94 THE METHOD OF DARWIN. 



on coral islands. 1 Time has shown that the 

 conclusions reached from this mapping of facts 

 is too general. The conclusions that volcanoes 

 are invariably absent from the areas which have 

 recently subsided or are still subsiding, and are 

 commonly present in areas that are rising or 

 have recently risen, that fringing reefs lie in 

 the areas of elevation and atolls in the areas 

 of subsidence, may not be accepted without im- 

 portant reservations. But whether time shall 

 ultimately substantiate or correct Darwin's con- 

 clusions, or shall even destroy some of them, 

 his classifications will always remain essential 

 to the study of coral islands. 



1 Structure and Distribution of Coral Islands, p. 189. 



