474 STJBMEEGED LAND-STJEFACES AT BAEEY. [Aug. 1 896.. 



27. On stjbmeeged Land-stjefaces at Baeet, Glamobganshtee. 

 By A. Steahan, Esq., M.A., F.G.S. With Notes on the Fauna 

 and Floea by Clement Reid, Esq., E.L.S., F.G.S. ; and an 

 Appendix on the Miceozoa by Prof. T. R. Jones, F.R.S., and 

 F. Chapman, Esq., A.L.S., F.R.M.S. (Communicated by per- 

 mission of the Director-General of H.M. Geological Survey. 

 Read March 25th, 1896.) 



Baeez Island lies off the northern coast of the Bristol Channel,, 

 about 7 miles south-west of Cardiff. Until the year 1884 it was 

 separated from the mainland by the tidal estuary of the Cadoxton 

 River on its northern side, and on the east by a tract covered at high 

 water, but in which the solid rock cropped up through the tidal 

 deposits at frequent intervals, the most conspicuous prominences 

 being known as the Coston, Mark, Bendrick, and Black Rocks. 



In the absence of evidence to the contrary we may assume that 

 the course followed by the river at this time was its original course, 

 for it falls into the general south-westerly direction of the rest of 

 the valley. The numerous outcrops of rock, moreover, in the tidal 

 area east of the island make it unlikely that there was an outlet 

 here of sufficient depth to drain the marshes behind it. This tidal 

 area seems to have been one of three low cols in the water-parting 

 on the southern side of the Cadoxton Yalley. One of these would 

 separate the western part, or the Little Island, from the main part 

 of Barry Island were it not for a ridge of blown sand ; another 

 occurs east of Hayes Farm, where, as shown by the Ordnance level 19 

 given on the map facing this page, the alluvial marsh of the Cadoxton 

 River closely approaches the present coast. The third, presumably 

 a trifle lower than either of the others, was submerged during the 

 subsidence of the land, of which proof will be given in the following 

 paper. It was therefore as a direct consequence of this subsidence 

 that Barry was separated from Hayes Farm and became an island, 

 while a slight increase in the extent of the movement would have 

 given us three islands instead of one as the result of the submer- 

 gence of the old water-parting. 



As to the date of the insulation of Barry, it will be seen subse- 

 quently that it had not taken place before Neolithic times ; on the 

 other hand the island is, I believe, referred to as such in the earliest 

 historical records. 



In 1884 the Barry Docks were commenced. The river was 

 diverted from its ancient course and carried to the sea by an 

 artificial cut east of the gap referred to. Its former valley on the 

 northern side of the island was partly filled up and partly excavated 

 to form a dock, the entrances to which were made in the tidal area 

 east of the island, the sea being excluded by a wall. Subsequently 

 a second dock (shown on the map as • Barry Docks Extension ') was 

 commenced in this reclaimed area, and it was the excavation made 



