Vol. 52.] AT BARKY, GLAMORGANSHIRE. 489 



as he had during many years past paid considerable attention to the 

 submerged forests and peat-beds which have been found along the 

 Welsh coast of the Bristol and St. George's Channels. Some sub- 

 merged forests were noticed by the 'historian Giraldus Cambrensis 

 in the 12th century, at spots now constantly covered by the sea ; 

 and there is other evidence to show that great encroachments have 

 taken place in comparatively recent times. When the forests which 

 contained oak, alder, birch, etc. as their chief trees, grew on the 

 plains and in the valleys exposed to the westerly winds, it is clear 

 that the sea must have been at a considerable distance ; and it seems 

 more than probable that the Bristol Channel was then mainly a tract 

 of marshy plains traversed by important rivers. The trees in the 

 forest at Whitesand Bay, Pembrokeshire, which he examined many 

 years ago, had been deeply rooted in the underlying Boulder Clay, 

 and many antlers of the red deer, a jaw of the brown bear, and a 

 well-worked flint-flake were found by him in the peat-bed. Possibly 

 the oldest of the beds referred to by the Author might belong to 

 the same period, but the others appear to be of more recent date. 



Mr. Codrington said that the old beds of the river and ' pills ' along 

 the coast indicated subsidence at least as great as 55 feet. The rock- 

 bed of the Wye at Chepstow was 42 feet below low water, having 

 40 to 50 feet of mud over it containing nuts, leaves, and oak-timber. 

 In Milford Haven the rock-bottom of My land Pill was 45 feet below 

 low water, and that of the channel of the pill at Milford Docks 

 deeper still, with 40 to 50 or 60 feet of mud. There were many 

 such sections. 



The Author replied to Prof. Jones that he had acknowledged the 

 occasional invasion of the estuary by salt or bra.ckish water, implied 

 by the presence of foraminifera and some of the ostracoda. At the 

 same time the entire absence from the estuarine beds of the marine 

 shells which abounded in the overlying deposits was significant. 

 The hypothesis of the President, that when the lowest peat was 

 being formed the Bristol Channel was dry land traversed by a 

 winding river had occurred to him also, and it was with this in mind 

 that he had pointed out the existence of a deep narrow channel about 

 2 miles south of Barry. The relative levels of the peat-beds and the 

 sea could not be explained by mere encroachment of the latter. No 

 glacial deposits occurred either in the excavations or in the immediate 

 neighbourhood. The existence of the numerous buried valleys 

 referred to by Mr. Codrington, and the general character of the 

 deposits filling them, were well known to him. They had been 

 explored, however, chiefly by borings, and he attached greater value 

 to observations made on such deposits in situ, and to specimens 

 collected with every precaution against the mixing of the fauna and 

 flora of different beds, than to the mangled debris brought up by 

 the boring-tool. 



2i2 



