60 Proceedhigs of Philosophical Societies. [July, 



ture to portions of the stone which are connected with the 

 hardest form of the rock found in the vein, together with the 

 curious intermixture of the compact with the stalactitical varie- 

 ties, are clearly referable to the same origin, and appear to have 

 been produced by a similar process of argillaceous infiltration. 



A paper, by the same author, entitled " Remarks on the 

 Columnar Structure of Trap Rocks," was then read. 



In following the track of the streams of lava from their 

 sources, we are struck with the increasing solidity which they 

 assume in the course of their descent, their texture becoming 

 gradually less and less porous, until, as we approach the val- 

 ley which forms the limit of their progression, they put on 

 almost invariably the character of a compact basalt, but it is 

 only where the lava is cut through by the stream which occupies 

 the lowest part of the valley that it assumes a columnar struc- 

 ture. This arrangement has been ascribed by the late Mr. 

 Faujas St. Fond to the sudden congelation of the liquified mass 

 by falling into water ; but in opposition to this opinion, it may 

 be observed, first, that sudden congelation instead of being 

 favourable to a regular arrangement among the particles of mat- 

 ter, proves, as far as we know, almost destructive of it; secondly, 

 that there are numerous instances of columnar trap being met 

 with in situations where no collection of water can be supposed 

 to have existed ; as, for insti^nce, the columnar clinkstone por- 

 phyry on the summit of the Mozen, the highest mountain of the 

 Vivarais ; and thirdly, that it would become necessary if this 

 hypothesis were admitted, to attribute an igneous origin to gra- 

 nite and other rocks which exhibit occasionally the same pris- 

 matic arrangement. It appears more probable, that the natural 

 structure of the stone has been developed by the continual 

 action of the stream, which, in the course of ages, has cut itself 

 a passage through the existing materials of the rock, though, 

 perhaps, it would be going too far to assert that the columnar 

 arrangement is always referable to this cause. In confirmation 

 of this conjecture, it may be observed, that the rock, at the foot 

 of which the town of Murat (Dep. of Cantal in Auvergne) is 

 situated exhibits some remarkably regular basaltic columns, 

 which, towards the summit, are vertical ; but as they descend, 

 appear to be bent in conformity with the slope of the hill. The 

 curve which they describe still increasing, they become by 

 degrees quite horizontal ; so that the extremities only of the 

 prisms jut out from the side of the rock ; the interstices between 

 the columns being filled up by a species of trap much looser in 

 its texture than that of which the columns themselves are com- 

 posed. Again at the village of Prentigarde, about half a league 

 from the Bains de Mont Doi, in the Dep. Puy de Dome, the 

 trachybe or porphyry formation is surrounded by a basalt of a 

 very compact character, and of both these rocks a good section 

 is exhibited at the cascade of Querceiul, where a small mountain 



