OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 
321 
lenticular cell, whilst in others there are several cells arranged in a vertical series, 
the septa between these cells being square at right angles to their longer axes. The 
section represented in fig. 6 having been made at but a little distance from the 
medulla, we see at c one of the primary rays separating the primary wedges. It 
differs from the more numerous secondary ones in consisting of more than one vertical 
line of cells. As we approach the pith the number of such vertical series increases, 
whilst at the periphery these primary medullary rays become almost undistinguishable 
from the ordinary or secondary ones. 
Notwithstanding the large number of specimens of this plant contained in my 
cabinet, I never succeeded in finding a solitary fragment in which the bark is 
preserved. Mr. Butterwortii fortunately had better success. He met with the 
specimen represented in fig. 7. In this example the vascular zone is surrounded by a 
very thin layer of parenchyma (fig. 7, d), rarely displaying more than three or four 
cells in its thickness ; more generally there are but two. The structure is obviously 
little more than an epidermal layer. At fig. 7, d' we have an indication of the loose 
way in which this layer was attached to the vascular zone, and which probably 
explains the excessive infrequency with which it is found in its normal position. 
This specimen further indicates the existence of concentric lines, suggestive of 
successive interruptions to the continuous growth of the vascular zone. 
I have already observed that these stems were branched, such secondary branches 
ascending from the primary one at an oblique angle. In fig. 2 the section has 
intersected the base of one such branch in a very oblique manner. In the oblique 
transverse section, fig. 4, I endeavoured to incline the section so as to intersect the 
branch longitudinally. Though almost successful I was not completely so, since I did 
not quite trace the pith, a, of the branch into connexion with that (a) of the main 
axis. It was sufficiently so, however, to demonstrate the continuity of the one with the 
other. In fig. 2 these medullary cells are intersected at a near the base of the branch. 
I have obtained no clue whatever to the plants to which these branches belonged. 
It is clear that structurally the individual woody wedges exhibit no difference 
whatever from those of a Calamite. Neither transverse nor tangential sections of the 
one can be distinguished from similar sections of the other. Then- large primary 
medullary rays also exactly resemble those of Catamites. But the absence of 
articulations and the consequent absence of the alternation of the woody wedges in 
passing from one internode to another at once distinguish the two types. Further, 
the absence from Astromyelon of the remarkable internodal canals of Catamites, 
and also the difference in the medullary tissues of the two plants, constitute further 
distinctions between them. In Astromyelon the medulla was not fistular. The two 
exceptions, seen in figs. 4 and 5, are the only ones I have met with, and these may 
have been the result of desiccation rather than of a characteristic tendency to the 
formation of a fistular pith. That the growth of the vascular zone was exogenous is 
obvious enough. Though consisting wholly of barred vessels, the bundles are as open 
2 T 2 
