GEOLOGICAL OBSERVATIONS IN FIJI. 95 



limestone which ha\'e been eroded t(i fairly even surfaces by atmos- 

 pheric solution, and later submerged. 



5. Areas of subsidence in Fiji have undoubtedly been compen- 

 sated by areas of uplift. In the latter areas shoals or islands have 

 been formed, on or about which coral reefs have developed. 



6. In Fiji, the Darwinian order of fringing reef — barrier reef 

 — atoll is not generally the result of progressive subsidence. On the 

 other hand, the association of the three reef forms is due to a series 

 of irregular movements during which platforms are developed by the 

 integration of the above-mentioned processes. 



7. The detailed history of the islands of Fiji shows that they have 

 not been stable during recent geological time, and an adaptation of 

 Darwin's theory may apply to many of these islands. Yet it is not 

 safe to assume its universal application. 



In conclusion, it may be stated that in Fiji as elsewhere (Cp. 

 Vaughan, 1914, p. 31) the modern reefs are developing on basements 

 which are antecedent to the reefs. Most of the older writers postu- 

 lated that the reefs rested on wave-cut benches, without telling how 

 such benches could be carved in spite of the protection of the reefs, 

 or why the reefs should be absent. The Glacial-control theory solved 

 the problem by assuming that the reefs were removed during the 

 Glacial period. The data assembled by Daly (1915) and Vaughan 

 (1916) convince the writer that Pleistocene benches exist very gener- 

 ally throughout the coral seas. Nevertheless, the platforms in Fiji 

 are much more modern in their development. The writer was unable 

 to discover in these islands any evidence of Pleistocene wave-cut 

 platforms. 



