bulbils on the shoots. These are vegetative processes which give rise to new individuals when they tumble to the ground.

It is of interest to consider the general characteristics of the members of the Fern tribe. As a rule the stem is either in the nature of a short underground process bearing a rosette of leaves, as in the case of the Male Fern and Hartstongue, or there is a horizontal stem more or less below the surface of the soil, such as is to be seen in the case of the Bracken Fern and the Polypody. Sometimes the stem assumes the proportions of a trunk, but these Tree Ferns only occur in the tropics. Where the stem of the Fern is upright it is properly termed a caudex, whilst in its horizontal form it is spoken of as a rhizome. There is actually some doubt as to the real nature of the frond of the Fern. Some botanists are inclined to believe that it is not really a leaf at all, but is a modified stem structure. Those who hold this view consider that the curious scaly structures so common amongst Ferns are really the leaves of the plant. Here the matter must be left on the present occasion, as it is proposed to use the terms leaf and frond as meaning the same thing.

An outstanding feature in the case of most Ferns is the remarkable manner in which the fronds are subdivided. In the case of the Male Fern it is seen that the upper part of the stalk, or rachis, as it is called, bears two rows of leaflets. These leaflets are properly referred to as pinnæ. When the leaflets are subdivided the divisions are spoken of as pinnules. These pinnules may be deeply lobed, and when this is the case each lobe is called a segment. In very large fronds the pinnules are again divided; the frond is then said to be tri-pinnate. Sometimes towards the top of the pinnæ or the frond the divisions become less pronounced; this character is designated pinnatifid. It should be noticed that the lower portion of the stalk, on which there are no pinnæ, is called a stipes. Of course in some cases, as with the Hartstongue, the leaf is quite undivided, without even any very pronounced indentations on the margin.

The unrolling of the Fern frond is a very beautiful process. Where the leaf is not divided in any way the process of expanding resembles the uncoiling of a watch-spring. Even where there are divisions the unrolling goes forward in the same manner with each subdivision, even down to the lobes. This particular mode of unfolding is called circinate. The texture of the leaves of Ferns is mostly thin and delicate, so that apart from some exceptions the foliage is not able to withstand the action of dry air. A notable feature with a large number of Ferns is the length of time which the leaves take to develop. The fronds of the Male Fern, for instance, start in the bud at least two years before they actually unfold. An examination will show that the roots of the Male Fern spring from the frond bases. It will be found that the position of the roots is the same in all Ferns.

With all Ferns the production of spores is confined to the leaves. In many instances there is no distinction between the fertile and the barren leaves. The stem does not start at once to produce leaves bearing the sporangia or spore cases. Thus, in the very young Fern the fronds are always barren; as the stem becomes older, fertile fronds are produced. In some cases the sporangia are borne on distinct leaves, as in the case of the Hard Fern, or on special parts of the leaves, in the manner to be seen in the Royal Fern. The difference in such cases is not really a very important distinction. A careful examination of the fertile portion of a Royal Fern frond will show a small amount of green tissue, or mesophyll, as it is called, at the lower portion of the pinnæ. Actually the fertile leaf, or part of a leaf, is similar to the barren portions, save that it produces a much reduced amount of green tissue or, in some cases, perhaps none at all.

In general appearance the Club Mosses bear a resemblance to the true Mosses, and hence the popular name, which is certainly rather misleading. With these plants the leaves are small and almost bristle-like, and are gathered closely round the stem. In many of the Club Mosses a large part of the stem lies closely along the ground, and from this at intervals roots are sent down into the soil and leafy shoots rise upwards. The sporangia are produced on special leaves, which are usually gathered together in the form of cones.

Although they vary somewhat in size, all the Horsetails are striking plants. Here there is a branching underground rhizome from which arise the aerial stems. The most distinctive feature of the plant are the whorls of smaller branches which arise from the joints of the main stem. These carry on the work which is usually assigned to the foliage of the average plant,—that is, the assimilating of carbon from the carbonic acid of the atmosphere. The real leaves of the Horsetail are much reduced in size, and take very little part in the work of nutrition. We shall find them at the joints of the stem as rings, each collection forming a kind of sheath. The leaves, which are usually of the same number as the branches, show no sign of their individuality, save in the little projecting teeth. In some species the fertile shoots, which appear in the form of cones, are produced specially. These appear in the spring before the ordinary vegetative growths, and are quite destitute of chlorophyll. In other species the normal green shoots are fertile at the termination. The sporangia are borne on curious scale-like leaves, a large number of which go to the making of a cone.

One or two aquatic plants, which belong to the Vascular Cryptogams, call for comment. The Pillwort is a singular plant not uncommon in damp situations. The leaves of this plant are narrow, and the spores are produced in curious rounded processes. The Water Fern (Azolla) is an introduced plant which sometimes grows abundantly on lakes in the South of England. Both the before-mentioned plants are allied to the Ferns. The Water Club Mosses (Isoëtes) are represented in this country by a species commonly known as the Quillwort. This plant grows in lakes, and is easily recognized by its quill-like foliage.

Owing to the large number of species a somewhat elaborate classification is necessary in the case of Ferns. In distinguishing the different families, the manner in which the collections of spore cases, known as sori, occur, as well as the features which the individual sporangia present, are important guides. The actual position of the sorus on the leaf, the presence or absence of a covering (indusium), are also distinctive features, both in the families and sub-families. When the individual sporangium is examined it is found that there is often present an annulus, a special ring of cells which plays an important part in the rupturing of the case. The extent of this ring or (as sometimes happens) its absence will alike be a decisive factor in fixing the family to which a species belongs. In some families a prominent feature is the fact that the sporangium has little or no stalk, although this is the exception rather than the rule. For a more complete description of the sporangium of the Fern the reader is referred to a succeeding chapter. It is certainly helpful to a study of these beautiful plants to try to fix in the mind the families, and their characters, of the order Filices. In all there are eight families belonging to the Fern tribe. These are given in the order in which they occur in technical books.

1. Hymenophyllaceæ.—The Filmy and Bristle Ferns. This family includes some of the simplest kinds of Ferns. There are only three representatives in the United Kingdom. These are Hymenophyllum tunbridgense, H. Wilsoni, and Trichomanes radicans. The two