GEOLOGICAL OBSERVATIONS IN FIJI. 91 



true, all evidence (if a former harrier reef or atoll would soon be de- 

 stroyed after uplift, and it is unnecessary to assinne a considerable 

 benching of the atoll in ordef to expose the inner lagoon limestones. 

 The writer believes there is more chance for the preservation of the 

 rounded coral heads growing near the centers of lagoons, where wave- 

 work is less vigorous, than for the preservation of the reef coral at the 

 edge of an atoll. 



In spite of the lack of conclusive evidence the writer is inclined to 

 believe that such islands as Ivambara and Fulanga are elevated atolls. 

 Further grounds for this belief are as follows: — 



1) It is impossible to believe that the coralliferous limestones 

 merely veneer older, non-coralliferous limestones, since the latter are 

 nowhere exposed in spite of profound erosion. 



2) The fossil corals occurring in the elevated limestones are identi- 

 cal with living species which are now constructing atoll-reefs. 



3) The atoll form of the islands is too strongly marked to be merely 

 the result of differential solution. 



4) The normal limestone-coral banks of the present day are higher 

 at their center than at their edges and hence, if they were elevated, 

 would not have interior depressions. 



Ancient atolls are therefore preserved in the uplifted limestone 

 islands of Fiji; likew^ise an ancient barrier reef is recognized in the 

 elevated reef of Lakemba. Most important of all is the fact that in 

 all known cases the older limestones rest unconformably on an eroded 

 basement. 



Daly, however, (1915, pp. 199-200) infers that certain islands of 

 Fiji which have the atoll-form were developed on Pleistocene wave- 

 cut benches in post-Pleistocene time, and were subsequently elevated. 

 Kambara and Fulanga belong in this class. Such a conception would 

 divide the cycle of the older limestones into two parts and would 

 negative the view that the elevated atolls developed as a result of the 

 subsidence of their eroded basement. In the older division would be 

 placed the limestones planed by Pleistocene wave-cutting; in the 

 younger division the post-Pleistocene atoll-limestones would be found. 

 Accepting this hypothesis, an unconformity should exist between the 

 two limestone series. If such an atoll were ele^•ated over 40 fathoms, 

 the depth of a stable. Pleistocene, wave-cut bench below the present 

 sea-level according to Daly, the unconformity should be exposed, if 

 the outer reef-face were sufficiently eroded. The atoll of Fulanga has 

 been elevated more than 45 fathoms, that of Kambara more than GO 

 fathoms; yet no evidence of an unconformity was seen in the re- 

 treating sea-cliffs of these islands. 



