OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 
463 
vascular tissues which prevail in the mineral charcoals so abundant in the coals of 
the Carboniferous rocks. 
All the specimens now described are from the Halifax beds ; but I have received 
from Mr. Butterworth a specimen from Oldham in which the bark, though in a 
very imperfect state, presents the essential features of the Halifax examples. 
The above descriptions show that we either have one very variable plant charac¬ 
terised by a bark in the middle of which are numerous large lacunae, separated from 
one another by vertical radiating cellular laminae, or that we have two or more distinct 
plants which have the same peculiar form of bark, but which differ in the organization 
of their central vasculo-cellular axes. I am inclined to regard the former of these 
alternatives as affording the true interpretation of these specimens. I think there 
can be no doubt respecting one conclusion suggested by the peculiar structure of the 
bark, viz.: that it indicates a plant of more or less aquatic habits. A similar structure 
is found in several flowering plants such as Myriophyllum and the petioles of Aponon- 
geton. It reappears in the Marsilece and Pilularia, amongst Rhizocarps, and it is not 
essentially dissimilar from that of the living Equisetums. That the plant was a 
Phanerogam is most improbable. It differs alike from the recent Equisetums and the 
fossil Calamites in the entire absence of nodal joints and medullary phragmata. The 
question suggests itself, may it possibly be a representative of the Marsileaceae ? 
I have examined many living species of Marsileci, and find that on making transverse 
sections of their rhizomes, not too far from their growing tips, we obtain results not 
dissimilar from those seen in fig. 7. We constantly obtain three different, but organi¬ 
cally united, sections. One of these is that of the main stem, in which a horseshoe¬ 
shaped vascular bundle approaches so nearly to a perfect circle as almost to enclose a 
central cellular axis. Another is that of the base of a leaf-petiole. In this the fibro- 
vascular bundle is V~ s h a P ec l as hi the secondary petioles of so many ferns. The third 
is a root-section in which the vascular bundle is central, and enclosed within those 
circular zones of cortical cells that are so common in the root-structures of Crypto¬ 
gams. The existence of three such dissimilar structures in one section of a stem sug¬ 
gests the possibility of a similar explanation being applicable to the plant under 
consideration. The absence of the exogenous zone from the recent Marsilece, whilst it 
i . . ' & . . . ... 
is conspicuously present in our fossil, does not militate against my suggestion, since m 
this, as well as in other features, the Astromyelon Williamsonis does not diverge more 
widely from the living Marsilece than the arborescent Lycopods of the Coal-measures 
do from their dwarfed living representatives, or than the Equisetums do from the 
Calamites. 
Fig. 15 represents a section of a small organism from Halifax, ’02 of an inch in 
diameter, for which I am indebted to Mr. Spencer. It may possibly belong to the 
plant just described, but this is doubtful. It consists of a central area,, which is 
divided by a crucial arrangement of cells, a, into four lacunae, b, The rest of the 
organism consists of a mass of parenchymatous disarranged cells of various shapes and 
3 o 2 
