EXTERNAL CHARACTERS 629 
In 1884 I formulated a theory of origin of the “ phyllopodium,” or rachis 
of the leaf, chiefly based upon comparative study of the leaves of Ferns.! 
It was pointed out how in an enlarging dichotomous system a main rachis 
asserts itself as a supporting organ among parts similar in origin and structure 
to itself. This theory of genesis of the Fern-leaf may now be restated as 
follows: the Fern-leaf was originally a limited structure of flattened form, 
endowed with growth at its distal end: this growth was conducted with fan-like 
segmentation, but it was apt to be localised at points which diverge dichoto- 
mously. Sometimes the margin remained entire, and the fan-like expansion 
is then traversed by dichotomising veins: it is not improbable that this is 
really a secondary condition of condensation of a branch-system. More 
commonly the margin grows out dichotomously, the veins following, and 
a fan-like forking is produced such as is actually seen existent in many 
Ferns. But frequently with the enlargement of the branch-system the 
equality of the forking was not maintained: certain branches took the 
lead, and a sympodial development resulted in a rachis being produced, 
as a strong support for the rest, though primarily it was of similar origin 
with them. It is but a slight modification which would establish the rachis 
thus initiated as the definite “phyllopodium,” upon which the earlier, and 
sometimes also the later branches would arise monopodially, being lateral 
from the first: and thus a Pecopterid-type of leaf would result. The facts 
certainly indicate that such a transition has been effective in descent, though 
it may be a question whether all pinnate types, for instance the Marattia- 
ceae, originated in this way. Lastly, it may be noted that the winged 
structure, so prevalent in Fern-leaves as lateral lines leading even to the 
base, still indicates the ultimate origin from a flattened expansion: the 
margins may often still be traced in this way even where the petiole is 
almost cylindrical in form. 
A theory of the whole shoot based upon similar progressions was also 
suggested in the same Memoir in 1884, viz. that just as the phyllopodium 
’ gradually asserts itself as a supporting organ among structures of similar 
origin and structure to itself, so also the stem may have gradually acquired 
its characters by differentiation of itself as a supporting organ from other 
members similar to itself in origin and development. A similar idea has been 
subsequently expanded into Potonié’s theory of origin of the Fern-shoot.? 
There seems to be no sufficient foundation in fact for its acceptance. In 
the first place, there is in Ferns no known case where the axis and leaf 
appear as the two branches of a dichotomy, so that the suggestion is purely 
hypothetical: it is based only on analogy with what is learned from the 
comparative study of the leaf. The strength of the argument referring 
the branching of Fern-leaves to an original dichotomy rests on the fact 
that that mode of branching commonly appears at the apex, and is specially 
apparent in the lateral branchings. There are no such examples showing 
1 Phil. Trans., 1884, part ii, pp. 604-5. 
2 Palacophytologie, pp. 156-159. 
