part 3] lower palaeozoic of abthog— dolgelley, 



315 



great Bala Fault near Drwsynant (6 miles above Dolgelley), 

 where there is an old valley right over the watershed into the 

 valley of the Dee and Bala Lake. 1 



Near Dolgelley, however, the Wnion Valley suddenly leaves the 

 main fault-line (which is continued by the Ceunant Gorge) and 

 follows a winding course for some 2 miles towards Penmaenpool 

 (fig. 7, p. 310). This winding valley must cross the line of the 

 Derwas Fault (with its continuation up the (xanllwyd Valley) and 

 •also the line of the Dolgelley Fault, which is probably continued 

 north-eastwards, swinging round in the same manner as, and 

 parallel with, the Derwas Fault. In neither case, however, is the 

 course of the Wnion Valley affected by the fault. Thus the lower 

 part of that valley and its continuation — the head of the Maw- 

 ddach Estuary — are independent of the faulting. This local 

 absence of direct relationship finds a striking parallel some 15 miles 

 away to the south, in the case of the Dovey Valley and its estuary. 

 The course of the Dovey Valley has been shown by Prof. O. T. 

 Jones & Mr. W. J. Pugh " to be independent of the faulting as 

 regards the stretch which lies near and just below Machynlleth, 

 although it is actually crossed by strike-faults. On the other 

 hand, the higher reaches of the Dovey Valley, like the corresponding 

 parts of the Wnion Vallej^, are determined by the faulting. It is 

 probably more than a coincidence that in each case the only im- 

 portant town of the valle}^ — Dolgelley in the one case, Machynlleth 

 in the other — should stand just at the point where the valley 

 leaves the fault-line. 



V. Comparison with other Abeas. 



(1) The Upper Cambrian. 



The sequence of the Upper Cambrian rocks, as developed in the 

 Arthog-Dolgellev district, compares almost exactly with that in 

 the Arenig 3 and Tremadoc 4 districts. The developments of the 

 Ffestiniog and Dolgelley Beds are so similar in the different areas 

 that no further comment is needed. In regard to the Tremadoc 

 Slates the comparison is also very close, as is seen from the 

 following synopsis : — 



Arthog. 



Upper Pencil-Slates. 



(Upper Dictyonema Band.) 



Asaphellus Flag-? & Shales. 



Lower Pencil-Slates. 

 Lower Dictyonema Band. 

 Niobe Beds. 



Arenig. 



Absent in part. 



Shwmardia Beds. 

 Asaphellus Flags. 

 Bfillero2)lion Beds. 

 Dictyonema Band. 

 Niobe Beds. 



Tremadoc. 



Garth-Hill Beds. 



Penmorfa Beds. 

 Portmadoc Flags. 

 Moelygest Beds. 

 Dictyonema Band. 

 Niobe Beds. 



1 P. Lake, Geol. Mag. 1900, p. 141. 



2 Q. J. G. S. vol. lxxi (1915) p. 378. 



:i W. G. Fearnsides, Q. J. G. S. vol. Ixi (1905) p. 614. 



4 Id. ibid. vol. lxvi (1910) p. 155. 



