78 THE LESSON OF EVOLUTION 



Permian. Some of the plants e.g., Psilophyton and 

 Ptilopliyton cannot be placed in our modern classifi- 

 cation, for they are unlike anything living ; while, at 

 the same time, they are not sufficiently well known to 

 warrant new orders being made for their reception. 

 Sir J. W. Dawson's opinion that the RJiizocarpece 

 existed in abundance in Carboniferous times has not 

 been confirmed by later observers ; and Spkenophyllum 

 has been placed in a special class, connecting the 

 Equisetaceae with the Lycopodiacese. In fossil ferns 

 the fructification is rarely preserved, so that the 

 arrangement of the venation of the fronds has to be 

 largely used in their discrimination ; and this makes 

 their classification doubtful. Most of the Carboni- 

 ferous ferns look much like modern ones, and some 

 certainly belong to living families ; but Megaphyton 

 was a remarkable tree-fern with only two rows of 

 large fronds, one on each side of the stem. Sigillaria 

 connects the vascular Cryptogams with the Gynmo- 

 sperms ; for it has the fruit and leaves of a Lycopod, 

 while the internal structure of the stem closely 

 resembles that of the Coniferse; and Lyginodendron 

 connects the ferns with the Cycads. 1 



The forests were formed by tall spore-bearing trees, 

 chiefly Sigillaria, with its unbranched stems clothed 

 with long grass-like leaves ; Lepidodendron, with rough 

 branched stems ; and many tree-ferns. In the swamps 

 were numerous dense clumps of Catamites and Annul- 

 aria, with hollow, reed-like stems, sometimes a hundred 

 feet in height, and distinct whorls of needle-shaped 

 leaves. On the uplands Gynmosperms grew abun- 

 dantly. Dadoxylon was perhaps related to the fan- 

 1 Philosophical Transactions, Series B. } vol. clxxxvi. p. 765. 



