(&) Botryopterideae.—Botryopteris cylindrica has a long thin stem 2-5cm. in diameter bearing leaves at wide intervals and branching dichotomously but other plants belonging to the same family show considerable differences in habit. There is a simple stele consisting of a solid strand of tracheids with the smaller and first differentiated elements in the centre. The petiole is at its base very like the stem in structure. The sporangia differed from those of the Zygopterideae in being small and pear-shaped with a multiseriate annulus on one side only. The other members of the group have a more complex structure. The similarity in structure between the petiole and the stem of Botryopteris cylindrica is a primitive character and fits in with the suggestion that the large type of leaf or frond is derived from a subordinated lateral branch of the shoot. The Botryopterideae as a group show several points of comparison with the living Osmundaceae, repre sentatives of which family are found in the Permian. It is also possible that the Ophioglossaceae have descended from the same stock.
Osmundaceae.—Petrified stems from the Permian are known which give satisfactory proof of the existence of plants closely agreeing in many anatomical details with Osmunda (royal fern). In Thamnopteris the woody cylinder was solid but the central part was like that in the Zygopterideae made up of parenchyma and small tracheids. There is evidence to be drawn from closely re lated fossils that the type of stele found in Osmunda has been derived in the course of descent from a solid stele such as that found in Botryopteris, Zygopteris and Thamnopteris suggesting intermediate steps in the process. Like Osmunda these Permian fore-runners had short stems with densely crowded fronds and adventitious roots. Fructifications of Palaeozoic Osmundaceae have not yet been recognized with certainty but Discopteris, an Upper Carboniferous form, has perhaps some claim to recognition.
Marattiaceae.—Petrified stems have been found which in struc ture are so very like those of the Marattiaceae that many bot anists are inclined to regard them as proof of the existence of that group in the Palaeozoic. Psaronius, a form-genus of tree ferns ranges from the base of the Upper Carboniferous upwards and is quite common in the Permian. The stem has a complex vascular system, a polycylic dictyostele, except in Psaronius Renaultii, the Carboniferous species which has a colenostele. In these respects a close comparison is possible with the living Marattiaceae. In some species of Psaronius the fronds were attached alternately on opposite sides of the tree so that there were two vertical rows. In others there were several vertical
series and as many as four fronds might be attached at the same height from the ground. There was no secondary thickening but the stems were buttressed by a considerable development of ad ventitious roots forming a fibrous investment round them and giving them considerable rigidity. Several types of fertile fronds of Carboniferous plants are known which, though probably for the most part Pteridosperms, are fern-like and bear microsporangia grouped in synangia somewhat like those of living Marattiaceous ferns. Asterotheca has large multipinnate fronds which in the sterile condition would be grouped in the form-genus Pecopteris. The sporangia are united in synangia (fig. 13 A and A') which are attached to small protuberances on the surface of the frond. In Scolecopteris the synangium has a short stalk with a central vascular strand. In the closely allied Acitheca the synangium has no stalk but a vascular strand is present in the central column of tissue round which the sporangia are joined and each sporan gium terminates in a sharp bristle. In Ptychocarpus the sporangia are more intimately united and not so pointed. Danaeites an Upper Carboniferous type has long linear synangia, very like those of the living Danaea, with two rows of about to closely packed sporangia each of which opened by a terminal pore. In view of the fact in the Palaeozoic is also slight ; Oligocarpia, an Upper Carbon iferous fern, shows in the form and grouping of sporangia close resemblance to the living Gleichenia. The fact that the annulus is sometimes formed of more than one series of cells led Kidston to doubt its relationship to the Gleicheniaceae ; but sporangia are sometimes found with more than one series in Gleichenia so that this objection is not completely justified. Chansitheca from the that some fructifications, known to belong to Pteridosperms, were in the form of synangia it is as yet uncertain whether most of the plants included in Asterotheca, Scolecopteris, etc., are ferns or pteridosperms. The evidence for the existence of Marattiaceae in the Palaeozoic rests primarily on the Marattiaceous structure of the Psaronius stems.
Schizaeceae. The sporangia of Senftenbergia (fig. 13, C), an Upper Carboniferous genus, like those of the living Schizaea have an apical group of thick-walled cells constituting an annulus. An apparently closely allied genus Klukia of Jurassic age would seem to form a link between Senftenbergia and Schizaea as regards the sporangial structure.