938 Leer — Notes on the Anatomy and Morphology of 
We have here then a connexion established between the medullary bun- 
dles of the stem and the internal phloem groups of the petiole ; a con- 
nexion which very strongly upholds Mr. Worsdell’s opinion that internal 
phloem groups represent medullary bundles which have lost their xylem. 1 
On the lower side of the arc-shaped bundle is an irregular layer of 
cells similar in appearance to the cells of the ground tissue, but contain- 
ing starch in large grains (c, Text-fig. 3). This layer forms a definite 
sheath on the lower side of the vascular bundle. Much smaller starch 
grains may be present in the other cells of the ground tissue, but in no 
definite layer or group of cells. The ground tissue consists of loose spongy 
parenchyma. There are a few scattered laticiferous cells running up the 
petiole. 
As the arc-shaped bundle of the petiole passes up the midrib it gives 
off successive branches from its free edges ; these branches pass off into 
the lamina. In transverse section the structure of the midrib is seen to be 
the same as that of the petiole, except that the vascular bundle is smaller 
owing to its branching. 
4 . The Morphology of the Protuberances and Spines. 
The morphology of the spines and protuberances is not clear. Little 
morphological importance can be attached to the arrangement of the 
vascular tissue of the spine, which might occur in either a stipule or a 
stem. Thus the stipules of Acacia have a structure similar to that seen 
in these spines. The course of the vascular bundles in the protuberance 
has been already described, but it is advisable to restate the facts briefly 
before discussing the morphology. Several vascular bundles leave the 
ring and run out through the protuberance in the form of a hollow 
cylinder, which later divides into two portions, the inflorescence supply 
and the leaf-trace (Text-fig. 4). Running up through the tissue enclosed 
by the hollow cylinder of ring bundles are a number of medullary bun- 
dles. A few of these join on to the cylinder of ring bundles, rarely 
branching off again, but the majority run straight up and separate into 
three portions, which supply the three spines. We see, therefore, that while 
the vascular supply of the inflorescence and leaf is derived from the bundles 
of the ring, together with some medullary bundles, that of the spines 
is derived entirely from the medullary bundles. It is known that medullary 
bundles generally originate from leaf-trace bundles which, after running 
down one or more internodes as part of the normal ring, turn into the pith. 2 
The material available has not been sufficient to determine whether or not 
the medullary bundles of Pachypodium originate in this way. 
1 I wish here to express my thanks to Mr. Worsdell for his valuable suggestions and assistance 
in this question. 
2 Scott and Brebner (4). 
