248 UNDERWOOD: REVIEW OF THE GENERA OF FERNS 
We have spoken of the later Hookerian system as unjust be- 
cause it ignores in too many cases the claims of prior publications ; 
as unnatural because it associates together in the same genera 
forms of growth, that have no natural association or alliance what- 
ever, but are thrown together because they possess the accidental 
peculiarity of some such secondary or trivial character as “no 
involucre " or “ sori following the veins and like them free, forked 
or anastomosing," thus making of genera unholy alliances of un- 
related entities instead of natural groups of closely associated 
species. One has only to compare the heterogeneous assemblages 
of plants arrayed in Synopsis Filicum under such generic names as 
Gymnogramme, Acrostichum, Polypodium, Dicksonia, and Davallia, 
to appreciate the unnatural character of these generic concepts. 
The members of a natural genus should resemble one another 
sufficiently to enable one to attribute to them a monophyletic origin. 
We have spoken of the system as unscientific for the same reason, 
added to the fact that it has largely ignored anything except the 
superficial leaf form and soral arrangement of the sporophyte in 
the separation of genera and the determination of affinities, leav- - 
ing in the background the biological characters of the stem, 
habit alliances, and the subject of venation, so important in the 
study of affinities in any modern sense. In short, the whole later 
Hookerian system of genera is the natural result of a too exclu- 
sive study of herbarium sheets and a convenient method of rapidly 
"pigeon-holing" a lot of plants that must be named for corre- 
spondents, rather than a logical scientific study of the living fern 
world and the interrelations of its diversified forms. 
If we are to have genera simply as a matter of convenience for 
naming plants, then this system is, perhaps, as good as any that 
has been proposed ; but if we are to look upon genera as natural 
groups of allied species, then we must seek for a more exact rep- 
resentation of the fern world, holding ever in view the fact that 
in many cases where the links of the evolution are still in existence 
certain genera like some species will necessarily intergrade. It 
must further be borne in mind that no system founded with Lin- 
naean concepts of species can fit an organic world of progressing or- 
ganisms formed ona Darwinian or Lamarckian plan. 
No one can question two facts stated by Hooker, namely, that 
