CHAP. XV.] OF CONTINENTS AND OCEANS. 331 



land as an instance by doubting whether it can truly be 

 called an oceanic island. To this Mr. Crosby replies that 

 "it is difficult to see how it can be differently classified, 

 since the ocean between it and Australia is one thousand 

 miles broad and three miles deep." He proceeds, more- 

 over, to show that there are other exceptions : Spit zber gen 

 may fairly be called an oceanic island, as it is more than 

 400 miles from the nearest part of Europe, but it consists 

 largely of Palaeozoic and Mesozoic rocks. The Philippine 

 Islands again are certainly extra- continental, being sur- 

 rounded on all sides by very deep water, yet rocks of 

 Secondary age occur in them. Lastly, stratified rocks, both 

 of Palaeozoic and Mesozoic age, are stated to occur in New 

 Caledonia, which is separated by 700 miles of deep water 

 from the nearest land (Australia). Besides these there are 

 other reported occurrences of pre-Tertiary rocks on oceanic 

 islands which require confirmation. 



Finally, Mr. Crosby observes, " the oceanic islands are 

 of course merely the tops of submerged mountains, and it 

 is only with the highest points of the continents that they 

 can be properly compared. Now supposing the existing 

 continents were submerged to a depth of 15,000 feet, what 

 would be the geological character of the land remaining 

 above the sea ? Palaeozoic and Mesozoic rocks would pro- 

 bably be about as scarce in it as in modern oceanic islands. 

 As a rule the loftiest mountains of the globe are composed 

 of eruptive rocks, and in many cases they are extinct or 

 even active volcanoes ; although the main mass of every 

 mountain system is formed of stratified formations." ] 

 This being so, it is only to be expected that the mountain 

 tops of a submerged continent should generally consist of 

 volcanic rocks, and even these are often covered and con- 

 cealed by the recent growth of coral-rock. 



From the above considerations it is clear that the case 

 1 " Geol. Mag./' Dec. 2, vol. x. p. 251. 



