IN 1836 Sedgwick and Murchison described the existence in Devonshire of a series of rocks bearing fossils intermediate in character between those of the Upper Silurian series and those of the Carboniferous Limestone. This was done with the assistance of Mr. Lonsdale in all the paliontological part of the question, in which the argument chiefly lay. On these and certain stratigraphical grounds, it was considered that they are the equivalents of the Old Red Sandstone of the centre of England and of Scotland, and the name DEVONIAN being applied to them, the terms Devonian and Old Red Sandstone are generally considered as equivalents in point of geological time.

According to the late Professor Jukes, the lowest strata of the Barnstaple Bay district in North Devon consist of red sandstones and conglomerates, similar to those of part of the Old Red Sandstone of Ireland, and not unlike that of the Mendip Hills. This, taken in connection with the resemblance of the overlying strata to the lower Carboniferous rocks of the south of Ireland, led him to consider the chief part of the Devonian rocks of Devonshire to be of Carboniferous age. To this conclusion he was led partly by paleontological considerations into which I cannot here enter. The opposite

[100 Devonian Strata.]

opinion, that the Devonian strata are in the main the marine equivalents of the Old Red Sandstone, continues to be generally held. Till a new survey of Devonshire helps to settle the question I give the usual reading of the history of the Devonian strata, though I think it probable that Jukes will turn out to be correct in questioning the right of the Devonian strata to the conventional name of an independent series.

In Devonshire the strata have been divided into Lower, Middle, and Upper Devonian. The Lower chiefly consists of slaty beds and green and purple sandstones, with many Brachiopoda of the genera Chonetes, Orthis, Spirifera, &c. The Middle group, which includes the Plymouth limestone, contains numerous corals, the most common genera of which are Acervularia, Alveolites, Cyathophyllum, Favosites, Petraia, Strephodes, and the sponge Stromatopora. With these are found Encrinites, Spirifers, Atrypœ, Leptœnœ, Productœ, Rhynchonellœ, Stringocephali, and Calceola (C. sandalina)—the last a genus peculiar to the Devonian rocks. Many Lamellibranchiate molluscs also occur, together with Gasteropoda of the genera Euomphalus, Loxonema, Machrocheilus, Murchisonia, Pleurotomaria, Turbo, &c. Also many Cephalopoda of the genera Clymenia, Cyrtoceras, Orthoceras, Goniatites and Nautilus. The last two are unknown in the British Silurian series, though Nautilus occurs in the Upper Silurian rocks of North America. The Goniatite may be roughly said to be intermediate in structure between the Nautilus and Ammonite. The latter does not occur in Palæozoic strata. A few Trilobites are found in the British Devonian rocks, and various Crinoids. The Upper Devonian group contains land plants and many shells, some of which are

FIG 25.
Devonian Fossils
Group of Devonian fossils (marine); and two from Old Red Sandstone, the Cyclopteris and Anodonta (land and fresh-water), see p. 115.
[102 Devonian Strata.]

identical with those found in the Lower Carboniferous Limestone shales.

There is in England a considerable diminution in the number of Devonian fossils when compared with those of the Silurian rocks. Thus about 1,500 species of Silurian fossils are named, while of marine Devonian we have under 400 species, and adding those of all kinds in the freshwater strata of the Old Red Sandstone, 535 species. Of corals, 11 of the genera only are also Silurian. Of Echinodermata, there are 10 genera and 21 species, only 3 of the genera being also Silurian; Crustacea, 13 genera, 35 species, 5 of the genera being also Silurian, including those found both in the Devonian rocks and the Old Red Sandstone. In the latter no Trilobites occur, but only Crustacea of the genera Eurypterus (6), Pterygotus (4) (fig. 26), Stylonurus (7), while in the Devonian formations of Devonshire we find 5 genera of Trilobites:—Bronteus (B. flabellifer) Cheirurus 2, Phacops 6, Homalonotus 2, and Harpes 1, all being genera common in the Silurian strata, though the species are distinct. Twelve of the Devonian genera of Brachiopoda occur in Silurian rocks, but of 96 Devonian species few pass downwards, and these are doubtful. The most prevalent genera of Brachiopoda are Athyris, Atrypa, Cyrtina, Orthis, Rhynchonella, Spirifera, Streptorhynchus, and Terebratula. Species of the genera Leptœna and Pentamerus decline in numbers, while Orthis, Rhynchonella, and Spirifera are much increased. Of 21 genera and 60 species of Lamellibranchiate molluscs, the species are all, or almost all, distinct from those of Siluria, while only 6 of the genera are the same. The most prevalent forms are Aviculopecten (10), Pterinea (9), Uucullœa (7), and Ctenodonta (7).

