OTHER CYCADOFILICES 473 



these seeds cannot, therefore, be distinguished from 

 those of the Cordaiteae, to be considered in the next 

 chapter. The presence of a pollen-chamber and of a 

 well -developed vascular system are characters which 

 the seeds of the Pteridosperms share with those of the 

 Cordaiteae, Cycadaceae, and Ginkgoaceae. 



As regards the male organs our information is 

 too limited to allow of any generalisation ; so far as 

 appears at present the microsporangia of the- Pteri- 

 dosperms did not differ essentially from the isosporangia 

 of Marattiaceous Ferns. 



The anatomical characters have already been dis- 

 cussed at some length, and further reference will be made 

 to them in the concluding chapter. There is probably 

 no constant distinctive character in the structure of 

 either stem or leaf ; but taking the sum of the anatomical 

 characters the Pteridosperms, so far as we know, appear 

 to have retained much more of a Fern-like organisation 

 than any of the true Gymnosperms. As we have 

 already seen, it is probable that both Medulloseae and 

 Lyginodendreae were derived from some simple 

 protostelic Fern-type, Sutcliffia on the one hand and 

 Heterangium on the other showing a convergence in 

 that direction. 



Other Cycadofilices 



There are a certain number of fossil plants which 

 cannot yet be classed as Pteridosperms because we have 

 no evidence as to their fructification, but which, from 

 their vegetative characters, appear to belong to the 

 same great class of plants. For such fossils, Potonie's 



