KESULTS OF PAL^EONTOLOGICAL RESEARCH 63 



(2) The G-ymnosperms are themselves very old 

 forms : judging only by the palseontological remains, 

 they belong to the oldest flora. Of a genetic connec- 

 tion with ferns or other plants there is naught to be 

 said. 



The new class of ( Pteridosperms ' of the Carboni- 

 ferous formation changes nothing in this respect, 

 since despite the fern-like foliage they form no inter- 

 mediate link between Ferns and Gymnosperms but, 

 as seed-bearing plants, are pure Gymnosperms. Even 

 to-day we have, among the Cycads, which are true 

 Gymnosperms, forms with fern-like foliage ; and one 

 Cycad (Stangeria paradoxa) (a species now existent) 

 ranked long as a fern, until its flowers were discovered. 1 



The determination of the systematic classification 

 must, even according to Potonie, 2 who is a strong 

 advocate of transitional forms, be made dependent 

 upon the discovery of the seed or of the spore heaps. 

 If seed be found then the foliage and stems appertain- 

 ing thereto are those of true seed-bearing plants, in 

 the other case they are true ferns. He himself grants 

 that agreement in many anatomical characters thick- 

 ness of growth, venation, form of leaf may all be 

 attributed to ' adaptation to the same mode of existence/ 



(3) Eepresentatives of the true flowering plants 

 or Angiosperms c . . . appear at first in the Chalk 

 and in forms of such high organization as to agree 



1 Gothan, p. 35. 



2 Engler and Prantl : Die Naturlichen Pflanzenfamilien, I, Part iv, 

 p. 789. 



