chap. vi.J GEOLOGY. 131 



Mr. Bennet* observed a pumice-bearing tufa in tbe in- 

 terior of tbe curral, near the bottom, at tbe depth, that is, of 

 several thousand feet beneath the summits of the walls of that 

 great central crater. These consist of a series of tufas and 

 conglomerates which rise, pierced by dikes, and surmounted 

 with crests of basalt, to a height above the sea of five and 

 six thousand feet; if, therefore, the bed observed by Mr. 

 Bennet is identical with that which appears on the southern 

 coast, it is certain that this lies low in the series of the rocks 

 of Madeira. 



Mr. Smith has remarked that the seams of pumice " often 

 contain portions of heavier volcanic products, as. cinders or 

 scoria?, dispersed without regard to gravitation, proving that 

 the various materials could not have been deposited under the 

 sea, because in water they would instantly have separated ac- 

 cording to their respective weights ;" and he proceeds to con- 

 clude that " the volcanic products of the island, being 

 subaerial, the curral is not a crater of elevation, though it 

 agrees with the characters assigned to such craters ;" and he 

 is further induced to "infer, from the resemblance of the 

 Curral das Freiras to the more ancient portions of Teneriffe 

 and the other Canary Islands, said to be craters of elevation 

 raised from beneath the level of the sea, that a wrong con- 

 clusion has been drawn respecting them." The supposition, 

 however, on which this large inference was grounded seems to 

 be without foundation. A single shower of materials of un- 

 equal specific gravities might possibly be completely sifted by 

 falling through water ; but in case of consecutive showers no 

 such effect would follow, since the different rates of descent 

 would be variously compensated by the different times of 

 ejection : but, in point of fact, the nodules of pumice do 



* See Geol. Trans., vol. i. p. 391, No. 17. 



