308 STKATIGKAPHICAL GEOLOGY 



All the coal-seams in South Wales undergo a change of character 

 as they are followed from east to west, the seams which are 

 bituminous house-coals in the east becoming hard steam-coals in the 

 central parts of the basin and passing westward into stone-coal or 

 anthracite. It is also found that the coals of the upper measures 

 are more bituminous than those of the lower. 



In the Bristol coalfield a thickness of about 2000 feet is found, 

 although the summit is nowhere seen ; it is divided into two parts, 

 the lower portion or Farrington Group occurring in both the 

 northern and southern basins, the upper portion or Radstock Group 

 only coming into the southern or Somerset basin. 



The small coalfield of the Forest of Dean calls for some notice, 

 because previous correlations of the beds with those of the South 

 Wales and Bristol basins have been altered by a recent study of 

 the fossil plants. It was supposed that there was an upper set of 

 Coal-measures underlain by an equivalent of the Pennant grits. 

 It now appears, however, that all the productive Coal-measures 

 belong to the upper stage, and that the supposed Pennant grits 

 are merely a set of massive sandstones in the lower part of this 

 stage, which thus consists of the following members : 



Feet, 



Upper shales and coals (10 seams) ..... 1480 

 Thick sandstones with 3 coal-seams .... 850 

 Trenchard sandstones and 2 coals 77 



2407 



These beds are directly underlain by the so-called Millstone Grit, 

 and consequently there is no representative of either the Middle 

 or Lower Coal-measures. 



6. Devon and Cornwall 



The " Upper Carboniferous " or Westphalian Series of this area 

 has hitherto generally been called the Culm-measures, but this term 

 has become misleading and should be dropped, for they are simply 

 barren Coal-measures and are not the equivalents of the strata 

 known as Culm-measures in Central Europe. 



According to Mr. Ussher r the series is divisible into two parts 

 or stages, which he called Middle and Upper Culm-measures, but 

 this is not confirmed by Mr. Arber who regards the whole 

 succession of beds as forming a single lithological division, the 

 so-called "Eggesford Grits" being only a local variation of type 

 which he has recognised in several localities. 8 



Mr. Arber describes the series as consisting essentially of fine- 

 grained sandstones or grits and shales in alternating bands. The 

 beds are highly flexurecl and broken by faults, many of the sharp 



