The central axis is thus shown to be a cylinder composed of bundles of vessels, disposed in a radiating manner, and separated from each other by medullary rays; the whole constituting a ligneous zone resembling that of Sigillaria (see _Lign. 35_); but the inner circle of medullary tissue seen in the latter is altogether wanting. This difference is similar to that observable in the stems or branches of a dicotyledonous tree, in which the woody cylinder is associated internally with bundles of medullary tissue, and the roots of the same tree that are destitute of them. Part of a vascular bundle from the woody tissue of a Stigmaria, seen by a high power and transmitted light, is figured _Pl. V. fig. 6_; the smooth interspaces are composed of cellular tissue. [Illustration: Lign. 38. Erect Stem of a Sigillaria, with roots. _Coal Mine, near Liverpool._ _a._ The trunk of the tree, traversing a bed of Coal. _b._ The roots (_Stigmariæ_) spreading out in the Under-clay. ] [Sidenote: ERECT SIGILLARLA WITH ROOTS.] This opinion of M. Brongniart was confirmed by the discovery, in 1843, at St. Helen's, near Liverpool, of an upright trunk of a Sigillaria, nine feet high, with ten roots eight or nine feet in length, still attached, and extending in their natural position. _These roots are undoubted Stigmariæ of the usual species, S. ficoides; and the radicles, formerly considered leaves, are spread out in all directions, to the extent of several feet._[103] To the sagacity and perseverance of Mr. Binney, of Manchester, science is indebted for this important discovery; the same gentleman laid bare, on the floor of the mine at Dunkinfold, near Manchester, a large erect trunk of a Sigillaria, with numerous Stigmariæ roots. [103] From a communication to the British Association at Cork, 1843. In the Pictou coal-field of Cape Breton, in Nova Scotia, similar facts have been brought to light; the remarkable phenomena existing in that locality, of successive carboniferous deposits containing scores of erect trees with roots spreading into their native soil, presenting peculiar facilities for verifying the observations made in England. In an interesting memoir on the coal-fields of Nova Scotia, Mr. Richard Brown has given a detailed account of numerous examples of stems of Sigillariæ, and of Lepidodendra, (a tribe of gigantic club-mosses of which we shall presently treat,) with the roots attached; these roots, in every instance, had the characters and structure of Stigmariæ. In one instance, the stem of the tree was broken off close to the roots, and the hollow cylinder of bark was bent down and doubled over by the pressure of the surrounding mud, so as effectually to close up the aperture, and leave only a few irregular cicatrices converging near the apex; this fossil explains the true nature of the "dome-shaped" plant figured in the Fossil Flora, and in Dr. Buckland's Essay.[CZ] [104] See Pictorial Atlas, p. 200: and Petrifactions, pp. 37, 38.