244 Philip Lake — Bala Lake and liiver System. 



the Eden across the valley and over the opposite hills, it leads u& 

 into Afon Dulas. The Dulas maintains the same direction until 

 it enters the valley of the Dyfi, where it also is deflected to the 

 rio-ht. If we neglect this deviation and proceed in the same line, 

 we come again upon another stream flowing in the same direction, 

 namely, Afon Eheidol. This river runs southward as far as Devil's 

 Bridge, where it in turn is suddenly deflected at right angles, so 

 as to enter the sea at Aberystwith. But our line is continued still 

 further by the upper part of the Ystwyth and the valley of the Teifi. 



So exactly are all these rivers — the Eden, Afon Dulas, the upper 

 parts of Afon Eheidol, and of the Ystwyth and the Teifi— in the 

 same line,^ that the valleys to which they belong seem all to be 

 portions of one great valley, the head of which probably lay as 

 far north as Blaenau Ffestiniog. This valley, according to my 

 views, was subsequently broken up by the formation of the transverse 

 valleys belonging to the N.E.-S.W. system. 



Similarly, on the south-east side of the Bala fault, we find the 

 continuations of Afon Lliw and Afon Llafar in the river which is 

 now called the Vyrnwy ; of Afon Try weryn in the Eiver Tanat ; 

 of the Ceirw (probably) in the Ceiriog ; and of the Alwen, in the 

 Dee below Corwen. 



Another important stream south of the Bala fault is the Banw, 

 and possibly the original head-waters of this are represented by 

 the upper part of the Dyfrdwy. 



The Banw, the Vyrnwy, and the Tanat, all enter the sixth of 

 the N.E.-S.W. valleys given in the list on a previous page. The 

 Banw and the Vyrnwy are deflected and flow along this valley ; 

 but the Tanat maintains its original course and even crosses the 

 next transverse valley. The original continuation of the Banw 

 is probably to be looked for in the lower part of Afon Ehiw and,^ 

 across the Severn, in the Onny Eiver. 



If we assume that the Eden, Lliw, etc., were the upper portions 

 of the valleys which I have endeavoured to trace, we must conclude 

 that before the transverse valleys were formed — those which now 

 run from north-east to south-west — a radiating system of drainage 

 was established, the centre of which lay in the high ground at and 

 near the present source of the Conway.'^ 



This appears also to be the centre from which radiate the principal 



1 It -will be noticed that this line is almost exactlj- parallel to the general trend of 

 the coast of Cardigan Bay, a fact which seems to point to some community of origin. 



* If the radial valleys existed before the formation of the transverse valleys, we 

 might expect that where the lines joining the remnants of the radial valleys cross 

 the intervening ridges, there should be a col or depression ; and this is, " indeed, 

 very commonly the case. Moreover, opposite each important stream where it enters 

 a transverse valley, there is generally a little valley running down fi'om the col and 

 entering the transverse vaUey on the opposite side. For example, opposite the 

 point where Afon Lliw enters the valley of the Bala fault, a valley runs in from 

 the col called Bwlch-y-pawl, which is situated at the bead of the Eiver ^'yrnwy. 

 Opposite Afon Tryweryn is the valley of the Hirnant, and from this we pass across 

 a col into the head of Afon Tanat. Similarly, in tlie line of the Ceirw we find 

 Afon Trystiou entering the valley of the Dee from the opposite side ; and from the 

 head of Afon Trystion a decided depressiou leads us to the head of the Ceiriog. 



