478 MR MILNE ON THE GEOLOGY OF ROXBURGHSHIRE. 



These dykes appear to indicate the production, by some cause or other, of 

 very extensive cracks in the earth's crust, which were afterwards filled up by 

 igneous matter injected from below. But for such cracks, the whin dykes de- 

 scribed in the former part of this Memoir would not have existed. 



I am aware that on this point there may be difference of opinion ; as in geo- 

 logical treatises these dykes are generally explained on the supposition that the 

 igneous matter, by being forced up, produced the crack. But in opposition to this 

 view, I submit, (1.) That if the igneous matter was forced up through the sedi- 

 mentary stata, where there was no previous rupture, the edges of these strata 

 would, in almost all cases, have been turned upwards on each side of a dyke. 

 Now, this is not the case in any of the dykes of Roxburghshire ; and though I have 

 seen elsewhere, cases where the strata were inclined upwards to the dyke, and also 

 where they have dipped down towards it, these rare cases can be explained by 

 subsequent vertical movements of the strata on one side or other of the dyke. 

 (2.) I have to observe, that if igneous matter was erupted through strata, where 

 there was no previous fissure, it would all flow out at the place where it first got 

 vent, and would never form a narrow dyke only 20 feet wide (which is the average 

 width of the Hawick basaltic dyke), and intersecting the country in a line very 

 nearly straight. 



On these two grounds, it seems to me perfectly clear, that at or shortly before 

 the eruption of the greenstones and basalts, some great convulsions took place, by 

 which the earth's crust was rent and ruptured, just as it had been at a former 

 period, and that through these rents portions of the earth's molten nucleus were 

 again ejected. 



If the theory of ELIE DE BEAUMONT, before referred to, be well founded, that 

 the elevation of mountain chains, and the protrusion of the ancient trap rocks, is 

 caused by changes in the shape or volume of the earth's nucleus, then the same 

 theory seems sufficient to explain the production of rents, and the injection of 

 these rents with igneous matter at a later epoch. Assuming that these last con- 

 vulsions may be accounted for by changes in the earth's nucleus, it is deserving 

 of remark, that these changes were not precisely similar to those which had pre- 

 viously occurred ; and, moreover, that the substance of the nucleus itself must 

 have undergone some alteration. At least, the matter erupted from it is ex- 

 tremely different, there having been felspathic rocks in the first instance, green- 

 place the dyke appeared to run in. He has returned the map to me, having marked on it the places where 

 he recognised the dyke. From his account, it appears to run by Clennel, Borrowden, Whittle, Dibden, 

 Framlington, and Acklington. This last point is about seven or eight miles from the sea, and beyond it 

 MATHESON did not proceed in his search. Though the dyke is reported by him to present a very variable 

 direction and width its average direction and width seem to agree with its character in these respects in 

 Roxburghshire. The dyke at Howick, mentioned by Mr WOOD, cannot therefore be the Hawick dyke, 

 though it runs parallel with it, and about twelve miles to the north. 



