_ central cylindrical column (connective), along the antero-lateral aspects of which are at tached two larger cellular masses; the outer surface of each is marked by a vertical furrow, indicating its division into two halves, which are the rudirnents of the future kezili. In each half a single axile vertical column of cells soon becomes distinguished from those surrounding them by their greater size and granular contents. In each of these cells the nucleus disappears, and is replaced by two others, this being followed by a division of the cell contents (primordial membrane), which results in the formation of a new cell round each nucleus. By the repeti tion of this process a mass of cells — the parent-cells of the spores—is fonned, which occupies the centre of each rudimentary loculus. The next change observed is the thickening of the walls of the parent-cells by gelatinous deposition on the interior surfaces. This is followed in all of them by disappear ance of the nucleus, and consequent division of the contents of the cell (primordial mem brane) into two portions, each surrounding a new nucleus. These, however, are only trans itory formations, and are soon succeeded by four permanent nuclei, which are placed towards the four angles of a regular tetrahe dron, each invested with a primordial sac con taining a granular mucus, on the surface of which is soon secreted a gelatinous layer. In this tnanner the parent cell is divided into four compartments —the so-called special parent cells of the pollen grains. Within each com partment is now formed a new cellulose mem brane on the surface of the primordial utricle. This is transformed into a resistant and co loured tegument, which is the outer mem brane of the pollen grain, and exhibits various projections of its surface, which differ ac cording to the species.
109. While these changes are taking place in the central mass of each loculus, the tissue forming its wall is transformed into a capsule of three distinct cellular layers. The inner layer consists of radiating prismatic cells, and is soon absorbed. The cells of the second layer are distinguished by their containing at first numerous starch granules, and afterwards by the deposition of spiral fibres on the inner surfaces of their walls. These are usually dice-shaped cells arranged in concentric layers. The external or epidermic layer consists of tabular cells in contact by their edges.* 110. Review of the analogies which present themselves in the history of the development of the reproductive organs of the higher Cryptoganzia and of the Phanerogamia.—The families in question are distinguished by the presence of what is called " sexual reproduction" from all others. It is true that among the Characem, Conju gatm, Vaucheriacem, and Desmidem, the con currence of two dissimilar parts is neeessary for the development of the germ ; but in them the phenomena do not present themselves in such strict conformity to law, and the anato mical relations of the germ to the organ which contains it are not nearly so complicated as in-the plants under our consideration. Taking the sexual germ as our starting point, in com paring the history of the development of the phanerogamous with that of the cryptogamous plant, the following analogies present them selves :— 111.-1. Analogies exWing-between the ovule, the anther, and the sporangiunz — In Zostera marina the termination of a stem destined to bear reproductive organs, is broadened out in the form of a spatula, concave on one side, convex on the other. On the concave sur face are observed, early in the development, two vertical series of papillm—one on each side of the middle line— which are the rudi ments of the organs which support the ovules and anthers. In each series, the two kinds of
rudiments are arranged alternately in such a manner, that an ovule in one series is always on the same level with an anther in the other. The rudimentary organ which is destined to contain the ovule, commences as an imperfect ring of cellular tissue, on the inside of which is seen a little round projection —a bud in the axil of a leaf. From this projection is developed the ovule, with its teguments, just as described in Orchis Morio. The axis of its nucleus is occupied by a vertical series of cells. Of these the uppermost enlarges, and becomes detached from its neighbours, so as to form the embryo-sac:I- If we compare this process with what occurs in Selaginella, we find in each case, a cell belonging to the stem in the axil of a modified leaf, which transforms itself into an axial organ. In each case one of the central cells enlarges and becomes dctached —in Selaginella, to form the mother cell of four spores—in the phanerogamous plant, to become the embryo-sac.
112. The exact correspondence, step for step, which exists between the development of the anther, and that of the sporangium, will be best seen by successively comparing the de scriptions contained in § 89 and § 108. It is rendered still more striking when we consider the very remarkable variations which present thernselves in the structure of the anther among the Phanerogamia themselves ; as e . g. in Zostera, among the Orchidacem, and other examples for the description of which space is wanting. The contemplation of these analo gies leads us to remark how little relation there seems to be as respects the organs under con sideration between the morphological import of the rudiment and its development. The ovule of Zostera is an axial organ, originating in the axil of a modified leaf ; its analogue in development, the anther, is itself a bilateral foliar organ. The sporangium of Equisetum seems to originate as a leaf,— that of Selagi nella, as an axis in the axil of a leaf.
Analogy between the embryo-sac, the pollen cell, and the parent cell of four spores — In approaching this, the most difficult part of our inquiry, we must refer to the Coniferm, as holding in so many respects an intermediate position. Of those stages of the development which precede the act of impregnation in Se laginella, the first, namely, the division of the parent cell into four compartments, and the formation ofa spore in each,is entirely wanting in the Coniferm. The prothallium— under standing by the term the organ of which the archegonia form a part—is represented by the corpuscula, between which and the archegonia, the resemblance in structure is very striking.— The difference in the mode of origin of the germ-cell, on the other hand, is no less re markable. " Among the Cryptogamia there is," says Ilofmeister, " only one germ-cell which completely fills the central cell of the arche gonium, while in the Coniferm, very numerous germ-cells swim in the central cell of the cor pusculum, of which one only, applied against its lower end, is fecundated." In the gymno spermous Phanerogamia, all the steps of de velopment which intervene between the parent cell and the germ, disappear ; the latter origi nating altogether independently at the upper end of the embryo-sac. As the transforma tion of the germ-cell is the most important element in the process of development, it presents the greatest degree of constancy. It always commences by the formation of one or more septa, the direction of which, in rela tion to that of the first axis of growth, is transverse or nearly so.