8u8 Dahj — l^leistocene Glacialion and Contl Reef Problem. 



iiient of the sinooth lioors, on which the existing reefs stand, 

 is not to be ascribed to Tertiary marine erosion. In fact, it is 

 an ojKMi question whether tlie atoll archipelaii'oes do not repre- 

 sent a highly exceptional set of conditions, sucl> as have 

 very seldo(n prevailed in tlie liistory of the globe. According 

 to the hypothesis here ])roposed, the essential conditions have 

 included a hollowing (negative) movement of sea-level with 

 subsequent heaping (positive) movement of sea-level, each 

 movement being connected with that rare phenomenon, general 

 glaciation. Atoll-forming ])eriods may have been at least as 

 rare as ])criods of such glaciation, and it is possible that only 

 since the late Pleistocene has the first atoll archipelago come 

 into existence. Single small atolls might have been formed 

 at any time since the reef-building corals were evolved, but 

 such groups of atolls as the Maldives, with their peculiar relft- 

 tion to the underlying plateaus, seem to demand special shifts 

 of level. For explanation of existing atolls and barriers the 

 preference is here given to a moderate tumular movement of 

 the ocean, rather thai) to the enormous crustal displacements 

 implied in the Darwin-Dana hypothesis. CoiTelating ice-caps 

 and coral reefs, we use the great discover}^ of Louis Agassiz to 

 support a principal conclusion of Alexandei' Agassiz. To the 

 father, a zoologist, geology owes the glacial theory ; to the son, 

 a zoologist, geology owes a matchless collection of facts, which 

 not only illuminate the theory of coral reefs, but also pro- 

 foundly affect the problem of crustal deformation in the 

 oceanic areas. 



