37 



geological horizon. The termination " ites" or some similar 

 widely adopted ending, is in some degree a safeguard ; it at least 

 shows at once that the species is a fossil, and not a recent fern, 

 and, therefore, as experience has taught, to be accepted cautiously. 



To retain all the old names in spite of advance in scientific 

 knowledge would be at once a want of recognition of palaeobotanical 

 progress, and a serious obstacle to phylogenetic investigations. 

 The better plan, and one less likely to add unsound links in the 

 chain of evidence on which genealogical diagrams are attempted, 

 would be to retain the older system of nomenclature, proposed by 

 Brongniart and added to or modified by subsequent writers, until 

 such specimens are forthcoming which supply us with fertile fronds 

 or pinnae, and thus render possible a comparison with recent forms 

 based on a surer foundation. 



In Hooker's well-known paper " On the Vegetation of the 

 Carboniferous Period, as compared with that of the present day," l 

 the difficulties of determining fossil ferns are brought home in a 

 clear and striking manner by one well qualified to speak on the 

 characters of recent ferns. Many of these difficulties are recognized 

 by all who have even the most superficial acquaintance with 

 palaeobotanical herbaria, but it is probable that had more attention 

 been paid to Hooker's valuable cautions our lists of synonyms 

 would not have reached that length and variety which they too 

 frequently show. 



There are a number of fossil ferns from rocks of different ages, 

 named neither on account of their venation nor fructification 

 characters, but on certain morphological characteristics which are 

 easily recognized and typical of some peculiar modern genera. 

 No doubt such a basis for a generic title may often be trustworthy, 

 and exceedingly useful as a means of arriving within measurable 

 distance of natural affinity. In such cases, therefore, where we 

 have striking morphological features to guide us, we may with 

 advantage make use of them without waiting for the further 

 evidence of sori or sporangia. 



In the works of Stur 2 and Zeiller 3 on Carboniferous ferns we 

 have admirable proofs of the better material which palacobotanists 



1 Mem. Geol. Surv. vol. ii. pt. ii. 1848. 



2 Loc. cit. 



3 Bassin Houiller de Valenciennes. 



