122 UNDERWOOD: OurR GENERA 
2. Habit and growth characters of stem. 
3. Position of sori in relation to veins. 
4. Indusial characters. 
The venation of the fern, forming as it does a part of the 
primary anatomical structure of the plant, must furnish characters 
which are subject to less variation than those which pertain either 
to the location of the sori on the veins or to the character of the 
indusial covering of the sorus, which at best is only a member of 
the epidermal system, normally a flattened trichome, and, like 
terminal or peripheral members generally, is subject to variation to 
a greater extent from external causes. While accidental conditions 
of united veins occur in certain free-veined species, and in certain 
genera there exist species closely related by habit and structure, 
in some of which the veins are normally free and in others the 
veins anastomose, still the characters of free or anastomosing veins 
follow, in general, very clearly defined types and can fairly be taken 
to represent one of the most fundamental elements in the determi- 
nation of generic groups. 
The determination of generic characters having been accom- 
plished, the question of generic names is an entirely different prob- 
lem. This is a direct corollary of the proposition that a genus is 
a group of related species. The botanists of the middle period of 
the last century proceeded without rule; priority was not consid- 
ered, and in taking up names one followed one fancy, and another 
a wholly different one, which resulted in the same name being ap- 
plied to widely different aggregates. Since the attempt has now 
been made to adjust nomenclature by rule rather than by individual 
fancy it has become necessary to fix upon certain species as genetc 
types * with which the generic name will inseparably stand as valid 
as 
in a wholly different sense. In /Historia Filicum, for example, John Smith Oy 
term ‘‘type’’ in the sense of the most common or well-known species included in his 
conception of the genus, whether it formed a part of the genus as originally proposed = 
not. Thus his type of Mephrodium is Aspidium molle Swz., which was not 4 pees 
the genus as originally founded by Richard, and is generically distinct from any species 
it did contain. The ‘‘ biological type’’ or center of variation and divergence 1S also @ 
wholly different matter and one that can never be fully determined until after the last 
species of any generic group has been described. The nomenclatorial type is the en 
considered here and, so far, it has only been definitely settled that this type must 
a species mentioned in connection with the original publication of the genus. 
