ploy, supcrccdes the npc(^st>ity of fcelfiig nny wondt r at itg cx- 

 tciifc or gi'audeur. 



V. »It must he obseryed that the reef ilseTf in the three classes agree 

 »in structure , even in the most minute details. The (lifTerence entirelr 

 »iies in the absence or prese.nce of neij^hbourinj; land, and the rtdative 

 »posilion wliicli the reef bears to it. In the two last nientioncd classes, 

 »lhere is one difiiculty in understanding their origin, which raust he poin- 

 »ted out. Since the time of Dampier it has been remarked that high 

 j>land and deep seas go togelher. Now when -we see a nuraber of raoun- 

 »tai!ious islands, coming abruptly down to the sea shore , we must suppose 

 » the strata of 'which they are composed are continued with nearly the same 

 »inclination benealh the water, but in such cases where tlie reef is distant 

 »several miles froin the coast, it will be evident upou a little consideration, 

 »that a line drawn perpendicularly down to the solid rock, on which the 

 5>reef must be based^ very far exceeds that sinall limit, at which the efti- 

 »cient lainelliform corals exist." 



That higli lands and decp seas go togetlier, was observed 

 by the jNTortlimen , some thousand years before Dampier was born 

 but Mr. Dauwin, an enligbtened modern geologist, should ]iave 

 remarked iirstly, that low, as well as high lands, terminate pre- 

 cipitonsly in tlic occan, however gradually slopfng down to 

 the coast and tlience under the sea to the ver2;e of soundin2:s. 

 The margin there becomes abrupt and precipitous, a fact indi- 

 cative of the conclusion, which is otherwise corroborated , that 

 the beds of the ocean have formerly broken away from the 

 submarine 'cru st and subsided bodily, towards the centre of the 

 globe. The observation already made, with respect to the effi- 

 ciënt agents in forming a coral reef, might suffice for setting 

 aside the difïiculty which Mr. Darwin propounds, butthestrong 

 probability of the suppositions made above, respecting the ba- 

 scmcnts of the circular and prolonged reefs being correct, does 

 tliat very fully. Moreover Mr. Dakwin forgets the fact, that 

 in mountains of the supermarine portions of the earth's crust, 

 the strata or outline formed by their edgcs, scldom or never, 

 lic in one uniform dip from the tops to the bases, but gra- 

 dually become more and more horizontal as tliey ncar the plain 

 or valley adjoining to the muuntains; of coursc in volcanicformcd 

 clcvations ihovo is not, nor can bc, any rc£fular strata or any, 



