17 



XYhile some of the excesses of the systems of Fee and Moore are 

 not accepted, many of the really scientific aspects of the systems of 

 Presl and John Smith have not been incorporated, and, as a whole, 

 it lacks most what a master would have put into it homogeneity 

 and consistency of treatment. For a system that is supposed to 

 proceed from low to high in an evolutionary- way, the order of ar- 

 rangement of the larger groups is surely peculiar, as may be seen 

 by the following : 



Family HYMENOPHYLLACEAE 

 Family CYATHEACEAE 

 Family POLYPODIACEAE 



1. IVoodsieae 



2. Aspidieae 



3. Oleandreae 



4. Davallieae 



5. Asplenieae 



6. Pterideae 



7. Viitarieae 



8. Polypodieae 



9. Acrosticheae 

 Family PARKERIACEAE 

 Family MATONIACEAE 

 Family GLEICHENIACEAE 

 Family SCHIZAEACEAE 

 Family OSMUNDACEAE 



The family OPHIOGLOSSACEAE is removed to a separate order, 

 as is also the family MARATTIACEAE. 



\Yhile there is long likely to be two schools of belief regarding 

 the relative position of the eusporangiate and leptosporangiate 

 forms, it is quite evident that were the groups of the Polypodiaceae 

 inverted, they would come nearer representing an ascending series, 

 and some of the later families are surely simpler. As to these 

 major groups themselves, there is little fault to be found ; possibly 

 Platycerium might be held by some to a more distinctive rank than 

 simply a mere member of the tribe Acrosticheae, but in the main 

 the separation into families is logical and scientific. With the ex- 

 ception of Parkeriaceae, the names are well chosen ; this one is un- 

 fortunate as based on one of the many synonyms of Ceratopteris, 

 while the name Ceratopteridaceae which is possibly less euphonious 

 has the double merit of being distinctive, and (in one of its forms 

 at least) more ancient. Some of the commendable features of the 

 work, so far as they pertain to our American ferns, may be noted 

 in brief form : 



