[Ill.u.s.tration: Fig. 367. Surface of another individual of same species, showing form of tubercles. (Foss. Flo. 34.)]

[Ill.u.s.tration: Fig. 368. _Stigmaria ficoides_, Brong. One fourth of nat. size. (Foss. Flo. 32.)]

_Conifers._--The coniferous trees of this period are referred to five genera; the woody structure of some of them showing that they were allied to the Araucarian division of pines, more than to any of our common European firs. Some of their trunks exceeded 44 feet in height.

_Endogens._--Hitherto but few monocotyledonous plants have been discovered in the coal-strata. Most of these consist of fruits referred by some botanists to palms. The three-sided nuts, called _Trigonocarpum_, seven species of which are known, appear to have the best claim to rank as palms, although M. Ad. Brongniart entertains some doubt even as to their being monocotyledons.

_Exogens._

The entire absence, so far as our paleontological investigations have hitherto gone, of ordinary dicotyledons or exogens in the coal measures, is most remarkable. Hence, M. Adolphe Brongniart has called this period the age of acrogens, in consequence of the vast preponderance of ferns and _Lepidodendra_.[316-A] Nevertheless, a forest of the period, now under consideration, may have borne a considerable resemblance to those woody regions of New Zealand, in which ferns, arborescent and herbaceous, and lycopodiums, with many coniferae, abound.

The comparative proportion of living ferns and _Araucariae_, in Norfolk Island, to all the other plants, appears to be very similar to that formerly borne by these tribes respectively in a forest of the coal-period.

I have already stated that Professor Goppert, after examining the fossil vegetables of the coal-fields of Germany, has detected, in beds of pure coal, remains of plants of every family hitherto known to occur fossil in the coal. Many seams, he remarks, are rich in _Sigillaria_, _Lepidodendron_, and _Stigmaria_, the latter in such abundance, as to appear to form the bulk of the coal. In some places, almost all the plants are calamites, in others ferns.[316-B]

_Coal, how formed--Erect trees._--I shall now consider the manner in which the above-mentioned plants are imbedded in the strata, and how they may have contributed to produce coal. "Some of the plants of our coal," says Dr. Buckland, "grew on the identical banks of sand, silt, and mud, which, being now indurated to stone and shale, form the strata that accompany the coal; whilst other portions of these plants have been drifted to various distances from the swamps, savannahs, and forests that gave them birth, particularly those that are dispersed through the sandstones, or mixed with fishes in the shale beds." "At Balgray, three miles north of Glasgow," says the same author, "I saw in the year 1824, as there still may be seen, an unequivocal example of the stumps of several stems of large trees, standing close together in their native place, in a quarry of sandstone of the coal formation."[317-A]

Between the years 1837 and 1840, six fossil trees were discovered in the coal-field of Lancashire, where it is intersected by the Bolton railway.

They were all in a vertical position, with respect to the plane of the bed, which dips about 15 to the south. The distance between the first and the last was more than 100 feet, and the roots of all were imbedded in a soft argillaceous shale. In the same plane with the roots is a bed of coal, eight or ten inches thick, which has been ascertained to extend across the railway, or to the distance of at least ten yards. Just above the covering of the roots, yet beneath the coal seam, so large a quant.i.ty of the _Lepidostrobus variabilis_ was discovered inclosed in nodules of hard clay, that more than a bushel was collected from the small openings around the base of the trees (see figure of this genus, p. 313.). The exterior trunk of each was marked by a coating of friable coal, varying from one quarter to three quarters of an inch in thickness; but it crumbled away on removing the matrix. The dimensions of one of the trees is 15-1/2 feet in circ.u.mference at the base, 7-1/2 feet at the top, its height being 11 feet. All the trees have large spreading roots, solid and strong, sometimes branching, and traced to a distance of several feet, and presumed to extend much farther. Mr.

