been almost entirely succulent, and capable of being squeezed into a small compass during partial decomposition. This squeezing process must have been conducted on a grand scale, both during and after the formation of separate beds, and each bed in succession was probably soon covered up by muddy and sandy accumulations, now alternating with the coal in the form of shale and grit-stone. Sometimes trunks of trees caught in the mud would be retained in a slanting or nearly vertical position, while the sands were accumulating round them; sometimes the whole would be quietly buried, and soon cease to exhibit any external marks of vegetable origin.*

To relate at full length the different processes, and the gradual superposition of one bed upon another, by which at length, and by slow degrees, the whole group of the coal-measures was completed, would involve far too much complication of detail to be described in a few pages; and when it is remembered that the woody fibre, when deposited, had to be afterwards completely changed, and the whole character of the vegetable modified, before it could be reduced to the bituminous, brittle, almost crystalline mineral now dug out of the earth for fuel, it will rather seem questionable whether the origin of coal was certainly and necessarily vegetable, than reasonable to doubt the importance of the change that has taken place, and

* There can be no reasonable doubt, judging from the analogy of existing vegetation, that some beds of coal may have been derived from the mass of vegetable matter present at one time on the surface and submerged suddenly. It is only necessary to refer to the accounts of vegetation in some of the extremely moist warm islands in the southern hemisphere, where the ground is occasionally covered with eight or ten feet of decaying vegetable matter at one time, to be satisfied that this is at least possible.

the existence of extraordinary means to produce that change. Nothing, however, is more certain than that all coal was once vegetable; for in most cases woody structure may be detected under the microscope, and this, if not in the coal in its ordinary state, at least in the burnt ashes which remain after it has been exposed to the action of heat, and has lost its bituminous and semi-crystalline character. This has been too well and too frequently proved by actual experiment, to require more than the mere statement of the fact.

The principal vegetable remains, the study of which conducts the botanical fossilist to a knowledge of the trees and 'shrubs that clothed the land during this period, are not met with in the substance of the coal itself, but in the shales and sandstones so abundantly mixed with the coal. These beds have received, and occasionally retained, the fragments that have been deposited in them; and from such fragments, consisting of leaves and small branches, portions of the larger branches, or even the entire trunk, and occasionally, also, internal casts or markings of the surface of the fruits of various trees, conclusions have been arrived at and analogies drawn expressing the relation they bear to existing plants. Now if, in all cases, the solid substance of the trunks and the perfect outline of the leaves and fruits had been preserved, the botanist might fairly have been expected to explain the general character of the vegetation thus exemplified; but unfortunately this is not the case. In most instances, even those which seem most favourable, it is only the cast of the tree in sandstone, the impression made by the outside or inside of the bark, and a little of the

substance of the bark turned into coal, that remains; and the shape even of the trunk has been often completely lost by crushing; thus showing, indeed, one peculiarity indicative of the great natural group of trees to which the fossils belonged (the Monocotyledons, or Endogens), but at the same time almost precluding the possibility of comparison with the few recent plants of the same tribe which appear to resemble them in general form or structure.

Still, in spite of the difficulties with which the subject is surrounded, and in spite of the perplexing confusion which might well alarm the botanist only acquainted with the ordinary marks of distinction exhibited by plants, a great deal has been done towards determining the general nature of the flora of the islands of the carboniferous period. It is remarkable, that, in the first place, this flora is found to be, to a great extent, uniform in all parts of the globe from which carboniferous fossils have been obtained; * and, in the next place, that, if we wish to compare this ancient flora with those which bear resemblance to it at present, either in the general preponderance of particular plants, or in the total absence of others, we must leave entirely the northern latitudes and the northern hemisphere, and transport ourselves to the islands in the neighbourhood of our antipodes, where New Zealand and the southern part of Australia, together with an innumerable multitude of small islands, form almost the only land that now exists in the vast area between the tropic of Capricorn and the South pole.

Namely, the whole of western, northern, and eastern Europe, North America from Alabama to Melville Island, various districts in Asia, eastern Australia, and Van Diemen's Land, and (probably) the Asiatic islands.

But although in such distant parts of the world we really have a somewhat similar group of plants to that of the coal-measures, and the dark-tinted ferns do now, as they once did in the northern hemisphere, take the place of our cheerful grasses, and even grow in rank luxuriance into forest trees, being associated with palms and some peculiar pines, there is probably after all but little true resemblance; nor can the Geologist feel satisfied that the condition of things was the same formerly in England as it is now in the islands of the southern sea. One of the most remarkable peculiarities of the coal fossils certainly is, however, the singular preponderance of the tribe of ferns already alluded to, and the great variety of form in which plants of this kind are developed. Of these forms, the annexed figures (26, 27) represent two that

Fig. 26

Fig. 27

PECOPTERIS.

ODONTOPTERIS.

are common and highly characteristic. They are found sometimes in isolated fragments, in sand or shale, but are sometimes so very abundant, that the whole mass

seems to have been formed rapidly in association with such vegetable remains.

Besides these two forms, there are, however, many others; as, for instance, the Sphenopteris (fig. 28), and the Neuropteris* (29); both of them common, and, in all probability, belonging to the group of arborescent ferns, and growing in wild luxuriance on stems of greater or less altitude. Some notion

may be formed of the peculiar character of such vegetation by referring to the frontispiece at the beginning of this volume, where it has been endeavoured, by combining existing and analogous forms with some restored forms of extinct plants, to communicate a notion, however vague, of the flora of the coal period.

Besides the arborescent ferns, then growing to a great size, I might also notice the gigantic proportions of other plants, whose modern representatives

* These names are all derived from the peculiar form of the leaf and its venation, in combination with the Greek word πTEρiç (pteris), a fern. The derivatives are respectively TEкw (peco), to comb; odovç gen. odovτος (odontos), a tooth; σφην gen. σφηνος (sphenos), a wedge; and νευρον (neuron), a nerve.