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



student of the coral-reef problem, and especially by those who question 

 the correctness of the Glacial-control theory, into which the change of 

 ocean level enters as an important factor. While recognizing the pro- 

 priety of this opinion, I have excused myself from the labor involved in 

 adopting it, because an independent observational test of the general 

 value of ocean lowering seems more desirable than a numerical recalcula- 

 tion of its measure. The desired test is found in the depth of several 

 submarine banks, from the central part of which reefiess and c'lift islands 

 rise. 



The survival of the clif t central islands proves that the banks are in part 

 at least due to abrasion, and that the truncation of the islands was incom- 

 plete. Hence the depth of the rock-floor in such banks at the base of their 

 cliffs would give a good indication of the position of ocean level when 

 their abrasion took place. The rock-floor depth can not be directly deter- 

 mined, because some aggradation of the banks has been accomplished 

 since abrasion ceased. Aggradation from the surviving islands can not, 

 however, be of great amount, because the cliffs still plunge down below 

 the sea surface and are as yet little benched at present sealevel. Whatever 

 aggradation has taken place ought to be less and not greater than the 

 aggradation of reef-rimmed banks, where organic sediments are supplied 

 in relative abundance and where the supplied sediments are largely re- 

 tained on the saucer-shaped surface. 



jSTow if we examine charts of the banks around Tutuila (Samoa), and 

 especially around the Marquesas Islands, the clift spur ends of which are 

 reef -free though the islands lie near the equator in the eastern Pacific, and 

 also around a number of small and clift residual volcanic islands on the 

 border of the coral seas, such as certain northwestern members of the 

 Hawaiian group, and Norfolk Island between Australia and New Zealand, 

 the depth of the inner, cliff -base margin of the banks is found to be about 

 20 fathoms, although their outer border may be twice as deep. The aggra- 

 dation of the extratropical banks, where coral growth is scanty or want- 

 ing, is probably less than that of the intertropical banks. Hence if the 

 extratropical banks were cut during the lower stand of the Glacial ocean, 

 the measure of 30 or 40 fathoms given by Daly as the amount of ocean 

 lowering may be taken as liberal, and certainly as not erring in the way 

 of being too small. 



Several interesting inferences follow. First, it does not seem possible 

 that the removal of aggrading sediments would bring about an agreement 

 between the smaller depths of the aggraded surface over the undoubted 

 rock platforms around the clift residual islands of the extratropical seas, 

 and the greater depths over the center of the very hypothetical rock plat- 



