structure that a zoologist has not so difficult a task before him in restoring, even from imperfect materials, the general aspect of an extinct animal.

I state these difficulties which face us at the very threshold of our investigations, not to magnify the work before us this evening, but to account for the comparatively little progress that has been made in the interpretation of extinct floras, and for the great diversity of opinion that exists among botanists as to the systematic position of numerous fossil plants; and further to account for the very large number of genera and species which have been established on imperfect and fragmentary materials, the systematic position of which is consequently indeterminable.

The progressive accumulation of observations, and the more careful preservation of instructive specimens in local and private museums are supplying the means of dealing with fossil botany after a different method. The most important recent advances in this science have been made in uniting the separate fragments,-roots and stems, leaves and fruits,-described under different names and placed in different and often widely separated genera, so as to build up vegetable individuals, the systematic position and affinities of which can be understood.

These observations are specially true in regard to the vegetation of the Coal Period. Little information has been obtained from the vast stores of the carbonized remains of the plants of this period which are ever being brought under the inspection of man in the form of coal, for this material is so completely altered as to be almost destitute of structure. The best preserved plants occur in the beds of shale which accompany the coal, or are obtained from earthy nodules in the coal itself, which injure its marketable value, and are consequently got rid of by the miners.

We may at once set aside that great division of the vegetable kingdom with which we are most familiar, comprising all plants that have true flowers and seeds, and confine our attention to the more obscure cryptogamous plants which are destitute of flowers, and for seeds have bodies of much simpler structure called spores. The cryptogams are either wholly cellular in their composition, like the mosses and sea-weeds, or they are composed partly of cells and partly of vessels, like the ferns and club-mosses.

If we except some supposed Alga, no traces of true cellular plants have been hitherto detected in the Coal-measures. The long-continued maceration to which the coal plants were subjected when the beds composed of their remains were forming on the surface of the earth, and the subsequent changes they have undergone, have reduced to one common structureless mass the varied vegetation of which the coal is composed. One of the first results of these operations would be the disappearance of the cellular plants, which under the then existing very favourable conditions must have abounded; just as the soft cellular parts are almost always destroyed of those specimens which have been so favourably situated as to have their vascular tissue preserved.

Excluding then the cellular cryptogams, we may shortly consider the classification and structure of the vascular forms. They are divided into four groups, all of which are represented in the indigenous Flora of Britain.

I. Ferns (Filices). Polypody, Brake, Spleenwort, etc. II. Horse-tails (Equisetacea). Horse-tail.

III. Club-mosses (Lycopodiaceae). Club-moss and Quill-wort. IV. Pill-worts (Marsileaceae). Pill-wort

I. The FERNS have a rhizome which creeps below or upon the surface of the ground, or rises into the air like the trunk of a tree. This trunk in some species attains a great height; it is nearly uniform in diameter throughout its whole length, and is covered with the symmetrical and regularly-arranged markings of the stalks of the old leaves. Internally it is composed of a central cellular pith surrounded by a cylinder of scalariform tissue, and this is invested by a cortical cellular layer or bark.

The woody cylinder is composed of simultaneous vascular bundles, which originate and are completely developed at the same time; there is consequently no addition to it from subsequent growth. It is penetrated by large open meshes, each of which permits the passage of the vascular bundles that supply a leaf, accompanied with a certain amount of cellular tissue from the medulla which occupies the centre of the mesh.

The leaves, which are very variable in size and form, not only perform the functions of ordinary leaves, but also bear the fruit, and are hence called fronds. The fruit is produced in clusters on the back or margin of the fronds; each cluster contains many sporangia, and each sporangium numerous uniform spores.

Though there is a great diversity in the size of the plants of this order-from the humble Wall Rue to the giant Alsophilas,-there is a remarkable uniformity in the size of the spores.

When the spore germinates it bursts through the outer membrane and puts forth a tubular prolongation, which increases by cell-multiplication until a small green leaf is produced, called the prothallus, on the under-surface of which two kinds of glandular-like bodies are developed the one, the antheridia, containing numerous cells with spermatozoids, the other, the pistillidia, one of which when fertilized develops into a true fern.

II. The HORSE-TAILS have slender, hollow, and jointed stems. Each joint terminates in a toothed membranous sheath, composed of leaves reduced to this elementary state. Whorls of branches and branchlets are given off at the joints in some species.

The fruit is produced in terminal cones composed of numerous stalked peltate scales, each of which bears on its under-surface a circle of sporangia filled with numerous uniform spores. The spores have a spiral covering, which, when they are ripe, breaks up into four clavate threads called elaters, which are remarkably hygrometric. The spores germinate like those of ferns.

III. The CLUB-MOSSES have solid stems composed of an axis of spiral vessels, surrounded by a thickish cortical cellular layer. The

leaves are simple, and arranged spirally on the stem. The branches are irregular and dichotomous.

The fruit is produced in terminal cones composed of imbricating scales. Each scale bears on its pedicel a small sporangium full of spores. In Selaginella two kinds of spores exist. The one, called microspores, produces spermatozoids; the other, macrospores, germinates, and forms a prothallus on which pistillidia appear; and these, when fertilized by the spermatozoids of the microspores, grow into perfect plants. In Lycopodium microspores only have been seen, and the process of its germination is still unknown.

