ORGANIZATION OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 751 
sufficiently good to enable us to decide, with certainty, whether the bundles in the 
outer part of their course are collateral or concentric. In the original Burntisland 
specimens a certain amount of tissue has usually perished all round the xylem of the 
bundles in the cortex, and there is no means of telling whether the whole of this 
delicate tissue consisted of phloém. The Dulesgate specimens are somewhat better 
preserved in this part, so that in some cases it is possible to see that the phloém of 
the cortical leaf-traces is limited to the external side of the xylem.* 
We may, therefore, regard it as probable that the collateral structure shown in 
fig. 26 was maintained throughout the course of the bundle in traversing the stem. 
Although the leaf-trace bundles sometimes became somewhat lobed (as seen in 
transverse section) in passing through the cortex, yet we have no clear evidence that 
they ever divided into twin bundles as in Lyginodendron. 
Our knowledge of the primary structure of the stele and vascular bundles of H. 
Grievit may be summed up as follows: 
1. The stele consisted of a central mass of primary xylem (made up of tracheides 
and conjunctive parenchyma) surrounded by a zone of phloém. The whole was 
enclosed by a well-marked pericycle. 
2. The primary wood may be divided into the peripheral xylem-strands, which are 
continuous with the leaf-trace bundles, and the metaxylem. Each of the peripheral 
strands had the same mesarch structure as the perimedullary strands of Lygino- 
dendron, or as the xylem of the foliar bundles in Cycadee. 
3. The leaf-trace bundles were collateral in structure on leaving the central cylinder 
and probably remained so in passing through the cortex. 
4, The primary tracheides, with the exception of the protoxylem and adjoining 
elements, possessed numerous bordered pits. 
4, The Secondary Tissues. 
It is not uncommon to find stems of H. Grievii which retain the primary structure 
unaltered, such as the stem represented in Plate 26, fig. 21, where the main axis is 
quite without any secondary tissues, though, curiously enough, they are present to a 
small extent in the branch. The majority of the stems, however, have a secondary 
zone, which is often of very unequal thickness on different sides of the same stem, 
and is on the whole less developed than in Lyginodendron. The secondary wood is 
very sharply marked off from the primary ; this is owing partly to the regular radial 
arrangement of the secondary elements and partly to their smaller size (see Plate 26, 
fir, 24; also Wituramson, “Organization,” Part IV., Plate 28, fig. 30). In cases 
where the secondary wood attained any considerable thickness, the size of its elements 
increased towards the exterior. 
* See especially the four transverse sections O.N. 1915-1915 C. 
