CORAL REEFS. APPENDIX. • 231 



dition of very slow elevation or absolute rest without any con- 

 spicuous break " (p. 263). If we assume that this arrest of 

 movement, passing into a condition of absolute rest, had still 

 further progressed into one of subsidence, I think we will then 

 be able to understand all the special (and apparently antago- 

 nistic) features of the region to which Prof. Semper refers, and 

 in a manner much more satisfactory than is oflfered by the ex- 

 planation of the distinguished German naturalist. It seems 

 to me that the condition here is almost precisely what we find 

 in the Bermudas: a coral-made land, which had been elevated 

 to some little height above the sea, undergoing waste and de- 

 struction through subsidence. This phase in the history of 

 Bermuda is so clear that there can be.no question concerning 

 it. In the Bermudas we have also a near-lying reef on the one 

 side (likewise the weather side) and a far-off" reef on the other, 

 with an intermediate body of water of some 50-60 feet depth. 

 But Prof. Semper himself gives data which lead one to sup- 

 pose that subsidence has in fact taken place. The biting out 

 or undercutting of the limestone plateau to which he calls at- 

 tention (p. ^54), and which may be paralleled with the similar 

 process on the south shore of the Bermudas, surely argues 

 much more strongly in favor of subsidence than of elevation ; 

 it certainly seems impossible for a rock to be at the same 

 time building up and breaking down. But further. Prof. 

 Semper informs us that it " is certain that the enclosed island 

 of Babelthuap was formerly much broader than it now is " 

 (p. 270). This condition is scarcely compatible with anj' theory 

 of elevation, and more nearly accords with the assumption of 

 waste through oceanic encroaches permitted by subsidence. 

 Indeed, the author himself seems not to have been convinced 

 that there was no subsidence, since he asserts that " we are 

 obliged, under all the circumstances, to assume the co-opera- 

 tion of some other force besides subsidence when endeavoring 

 to explain the peculiar formation of the northern reef, but still 

 without wholly excluding the effects of subsidence " (p. 253). 



Prof Semper rejects the hypothesis that the lagoon-ring of 

 an atoll is formed through accelerated marginal growth (p. 

 227) ; he attributes the lagoon-basin to decay and scour. 



