460 MR MILNE ON THE GEOLOGY OF ROXBURGHSHIRE. 



greywacke strata, forming, indeed, very nearly half a right angle with them. If, 

 therefore, it occasionally takes a course at right angles to the greywacke strata, 

 it must, in order to preserve its average direction, also occasionally take a course 

 parallel with them ; and, accordingly, it has been shewn that this is the case. 

 Nor is it difficult to understand, why this irregular course should have been taken, 

 if the dyke only fills up a rupture of the earth's crust. This rupture would 

 follow the lines of least resistance, and these, it is plain, must be either parallel 

 with, or directly across, and not obliquely across the strata. In the less com- 

 pact formation of red sandstone, there would not be the same difficulty of rup- 

 ture; and hence within its precincts, the course of the dyke is not so irregular as 

 in the greywacke formation. 



This dyke varies in composition, not merely in different parts of its course, 

 but even at any one spot, according as the specimens are taken from the sides or 

 the centre. At the sides the trap is fine granular, almost approaching to clink- 

 stone, and occasionally it is vesicular ; whereas in the centre the texture is coarse 

 but compact, and highly crystalline. The crystals are sometimes pretty large, con- 

 sisting chiefly of quartz, and more rarely of glassy felspar. These differences of 

 structure can be readily explained by the difference in the rates of cooling in the 

 different parts of the dyke. 



On the south bank of the Tweed, a little above Merton House, there is a mass 

 of greenstone, which has upraised and hardened the contiguous sandstone strata, 

 and a portion of them may be seen entirely enveloped in the trap. This rock 

 has all the appearance of a dyke, though of this there is (as I have already men- 

 tioned) no certainty, in consequence of its not being traceable for any great 

 distance. 



A little below the Manse of Castleton, on the south bank of the Liddell, there 

 is a mass of greenstone about 20 or 30 yards wide, which has upraised the strata, 

 and appears to be a dyke, running about west-north-west. At Larriston lime quar- 

 ry, a similar dyke may, I understand, be seen running in the same direction. It 

 is marked on the map. 



There is one other subject which ought here to be noticed, as common both to 

 stratified and unstratified rocks. I allude to the joints which intersect them. 



This is a point which of late has been attracting a good deal of attention, 

 and not more than it seems to deserve. But it is one which can be properly 

 worked out, only after an immense accumulation of observations, and a careful 

 classification of them, which in this county I have not been able to make, and 

 which I regret the more, as the observations I have made are very encouraging. 



At the Limekiln edge, the principal joints form fissures no less than 5 inches 

 in width, and are filled with a fine yellow clay. The minor joints at right angles 

 to these, form fissures about 4 inches wide. The former run north 55 east, the 

 latter north 35 west. In the limestone of Penton Linns (near Canonby), the prin- 



