Crosby.] 478 [November 7, 



Granting, however, for the sake of the argument, that the older 

 fossiliferous formations would be left above the water in some 

 cases ; if the islands of this class were large, they are fairly repre- 

 sented in the modern ocean by New Zealand, New Caledonia, 

 and Spitzbergen, and, if small, by the Seychelles, Salomon, Mar- 

 quesas and Cape Yerde Islands, and Kerguelen Island. The 

 smaller stratified islands, however, would usually be short-lived, 

 being destroyed by erosion. Volcanic and coral islands, on the 

 contrary, are constantly growing and making good the loss by 

 erosion. Submarine volcanoes suffer no erosion until their sum- 

 mits reach the surface of the water ; and their growth is mainly 

 vertical, since the water must ordinarily prevent the lava from 

 flowing far from the outlet or crater. Consequently, if a conti- 

 nent, the stratified summits of which are high and the volcanoes 

 low, is submerged, the former will be soon swept away by ero- 

 sion, and the lavas ejected by the latter will be piled up, monu- 

 ment-like, until they reach the surface, when, although erosion 

 checks the upward growth, its ravages are constantly made good, 

 by fresh outflows of lava. 



In the opinion of the writer, these considerations materially 

 diminish the surprise which one feels on first observing that the 

 oceanic islands are mainly volcanoes or coral reefs. For in no 

 other class of islands do we find those elements of growth which 

 enable them to keep pace with the increasing subsidence and to 

 make good the encroachments of the sea. An active volcano can- 

 not be permanently submerged, and the same is true of a coral 

 island, provided the subsidence goes on slowly enough. In short, 

 nearly all the larger oceanic islands do embrace considerable 

 masses of the older stratified formations; and the fact that the 

 smaller ones do not as a rule is satisfactorily explained by a com- 

 parison with the highest points of existing continents, and a due 

 consideration of the facts that small stratified islands would nec- 

 essarily be short-lived, and if submerged only 100 feet might 

 remain forever unknown, and that the volcanic and coral islands 

 cannot usually be either submerged or worn away, possessing a 

 power of growth which makes them eternal. 



If the oceans are permanent, the vast archipelago of Polynesia 

 is, beyond a doubt, one of the most inexplicable features of the 

 earth's surface. As Professor Dana has shown, we know by the 