[Old Red Sandstone. 103]

Megalodon is characteristic. Of the 13 genera of Gasteropoda, 9 are Silurian, but of 47 species, all are distinct. The most prevalent forms belong to the genera Euomphalus (6), Loxonema (8), Macrocheilus (7), Murchisonia (5) (there are 22 in the Silurian rocks), and Pleurotomaria 8. There are 5 species of Bellerophon, and 52 species of Cephalopoda, all distinct from Silurian species. Of 6 Devonian genera, only Orthoceras, Poterioceras, and Cyrtoceras are Silurian. The most prevalent species belong to the genera Clymenia (11), Cyrtoceras (13), Goniatites (10), this being their first appearance in the British strata, and Orthoceras (15), (there being 67 known species of this genus in the Silurian rocks).

It is stated that only about 10 per cent. of Upper Silurian fossils pass into the marine Lower Devonian strata. These two formations in England are, however, not found in contact, though they occur commonly enough in the regular order of succession on the Continent and in North America. About 10 per cent. of Lower Devonian fossils pass into the Middle Devonian, and about the same percentage from the Middle into the Upper. If this be true there may possibly be undiscovered unconformities between the subdivisions.

THE OLD RED SANDSTONE, as distinct from the Devonian rocks, is undoubtedly intermediate in age to the uppermost Silurian and the lowest Carboniferous strata. It is sometimes difficult to determine its precise limits either at its base or its top. It first received its name in contradistinction to the New Red Sandstone, the former occurring below, and the latter above the Carboniferous strata.

A vast triangular tract of Old Red Sandstone lies between the west coast of South Pembrokeshire, Bristol

[104 Old Red Sandstone.]

Channel, the south and south-eastern borders of the Silurian rocks of Caermarthenshire, Breconshire, Radnorshire, and Shropshire, and the long line of Carboniferous, Silurian, and New Red Marl strata that runs from Colebrook Dale to the Severn, east of Dean Forest. Fancy in your 'mind's eye' the Carboniferous rocks of the great South Wales Coalfield, and of Dean Forest, to be stripped away, and the whole of the region mentioned, of 120 by 90 miles in length and breadth, would consist entirely of Old Red Sandstone. The lower part is chiefly composed of beds of red marl and sandstone, with cornstones; and the upper, part contains strata of sandstone and conglomerate, forming the Beacons of Brecon, 2,860 feet high, these being the loftiest mountains in South Wales.

Cornstones are impure concretionary limestones, often imbedded 'in marl. In these, at the base of the series, near Ludlow, are species of Pterygotus and Pterichthys, and higher up, of Onchus and Cephalaspis, thus correlating them by fossils to the Old Red Sandstone of Scotland (fig. 26). Along the border of this formation, where the uppermost Silurian strata join the Old Red Sandstone, there is a gradual passage both palæontologically and in the colour and texture of the strata. The Eurypteri and Pterygoti chiefly belong to these passage-beds, and in the same strata at the very base of the Old Red strata, in which there are no mollusca, are species of fish of the genera Auchenaspis, Onchus (?), Pteraspis, Cephalaspis, and Plectrodus. The Silurian marine mollusca, in fact, quickly disappear where the beds begin to get red in colour, notwithstanding the perfect conformity of the two sets of strata in England and the borders of Wales, as, for example, in the neighbourhood of Ludlow. At Kington and

