636 FOURTH REPORT — 1834. 



email pebbles or gravel, which are agglutinated together by a 

 tufaceous paste or mud, having exactly the appearance of a 

 stream of hardened lava. This occurs in the middle of the 

 old red sandstone formation, on the banks of the Whitadder, 

 north-west of Dunse. 



3. There are a fevv examples of trap ejected after the deposi- 

 tion of the coal-measures, which in consequence are greatly 

 disturbed in its neighbourhood. The whole of Lamberton Hill 

 (near the sea coast) is an example of this, the coal-measures 

 which run along its base for about four miles on the shore being 

 now seen not only vertical, but even inverted to a considerable 

 extent. The trap here has risen up, and is so extensive as to 

 have upraised not merely the coal-measures on the one side, 

 but the grauwacke on the other, and completely obliterated the 

 old red sandstone group at this point, the only trace of it left 

 being a patch of conglomerate on the top of the hill. 



A few miles to the south of Berwick there is another mass 

 of trap, which forms the Kyloe hills, and from which a dyke 

 runs fifteen miles in a straight line towards Home Castle rock. 

 In the Tweed below Coldstream it thins out to nothing. The 

 dyke is a light -coloured greenstone. It varies in width, though 

 generally speaking it is broader near the Kyloe hills than at 

 its further extremity. 



The usual effects of trap in hardening the strata with which 

 it is in contact, are observable in this dyke and in those previ- 

 ously described. 



In some places there has been a slight overflow of the trap- 

 dyke into the softer strata in contact with it, as, for example, 

 the shales and coal, which could less easily resist the lateral 

 pressure of the confined current. 



There do not, however, appear, at any of the localities where 

 the dyke and the sedimentary rocks are seen in contact, to have 

 been any other changes effected on them. They are in no case 

 turned up on their edges, or altered in their general bearings. 

 But the case is widely different Nvith the trap-hills, all of which 

 have, wherever they are in contact with the trap, upraised the 

 adjoining rocks. This difference between the effect of trap-hills 

 and the effect of dykes may be explained by supposing that 

 they were merely currents of trap, which flowed into fractures 

 or rents previously existing across the country, caused, per- 

 haps, by the elevation of particular points by masses of trap 

 which have been pressed up from below. Such a rent was very 

 likely to be produced by the elevation of the Kyloe hills, and 

 the direction it took would naturally be towards some other 

 point where a similar disturbing and rending force existed. 