Hawkshaw, who has described these fossils, thinks that, although they were hollow when submerged, they may have consisted originally of hard wood throughout; for solid dicotyledonous trees, when prostrated in tropical forests, as in Venezuela, on the sh.o.r.e of the Caribbean Sea, were observed by him to be destroyed in the interior, so that little more is left than an outer sh.e.l.l, consisting chiefly of the bark. This decay, he says, goes on most rapidly in low and flat tracks, in which there is a deep rich soil and excessive moisture, supporting tall forest-trees and large palms, below which bamboos, canes, and minor palms flourish luxuriantly. Such tracts, from their lowness, would be most easily submerged, and their dense vegetation might then give rise to a seam of coal.[317-B]

In a deep valley near Capel-Coelbren, branching from the higher part of the Swansea valley, four stems of upright _Sigillariae_ were seen, in 1838, piercing through the coal-measures of S. Wales; one of them was 2 feet in diameter, and one 13 feet and a half high, and they were all found to terminate downwards in a bed of coal. "They appear," says Sir H. De la Beche, "to have const.i.tuted a portion of a subterranean forest at the epoch when the lower carboniferous strata were formed.[318-A]

In a colliery near Newcastle, say the authors of the Fossil Flora, a great number of _Sigillariae_ were placed in the rock as if they had retained the position in which they grew. Not less than thirty, some of them 4 or 5 feet in diameter, were visible within an area of 50 yards square, the interior being sandstone, and the bark having been converted into coal. The roots of one individual were found imbedded in shale; and the trunk, after maintaining a perpendicular course and circular form for the height of about 10 feet, was then bent over so as to become horizontal. Here it was distended laterally, and flattened so as to be only one inch thick, the flutings being comparatively distinct.[318-B] Such vertical stems are familiar to our miners, under the name of coal-pipes. One of them, 72 feet in length, was discovered, in 1829, near Gosforth, about five miles from Newcastle, in coal-grit, the strata of which it penetrated. The exterior of the trunk was marked at intervals with knots, indicating the points at which branches had shot off. The wood of the interior had been converted into carbonate of lime; and its structure was beautifully shown by cutting transverse slices, so thin as to be transparent. (See p. 40.)

These "coal-pipes" are much dreaded by our miners, for almost every year in the Bristol, Newcastle, and other coal-fields, they are the cause of fatal accidents. Each cylindrical cast of a tree, formed of solid sandstone, and increasing gradually in size towards the base, and being without branches, has its whole weight thrown downwards, and receives no support from the coating of friable coal which has replaced the bark. As soon, therefore, as the cohesion of this external layer is overcome, the heavy column falls suddenly in a perpendicular or oblique direction from the roof of the gallery whence coal has been extracted, wounding or killing the workman who stands below. It is strange to reflect how many thousands of these trees fell originally in their native forests in obedience to the law of gravity; and how the few which continued to stand erect, obeying, after myriads of ages, the same force, are cast down to immolate their human victims.

It has been remarked, that if, instead of working in the dark, the miner was accustomed to remove the upper covering of rock from each seam of coal, and to expose to the day the soils on which ancient forests grew, the evidence of their former growth would be obvious. Thus in South Staffordshire a seam of coal was laid bare in the year 1844, in what is called an open work at Parkfield Colliery, near Wolverhampton. In the s.p.a.ce of about a quarter of an acre the stumps of no less than 73 trees with their roots attached appeared, as shown in the annexed plan (fig.

369.), some of them more than 8 feet in circ.u.mference. The trunks, broken off close to the root, were lying prostrate in every direction, often crossing each other. One of them measured 15, another 30 feet in length, and others less. They were invariably flattened to the thickness of one or two inches, and converted into coal. Their roots formed part of a stratum of coal 10 inches thick, which rested on a layer of clay 2 inches thick, below which was a second forest, resting on a 2-foot seam of coal. Five feet below this again was a third forest with large stumps of _Lepidodendra_, _Calamites_, and other trees.

[Ill.u.s.tration: Fig. 369. Ground-plan of a fossil forest, Parkfield Colliery, near Wolverhampton, showing the position of 73 trees in a quarter of an acre.[319-A]]

In the account given, in 1821, by M. Alex. Brongniart of the coal-mine of Treuil, at St. Etienne, near Lyons, he states, that distinct horizontal strata of micaceous sandstone are traversed by vertical trunks of monocotyledonous vegetables, resembling bamboos or large _Equiseta_.[319-B] Since the consolidation of the stone, there has been here and there a sliding movement, which has broken the continuity of the stems, throwing the upper parts of them on one side, so that they are often not continuous with the lower.

From these appearances it was inferred that we have here the monuments of a submerged forest. I formerly objected to this conclusion, suggesting that, in that case, all the roots ought to have been found at one and the same level, and not scattered irregularly through the ma.s.s.