[Old Red Sandstone. 105]

south of Builth, where true passage-beds occur, the ordinary shells of the Upper Ludlow rocks become far less numerous, and are almost all of small size, including species of Modiolopsis and Modiola, Lingula cornea, Platychisma helicites, a small Discina, a small Theca, a few small Crustacea, of the genera Leperditia, Cytherellina, &c. The water was freshening and getting unfitted for marine life.

The remains of Cephalaspis Lyellii (fig. 26) are occasionally found all through the Old Red Sandstone of this large area. The absence of marine shells and the nature of the fossil fishes of the Old Red Sandstone long ago led Mr. Godwin-Austen to infer that the formation was deposited, not in the sea, as had always been asserted, but in a great fresh-water lake, or in a series of lakes. In this opinion I thoroughly agree, for the nearest living analogues of many of the fish are the Polypterus of the African rivers, the Ceratodus of Australia, and in less degree the Lepidosteus of North America. The red colour of the rocks also helps to the same conclusion. Each grain of sand and marl is red, because it is encrusted with a thin pellicle of peroxide of iron, which could not have been deposited from mere solution, as a crust enveloping each grain of sand at the bottom of a great open ocean; but if carbonate of iron were carried in solution into lakes, it might have been precipitated as a peroxide through the oxidising action of the air and the escape of the carbonic acid.1

1 There is no analogy between the coarse red sandstones and finer marls of the Old Red sandstone, and the very fine red ooze dredged from the deeps of the South Atlantic. The latter is a residue produced by the decomposition of Foraminifera, and in no way resembles the coarse mechanical strata of Old Red Sandstone.

[106 Physical Geography.]

The presence of land plants in the very uppermost Silurian strata, as, for instance, near Ludlow and May Hill, indicates the neighbourhood of land. The physical geography of the area was rapidly changing, marking the beginning of an evident Continental epoch. The subject is of so much importance, and when first propounded was considered to be such a dangerous innovation on established views, that I shall give the reasons in some detail, making use for that purpose of passages from my memoir 'On the Red Rocks of England, of older date than the Trias,' published in the 'Quarterly Journal of the Geological Society,' in 1871.

The circumstances which marked the passage of the uppermost Silurian rocks into Old Red Sandstone seem to me to have been:—First, a shallowing of the Silurian sea by accumulation of sediment, aided by slow upheaval, which gradually produced a great change in the physical geography of the district, so that the old marine area became changed into a series of mingled fresh and brackish lagoons, which finally, by continued terrestrial changes, were converted into lakes; and the occurrence of a very few genera or even species of fish and Crustacea, common both to the fresh and brackish or even salt waters, does not prove that the Old Red Sandstone is truly marine. At the present day, animals that are commonly supposed to be essentially marine, are occasionally found inhabiting fresh water. In the inland fresh lakes of Newfoundland, seals are common. They breed there freely, and never visit the sea. The same is the case in Lake Baikal in Central Asia, and it is on record that the inhabitants of the shores of the Sea of Aral, now brackish, were in old times clad in sealskins got from the seals that inhabited those

[Physical Geography. 107]

waters; and though these facts bear but slightly on my present subject, seals being airbreathing Mammalia, yet in some of the lakes of Sweden ordinary marine Crustacea are found. This may be accounted for in the way that I now attempt to account for similar peculiarities in the Old Red Sandstone strata. These Swedish lake-areas were submerged after the close of the Glacial epoch; and being deep basins (scooped out in a manner which I shall explain in a later chapter), while the land was emerging by upheaval, and after its final emergence, the salt water of the lakes freshened so slowly, by influx of rivers, that some of the creatures inhabiting it had time by degrees to adjust themselves to new and abnormal conditions.

