408 Scott . — On Peduncle of Cycadaceae. 
features of interest. They resemble the sporophyll- traces in 
the fact that each strand, after branching off from the main 
ring, often divides into two branches, which reunite higher 
up 1 . In many cases these cortical bundles are altogether 
cauline, not passing out into any kind of appendage. Their 
course has been carefully followed in series of successive 
sections, from several peduncles, and it has been proved that 
they sometimes end blindly in the cortical parenchyma, and 
sometimes turn inwards again, to reunite with the main 
vascular bundles of the stele. Probably all the cortical 
strands, which occur in the lower part of the peduncle, are of 
this cauline nature. They possibly represent the vestiges of 
a former leaf-trace system, which, now that the greater part 
of the peduncle is naked, no longer has any leaves to supply. 
Concentric Bundles. In some cases the cortical strands 
show a typically concentric structure, which they retain for 
a considerable distance. The longitudinal course of these 
concentric bundles has been followed in several instances, and 
one of them was traced through the greater part of the length 
of a peduncle. Low down, in the region of irregular 
orientation, a horse-shoe bundle was seen leaving the main 
ring. As it ascended, the curvature increased, until the 
structure became truly concentric, with xylem inside and 
phloem outside. The bundle remained concentric, but 
diminished in size, until, in the upper, regular, part of the 
peduncle, it turned in again, coming into contact with one of 
the main bundles, where it once more increased in bulk, but 
retained perfectly concentric structure. 
The transverse sections of two such bundles (from different 
specimens) are shown in Figs. 8 and 9, from which their strictly 
concentric structure, with central protoxylem, is evident. 
The parenchyma in the centre of the strand is no doubt 
homologous with the circummedullary tissue accompanying 
the normal bundles. 
These concentric cortical strands are of very inconstant 
1 Cf Solms-Laubach, l.c. p. 213. 
