LEPIDODENDRON. 75 



disappears, and is succeeded at first by indistinct holes of a circular form, or projecting 

 circular knobs, which soon pass into distinct areolae with an elevated point in their centre, 

 and radicles radiating from the root in all directions. When this state of the root is 

 attained, the gradual decrease in bulk, which has gone on from the base of the Sigillaria 

 to this point, ceases, and the Stigmaria root runs to the length of twenty feet and upwards 

 with about the same size. Its absolute termination is not often seen ; but, from specimens 

 found, it is generally believed to have an obtuse club-shaped end. Certainly I have 

 never seen it dichotomize into smaller roots, although other observers may have been more 

 fortunate. 



Professor Williamson, in his ' Observations' previously quoted, when speaking of the 

 structure of the medulla of Stigmaria, says that it possessed " a cellular pith without any 

 trace of a distinct outer zone of medullary vessels, such as is universal amongst the Lepido- 

 dendra." I have no doubt that he has found in numerous small stems, obtained from the 

 "Upper Foot Coal" near Oldham (a locality which I first observed, and pointed out to others, 

 but from which time I have never been able to obtain any more specimens for myself), 

 evidence of the cellular pith he alludes to. This cellular tissue is not the ordinary paren- 

 chyma usually forming piths, but the " orthosenchymatous" tissue common as a medulla 

 to Lepidodendron, Halonia, and Calamodendron (?). This structure (in the locality above 

 named) is found to prevail in many stems possessing radiating woody cylinders of barred 

 tubes, penetrated by medullary rays, and varying in diameter from one twentieth of an 

 inch to two inches. All these specimens, from the similarity of their structure, might be 

 taken to be Stigmaria ; but we seldom find their external characters so well preserved as 

 to recognise them as identical with the large roots found connected with Sigillaria stems. 

 From specimens in my possession it is almost certain that there are fossil roots, having all 

 the outward characters of Stigmaria, some of which have their medulla composed of the 

 cellular tissue alluded to by Professor Williamson ; whilst others have a medulla com- 

 posed of a combination of transversely barred vessels and cellular tissues, or what I have 

 ventured to term " orthosenchymatous" tissue, as shown in my specimens of Sigillaria 

 vascularis and the Stigmaria figured and described by Professor Geoppert. It is also 

 probable that these roots may have belonged to distinct, but allied, trees. These two 

 different kinds of medulla appear to have existed in Lepidodendroid trees ; the former in 

 Lepidodendron Harcourtii, and the latter in my Lepidodendron vasculare. 



It is scarcely necessary here to allude to the great number of genera and species of 

 fossil plants which have been erected on imperfect and ill-preserved specimens, only 

 exhibiting some of their external characters; but even amongst these, few persons could 

 determine a compressed Stigmaria jicoides from a Halonia regularis, except where the 

 latter was observed to dichotomize. When we examine the structure of the vascular 

 bundles traversing the stem from the central woody axis to the leaves or rootlets, we have 

 difficulty in distinguishing those which belong to Sigillaria and Lepidodendron from those 

 of Stigmaria and Halonia, only the latter are considerably larger. In fact the Stigmaria, 