The little Quillwort (Isoetes) which grows at the bottom of most of our mountain lakes, agrees with Selaginella in having two kinds of spores; but it differs from the true club-mosses in its habit and in the structure of the stem. Like Welwitschia it never increases in height; but this is even more remarkable in the Quill-wort than in Welwitschia, seeing that in it there is, as long as the plant lives, a continual development of nodes with their foliar appendages going The axis of the stem is composed of cellular tissue. This is surrounded by a vascular cylinder, which grows, as in exogens, by the addition of external layers, there being in this plant a true cambium layer outside the wood, a structure unknown in other cryptogams.

on.

IV. No plants allied to the PILLWORTS have hitherto been detected in the Coal Measures; we need not, therefore, be detained by an examination of their structure and development.

In examining the palæozoic cryptogams of the Coal Forests, I will follow the same order as that in which we have glanced at their living representatives.

I. The FERNS need not long occupy our attention. They were very abundant, though as a rule they were humble herbaceous plants. Arborescent stems are extremely rare-only two undoubted species have been met with in Britain. The numerous known forms have either grown on the earth, or, as is very probable, been Epiphytes. Fructification is rare; in the few cases in which it has been found it agrees with that of recent ferns. Occasionally young fronds exhibiting circinnate vernation have been met with, showing that this method of unrolling the frond was as characteristic of the ferns of that period as it is of those of the present.

The fern is a remarkably stable type of vegetation. The earliest forms, like the Cyclopteris Hibernica of Forbes from the Old Red Sandstone, agrees in all comparable points with the recent plants; and throughout all the intervening space no divergence in any point of importance has been detected.

II. No group of fossil plants can more fully illustrate the imperfect materials with which the palæontological botanist has to deal than that group which I have united under the name Calamites.' The various parts of the plant-the root, the stem, the leaves, and the

1 On the Structure of the Fruit of Calamites. Seeman's Journal of Botany, vol. v. (1867), p. 349, Pl. 70.

fruit-have been formed into numerous genera, which have been referred to widely different positions in the vegetable kingdom.

Considerable diversity of structure is to be found in those stems which are referred to Calamites. I shall ask your attention to one of these forms which I have described,' and which is beautifully illustrated by a series of drawings, recently published, of specimens in Mr. Binney's collection. This stem was composed of a central medulla surrounded by a woody cylinder, composed entirely of scalariform vessels and a thin cortical layer. The medulla penetrated the woody cylinder by a series of regular wedges, which were continued, as delicate laminæ of one or two cells in thickness, to the cortical layer. The cells of these lamina were not muriform; their longest diameter was in the direction of the axis. The wedges were continuous and parallel between each node. As the axial appendages were produced in whorls, the only interference with the regularity of the tissues was by the passing out through the stem at the nodes of the vascular bundles which supplied these appendages. As the leaves of each whorl were (with one or two exceptions) opposite to the interspaces of the whorls above and below, there was also at each node a rearrangement of the wedges of vascular and cellular tissues.

The stem is described as having been fluted on the outer surface. This error had its rise in the specimens examined, being only casts in the amorphous substance of the rock of the medullary cavity, surrounded by a thin film of coal representing the cylinder of wood. On the death of the plant, the cellular medulla decayed, while the woody cylinder was still able to retain its original form. The hollow interior was filled with some of the mud or sand in which the plant was buried. In the course of time this offered greater resistance to the pressure of the beds above than the originally hard cylinder of scalariform tissue, now softened by the moisture in which it had so long lain: the more indurated amorphous axis on pressure necessarily produced its characteristic ridges and furrows on the smooth outer surface of the film of coal. This coal is described as the cortex or bark, and stems exhibiting only the rocky casts of the medullary cavity are called decorticated specimens; but, besides the cortical layer, they have also been deprived of all that remained of their woody tissue.

The stem terminated below somewhat suddenly in a blunt cone, the internodes of which were slightly developed; and from the nodes were given off whorls of large roots, which again gave off innumerable branching rootlets (Pinnularia).

The stem or main axis was simple, supporting numerous branches arranged in whorls, which again produced numbers of whorled leaves. Three different forms of leaves have been formed into as many genera. When the structure of the fruits associated with them is better known, by the discovery of better preserved specimens, it is possible they may be found to constitute three genera, but there are no characters

1 On the Structure and Affinities of Lepidodendron and Calamites. Trans. Bot. Soc. Edin. vol. viii. (1866) p. 495, Pl. 8 and 9.

2 On the Flora of the Carboniferous Strata, Part I. By E. W. Binney. Palæont. Soc. vol. xxi., 1868.

possessed by the leaves which prevent them belonging to one welldefined genus.

6

1

PLATE I.-FOLIAGE AND FRUITS OF CALAMITES.

1 and 2, Asterophyllites; 3 and 4, Annularia; 5 and 6, Sphenophyllum.

The simplest form of leaf (Asterophyllites) is slender and linear, with a single nerve. This can scarcely be separated from the form to which the name Annularia has been given, and which differs chiefly in having a larger amount of cellular tissue spread out on either side of the midrib. This form has a different aspect in the fossil state from the other, for its whorls of numerous broad leaves are spread out on the surface of deposition, while the acicular leaves of Asterophyllites have penetrated the soft mud, and are generally preserved in the position they originally occupied to the supporting branch. third form (Sphenophyllum) consists of whorls of wedge-shaped leaves with one or more bifurcating veins. They occur like those of Annularia, spread out on the surface of the shale.

The

The plan of arrangement of the three forms is the same, and fruits are found associated with them which have the same general appearance; but they are so ill preserved that their internal struc