﻿Vol. 64.] BALA. AND LLANDOVERY EOCKS OP GLYN CEIRIOG. 



581 



upper series, it is not precisely coincident, but dips at a somewhat 



steeper angle, so that in passing down the boundary-plane we pass 



into continuously lower beds. 



The beds of the lower series, on the other hand, dip at a slightly 



steeper angle than the boundary, so that the plane of separation is 



related to the beds themselves somewhat in the fashion shown in 



fig. 12. Such a plane cannot be a plane of normal deposition. It 



might, indeed, be a plane of 

 Fig. 12. unconformity with accom- 



panying overlap ; but in 

 that case the beds above 

 would necessarily be mar- 

 ginal deposits, and of this 

 we see no evidence. 



Whatever the surface of 

 separation may be, it can 

 easily be traced from Nant 

 lorwerth to the neighbour- 

 hood of Hafod-y-gareg ; and 



its outcrop shows clearly that, for a considerable distance, it is 



nearly a true plane. 

 The hypothesis of unconformity being almost excluded by the 



character of the beds, we conclude that this surface of separation is 



a fault of low dip ; and the 

 question now arises whether 

 the motion of the upper 

 beds relatively to the lower 

 has been up or down this 

 plane — whether, in fact, 

 the fault is a thrust-plane 

 or what Dr. J. E. Marr 

 has called a lag-fault. 



It is often assumed that 

 a thrust-plane necessarily 

 brings lower beds upon the 

 top of higher, and that a 

 fault that brings higher 

 beds upon the top of lower 

 must be of a different 

 nature. But this is not 

 necessarily the case ; the 

 effect of a thrust-plane is 

 determined by the relative 

 dips of the thrust-plane and 

 of the beds themselves. 



If both beds and fault 

 dip in the same direction, 



then, provided that the dip of the beds is lower than that of the 

 fault, an overthrust will bring older beds upon the top of newer 

 (fig. 13). But if, on the other hand, the dip of the beds is steeper 



Pig. 14. 



