272 Geological Society, 



PlIOCEEDINGS OF LEAKNED SOCIETIES. 



GEOLOGICAL SOCIETY. 



December 6th, 1922.— Prof. A. C. Seward, Sc.D., F.E.S., President, 

 and afterwards Mr. K. D. Oldham, Vice-President, in the Chair. 



The following communication was read : — 



' On a Collection of Fossil Plants from the Falkland Islands.' 

 By Albert Charles Seward, Sc.D., F.E.S., Pres.G.S., and John 

 Walton, B.A. 



The collection of fossil plants submitted to the Authors for 

 examination was made by Dr. H. A. Baker at several localities in 

 East and West Falkland, and in Speedwell and Greorge Islands 

 (south of East Falkland). It includes a few fi-agments of 

 Lepidodendroid stems too imperfect for specific determination : an 

 examination of the specimens and a comparison of some other 

 plant-remains, previously described by Dr. T. Gr. Halle, with plants 

 from other countries lead the Authors to suggest a Devonian age 

 for the oldest plant-bearing beds. 



Numerous examples of Glossopteris leaves were collected, espe- 

 cially in Lafonia, and these are referred to Glossopteris indica 

 Schimper and G. hrowniana Brongniart, species which are not 

 confined to one geological series in the Grondwana System. Many 

 specimens of Equisetaceous stems were also obtained from the 

 Glossopteris Beds : of these several are clearly identical with 

 Falkland examples described by A. Gr. Nathorst and by T. Gr. Halle, 

 while others are compared with an Upper Triassic or Rhsetic 

 species Neocalamites carrerei (Zeiller), The examination of 

 some well-preserved wood from Choiseul Sound enables the Authors 

 to amplify the account given by Halle, of wood which is closely 

 allied to that discovered by Dr. Baker. A comparison of petrified 

 wood, most of which has been assigned by various writers to the 

 genus Dadoxylon, from different parts of Grondwanaland, points to 

 the prevalence in the southern botanical jDrovince of trees differing 

 in certain anatomical characters from contemjDorary plants in the 

 northern province. The present Authors' conclusion is that the 

 Pernio- Carboniferous flora as a whole, so far as it is possible to 

 base an opinion on the few species represented, agrees most nearly 

 with the Damuda and Beaufort Series of India and South Africa 

 respectively. The stems compared with Neocalamites favour a 

 reference of the beds at Cygnet Harbour and Egg Harbour to 

 a somewhat higher position ; and, on the other hand, the leaves 

 described as Glossopteris indica Schimper (cf. G. decipiens Feist- 

 mantel) from North Arm, although they represent a type which 

 has a wide range both in space and in time, suggest a possible 

 correlation with the Ecca Series of South Africa and the Talchir 

 Series of India. 



