36 FILICIDE. 



distinguish certain types of fern fronds, separated from one another 

 by the characters of their nervation; these have been, and still 

 are to a large extent, the accepted terms which are used in 

 speaking of Palaeozoic specimens. They are accepted on the 

 understanding that in making use of such names we are merely 

 admitting our imperfect knowledge, and, as the only possible 

 basis of classification, make use of a system which is thoroughly 

 artificial. Ettingshausen, whose works on the venation of the 

 vegetative organs of plants are so widely known, attempted a 

 classification of living ferns on the same lines as those which 

 are followed in dealing with fossils. 1 That such a system of 

 arrangement rests on a foundation utterly insecure has been 

 pointed out by Stur in his classic work on "Die Carbon-Flora 

 der Schatzlarer Schichten." 2 He shows how such a genus as 

 Polypodium affords examples of eleven of Ettingshausen's venation 

 types, and how the same genus has three types in common with 

 Acrostichum, four with Pteris, four with Asplenium, and six with 

 Aspidium. Then again the venation type Sphenopteris occurs in 

 twelve genera and three families. Hence it must be admitted 

 that the genera which are based on characters of venation alone 

 are essentially provisional, and, if recognized as such, are of 

 extreme value until increased knowledge places us in a position 

 to determine the family to which a fossil fern belongs. 



The custom of giving recent generic names to fossil ferns is 

 one which several writers have frequently adopted in dealing 

 with Mesozoic and Tertiary plants. This practice, I am inclined 

 to think, has been followed too commonly ; and the result has 

 been that among the more modern fossil ferns we find a large 

 number of species spoken of by the names of living genera, to 

 which they have little or no claim to relationship. In speaking 

 of fossil algae attention has been drawn to the great danger which 

 necessarily accompanies this use of modern names ; the same 

 remarks apply in the present instance with equal weight. Surely 

 there ought to be good evidence at our command before a fossil 

 fern is designated by such names as Dicksonia, Thyrsopteris, 

 Aspidium, etc., and thus presumably an authenticated occurrence 

 put on record of any of these genera at a certain locality and 



1 Farnkrt. Jetztwelt. Vienna, 1865. 



2 Abh. k.-k. ereol. Reichs. vol. xl. Abth. i. 1885. 