Again, we may suppose a set of circumstances such as the following:—If, by changes of physical geography of a continental kind, a portion of the Silurian sea got separated from the main ocean, more or less like the Caspian and the Black Sea, then the ordinary marine conditions of the 'passage beds,' accompanied by some of the life of the period, might be maintained for what, in common language, seems to us a long time. The Black Sea was once united to the Caspian, and the Caspian to the Aral, forming one great inland sea, which under varying physical conditions, has more than once changed its form and extent. At all events, since its separation from the Black Sea the Caspian has been simply a great brackish lake. The Black Sea is now steadily freshening; and it is easy to conceive that by a geographical change, such as the upheaval of the

1  For much valuable information on this subject, see 'Annals and Mag. of Nat. Hist.,' third series, vol. i. 1858, p. 50, 'On the Occurrence of Marine Animal Forms in Fresh Water,' by Dr. E. von Martens: translated by Mr. W. S. Dallas.

[108 Physical Geography.]

phosphorus, it might be converted into a fresh lake, if the supply of river water were sufficient to overbalance evaporation and secure an overflow. At present a great body of salt water is constantly being poured out through the Bosphorus, and its place taken by the fresh water of rivers. Owing, however, to the uncongenial quality of the freshening water, some of the Black Sea shells are strangely distorted as shown by Edward Forbes.

Or if we take the Caspian alone as an example, we have an inland brackish sheet of water, with a present area of 178,866 square miles, the surface of which is 83 feet below the level of the Black Sea. This, according to accepted zoological and physical views, was once united by a narrow strait with the North Sea. Changes in physical geography have taken place of such a kind that. the Caspian is now disunited from the ocean, while its waters are still inhabited by a poor and dwarfed marine molluscan fauna, and by seals. If by increase of rainfall the Caspian became freshened, the loss of watery evaporation not being equal to supply, it would by-and-by, after reaching the point of overflow, be converted into a great fresh-water lake, larger in extent than the whole area now occupied by the British Islands and the Irish Sea. It is even conceivable that the great area of inland drainage of Central Asia, now holding many salt lakes, might in the same manner be so changed that all its lakes would become fresh and widened in extent, thus occupying areas larger than all the Old Red Sandstone of Europe. Under these circumstances, in the Caspian area we should have a passage more or less gradual from imperfect marine to perfectly fresh-water conditions, such as I conceive to have marked the advent of the Old Red Sandstone. When the whole area was fairly separated from the sea,

[Physical Geography. 109]

the sediments might by degrees get into a condition to get coloured red in the manner previously mentioned. We have a case in point in an old inland sheet of water, as shown by the Red Marls of the extinct Miocene lakes of Auvergne in Central France.

The uniformity of action here sketched may present a difficulty to some geologists, seeing that on the borders of South Wales the Upper Old Red Sandstone, over a large space, overlaps the lower strata till they lie directly on Silurian rocks, and the same is the case in parts of Scotland. But on consideration these circumstances do not present any real difficulty. If the great hollow in which the Dead Sea lies, were gradually to get filled with fresh water, and the whole by degrees became silted up, 1,300 feet of strata would be added above the level of the present surface, and the upper strata all round would overlap the lower, apparently much as the Old Red Sandstone strata do in Wales and the adjoining counties. If the Caspian and other parts of the Asiatic area of inland drainage got filled with fresh water, the same general results would ensue.

Like the recurrent circumstances that have attended the rise and falls of empires through all historical time, so geological history has often more or less repeated itself, somewhere or other on the surface of the earth; and in this modern phase of Asiatic physical geography, it seems to me that we may have, so far as it has gone, a repetition of events, which, with minor variations, have happened again and again, in old-world geological epochs, the history of which I shall byand-by have to record. The farther off geological records recede, like inscriptions in an unknown tongue, the more difficult are they to decipher; the nearer they come to our own day, they are often more easy to read.

