30 ME. D. DATIES OX THE ECOLOGX OF \Y0\. Ixxvii, 



4. The Ecology of the Westpkaliax cf/id the Lowee Paet 

 of the Staeeoediax Seeies o/'Clydach Vale aud Gilfych 

 GrocH (East GrLAiioEGA>'). By Dayid Davles, F.G-.S. 

 (Bead Xovember ITth, 1920.) 



[Plate II.] 



Contents. 



Page 

 I Introduction 30 



n. Ecology 33 



(a) The Proportion of Genera within each Class on Ten 



different Horizons. 



(b) The Eelative Proportions of the Various Classes 



on the Ten Horizons. 



III. The Proportional Distribution of Indiridual Genera in 



the Floral Assemblages of the Various Horizons 57 



IV. Deductions as to Physical Conditions from the Floral 



Analyses 63 



V. Stratigraphical Considerations : the Existence of a Break 



between the Westphalian and the Staffordian Series ... 68 



VI. Concluding Eemarks 70 



I. lyXEODUCTIOX. 



The evidence collected for this paper was obtained from the shales 

 overlYing the different seams of coal worked at Clvdach Vale and 

 Gilfach Goch. 



As a result of 25 years' close exam.ination of the plant- 

 remains in an extensive collection of fossils from the shales in the 

 AVestphalian and the lower portion of the Staffordian Series (com- 

 monly knoAvn in South Wales as the 'Lower Coal Measures'), it 

 was decided to compile a strict record of their distribution. The 

 extent of roadways open for examination amounted to 60 miles at 

 Clydach Vale, and 20 miles at Gilfach Goch. and these roadways 

 gave access to the nine different seams of coal. The evidence for 

 the tenth horizon was collected from the spoils brought out of 

 the Abero'orkv Collierv. Blaenclvdach. The shales overlvins? the 

 several seams of coal were closely examined in situ for ail the evi- 

 dence that thev mio-ht contain, and thev were afterwards sent to 

 the sui'face for a further and closer examination. Here each slab 

 of shale was split into small fragments, in order to obtain an exact 

 record of the number of plants in each block. A considerable 

 variation in this number was observed. 



Specimens of the plants were set apart to form a separate 

 collection relative to each horizon. In this manner, 2500 specimens 

 were brought together, representing 4'j,0u0 plants which had been 

 recorded in the field for ecological purposes. Prior to the method 

 of investigation just described, 1500 specimens of plants possessing 

 some real value Avere brought home, thu.s bringing the number 



of specimens in the collection up to JrOOO. I am indebted to 



