AND THE ADJOINING DISTRICT, BERWICKSHIRE. 45 



which are likewise filled with trap. One of these already referred to, runs in a 

 NNW. direction under the Knock hill, not however directhj under its centre, but to- 

 wards the west side, and has thereby thrown the mass of felspathic rock, of which 

 that hill consists, over to east, producing a partial retroflexion of the sandstone 

 strata on that side. In this respect, it exactly resembles the Cuniledge trap-dyke, 

 which, as formerly mentioned, has thrown the granitic mass of the Staneshiel hill 

 also over to eastward. Another striking point of resemblance between these two 

 dykes, is the occurrence of an insulated mass of otiion basalt, exactly in the line 

 of continuation of each. The basalt in the line of the Borthwick trap-dyke oc- 

 curs (in the channel of the burn which falls into the Whiteadder at Elm Cottage) 

 about 3 miles NNW. from the Knock hill. It consists of concretionary masses of 

 various sizes, up to a foot or more in diameter, composed of nuclei of bluish grey 

 basalt, coarse in the grain, and very hard and heavy, surrounded by concentric 

 coats of a tufaceous substance. There is a slip in connection with the line of fis- 

 sure, from which the felspathic rocks of the Knock hill, and afterwards the augi- 

 tic traps now under consideration, were erupted. This has caused a downcast of 

 the old red sandstone strata on the west side to an extent of 300 or 400 feet. 



At Oxendean Commonhaugh, a mass of basalt is wrought in three quarries. It 

 has been erupted through the old red sandstone, some of the strata of which are 

 seen resting on its surface, much broken and altered by the heat. The original 

 red of these beds has been changed to a dull purple, and the fragments into which 

 they have been shivered are much indurated. The trap immediately subjacent is 

 much debased by having absorbed, when in a molten state, a quantity of the de- 

 tritus of the sandstone. The result is a curious compound, which is neither trap 

 nor sandstone, but a mixture of both, and which gi-aduates insensibly into the 

 aqueous rock on the one hand, and the igneous on the other. In some places, 

 also, the pulverized sandstone has got into fissures in the trap, where it has 

 afterwards consolidated, presenting the singular phenomenon of veins of sandstone 

 in basalt. The sandstone of these veins exhibits vertical lamination. 



The Basalt of Dunselaw is similar to the above, with which it is connected. It 

 has also been erupted through the old red sandstones, which rise towards it 

 all along the south side, as shewn by excavations in the town of Dunse and neigh- 

 bourhood. A large block of metamorphic red sandstone, which is known by the 

 name of the '•' Covenanters' Stone,'" may be seen near the top of the hill. It seems 

 to be a portion of a stratum which has been detached and borne up on the sur- 

 face of the trap. It is hard and gi-anular, and dips to ENE. at a high angle. 



The trap of the Castleknowes runs from Dunselaw in an ENE. direction, and 

 probably joins the Cumledge trap-dyke at right angles. At the New Tile Works, 

 about a mile east from Dunse, it is a beautiful amygdaloid, containing nodules 

 of various minerals, coated with green earth. On the face of the bank south of 



VOL. XVI. PAET I. iVI 