[110 Old Red Sandstone.]

In North Wales narrow streaks of Old Red Sandstone here and there crop out between the Upper Silurian rocks, and the Carboniferous Limestone of Flintshire and Denbighshire, and the same with bands of corn stone forms the fine escarpment of Traeth Dulas in Anglesey, where the sandstone lies directly on Lower Silurian rocks.

In the northern counties also, at Kirkby Lonsdale, Sedbergh, and Kendal, and all along the base of the Carboniferous Limestone between Orton in Westmoreland, and Greystock Park in Cumberland, patches and a long line of Old Red Sandstones, mans, and conglomerates occur.

A broad belt of Old Red Sandstone crosses Scotland in a north-east direction between the Firth of Clyde and the Firths of Forth and Tay, and Stonehaven in Kincardineshire, and Montrose. Patches lie in Arran, Bute, &c. The whole lies unconformably on Lower Silurian clay slates, and dips to the south-east under the Carboniferous rocks that occupy the great central depression through which the Forth and Clyde chiefly run. On the south-east side of this broad undulating hollow the Old Red Sandstone again rises from beneath the Coal-measures with a general north-west dip, and skirting the Lammermuir Hills, strikes south-west into the sea south of Ayr. On the south side of the Lammermuir Hills, it again appears on the hills between St. Abb's Head and Hawick, dipping under the Carboniferous rocks that, without a break, stretch from Berwick to the neighbourhood of Derby.

North of Stonehaven detached patches of Old Red Sandstone occur on the mainland as far as Pentland Firth, beyond which it forms the greater part of the Orkneys, and small portions of the Shetland Islands.

[Old Red Sandstone. 111]

The first patch lies between Fyvie and Penman Bay in Aberdeenshire; the second forms both shores of Moray Firth and Dornoch Firth, and stretches a long way up the Great Valley of the Caledonian Canal, through which at one time I have no doubt it passed all the way to the Firth of Lorn, between Oban and the island of Mull; and the third forms the greater part of Caithness. On the west coast a large tract of hilly ground between the neighbourhood of Loch Awe, Oban, and Kerera is chiefly formed of Old Red conglomerate.

For the first compendious account of the Old Red Sandstone of Scotland the world is indebted to Hugh Miller, whose wonderful faculty of graphic description enabled him, unassisted, to describe the rocks and the remarkable forms of fish they contain, which till his time were almost altogether unknown. Something, however, still remains to be done, before the precise relations to each other of some of the parts of the Old Red Sandstone of Scotland are clearly established. The researches of Professor Geikie and other officers of the Geological Survey, have shown that, south of the Grampian Mountains, there is an upper set of strata, lying in basins unconformably on the lower Old Red Sandstone.

Conglomerate often lies at the base of any part of the series that rests directly on the ancient slates and gneissic rocks, and occasionally thick conglomerates are intercalated among the sandy strata on various horizons, as, for example, on Moray and Cromarty Firths. These beds are sometimes thin, and sometimes of enormous thickness. Some of these conglomerates are clearly volcanic breccias and ashy beds; as, for example, on part of the Ochil Hills, south of the Firth of Tay, and from thence stretching westward at

[112 Glacial Conglomerates.]

intervals to the Forth, near Bridge of Allan. The ordinary sedimentary conglomerates are frequently very coarse, containing both water-worn and subangular fragments of the underlying rocks from the waste (denudation) of which it has been formed. Some of the fragments I have observed of a yard in diameter, in the great band of conglomerate that lies at the foot of the Grampian Mountains, and others, true boulders, of equal size, on the north coast of Scotland, east of Strathie in Caithness. The Silurian gneiss of the Grampian Hills and of the Highlands in general, is much older than the Old Red Sandstone, and the same may be said of the strata of the Lammermuirs, both of which were disturbed and denuded before the deposition of the Upper Silurian rocks. Later denudations of the same rocks formed the vast conglomerate of Old Red rocks south of Dunbar.