I also imagined that the soil to which the roots were attached should have been different from the sandstone in which the trunks are enclosed.

Having, however, seen calamites near Pictou, in Nova Scotia, buried at various heights in sandstone and in similar erect att.i.tudes, I have now little doubt that M. Brongniart's view was correct. These plants seem to have grown on a sandy soil, liable to be flooded from time to time, and raised by new accessions of sediment, as may happen in swamps near the banks of a large river in its delta. Trees which delight in marshy grounds are not injured by being buried several feet deep at their base; and other trees are continually rising up from new soils, several feet above the level of the original foundation of the mora.s.s. In the banks of the Mississippi, when the water has fallen, I have seen sections of a similar deposit in which portions of the stumps of trees with their roots _in situ_ appeared at many different heights.[320-A]

[Ill.u.s.tration: Fig. 370. Section showing the erect position of fossil trees in coal sandstone at St. Etienne. (Alex. Brongniart.)]

When I visited, in 1843, the quarries of Treuil above-mentioned, the fossil trees seen in fig. 370. were removed, but I obtained proofs of other forests of erect trees in the same coal-field.

[Ill.u.s.tration: Fig. 371. Inclined position of a fossil tree, cutting through horizontal beds of sandstone, Craigleith quarry, Edinburgh. Angle of inclination from _a_ to _b_ 27.]

_Snags._--In 1830, a slanting trunk was exposed in Craigleith quarry, near Edinburgh, the total length of which exceeded 60 feet. Its diameter at the top was about 7 inches, and near the base it measured 5 feet in its greater, and 2 feet in its lesser width. The bark was converted into a thin coating of the purest and finest coal, forming a striking contrast in colour with the white quartzose sandstone in which it lay.

The annexed figure represents a portion of this tree, about 15 feet long, which I saw exposed in 1830, when all the strata had been removed from one side. The beds which remained were so unaltered and undisturbed at the point of junction, as clearly to show that they had been tranquilly deposited round the tree, and that the tree had not subsequently pierced through them, while they were yet in a soft state.

They were composed chiefly of siliceous sandstone, for the most part white; and divided into laminae so thin, that from six to fourteen of them might be reckoned in the thickness of an inch. Some of these thin layers were dark, and contained coaly matter; but the lowest of the intersected beds were calcareous. The tree could not have been hollow when imbedded, for the interior still preserved the woody texture in a perfect state, the petrifying matter being, for the most part, calcareous.[321-A] It is also clear, that the lapidifying matter was not introduced laterally from the strata through which the fossil pa.s.ses, as most of these were not calcareous. It is well known that, in the Mississippi and other great American rivers, where thousands of trees float annually down the stream, some sink with their roots downwards, and become fixed in the mud. Thus placed, they have been compared to a lance in rest; and so often do they pierce through the bows of vessels which run against them, that they render the navigation extremely dangerous. Mr. Hugh Miller mentions four other huge trunks exposed in quarries near Edinburgh, which lay diagonally across the strata at an angle of about 30, with their lower or heavier portions downwards, the roots of all, save one, rubbed off by attrition. One of these was 60 and another 70 feet in length, and from 4 to 6 feet in diameter.

[Ill.u.s.tration: Fig. 372. Section of the cliffs of the South Joggins, near Minudie, Nova Scotia.]

The number of years for which the trunks of trees, when constantly submerged, can resist decomposition, is very great; as we might suppose from the durability of wood, in artificial piles, permanently covered by water. Hence these fossil snags may not imply a rapid acc.u.mulation of beds of sand, although the channel of a river or part of a lagoon is often filled up in a very few years.

_Nova Scotia._--One of the finest examples in the world of a succession of fossil forests of the carboniferous period, laid open to view in a natural section, is that seen in the lofty cliffs bordering the Chignecto Channel, a branch of the Bay of Fundy, in Nova Scotia.[321-B]

