— 32 — 



be perhaps possible enoiigli to fiiid a better solution tlian that 

 brought forward by him. Their absence in the Eed Sea, may 

 be passed over for the present, with remarldug, that if Mr, 

 Dauwin cliooses, lie may go into, or upon ifs area, and find. 

 single, doublé and treble lines of barrier reefs of coral forma- 

 tion, running to great distances in parallelism with the coast 

 and having gateways here and tliere, througli ■whicli vessels 

 pass in or out, the outer barrier being in some places twelve 

 to fourteen miles distant from the general outline of tlie nearest 

 coast, in short, altogetlier entitled to take place with their 

 more lengthy brethren in the offiiig of N. Eq. Australia and 

 west side of New Caledonia , and if he goes to tlie position 

 of 18° N. lat. and 50° 50' E. long, he Avill (unless it has very 

 lately disappeared) , see , // a ring of coral , about two miles across, 

 //with breakers all round the outer edge, and apparently deep 

 //water inside," in short the very, and the only sort of thing, 

 which in so far,*as we can see, or lie has she^vn, might or 

 could result from the disappearance by gradual subsidence, of 

 one of his descendiiig isles, a lagoon reef, to wit. 



XXII. »ïo proceed with otir examination , to the -westward of tlie line 

 j>of subsidence, of ■vriiich Keeling Isiatid (Cocos) is ihe index, we have an 

 »aiea of elevalion. For, on the norlhern end of Ceylon and on the eas- 

 » lera shores of India, elevated shells and corals, such as now occur ia 

 » the neighbourinn; seas , have been observed; again , in the middle of the 

 »Indian ücean , the Laecadive, Maldive and Chagosline ofatolls, or lagoons, 

 Tshow a line of subsidence. The best characterized of these, namely ihe 

 » Maldive Islands extend in length four hiindred and eighly railes, 

 »with an average breadth of sixty. These alolls agree ia most respeels 

 »witli the lagoons of the PaciUc. They differ however ia several of thera 

 » being ciowded togellier, such liltle groups being separated from other 

 sgroups by profoundly deep cliannels. Now if we look in a cliart at ihtt 

 ■» prolongation of the reef tow.irds the norlhern end of New Caledonia, 

 «and ihen complete the work of subsidence, so as to conlinne producing 

 »lhe sarne results, we sliould have the original reef broken up into raany 

 spalches, each of which, from the vigorous growth of coral on the out- 

 »side, would have a constant tendency to assuiue a rounded forra. Every 

 »accidenlal break in the continuity of the firsl line would delermine a fresh 

 »circlc. Ia the case there forc of the Low or Dangerous Archipelago, ia 

 » ihe Pacific, I believe that the lagoon islacds, were moulded round the 