Some of these conglomerates possess a character which unmistakably marks them as glacial boulder clays. The stones are of all sizes, and not mere pebbles, and they are generally subangular, just like those of many of the boulder clays of the last Glacial Epoch. Like some of these boulder clays also, the stones are imbedded in a red marly paste, once unconsolidated clay, and in similar conglomerates in the Cumbrian region, scratched stones have been found in some cases unmistakably resembling those which are allowed by all to have had their markings produced by the agency of glacier ice. A bold man might even go further, for opposite the mouth of the valley of Ullswater, at the outlet of the lake, there are great heaps of angular boulder-conglomerate, culminating in the big mound-like hills of Mell Fell and the neighbourhood, the stones cemented in a marly base. It is an obvious fact to skilled

[Old Red Sandstone. 113]

geologists, well known to those of the Geological Survey, who mapped the ground, that some of the valleys of Cumberland are of older date than the deposition of the Old Red Sandstone, and standing on the ground it was impossible for me not to feel the idea, that Mell Fell and Little Mell Fell look like, and may be, the relics of an old moraine, shed from a glacier of Old Red Sandstone age, that flowed down a valley far older than that of modern Ullswater, and long before the special hollow in which the lake lies was formed. The mountains were much higher than now, for since then they have undergone an immense amount of denudation.

In Scotland fishes are more or less found in all the broader districts occupied by Old Red Sandstone, but it was chiefly in the north, in Caithness, arid on both sides of Cromarty and Moray Firths, that Hugh Miller made his wonderful discoveries of fossil fishes of many species. They are found generally in bituminous schists and flags with occasional nodular concretions, and lie in various minor horizons among red and variegated sandstones and conglomerates, which contain the remains of many fish of the genera Diplopterus, Coccosteus, Pterichthys, Diplacanthus, Osteolepis, Glyptolœmus, Dipterus, Holoptychius, Cephalaspis, besides crustacea, such as Pterygotus, and the small bivalve Estheria Murchisoniœ. (Fig. 26.)

Most of the genera belong to a sub-order of Ganoid fish called Crossopterygidœ by Huxley, from the fringelike arrangement of the fins, a sub-order several species of which are still living in the rivers of Senegal and in the Nile.

I have specially mentioned these circumstances, for the purpose of keeping before the mind of the reader the broad fact that the Old Red Sandstone, as a whole, is

FIG. 26.

Old Red Sandstone Fossils
Group of Old Red Sandstone fossils.

[Physical Geography. 115]

distinct in space, if not in time, from the marine Devonian strata, for in most books both are generally included under the term Devonian, and the ordinary reader makes no distinction between them. There is, however, this marked distinction, that one is of marine and the other of fresh-water origin, and therefore that the latter belongs to a broad Continental area, outside the shores of which our British Devonian beds were deposited, while in other areas, such as part of Russia, the intermingling of fresh-water and marine interstratifications seems to imply a set of estuarine conditions. That our Old Red Sandstone, to the very top, was of fresh-water origin is evident, not only by the presence of special genera of fish, but also in the rocks of Dura Den, of a fresh-water shell, Amodortta Jukesii, and of ferns, Adiantites Hibernicus and Cyclopteris, also Lepidodendron, &c. The shell proves fresh water, and the plants the vicinity of land. See Fig. 25, p. 101.

