239 
some Notes on the Leaf-trace in the Ferns . 
under none of which has C. fraxinea been placed. Both species were placed 
in Gymnogramme (but in different sections of the genus) in Hooker and 
Baker’s ‘ Synopsis Filicum \ Fee, at the time of proposing the generic 
name for Coniogramme fraxinea , proposed the name Dictyogramme for 
Diels’ C, japonica, This separation is certainly upheld from the result of 
examination of the pinna-traces of the two species. Notholaena affinis 
(Mett.), Moore, N. bonariensis (Willd.), C. Chr., N. distans , R. Br., 
N. hirsuta (Poir.), Desv., N. hypoleuca , Kze., and N. mollis , Kze., have all 
been placed in the genus Cheilanthes by Mettenius; they agree in having 
the marginal type of pinna-supply. Notholaena marantae and N. sinuata , 
which have the extra-marginal type of pinna-supply, have been placed in 
various genera, but never in Cheilanthes. In the genus Onychium all four 
species examined have regularly been grouped together (by Prantl—in 
Cryptogramme— by Hooker and Baker in the ‘ Synopsis Filicum ’, and by 
Christ in ‘ Die Farnkrauter der Erde ’). There is in this genus no explana¬ 
tion from the nomenclatural data which suggests an isolation of O. japonicum 
from the other species. 
In these notes on the ‘ aberrant ’ species I by no means wish to 
suggest a re-grouping based on their types of pinna-supply. The rule 
regarding the constancy of occurrence of one type of pinna-supply within 
a genus is not invariably applicable, yet in so large a number of the 
species examined does it apply, and in several critical genera so closely does 
a grouping based on the type of pinna-supply correspond with the grouping 
of species made on other grounds, while so many of the exceptions to the 
rule have been more or less problematical to systematists, that it would 
seem that the rule is founded on some real phenomenon and not on mere 
chance. It is the marked constancy of occurrence of one type of pinna- 
supply within a genus of Ferns which makes it appear probable that 
changes in those structural features of a Fern, customarily used for purposes 
of systematic classification (including generic delimitation), have been 
accompanied by alterations in the form and system of branching of the leaf- 
trace. These alterations are represented, in part at least, by the abandon¬ 
ment of one type of pinna-supply and the adoption of the other. It was 
this constancy which led me to use 1 the attractively concise, if not legiti¬ 
mate, expression, that ‘ systematic position ’ is one of c the factors which 
control the form of leaf-trace and its system of branching 
It may be noted that while the marginal or extra-marginal type of 
pinna-supply has been recorded as characteristic of certain species, the same 
type of pinna-supply is not found in connexion with every pinna of any 
Fern-leaf. In many leaves—indeed, in the majority examined—the 
ultimate pinnae, 2 and frequently several pairs below the tip of a leaf, 3 are 
1 Trans. Roy. Soc. Edin., vol. lii, pp. 24, 32. 2 Ann. of Bot., vol. xxvi, p. 251. 
3 Trans. Roy. Soc. Edin., vol. 1 , p. 360. 
