378 DR. D. H. SCOTT ON BOTRYCHIOXYLON PARADOXUM. 
observed has a diameter of about 0:4 mm.—roughly half that of the strand under 
discussion. It may be added that there is no clear sign of any cortex belonging to the 
strand, such as should be present if the organ were an endogenous one, like an adventitious 
root. Further, the marked curvature of the primary xylem-plate is not to be expected 
in a root. None of these arguments are, however, in themselves conclusive. 
So far as structure is concerned, the strand in question might quite well belong to a 
branch, comparable to the axillary branches of Ankyropteris scandens or Grayi. For 
example, the axillary stele from the latter species, figured in a recent paper (Scott, '19, 
pl. ii. phot. 10), agrees quite well, allowing for the difference between primary and 
secondary wood. The branch interpretation is, however, ruled out of court by the fact 
that the branching in Botrychioxylon was dichotomous (Pl. 39. fig. 10), the bifurcation 
taking place immediately above the level of the problematic strand. We cannot suppose 
that two different kinds of branching took place in the same stem within a few milli- 
metres of each other. 
There remains the leaf-trace interpretation. This finds support, first, in the mode of 
departure of the strand from the stele, shown in the section next below that which 
contains the strand in the cortex. Here, as we have seen, the structure is as nearly as 
possible the same as that shown at a corresponding stage of the emission of a leaf-trace 
in Ankyropteris corrugata ; it may also be compared in this respect with J/etaclepsy- 
dropsis duplex*. Another important argument is derived from the presence of a petiole 
in the next transverse section above that showing the strand (Pl. 37. fig. 4). The region 
of bifurcation shown in longitudinal section in Pl. 39. fig. 10 intervenes, but the vertical 
distance is small, the height of the intermediate piece from which the longitudinal 
sections were cut being only 3 mm. The petiole is on the same side of the stem as the 
strand, the oblique branch of the bifurcation having gone off on the opposite side. The 
petiole is not in continuity with the stem, but there can be no reasonable doubt that it 
belongs to it; the structure of the tissues is the same, except that the petiole, as 
we should expect, has a stronger mechanical construction, the middle cortex being 
differentiated as a somewhat sclerotic zone (Pl. 39. figs. 11 & 12). The appendage 
borne by the petiole is identical with the large aphlebie which occur on the stem and 
are described below. The petiole contains a singular vascular bundle resembling the 
Dineuron type (see below, p. 379). | 
There is a strong presumption that the petiole is the organ to which the strand shown 
in the lower section belongs, and that the strand is therefore a leaf-trace which becomes 
the petiolar bundle higher up in its course. The curvature of the primary xylem, with 
concavity outwards, is characteristic of Zygopteridian leaf-traces. Taking all the facts 
into consideration, I believe that this is the right interpretation, and that the strand in 
question is a leaf-trace with secondary wood T. On this view it is perfectly comparable to 
the stele of the stem; in both, only the central part of the xylem retains a primary 
character ; the outer zone has been formed by secondary growth and betrays its cambial 
origin in the radial seriation of its elements. In both, the considerable secondary 
* Gordon, '11*, pp. 174 & 183, figs. 7-16. 
T An apparently similar strand is incompletely shown in the tangential section of the cortex (2463). 
