where stomata of peculiar form occur in the epidermis, while subepidermal strands of sclerenchyma occupy the ridges. In the cortical tissue beneath each furrow a wide intercellular space is present running the length of the inter node, and called the vallecular canal. The central cylinder of the root, in which there are several xylem and phloem strands, has around it a two-layered endodermis, the inner layer of which appears to take the place of a pericycle. The sporangia are borne upon lateral outgrowths of the axis (the sporangiophores), which arise in whorls and are associated in definite strobili or cones (fig. 3, C); at the base of the cone an outgrowth of the axis like a rudimentary leaf sheath (the annulus) is present, Each sporangiophore (fig. 3 D) consists of a stalk expanding into a peltate disk of hexagonal outline; from the inner surface of the latter six to nine large sporangia hang parallel with the stalk. The single vascular bundle supplies a branch to the base of each sporangium. The latter arises from a number of superficial cells, the cells destined to form the spores being derived from a single one of these. A tapetal layer is derived from the cells surrounding the sporogenous group, and the arrest of a number of the spore-mother-cells further contributes to the nourishment of the remainder, each of which gives rise to four spores. The outermost layer of the cell-wall of the ripe spore splits along spiral lines, giving rise to the elaters; these two long strips of wall, attached by their middle points to the spore, tend to straighten out in dry, and close round the spore in damp air. They thus assist in the opening of the sporangium, which takes place by a slit on its inner face. Further, several spores will be likely to germinate together owing to their elaters becoming entangled; a fact of some importance, since the antheridia and archegonia, though occurring sometimes on the same prothallus, are more often borne on separate individuals. The prothalli contain abundant chlorophyll, and are dorsiventral. Those that bear the antheridia are the smaller, and are either filamentous, or flattened, and irregularly lobed. The antheridia are deeply sunk in the tissue; the spermatozoids consist of a spiral of two or three coils, the numerous cilia being attached to the pointed anterior end. The female prothalli, which are sometimes branched, consist of a thick cushion bearing thin, erect lobes, at the base of which the archegonia are situated. The necks of the latter are short, the central series of cells consisting of ovum, ventral canal cell and one or two canal cells. The half of the embryo directed towards the archegonial neck gives rise to the apex of the stem and a sheath of three leaves, the other half to the small foot and the primary root. The first shoots are of limited growth, being replaced by lateral branches, which gradually acquire the number of leaf-teeth characteristic of the species.
Fossil species, some of which attained a great size, are known, to which the name Equisetites is given, since they appear to be closely allied to the existing forms. Two other extinct genera, Phyllotheca and Schizoneura, may be mentioned here. Abnormal specimens of Equisetum in which the strobilus is interrupted by whorls of leaves are of interest for comparison with the fructiiication of Phyllotheca. The most important and best known of the extinct Equisetales are, however, the Calamites (see Palaeobotany: Palaeozoic). In the primary structure of the stem the Calamites present many points of resemblance to Equisetum, but secondary thickening went on in both stem and root. These plants, which appear to have grown in swampy soil, thus attained the dimensions of considerable trees. The leaves, which were of simple form (except in Archaeocalamites, where they forked), were inserted in whorls at the nodes; they were either free from one another or cohered by their bases into a sheath. The branches alternated in position with the leaves, and sprang from just above the insertion of the latter. Some of the branches terminated in cones, which present a general similarity to those of Equisetum. This similarity is closest in Archaeocalamites, an ancient type found in Upper Devonian rocks; in this the strobilus consists of peltate sporangiophores inserted in whorls on the axis. In the other Calamarian strobili known the whorls of sporangiophores are separated by whorls of bracts. In some the sporangiophores stood midway between the sterile whorls, while in others they approached the whorl above or below. There is a close resemblance between these sporangiophores and those of Equisetum, but as a rule only four sporangia were borne on each. Some Calamites were heterosporous, sporangia with microspores and megaspores being found in the same cone.
Our knowledge of the extinct Equisetales, full as it is with respect to certain types, does not suffice for a strictly phylogenetic classification of the group. The usual subdivision is into Equisetaceae including Equisetum and Equisetites (with which Phyllotheca and Schizoneura may be provisionally associated), and Calamariaceae, including Calamites and Archaeocalamites.
II. Sphenophyllales.—The two very distinct genera Sphenophyllum and Cheirostrobus, included in this group, are known only from the Palaeozoic rocks. Though the high specialization of this ancient group of plants renders the determination of their natural affinities difficult, indications are afforded by anatomy and the morphology of the strobilus.
In general appearance the species of Sphenophyllum (the remains of Cheirostrobus known do not allow of any idea of its habit being formed) present some resemblances to the Equisetales. The long, sparingly branched stem bore at the somewhat swollen nodes whorls of six to eighteen wedge-shaped or linear leaves, which did not alternate in successive whorls. Both the broader and narrower leaves may be more or less deeply divided, and both forms may occur on the same shoot. From the relation of the thickness of the stem to its length it may be inferred that the shoots of Sphenophyllum derived support from adjoining plants. Without entering into detail regarding the anatomy, it may be stated that secondary thickening took place in both genera. The single stele in the stem consisted of the phloem surrounding a solid central strand of xylem, the groups of protoxylem being situated at the projecting angles. In Sphenophyllum, in which the transverse section of the xylem is triangular, there were three or six protoxylem groups; in Cheirostrobus they were more numerous. The anatomy of the stem is thus very unlike that characteristic of the Equisetales, and presents essential points of resemblance to the Lycopodiales and especially to the Psilotales. The general morphology of the cones, on the other hand, suggests some affinity with the Equisetales. The cone of Sphenophyllum consisted of an axis bearing at the nodes whorls of bracts, united below into a sheath. The overlapping bracts afforded protection to the sporangia, which were borne on sporangiophores springing from the upper surface of the coherent bracts near their origin from the axis; two sporangiophores usually arose from each bract, and sometimes adhered to its upper surface for some distance. Each bent round at the upper end, and bore one or two sporangia on the side turned towards the axis. The mature sporangium had a wall of a single layer of cells, which were larger towards the