66 Dr. Mac CuUoch on the Concretionary and 



but divisible by transverse joints into very flat tables, and marked 

 by other peculiarities. 



In trying to explain the origin of this structure, it is to be re- 

 marked, that the appearance is limited to a small portion of exten- 

 sive beds which elsewhere preserve their natural characters ; and 

 that, in both, particularly in the sandstone, there is a simultaneous 

 change of the mineral character of the rock. The sandstone passes 

 into jasper ; that being evidently the case only where, from being 

 intermixed with clay and thus passing into shale, it is of a com- 

 pound nature. The simple shales that are found in it are indu- 

 rated ; and the purer sandstone is also hardened, so as to resemble 

 some of the varieties of quartz rock. These now are precisely the 

 changes that take place in similar sandstones, where they are found 

 in contact with trap rocks ; appearances so well known to all geo- 

 logists, that it is unnecessary to name any examples, except that 

 in Salisbury Craig near Edinburgh, and that described by myself 

 at Stirling Castle. 



It is well known that the masses of trap once incumbent on the 

 upper strata, are often entirely removed ; and those who know the 

 ground about Dunbar, are equally aware of the existence of nume- 

 rous detached portions of these rocks, which, there is every reason 

 to believe, have once been connected into a continuous mass. It is 

 not, therefore, unreasonable to suppose, that such a mass may have 

 once covered that portion of this sandstone, which has undergone 

 that change to jasper which, in other cases, these are known to pro- 

 duce. It is next to be seen whether any facts can be adduced to 

 prove, that the columnar structure was the consequence of the 

 same action. 



In Rum, the columnar sandstone actually lies beneath a mass of 

 trap ; so that the fact of their simultaneous presence, at least, is 

 proved. This, it is true, is as yet a solitary instance ; but here, 

 fortunately, direct experiment comes in aid of the supposition that 

 the action of heat has produced the columnar structure of sand- 

 stone. In the hearthstones of iron-furnaces, I have observed, that 

 the sandstones of which they are formed, become divided into 

 polygonal prisms, exactly resembling those of the natural prismatic 



