OVERLYING ANP TRAP ROCKS. 113 



The AMVGDALOIDS demand a few words, though the 

 origin of the imbedded nodules has been formerly 

 examined. In the analogous lavas, which are cavern- 

 ous and empty, or cavernous and filled with various 

 minerals, the caverns have been caused by the extri- 

 cation of air or steam during fusion ; while, under a 

 pressure preventing that extrication, there are pro- 

 duced solid, instead of scoriform lavas. In the 

 arnygdaloids, superficial empty cavities sometimes re- 

 sult from the waste of the included minerals under 

 exposure ; but, in other instances, they are deeply 

 situated in the mass, where neither air nor water 

 could have had access, or whence^ at least, the sup- 

 posed nodule could not have been removed. And as 

 the interior of such cavities is in all respects similar 

 to those of the volcanic scoria, while they are also 

 often elongated, and even contorted, under a corre- 

 sponding elongation or contortion in the whole mass 

 of rock, it is evident that these cavernous rocks have 

 not been subjected to a pressure capable of preventing 

 the extrication of air ; as I need not repeat that they 

 could acquire this character from no cause but fusion. 



It has often been said, by those who have written 

 so much on what they did not know, that the cavernous 

 structure was a test to distinguish between the traps 

 and the volcanic rocks. But it is not so, unless the 

 Little Cumbray be supposed the remains of a volcano. 

 Had even the rock of Edinburgh Castle been examined 

 by those who wanted no opportunities, this assertion 

 could never have been made. The Claystones of that 

 island are often as perfectly cavernous as the scoriform 

 lavas, and so ligbt as almost to float on water ; while 

 it is superfluous to say that the caverns have been 

 always empty : their elongated or tortuous forms and 

 interior varnished surfaces, moreover, rendering them 



VOL. ii. I 



