262 W. M. Davis — Shaler Memorial Study of Coral Reefs. 



bottom suffers a compensating uplift, the general sea level may 

 neither rise nor fall, except for the small changes mentioned 

 above. Thus the submergence of the island shores may be 

 about equal to the local subsidence of the ocean bottom, as 

 Darwin supposed. If the area of compensating uplift is in 

 part or in whole continental, the submergence will be some- 

 what less than the subsidence ; for example, if the subsiding 

 area is one-tenth of the ocean bottom, and the amount of sub- 

 sidence is 500 feet, while the compensating uplift is entirely 

 within a continent, the amount of submergence will be 450 

 feet. Hence, in so far as an economical method is a probable 

 method, Darwin's theory of regional subsidence of the ocean 

 bottom is a more probable cause of the submergence indicated 

 by barrier-reef islands than the alternative theory of a rise of 

 sea level as a result of an uplift of the ocean bottom elsewhere; 

 but such probability is only a first step toward proof. In my 

 own opinion final proof can not be reached in a problem of the 

 kind here discussed until we make undreamed of additions to 

 our present knowledge of the earth ; the most that can now be 

 done is, as above, to reach a fair degree of probability. A 

 second step in that direction may be made by considering the 

 diversity in the measure and the date of the submergences 

 which various reef-encircled islands have suffered in different 

 island groups and even within the same group. 



In considering this aspect of the problem we will return for 

 a moment to the supposition that the submergences indicated 

 by the embayments of barrier-reef islands and by the uncon- 

 formable contact of elevated reefs with their eroded founda- 

 tions are due to elevations of the ocean surface caused by 

 uplifts of the ocean bottom elsewhere ; and we will recall that 

 when a reef-island is thus submerged, all other islands and all 

 the continental coasts of the whole world must, as already 

 stated, suffer submergence of the same date, amount and rate, 

 except in so far as local movements cause departures from this 

 improbable uniformity. Local differences in amount and rate 

 of submergence in the coral-reef region may be produced if it 

 shares in varying measure the ocean-bottom uplift that causes 

 the rise of the ocean surface, or the compensating subsidence 

 that lessens the rise; but differences in the geological date of 

 submergence can be caused only by different submergences ; 

 and here the theory of a rising ocean surface encounters a seri- 

 ous difficulty. All the islands and all the continental coasts 

 that did not take part in an earlier submergence, during which 

 a now elevated coral reef was formed, must at that time have 

 been uplifted by the amount of the submergence ; later on, all 

 the islands and all the continental coasts that do not take part 

 in the submergence in association with which a new sea-level 



