OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 
237 
referred to. It will be observed that some of these cells are nearly cubical in shape, 
others more elongated ; some have square ends, others oblique ones ; but it is important 
to notice that towards the exterior of this cellular mass ( b ') the cells exhibit a strong 
tendency to become prosenchymatous. All these cells, where mineralization has not 
altered their structure, are more or less regularly barred. Where the convex inner 
extremity of each fasciculus of the woody zone encroaches upon the medullary sheath, 
this cellular layer almost disappears, though not altogether so. The large lenticular 
radii to which allusion has been made take their rise in this cellular tissue. My speci- 
mens show that the longer axis of each cell becomes suddenly deflected in the horizontal 
direction. That such is the case is shown, not only by their general aspect, but by the 
reversal of the direction of their transverse bars, which are now vertical, and not hori- 
zontal as before. Many of these deflected cells are perfectly muriform, but others are 
more or less prosenchymatous. In the immediate neighbourhood of the cellular tract 
there is a considerable disturbance of the parallelism of the small contiguous barred 
vessels, so that the origin of such of the latter as contribute to the formation of the 
foliar bundle is not easily traced; but, however originated, some of them accompany the 
deflected cells to constitute that bundle. In no case do any of the inner and larger 
vessels of the medullary sheath take any part in the formation of these bundles ; and my 
present impression is that all those which do so should rather be regarded as belonging 
to the innermost part of the woody zone than to the exterior of the medullary cylinder*. 
Whichever is the fact, I am convinced that these vessels are the exact equivalents of 
those furnishing the foliar bundles in the true Lepidodendra. These bundles were 
needed, in the very earliest stage of the growth of the young shoot, to sustain the deve- 
loping leaves ; and though at this stage of its development the woody zone was obviously 
represented in a very feeble manner, it nevertheless fulfilled its functions in contributing 
its quota to the foliar nutrition. But there remains to be explained the supposed absence 
of true medullary rays mentioned by Corda as characterizing his Diploocylon , but which 
were observed by Brongniart in his Sigillaria elegans. None of these writers were 
aware of the existence of barred or scalariform cells in tbe medulla; of these plants. 
Consequently when Corda found barred tissues running horizontally, not only in the 
large lenticular spaces separating the ligneous fasciculi, but also in the smaller ones 
separating individual laminae, he concluded that all these were necessarily bundles of 
barred vessels , and in consequence denied the existence of medullary rays. Since, how- 
ever, all the medullary cells of many of these Lepidodendroid plants (see figs. 1 & 3) are 
barred, it follows that some of those in other portions of the ligneous zone would, in 
all probability, be the same ; and such proves to be the case. In the example which I 
am now describing it is difficult in some places to say which are sections of fusiform 
cells, and which of parts of contorted vessels; but in a large number of specimens 
* Later researches amongst tire Burntisland plants have enabled me to clear up this very obscure point, and 
to determine that the vessels in question do belong to the outermost surface of the medullary cylinder. See 
Proceedings of Royal Society, vol. xx. p. 199. — May 7th, 1872. 
MDCCCLXXII. 2 1 
