92 A. Heard — The Petrology of the Pennant Series. 



This probably indicates that the long subsidence from Ordovician 

 times had finished, and that a crumpling of the geosyncline had 

 begun before Middle Coal Measure times. 



There is nothing in the Culm Measures to indicate the presence 

 of Transition or Upper Coal Measures,^ and it is possible that regions 

 similar to these were undergoing denudation during Middle, 

 Transition, and Upper Coal Measure times, contributing the quartz, 

 chert, chlorite, mica, badly weathered felspars, etc., to form the 

 Pennant Series. 



The strike of both the bedding and cleavage of the Delabole 

 Slates is north-west to south-east.^ An anticlinal movement with an 

 axis running approximately north-west — south-east would account 

 for the similar direction of cleavage strike, and at the same time 

 provide a shore line during early Coal Measure times, which would 

 supply the material for the sediments of the Pennant Series. 



The fact that the lowest horizon of the Pembrokeshire Coalfield 

 as indicated by the fossil flora is one of Middle Coal Measure times 

 (i.e. Lower Coal Measures are not represented by the Coal Series) 

 suggests that the whole of the Coal Series of South Wales was formed 

 subsequent to the crumpling of the geosyncline, which was initiated 

 towards the end of the Lower Carboniferous period. 



Conclusions. 



(1) The sandstones below the No. 2 Rhondda coal-seam, east of the 

 River Taff, are petrologically identical with the Pennant Sandstones. 

 The lower boundary of the Pennant Series, east of the Taff, in my 

 opinion, ought to be fixed, not at the No. 2 Rhondda, but at the 

 base of these sandstones. 



(2) The Pennants were deposited in a strongly reducing environ- 

 ment caused by decaying vegetable matter, probably under lagoon 

 conditions. 



(3) Their source was in a south-west (approximately) direction, 

 and probably originated during uplift in early Coal Measure times, 

 causing Carboniferous and Devonian rocks, and possibly some 

 Pre-Cambrian ridge to be denuded. 



(4) Up to the present I have been unable to obtain any satisfactory 

 means of zoning these rocks by lithological characteristics, although 

 when first mapping the area such a zoning a|)peared highly probable. 



I have to express my indebtedness to Professor A. H. Cox, Ph.D., 

 D.Sc, and to Professor 0. T. Jones, M.A., D.Sc, for their valuable 

 advice and criticism. 



1 Dewey & Flett, " British Pillow Lavas" : Gbol. Mag., 1911, p. 205. 



2 Howe, Geol. of Building Stones (Arnold), 1910, p. 299. 



