280 CORAL AND ATOLLS 



the surf edge of tlae burrier to the breccia rock of the lagoon 

 shore. Too much importance cannot be attached to this fact, 

 which is sometimes forgotten or not appreciated. The islands 

 are mere piles of debris upon this level platform. It is easy to see 

 that the lagoon margin of this platform is the first formed effort 

 of the waves upon the newly risen reef, and that once formed, 

 it has no power to alter its level in any way. The forces that 

 made it are retreating, and as it was when made by the waves 

 at the first gale that met the reef, so it must remain. The 

 extreme seaward edge of the barrier represents the very latest 

 effort of wave-building, and so we have from seaward edge to 

 lagoon shore a platform that represents the sea-building 

 throughout the entire history of the atoll. It is true that the 

 lagoon edge of the platform becomes covered with beach sand, 

 and that the intermediate part becomes covered with the 

 debris pile of the island ; but at any point the breccia stratum 

 may be exposed below its coverings. 



Since the breccia platform, when once completed by the 

 waves, and finally deserted by them, can never undergo any 

 further building, it follows that the inner margin of the 

 platform will certainly give the level of the reef, at the time 

 of its very first meeting with the waves. If the whole atoll 

 has sunk, then this inner edge of the platform must occupy 

 a lower plane than its outer edge. The sinking, if it did 

 happen, would not destroy its own traces, but would have a 

 very definite index in this lagoon edge of the breccia platform. 

 As a matter of fact, in this atoll, I have failed to find on 

 levelling across an island any difference between the level of 

 the barrier flats to seaward and the breccia platform exposed 

 on the lagoon shores. I therefore conclude that in the whole 

 long history of this atoll — from the time at which its reef 

 first came within the domain of the breakers to its present 

 advanced condition — no subsidence has taken place. Therefore 

 I reject " subsidence " as an explanation of the features of the 

 Cocos-Keeling atoll. 



Since there is no trace of subsidence to be found in this 

 atoll — nothing to be read in its history, or to be found in its 



