ioo Oliver. — On the Structure , Development , 
mechanically. Here, in the regions between the leaf-bases 
(i.e. right and left, supposing the leaves to run anteriorly 
and posteriorly), the cell-walls of the cortex are strongly 
thickened, in striking contrast to the thin-walled cortical cells 
of the internodes. This, no doubt, is to resist the pressure 
and strain occasioned by the continuous rippling of the water 
against the leaves. Further, there is present in the hypodermal 
layer of cells a meristem (cl., Fig. 52), by the activity of which 
fresh cortical cells can be added to lend, if necessary, addi- 
tional strength. In Fig. 52, a portion of a cortex at a node is 
given. The thick-walled cells are strongly pitted where they 
abut upon one another, but this I have not indicated. 
It is only at the nodes that the axial bundle-cylinder is 
branched. Here there is an anastomosis of equivalent elements, 
and two bundles are given off (one on either side, Fig. 57) to 
the leaves. The bundle (/. tr.) running to a leaf passes hori- 
zontally through the cortex of the stem and divides into three 
bundles, which lie in the same horizontal plane. This division 
into three takes place close to the central bundle-cylinder. 
The three bundles run undivided up the petiole, and have 
their elements collaterally arranged. Directly the leaf-bundle 
enters the axial-cylinder it forks, and its elements anastomose 
with the groups A and B (see Fig. 57). Immediately above 
or below the point at which the bundles run in, the normal 
ring of vessels is seen. The bundle which runs to the axillary 
bud originates from the main leaf-trace bundle just before it 
divides into three (/. tr.). In the horizontal section, which 
shows the insertion of the leaf-bundles, adventitious roots may 
be seen originating (adv. rt .) ; these are formed from the layer 
of cells next below the endodermis, and pierce through the 
cortex. They usually lie dormant, however, in the upper 
floating nodes of the stem. 
The Root . — I have in no instance been able to observe the 
adult primary root; no doubt it is of little importance, being 
early superseded by numerous adventitious roots. A transverse 
section of one of these latter shows, as in the stem, a small 
central cylinder, surrounded by a wide cortex. The axial 
