GLACIAL-CONTROL THEORY OF CORAL REEFS. 249 



enough for the marine abrasion assumed. A second important ground 

 for scepticism regarding their views is the depth of water on the plat- 

 forms, measuring so commonly 50 m. to 90 m. or more. The time 

 necessary for the sea to cut benches so far below its surface must be 

 so greatly extended that one's faith in the thesis is handicapped. The 

 difficulty is the more portentous because the older statement of the 

 abrasion hypothesis expressly or tacitly assumed the abrasion to have 

 occurred when the sea had its present level, and also that the earth's 

 crust, throughout nearly all of each coral-reef region, remained essen- 

 tially stable during the immense time implied in the cutting of very 

 wide benches at depths of from 50 to 90 m. Neither of these difficul- 

 ties exists for the Glacial-control theory. 



The subsidence theory attained its popularity because it explained 

 the surface topography of the reef-bearing archipelagoes so perfectly. 

 Just as truly does it fail to explain the submarine topography, as 

 Wharton and others have recognized. That theory has never been 

 presented in a thorough, qua7ititative way, account being taken of 

 the ivholc topography of each coral-crowned plateau. 



Perhaps no single fact more signally favors the Glacial-control 

 theory or more discredits the subsidence theory, than the observed 

 break of slope between reef and lagoon floor, which so commonly lies 

 50 to 90 m. below present sea-level. Compared to that elementary 

 physiographic fact, the criterion of the drowned valleys found in some 

 central islands must be considered as not so powerful in testing 

 theories. 



Only less significant is the general accordance in depth of the sub- 

 marine shelves outside the coral seas with the broader lagoon floors 

 and with the rimless banks in the coral seas themselves. Conviction 

 that most of these flats were developed by common agencies, in a 

 recent geological period, is hard to resist. No published statement of 

 the subsidence theory accounts for the accordance and no reasonable 

 appendix to that theory, sufficient in explanation, is in sight. 



Of special importance, too, is the topography of coral islands which 

 have been uplifted in post-Pliocene time. In apparently every case 

 the amount of uplift is directly proportional to a shallowing of the reef 

 lagoons. If the uplift has been as much as 60 m., the lagoon floor 

 has generally become dry land or a true lake. Within, or outside of, 

 the coral seas, coasts recently uplifted 60 m. or more, lack well-defined 

 shelves at the standard depth of from 35 to 90 m. The writer has 

 been able to find no exceptions to the rules just stated, but this excel- 

 lent test, so far affirmative of the Glacial-control theory, needs further 

 investigation. 