When all the foregoing statements are fairly considered, it seems to me that we obtain sufficient material from which to form a conception of the physical geography of our area during the deposition of the Old Red Sandstone; as follows:—

In a mountainous region of which the Scandinavian chain formed part, the lakes of the Old Red Sandstone epoch lay; for patches of these strata opposite Scotland, and bordering the sea, lie on the Norwegian coast. What was the extent of the Great Lake in which the central Scottish strata were deposited I am unable to say, for they strike out to sea in the Firth of Clyde on the west and to the North Sea on the east coast, forming a stretch of country 100 miles in length by about 60 in breadth. Whether or not, the Old Red Sandstone of

[116 Physical Geography.]

this area was originally united to that which borders the Firths of Moray and Dornoch, and from thence on to the sandstones of Caithness and the Orkneys, I cannot tell, though it has been usually stated that the eastern side of the Lower Silurian rocks and the Grampian heights were continuously fringed by Old Red Sandstone. It seems to me, however, to be not unlikely, that as the great Grampian range south of the Dee even now attains to heights of about 2,000 feet in Kincardineshire, in older times, having suffered much less from denudation, they were higher than now, stretched further east, and possibly formed an effectual barrier between two lake-areas in which Old Red Sandstone was deposited. But even if the red sandstones of the whole of Scotland were once united to those of the coast of Norway, in one continuous stretch of inland water, it is not without parallel in the living world, for the brackish Caspian lake occupies a larger area, and it has been said that even in historical times the Caspian was larger than now. The great fresh-water lakes also of North America, from Lake Superior to Lake Erie, exclusive of Ontario, occupy an area far larger than the whole of Scotland with all its islands. Three of these lakes, Superior, Michigan, and Huron, practically form one sheet of water, united by straits somewhat analogous to those of the Bosphorus and Hellespont; and the lowest of these, Lake Huron, is only 37 feet below the level of Lake Superior, while Erie is 36 feet lower than Lake Huron, with a distance of more than 70 miles between them, part of which is occupied by Lake St. Clair.

When we try to realise the kind of scenery of this old period, we are led to something of this kind. The lake or lakes, was or were, more or less encircled by high

[Physical Geography. 117]

mountains, and on the banks, and perhaps as occasional islands, volcanic cones disgorged streams of lava and discharged showers of ashes and stones, to be interstratified with the ordinary sediments, in a manner analogous to that which accompanied the deposition of the Miocene strata in Auvergne and other areas in what is now central France. At the same time, from the lofty mountains that now form the Highlands, but higher then, glaciers descended into the water, and fleets of icebergs floated hither and thither, and, melting, dropped their moraine matter to intermingle with other sediments, while further south, in Cumbria, similar glaciers descended from the ancient mountains, higher and different in form from those of modern date in the same area.

In a region still further south, we come to the lake in which the Old Red Sandstone of South Wales and the adjoining counties was deposited. These strata certainly spread further north and west than the edge of the main mass does now, a fact shown by the large outliers by Presteign, Clun, and Bettws Crwyn in Montgomeryshire. Making an allowance for this extension, the lake must have been not less than about 100 miles in length, by a breadth varying from 70 to 100 miles, for traces of Old Red Sandstone have been proved in deep borings through the coalmeasures at the south end of the South Staffordshire coal-field. Away in the distant west, rose the lofty mountains formed in part of the far more ancient Lower Silurian rocks of North Wales, but no contemporaneous volcanic rocks are anywhere found among the Old Red Sandstone strata that were deposited in the adjacent lake, the eastern shores of which were, I think, low and unimposing.

[118 Plants.]

Respecting the vegetation of the country there is little to say, for the ferns and lepidodendrons afford but feeble and fragmentary evidence. It may have been that the whole region stood at a comparatively high and bleak level, or it may be that the plant-bearing localities remain to be discovered. This, however, we know, that in North America, in equivalent strata, there lie buried the remains of a large and luxuriant flora, which generically has close affinities with that of the succeeding Carboniferous Epoch. Such plants as we have lie, some at the base, and others near the top of the British Old Red Sandstone, which, indeed, in some areas gradually merges into the Carboniferous strata, that, under varying marine and wide-spread terrestrial conditions, form the next stage of one long Continental Epoch.