there is only the one terminal, functional sporangium. The microspores are united by means of hardened protoplasm into one or more masses, while the solitary megaspores have a more or less complicated episporium.
|(Reduced. After Bischoff from Strasburger's Lehrbuch der Botanik.)|
|Fig. 10.—Salvinia natans.|
|A, From the side.B, From above.|
2. The Marsiliaceae also include two genera, Marsilia and Pilularia, the latter of which is found in Britain. The plants grow as a rule in marshy places, though some species of Marsilia are xerophytic. The creeping stem produces roots from the ventral surface and leaves from the dorsal surface; the leaves when young are circinately coiled. The leaves are simple and linear in Pilularia, but in Marsilia bear a pinnate four-lobed lamina. The highly specialized sporocarps are borne on the basal portions of the leaves, as a rule singly, but in some species of Marsilia in numbers. The development of the sporocarp shows that it corresponds to a pinna, although when mature it may appear to occupy a ventral position in relation to the vegetative portion of the leaf. It has a complicated structure in both genera; in Pilularia its shape is nearly spherical, while in Marsilia it is elongated and bean-shaped. The sori are developed in depressions and are thus protected within the resistant outer wall of the sporocarp. There are usually four sori in Pilularia, while in Marsilia they form two longitudinal rows. Each sorus includes both microsporangia, with numerous spores, and megasporangia, each of which contains a single megaspore with a complicated wall. Enclosed within the sporocarp they can endure a period of drought, but on the return of moist conditions are extruded from the sporocarp by the swelling of a special mucilaginous tissue and the spores become free. The development of the prothalli is in general similar to that of the Salviniaceae, though the resemblance may be homoplastic. The stem in the less reduced forms is solenostelic with sclerenchymatous ground tissue occupying the centre of the stele.
In the absence of direct evidence from Palaeobotany, and bearing in mind the modifications associated with adaptation to an aquatic life in other plants, the recognition of any more definite affinity for these heteros porous ferns than that indicated above appears to be inadvisable. Further evidence is necessary before they can be removed from such a position of convenience as is assigned to them here and placed in proper relation to the series of the Filicaceae.
The several phyla of Pteridophyta having now been briefly described, their relationship to one another remains for Phylogeny. consideration. The available evidence does not suffice to solve this question, although certain indications exist. In the earliest land vegetations of which we have any sufficient record specialized forms of Equisetales, Lycopodiales, Sphenophyllales and Filicales existed, so that we are reduced to hypotheses founded on the careful comparison of the recent and extinct members of these groups. In this connexion it may be pointed out that the fuller study of the extinct forms has as yet been of most use in emphasizing the difficulty of the questions at issue. It has thus led to a condition of uncertainty as regards the relationship of the great groups of Vascular Cryptogams, in which, however, lies the hope of an ultimate approach to a satisfactory solution. The study of the Sphenophyllales, however, as has been pointed out above, appears to indicate that the Equisetales and Lycopodiales may be traced back to a common ancestry. As to the relationship of the Filicales to the other phyla, evidence from extinct plants appears to be wanting. If, as has been suggested by Bower, the strobiloid types are relatively primitive, the large-leaved Pteridophyta must be supposed to have arisen early from such forms. The question cannot be discussed fully here, but enough has been said above to show that in the light of our present knowledge the main phyla of the Vascular Cryptogams cannot be placed in any serial relationship to one another.
It may even be regarded as an open question whether some of them may not have arisen independently and represent parallel lines of evolution from Bryophytic or Algal forms. This leads us to consider the question whether any indications exist as to the manner in which the Pteridophyta arose. It will be evident that no direct record of this evolution can be expected, and recourse must be had to hypotheses founded on the indirect evidence available. There appears to be no reason to doubt that the sexual generation is homologous with the thallus of a Liverwort, or of such an Alga as Coleochaete. It is with regard to the origin of the spore-bearing generation of the Pteridophyta that differences of opinion exist. This, though at first dependent on the prothallus, soon becomes independent. It may be regarded as derived from a wholly dependent sporogonium not unlike that of some of the simpler Bryophyta; the latter are assumed to have arisen from primitive Algal forms, in which, as the first step in the interpolation of the second generation in the life cycle, the, fertilized ovum gave rise to a group of swarm spores, each of which developed into a new sexual plant. On this view the origin of the sporophyte is looked for in the gradual development of sterile tissue in the generation arising from the fertilized ovum, and a consequent postponement of spore-formation. Certain green Algae (e.g. Oedogonium, Coleochaete), the Bryophyta, and the simpler Pteridophyta, such as Phylloglossum, have been regarded as illustrating the method of progression, though there is no reason to regard the existing forms as constituting an actual series. For a discussion of this view, which regards the alternation of generations in Pteridophytes as antithetic and the two generations as not homologous with one another, reference may be made to the works of Celakovsky and Bower. Although the antithetic theory is supported by many facts regarding the life-history and structure of the group of plants under consideration, it is quite possible that a stage in which the sporophyte was wholly dependent on the gametophyte may never have been passed through in their evolution. The spore-bearing generation may throughout its phylogenetic history have been independent at one part of its life, and have been derived by modification of individuals homologous with those of the sexual generation, and not by the progressive sterilization of a structure the whole of which was originally devoted to asexual reproduction. A number of facts regarding the Algae, and also those relating to such deviations from the normal life cycle as apogamy or apospory, may be regarded as lending support to this view, which, in contrast to the theory of antithetic alternation, has been called that of homologous alternation. Without entering further into the discussion of these alternative theories, for which the literature of the subject must be consulted, it may be pointed out that on the latter view the strobiloid forms of Pteridophyta would not necessarily be regarded as primitive relatively to the large-leaved forms, and also that the early stages of the origin of the sporophyte in the two cases may have proceeded on different lines.
Another question of great interest, which can only be touched upon here and may fitly close the consideration of this division of the Vegetable Kingdom, concerns the evidence as to the derivation of higher groups from the Pteridophyta. The most important positive evidence on this point indicates that the most ancient Gymnosperms were derived from the Filicales rather than from any other phylum of the Vascular Cryptogams. Extinct forms are known intermediate between the Ferns and the Cycads, and a number of these have been shown to bear seeds and must be classed as Pteridospermae. These forms will, however, be found discussed in the articles treating of extinct plants and the Gymnosperms, but their recognition will serve