In the annexed section (fig. 372.), which I examined in July, 1842, the beds from _c_ to _i_ are seen all dipping the same way, their average inclination being at an angle of 24 S.S.W. The vertical height of the cliffs is from 150 to 200 feet; and between _d_ and _g_, in which s.p.a.ce I observed seventeen trees in an upright position, or, to speak more correctly, at right angles to the planes of stratification, I counted nineteen seams of coal, varying in thickness from 2 inches to 4 feet. At low tide a fine horizontal section of the same beds is exposed to view on the beach. The thickness of the beds alluded to, between _d_ and _g_, is about 2,500 feet, the erect trees consisting chiefly of large _Sigillariae_, occurring at ten distinct levels, one above the other; but Mr. Logan, who afterwards made a more detailed survey of the same line of cliffs, found erect trees at seventeen levels, extending through a vertical thickness of 4,515 feet of strata; and he estimated the total thickness of the carboniferous formation, with and without coal, at no less than 14,570 feet, every where devoid of marine organic remains.[322-A] The usual height of the buried trees seen by me was from 6 to 8 feet; but one trunk was about 25 feet high and 4 feet in diameter, with a considerable bulge at the base. In no instance could I detect any trunk intersecting a layer of coal, however thin; and most of the trees terminated downwards in seams of coal.

Some few only were based in clay and shale, none of them in sandstone. The erect trees, therefore, appeared in general to have grown on beds of coal.

In some of the underclays I observed _Stigmaria_.

[Ill.u.s.tration: Fig. 373. Fossil tree at right angles to planes of stratification. Coal measures, Nova Scotia.]

In regard to the plants, they belonged to the same genera, and most of them to the same species, as those met with in the distant coal-fields of Europe. In the sandstone, which filled their interiors, I frequently observed fern leaves, and sometimes fragments of _Stigmaria_, which had evidently entered together with sediment after the trunk had decayed and become hollow, and while it was still standing under water. Thus the tree, _a b_, fig. 373., the same which is represented at _a_, fig. 374., or in the bed _e_ in the larger section, fig. 372., is a hollow trunk 5 feet 8 inches in length, traversing various strata, and cut off at the top by a layer of clay 2 feet thick on which rests a seam of coal (_b_, fig. 374.) 1 foot thick. On this coal again stood two large trees (_c_ and _d_), while at a greater height the trees _f_ and _g_ rest upon a thin seam of coal (_e_), and above them is an underclay, supporting the 4-foot coal.

[Ill.u.s.tration: Fig. 374. Erect fossil trees. Coal-measures, Nova Scotia.]

If we now return to the tree first mentioned (fig. 373.), we find the diameter (_a b_) 14 inches at the top and 16 inches at the bottom, the length of the trunk 5 feet 8 inches. The strata in the interior consisted of a series entirely different from those on the outside. The lowest of the three outer beds which it traversed consisted of purplish and blue shale (_c_, fig. 373.), 2 feet thick, above which was sandstone (_d_) 1 foot thick, and, above this, clay (_e_) 2 feet 8 inches. But, in the interior, were nine distinct layers of different composition: at the bottom, first, shale 4 inches, then sandstone 1 foot, then shale 4 inches, then sandstone 4 inches, then shale 11 inches, then clay (_f_) with nodules of ironstone 2 inches, then pure clay 2 feet, then sandstone 3 inches, and, lastly, clay 4 inches. Owing to the outward slope of the face of the cliff, the section (fig. 373.) was not exactly perpendicular to the axis of the tree; and hence, probably, the apparent sudden termination at the base without a stump and roots.

In this example the layers of matter in the inside of the tree are more numerous than those without; but it is more common in the coal-measures of all countries to find a cylinder of pure sandstone,--the cast of the interior of a tree, intersecting a great many alternating beds of shale and sandstone, which originally enveloped the trunk as it stood erect in the water. Such a want of correspondence in the materials outside and inside, is just what we might expect if we reflect on the difference of time at which the deposition of sediment will take place in the two cases; the imbedding of the tree having gone on for many years before its decay had made much progress.

The high tides of the Bay of Fundy, rising more than 60 feet, are so destructive as to undermine and sweep away continually the whole face of the cliffs, and thus a new crop of erect trees is brought into view every three or four years. They are known to extend over a s.p.a.ce between two and three miles from north to south, and more than twice that distance from east to west, being seen in the banks of streams intersecting the coal-field.

In Cape Breton, Mr. Richard Brown has observed in the Sydney coal-field a total thickness of coal-measures, without including the underlying millstone grit, of 1843 feet, dipping at an angle of 8. He has published minute details of the whole series, showing at how many different levels erect trees occur, consisting of _Sigillaria_, _Lepidodendron_, _Calamite_, and other genera. In one place eight erect trunks, with roots and rootlets attached to them, were seen at the same level, within a horizontal s.p.a.ce 80 feet in length. Beds of coal of various thickness are interstratified. Some of the a.s.sociated strata are ripple-marked, with impressions of rain-drops.

