472 STUDIES IN FOSSIL BOTANY 



few cases that we have any knowledge of structure. 

 Provisionally, the following diagnosis, agreeing with 

 that already given in Chapter X. (p. 356), may serve 

 the purpose of distinguishing the group from the other 

 Palaeozoic seed-plants. 



Male and female sporophylls little differentiated 

 from the vegetative foliage ; no cones formed. 

 Anatomy of either stem or leaf or both, of a 

 Filicinean type, as was also the habit. 

 The first part of this diagnosis requires a word of ex- 

 planation, as it appears inconsistent with M. Grand'Eury's 

 opinion that the reproductive organs were usually borne 

 on special inflorescences. The part of the sporophyll 

 immediately bearing the seeds or microsporangia may 

 be modified by suppression or alteration of the lamina, 

 but in many cases (N europteris , Aneimites, Pecopteris 

 Pluckeneti, Crossotheca) the fertile rachis forms part of 

 an otherwise normal foliage-leaf, and this may have 

 been so generally. Even where no pinnules have been 

 found in connection with the fertile rachis, the latter 

 resembles that of the vegetative leaf in its branching. 

 We may safely take it as a character of the Pterido- 

 sperms that their sporophylls were more like the foliage- 

 leaves than in any other group of Spermophyta ; in 

 some cases they are even less modified than in many 

 Ferns. 



No diagnostic character can at present be based on 

 the seed itself ; the radially symmetrical seeds of 

 Lagenostoma and Trigonocarpus are no doubt charac- 

 teristic of certain families of Pteridosperms, but 

 nothing is known of the structure of the platyspermic 

 type of seed found in Aneimites and Pecopteris Pluckeneti; 



