CORAL REEFS. APPENDIX. 219 



Mr. Guppy suggests, otherwise it would be almost impossible to 

 account for the nearly equal altitude (above the water) which 

 this formation holds in the different islands of the island group. 

 Indeed, the fact that by far the greater number of coral-islands 

 and reefs lie practically at the level of the sea, or but a few Imn- 

 dred feet above it in the case of f ringing-reefs, is an almost in- 

 superable objection to the theory which holds to the formation 

 of atolls through elevation ; the uniform line of position is 

 opposed to any law of chances which might be assumed to 

 govern a broad elevation. The same objection naturally does 

 not apply to a theory of upgrowth in a stable area any more 

 than it does in the case of a subsiding one. 



Mr. Guppy has, indeed, himself anticipated some of the ob- 

 jections to his own views, but it appears to me he has failed 

 to grasp their full significance. If, as it is claimed by the au- 

 thor, reef-building corals may thrive at a depth of 40, 50, or 

 60 fathoms, and if their structures are planted directly upon 

 deep-sea deposits, then manifestly the thickness of the coral- 

 made rock should be very much greater tlian has actually 

 been found to be the case. 



Mr. Guppy attempts to meet this difficulty by assuming [the 

 immediately reversed position] that reef-corals will be usually 

 confined to depths of less than 20 or 30 fathoms, and that the 

 " rapid sub-aerial denudation, to which these regions of heavy 

 rainfall are subjected, would be an important agency in the 

 thinning away of the raised coral formations" (Proc. Royal 

 Soc. Edinburgh, 1885-86, p. 890). This is surely begging the 

 question — indeed, it might be said, it is abandoning the main 

 proposition — since in the feeble development of the coral-made 

 rock the one vulnerable argument against the Darwinian hy- 

 pothesis was supposed to lie ; it is in this fact that the oppo- 

 nents of the subsidence theory have intrenched themselves. 

 Yet we have here the testimony of the only investigator in the 

 premises that the thinness of the rock in question is probably 

 not as thin as it is supposed to be; indeed, for any evidence 

 that has been brought forward to the contrary, the rock may 



