4 
A. G. Tansley. 
is, however, in a different position. The function of the petiolar 
vascular system is to supply the branches of the frond (pinnae) 
which come off right and left in one plane. This function is 
performed mainly by the sides of the leaf-trace curve, while the 
dorsal or central part of the curve runs on up the rachis, gradually 
diminishing in size as its lateral tracheae are drawn into the supply 
of successive pinnae. Meanwhile the gap on the ventral (upper) 
side of the petiolar strand is not closed, because there is no reason 
why it should be, since there is no demand for conducting tissue in 
that region. In a good many cases on the other hand, the petiolar 
system is entirely concentrated in the two lateral strands which 
directly supply the pinnae, the leaf-trace splitting into two shortly 
after leaving the stele (Fig. 55a), or even, in many of the more 
advanced types, leaving the stele as two separate lateral strands, 
so that the dorsal (abaxial) strand is not developed at all (Fig. 53a). 
The complications of the “horse-shoe type” of petiolar system, 
are, as we should expect, mainly elaborations of the lateral parts of 
the primitive strand. These elaborations fall under two heads. 
First is the incurving of the free ends of the leaf-trace arch , which 
begins in quite simple types (e.g., Gleichenia, Figs. 38a, 41a). In the 
extreme case the free ends become as it were spirally wound on 
each side within the general contour of the primitive curve. 
Matonia pectinata (Fig. 61y) shows a highly developed case of this 
elaboration, but it is also well developed in many Cyatheacese. 
Secondly we have the formation of lateral folds, i.e., of a bay in 
the centre of each side of the primitive curve. This modification, 
but slightly developed in the Hypolepis- type (Fig. 66, Ic), in which 
the bays are quite shallow, but in Pteris incisa var. integrifolia (Fig. 
65), and in Pteris aquilina (Fig. 76) the lateral folds are very deep. 
In many Cyatheaceas the two types of elaboration co-exist 
(Figs. 69, 97). 
It has been shown in previous lectures that both of these 
types of elaboration commonly occur in connexion with an internal 
cauline vascular system, which in some cases at least appears to be 
largely, though not primitively, a basipetal development into the 
stem of the complications of the petiolar system. 
It seems, though it is a point difficult to prove on the data at 
present available, that the elaborations of the petiolar system must 
have been developed in relation to an increase in the assimilating 
surface, either in the size or the number of the pinnse, or in both. 
This view is quite in accordance with the size of the fronds in the 
