212 THE BERMUDA ISLANDS. 



zone of 100-120 feet; under favorable conditions "they may 

 thrive in depths of 50 or 60 fathoms, and thus we can readily 

 explain tiie apparently abnormal depths inside some atolls and 

 barrier-reefs." 



7. Reefs grow out on their own talus. 



It will thus be seen that Mr. Guppy dissents from those who 

 hold to the theory of subsidence, but it can scarcely be said that 

 his facts are fully, or even largely, in accord with the substi- 

 tute theory of Mr. Murray ; nor can they be said to be opposed 

 to the requirements of the Darwinian hypothesis. Perhaps 

 the most important of Mr. Guppv's generalizations is that reef- 

 building corals can thrive at considerably greater depths than 

 has been generally supposed, reaching under favorable condi- 

 tions to fully three times the depth of the commonly accepted 

 limit. Indeed, if this condition can be proved to exist it would 

 naturally do away with much of the necessity for a belief in 

 subsidence, since it would (or could) explain one of the most 

 distinctive features of coral structures, the deep lagoons and 

 channels. But the evidence on this point is of a very unsatis- 

 factory nature. Sporadic growths of reef-building corals may 

 well be found in depths exceeding the so-called coral-zone, but 

 until it can be shown that anything like a reef-development 

 takes place in this greater depth, we are justified in restrict- 

 ing the coral-zone to the narrow limits which have been gen- 

 erally assumed by naturalists. Mr. Guppy, indeed, informs us 

 that "under favorable conditions, reef-corals may thrive in 

 depths of 50 or 60 fathoms " (p. 903), but this statement seems 

 to rest merely upon an antecedent statement (p. 887) that "off 

 the reef of Choiseul Bay I [the author] did not seem to have 

 reached this lower limit [of coral growth] in soundings of 40 

 fathoms." And does this indicated depth of 40 fathoms rest — 

 as it certainly seems to — on the fact that in a cast of 31 fathoms 

 the arming preserved a " rounded impression of the size of a 

 billiard-ball, the inner surface of which retained the prints of 

 small cells as if of a Porites" (Ann. Mag. Nat. Hist., June, 

 1884, p. 464)? 



