268 W. M, Davis — Skaler Memorial Study of Coral Reefs. 



tional changes due to the glacial period, uniform through all 

 the torrid seas, being superposed on the long-lasting changes 

 due to regional subsidence, which presumably varied in rate 

 and in amount from place to place. The effect of this com- 

 bination will be to retard submergence while the continental 

 ice sheets are forming, and to accelerate it while they are 

 melting away. During the time of retarded submergence, reefs 

 would tend to broaden by outward growth, and lagoons would 

 tend to be shallowed and filled by outwash from the land, by 

 local organic deposits, and by inwash from the reefs. Atolls 

 might thus be transformed into cays, and barrier-reefs into sea- 

 level flats. Then during the following time of accelerated 

 submergence, the reefs, especially while their corals are still 

 feeble because of a slow warming of the cooled ocean of a 

 glacial epoch, might not be built upwards in all their length 

 as fast as submergence is then taking place ; thus a platform, 

 incompletely bordered by a reef, which need not necessarily 

 rise from the platform edge, would result ; eventually a com- 

 plete reef might be established again, but that eventuality 

 may not yet be reached in all cases. Before these possi- 

 bilities can be regarded as probabilities, much study must be 

 given to them. The most doubtful element in the case is the 

 rate of subsidence and its relation to the rate of change of sea 

 level by glacial control. 



If instead of subsiding islands on a sinking ocean floor, we 

 consider stationary islands in an ocea,n that rises because of an 

 uplift in its floor elsewhere, the same results will follow from 

 combining therewith the changes of ocean level caused by 

 glaciation : or similar results may follow from various combi- 

 nations of all three processes : — local subsidence, changes of 

 ocean level due to climatic changes, and changes of ocean 

 level due to movements of the ocean bottom in other than 

 coral-reef regions. 



It thus appears that oscillations of sea level during the 

 glacial period may work most harmoniously with Darwin's 

 theory of subsidence, and produce significant though subordi- 

 nate modifications in its consequences ; hence to suppose that 

 glacially controlled oscillations of sea level have acted alone 

 during the formation of existing coral reefs seems to me 

 improbable as well as inadequate. It is true that in Daly's 

 exposition of the glacial theory, he announces that it " in no 

 sense excludes complications due to local warpings of the 

 earth's crust," but he adds that changes of this kind are 

 neglected in his argument because of the high probability that 

 they have not " been important enough, since the Tertiary, to 

 affect, by more than a fathom or two, the changes of level 

 assumed " (307). The reasons for this conclusion are not 

 stated, but they are perhaps associated with theoretical views 



