490 
PROFESSOR W. C. WILLIAMSON ON THE ORGANIZATION 
through the longitudinal canal (e). As in the other figures, a is the fistular cavity, 
h the medullary cells, g the vessels of the woody wedges which, in fig. 24, are seen 
to pursue their usual arching course as they cross the node (*’). At e we have the 
longitudinal canals of the superior of the two internodes intersected in each of the 
sections. 
In fig. 24 this canal passes out of the line of the section, but in fig. 23 it dilates, at 
the node, into one of the large triangular spaces seen in fig. 22. These, and similar 
sections which I possess from the same specimen, seem to indicate that, in it, the cellular 
medullary nodal diaphragm was not complete, but that the various internodal fistular 
cavities communicated freely with each other. 
The most important feature of these sections is seen in fig. 23, Z, where we find that 
the longitudinal fissures seen in fig. 22, Z, are the inlets to well-marked oblong passages 
filled with the same amorphous matrix as occupies the fistular cavities of the pith, and 
which are thus proved to have been open canals when the plant was entombed. In the 
memoir on Calamopitus already referred to, I entered at considerable length into the 
history of these canals, applying to them the name of verticillate medullary radii. But 
at the time when that memoir was written my imperfect specimens had not enabled me 
to discover either the fistular character of these calamitean stems or the peculiar process 
of medullary absorption which has taken place in them. Hence I stated that “ these 
radii appear to have been composed of the same tissue as the medulla itself, judging 
from the circumstance that the inorganic material with which they are filled is identical 
with that replacing the pith. They have most probably united the pith with the bark” 
( loc . cit. p. 163). It is now evident that the account which I then gave of the position 
of these curious rays in relation to the rest of the tissues, and which I further illustrated 
by a diagrammatic figure ( loc . cit. tab. 5. fig. 17), is in all respects correct, with the 
exception that they prove to have been canals formed, first by a rupture, and afterwards 
by an absorption of the cells of the primary medullary rays, and were not special pro- 
longations of the medullary parenchyma through those rays, as I deemed probable. 
Various attempts have been made to show that these canals originally transmitted special 
vascular bundles either to roots or to branches, which vessels were supposed to have 
become decomposed, admitting the inorganic matrix into the cavities formed by their 
disappearance. But all my specimens negative this idea. In the first place, these canals 
are not planted upon the node from which such branches would spring, but a little 
below it ; whilst immediately above them, and in their right place, directly over the 
node, we find the true lenticular bundles going to such branches, springing from the 
innermost vessels of the ligneous zone. This is demonstrated in fig. 25, which represents 
a tangential section like fig. 22, only passing through the external instead of the internal 
surface of the woody zone ; c, c indicate cellular primary medullary rays, f the woody 
wedges, i, i the node, with a verticil of lenticular structures (m) marking the position of 
the branches, whilst at l, passing through the cellular tissue , below and altogether dis- 
connected from the vascular structures , are the disputed organs. The sketch (fig. 25) 