Taking into account forty-one clays filled with roots of _Stigmaria_ in their natural position, and eighteen layers of upright trees at other levels, there is, on the whole, clear evidence of at least fifty-nine fossil forests, ranged one above the other, in this coal-field, in the above-mentioned thickness of strata.[324-A]

The fossil sh.e.l.ls in Cape Breton and in the Nova Scotia section (fig.

372.), consisting of _Cypris_, _Unio_ (?), _Modiola_, _Microconchus carbonarius_ (see fig. 375.), and _Spirorbis_, seem to indicate brackish water; but we ought never to be surprised if, in pursuing the same stratum, we come to a fresh or purely marine deposit; for this will depend upon our taking a direction higher up or lower down the ancient river or delta deposit. When the Purbeck beds of the Wealden were described in Chap.

XVIII., I endeavoured to explain the intimate connection of strata formed at a river's mouth, or in the tranquil lagoons of the delta, or in the sea, after a slight submergence of the land, with its dirt-beds.

In the English coal-fields the same a.s.sociation of fresh, or rather brackish water with marine strata, in close connection with beds of coal of terrestrial origin, has been frequently recognized. Thus, for example, a deposit near Shrewsbury, probably formed in brackish water, has been described by Sir R. Murchison as the youngest member of the carboniferous series of that district, at the point where the coal-measures are in contact with the Permian or "Lower New Red." It consists of shales and sandstones about 150 feet thick, with coal and traces of plants; including a bed of limestone, varying from 2 to 9 feet in thickness, which is cellular, and resembles some lacustrine limestones of France and Germany.

It has been traced for 30 miles in a straight line, and can be recognized at still more distant points. The characteristic fossils are a small bivalve, having the form of a _Cyclas_, a small _Cypris_ (fig. 376.), and the microscopic sh.e.l.l of an annelid of an extinct genus called _Microconchus_ (fig. 375.), allied to _Serpula_ or _Spirorbis_.

In the lower coal-measures of Coalbrook Dale, the strata, according to Mr.

Prestwich, often change completely within very short distances, beds of sandstone pa.s.sing horizontally into clay, and clay into sandstone. The coal-seams often wedge out or disappear; and sections, at places nearly contiguous, present marked lithological distinctions. In this single field, in which the strata are from 700 to 800 feet thick, between forty and fifty species of terrestrial plants have been discovered, besides several fishes and trilobites of forms distinct from those occurring in the Silurian strata. Also upwards of forty species of mollusca, among which are two or three referred to the freshwater genus _Unio_, and others of marine forms, such as _Nautilus_, _Orthoceras_, _Spirifer_, and _Productus_. Mr.

Prestwich suggests that the intermixture of beds containing freshwater sh.e.l.ls with others full of marine remains, and the alternation of coa.r.s.e sandstone and conglomerate with beds of fine clay or shale containing the remains of plants, may be explained by supposing the deposit of Coalbrook Dale to have originated in a bay of the sea or estuary into which flowed a considerable river subject to occasional freshes.[325-A]

[2 Ill.u.s.trations: Freshwater Fossils--Coal.

Fig. 375.

_a._ _Microconchus carbonarius_.

_b._ var. of same; nat. size, and magnified.

Fig. 376. _Cypris inflata_, natural size, and magnified. Murchison.[325-B]]

In the Edinburgh coal-field, at Burdiehouse, fossil fishes, mollusca, and cypris, very similar to those in Shropshire and Staffordshire, have been found by Dr. Hibbert.[325-C] In the coal-field also of Yorkshire there are freshwater strata, some of which contain sh.e.l.ls referred to the genus _Unio_; but in the midst of the series there is one thin but very widely spread stratum, abounding in fishes and marine sh.e.l.ls, such as _Ammonites Listeri_ (fig. 377.), _Orthoceras_, and _Avicula papyracea_, Goldf. (fig. 378.)[325-D]

[Ill.u.s.tration: Fig. 377. _Ammonites Listeri_, Sow.]

[Ill.u.s.tration: Fig. 378. _Avicula papyracea_, Goldf. (_Pecten papyraceus_, Sow.)]