234 Mr J. D. Dana on the Origin of Continents. 



but not by the carbonate, when so used, that the dilute solu- 

 tion, as in the instance of distilled vinegar, is capable of retain- 

 ing sufficient carbonic acid to preserve the phosphate in a 

 state of solution. 



In conclusion, I would observe that, as the earths are so 

 widely diffused, constituting, with oxide of iron, the greater 

 portion of the superficies of our globe, inquiries into their 

 chemical action on each other can hardly be too minute or 

 extended, feeling persuaded, as I do, that, if zealously prose- 

 cuted, these will not fail to throw light on many phenomena 

 at present obscure, and to make us acquainted with the 

 causes of series of changes in geology, as to which at present 

 we have little more than conjecture. 



Barbadoes, March 22, 1847. 



On the Origin of Contiiteiits. By James D. Dana.* 



In a paper on the Volcanoes of the Moon, read before the 

 Association of Geologists and Naturalists, in September last,t 

 some suggestions were thrown out with regard to the Origin 

 of Continents, drawn from the condition of a cooling globe. 

 It was observed that the portions of the earth now consti- 

 tuting the great areas of land, were free, or nearly so, from 

 volcanic action, even in the Silurian period : while the oceans 

 appear to have been regions of eruption. Hence it was in- 

 ferred that contraction must have taken place to the greatest 

 extent over the parts now oceanic, just as any cooling sphere 

 becomes depressed on the side which cools last. This was 

 shewn to correspond with the actual history of our globe, 

 inasmuch as an increasing depth in the ocean cavity would 

 necessarily leave more and more land above water in suc- 

 cessive epochs, as accords wdth observations. It was ob- 

 served that the hypothesis was farther borne out by facts : 

 for while it appears that the land has, on the whole, been 

 increasing in extent, even through the tertiary era and sub- 



* Extracted from the American Journal of Science, vol. ni., second series, 

 t See Amer. Journal of Science, vol. ii., second series, 352, and also page 10 

 of this, the 43d volume of the Edinburgh New Philosophical Journal. 



