of islands the 

 fringing reefs are so narrow and the bayhead deltas are so small that the 

 submergence which lowered these islands to their present altitude must 

 be of more recent date than that which lowered other islands, where 

 the fringing reefs are already a mile or two wide and the deltas are well 

 developed, and surely of much more recent date than the submergence 

 which permitted the formation of the great fringing reefs, a mile or two 

 broad, on Yap, in the Caroline group of the North Pacific, or on Eodri- 

 guez, a lonesome island in the Indian Ocean; more recent also than the 

 submergence which determined the formation of the broad barrier reefs 

 of Mbengha in Fiji, of Borabora in the Society Islands, and of the Great 



