346 STUDIES IN FOSSIL BOTANY 



The possible affinities of Botryopterideae in other 

 directions than that of the Ferns will be considered 

 later on. 



II. Summary 



The general conclusion to which we are led by a 

 consideration of the evidence bearing on the existence 

 of Ferns in the Palaeozoic period is that this class was 

 then fairly well represented, though by no means holding 

 the dominant position formerly assigned to it. On the 

 one hand we have, as a characteristic group, the remark- 

 able family of the Botryopterideae, very different from any 

 of the more modern families of Ferns, though presenting 

 analogies with them in various directions. On the 

 other hand, we have the Marattiaceous type, much more 

 complex in anatomical structure, and with clear affinities 

 to the recent Ferns of that group. The doubts that 

 have lately been cast on the authenticity of Palaeozoic 

 Marattiaceae have been suggested by the similarity 

 between supposed Marattiaceous fructifications and the 

 pollen-bearing organs of certain Pteridosperms and 

 Cycadophyta. In the absence of anatomical evidence 

 we are doubtless often left in uncertainty whether a 

 given fructification is to be referred to the one category 

 or the other. The structure of the Psaronius stems, 

 however, appears conclusive as to the existence of a 

 considerable group of true Palaeozoic Marattiaceae, to 

 which, no doubt, many of the fronds bearing synangia 

 belonged. 



Mr. Arber's proposed group, the Primofilices, em- 

 bracing all the more primitive Palaeozoic Ferns, with the 



