34- Mr. W. Kemp's Observations on the latest 



tion it had been stationary at those levels for a longer period 

 than while it was emerging through the less-worn spaces be- 

 tween. It is a remarkable feature in those terraces, that they 

 can scarcely be traced but upon the north and south project- 

 ing shoulders of the hills, such as have been most exposed to 

 the sweeping currents from the west. The most distinguished 

 of those terraces is one upwards of 800 feet above the level of the 

 sea ; as a great number of our lesser hills considerably exceed 

 that height, it might be expected to be pretty generally marked ; 

 and not only is it so, but in a very remarkable degree. Ac- 

 cording to the situation of the hills, we have traced it from 

 where it was merely visible to where it was 300 feet broad ; in 

 many places it exceeds 100 feet, and everywhere it seems 

 chiefly scooped out of the solid rock. As we ascend to the 

 superior levels, the traces of each terrace become fewer and 

 more distant as they overtop the hills ; still upon some higher 

 eminences they are very well defined, as for instance where 

 they are so remarkable upon the north side of the Eildon hills. 

 From the first-mentioned terrace downwards they become 

 gradually less and less distinct, which is a further confirma- 

 tion of the theory in question, because the abrading action 

 would become gradually less powerful as the land towards the 

 west arose and checked the current. Owing to the detached 

 and rounded form of the hills, those shelves are nowhere of 

 any great length ; few of them exceed 300 yards, and many are 

 not so much. However, upon Ruberslaw, a high conical hill 

 about 6 miles west from the town of Jedburgh, there are two 

 terraces upwards of 800 paces in length by 30 in breadth, and 

 another 600 still broader. These are very beautiful, and in 

 some respects they are the finest in the district. To the east 

 is the valley of the Rule, and on the other side the ground 

 rises to a high ridge, extending eastward to the town of Jed- 

 burgh. Near the north summit of this ridge, a finely marked 

 terrace runs along its whole length, which is about 1^ mile, 

 and which correctly corresponds with one of those upon 

 Ruberslaw ; indeed so truly does it agree with the instrument, 

 that we can detect no deviation along its whole course. 



Let us conceive in the mind's eye an immense body of 

 rushing water sweeping along like a mighty river, would that 

 be sufficient to account for those terraces ? certainly not. We 

 cannot suppose that flowing water alone would run out those 

 shelves, as it would scour indiscriminately the surface over 

 which it swept, or rather it would act more powerfully upon 

 the lower depths by the superincumbent pressure of the water. 

 Then let us suppose that the sea was comparatively tranquil, 

 just as it is at present, being occasionally raised into fury by 



