363 



not only maintain their usual fossil distinctions, but exhibit lime- 

 stones of much greater thickness than in any other part of their 

 course. Mr. Murchisonhas also shown that rocks occupying a large 

 coast tract in Pembrokeshire, which from their mineral aspect had 

 been laid down as " greywacke ", consist of true coal-measures. 

 After noticing a ridge of intrusive rocks in Caermarthenshire, be- 

 tween the Towey and the Taf, as connected with certain great lines 

 of dislocation, he points out, in the Cambrian System of Pembroke- 

 shire, examples of the existence of two classes of trap rock, one 

 bedded or contemporaneous, the other amorphous and of posterior 

 intrusion. He further shows that the main directions of the stra- 

 tified deposits of this county are parallel to divergent zones of 

 trap. 



In another paper the same author states that he has lately disco- 

 vered to the north-west of Shrewsbury, proofs of an eruption of 

 trap posterior to the new red sandstone, and probably to the lias. 

 This line of fissure along which he has observed the new red sand- 

 stone affected for a distance of thirty miles is on the precise pro- 

 longation of a linear eruption in the Breiddin Hills, which he had 

 previously pointed out as having been in progress during and after 

 the epoch of the deposition of the Silurian strata. The more modern 

 trap is made up of a peculiar felspathic rock identical with some 

 of those at the great vent of eruption fifteen miles distant, where 

 they both alternate with and are intruded into the more ancient 

 deposits. 



It appears from these observations that volcanic operations were 

 renewed along the same line after a wide interval of time, showing 

 that we must be on our guard against inferring the synchronism of 

 coincident lines of derangement. The repetition also in the same 

 spot and at two distant periods of a trap identical in mineral cha- 

 racter is curious, and reminds me of an opinion lately mentioned to 

 me by Mr. Von Buch, that the composition of lava is often deter- 

 mined by that of preexisting volcanic rocks near the point of erup- 

 tion. Thus on two opposite sides of the same volcano, as on Tene- 

 riffe for example, a trachytic flow of lava will issue from a mass of 

 trachyte, and a basaltic flow from rocks of basalt. 



Mr. De la Beche has shown that the trappean rocks are associated 

 in such a manner with the new red sandstone of part of Devonshire, — 

 among other places, near Tiverton and Exeter, — as to indicate that the 

 trap and the sandstone were each in the course of formation at the 

 same period. Some beds of sand present every appearance of hav- 

 ing been of volcanic origin, and ejected from a crater, but the sand 

 became mixed with common detrital matter then in process of de- 

 position at the bottom of the sea. Numerous angular fragments, 

 some of them even one or two tons in weight, of quartziferous por- 

 phyry with a felspathic base, are intermingled with the conglome- 

 rates of the old red sandstone, and do not resemble any trappean 

 rocks discovered in place in this district. Mr. De la Beche conjec- 

 tures with much probability that these fragments were ejected from 

 volcanic vents, and that they fell upon the sand and pebbles then in 



