It only remains to be added, that whilst Eozooen is by far the most important organic body hitherto found in the Laurentian, and has been here treated at proportionate length, other traces of life have been detected, which may subsequently prove of great interest and importance. Thus, Principal Dawson has recently described under the name of Archoeosphoerinoe certain singular rounded bodies which he has discovered in the Laurentian limestones, and which he believes to be casts of the shells of Foraminifera possibly somewhat allied to the existing Globigerinoe. The same eminent palaeontologist has also described undoubted worm-burrows from rocks probably of Laurentian age. Further and more extended researches, we may reasonably hope, will probably bring to light other actual remains of organisms in these ancient deposits.

THE HURONIAN PERIOD.

The so-called Huronian Rocks, like the Laurentian, have their typical development in Canada, and derive their name from the fact that they occupy an extensive area on the borders of Lake Huron. They are wholly metamorphic, and consist principally of altered sandstones or quartzites, siliceous, felspathic, or talcose slates, conglomerates, and limestones. They are largely developed on the north shore of Lake Superior, and give rise to a broken and hilly country, very like that occupied by the Laurentians, with an abundance of timber, but rarely with sufficient soil of good quality for agricultural purposes. They are, however, largely intersected by mineral veins, containing silver, gold, and other metals, and they will ultimately doubtless yield a rich harvest to the miner. The Huronian Rocks have been identified, with greater or less certainty, in other parts of North America, and also in the Old World.

The total thickness of the Huronian Rocks in Canada is estimated as being not less than 18,000 feet, but there is considerable doubt as to their precise geological position. In their typical area they rest unconformably on the edges of strata of Lower Laurentian age; but they have never been seen in direct contact with the Upper Laurentian, and their exact relations to this series are therefore doubtful. It is thus open to question whether the Huronian Rocks constitute a distinct formation, to be intercalated in point of time between the Laurentian and the Cambrian groups; or whether, rather, they should not be considered as the metamorphosed representatives of the Lower Cambrian Rocks of other regions.

As regards the fossils of the Huronian Rocks, little can be said. Some of the specimens of Eozooen Canadense which have been discovered in Canada are thought to come from rocks which are probably of Huronian age. In Bavaria, Dr Guembel has described a species of Eozooen under the name of Eozooen Bavaricum, from certain metamorphic limestones which he refers to the Huronian formation. Lastly, the late Mr Billings described, from rocks in Newfoundland apparently referable to the Huronian, certain problematical limpet-shaped fossils, to which he gave the name of Aspidella.

LITERATURE.

Amongst the works and memoirs which the student may consult with regard to the Laurentian and Huronian deposits may be mentioned the following:[10]—

(1) 'Report of Progress of the Geological Survey of Canada from its Commencement to 1863,' pp. 38-49, and pp. 50-66. (2) 'Manual of Geology.' Dana. 2d Ed. 1875. (3) 'The Dawn of Life.' J. W, Dawson. 1876. (4) "On the Occurrence of Organic Remains in the Laurentian Rocks of Canada." Sir W. E. Logan. 'Quart. Journ. Geol. Soc.,' xxi. 45-50.' (5) "On the Structure of Certain Organic Remains in the Laurentian Limestones of Canada." J. W. Dawson. 'Quart. Journ. Geol. Soc.,' xxi. 51-59. (6) "Additional Note on the Structure and Affinities of Eozooen Canadense." W. B, Carpenter. 'Quart. Journ. Geol. Soc.,' xxi. 59-66. (7) "Supplemental Notes on the Structure and Affinities of Eozooen' Canadense," W. B. Carpenter, 'Quart. Journ. Geol. Soc.,' xxii. 219-228. (8) "On the So-Called Eozooenal Rocks." King & Rowney. 'Quart. Journ. Geol. Soc.,' xxii. 185-218. (9) 'Chemical and Geological Essays.' Sterry Hunt.

The above list only includes some of the more important memoirs which may be consulted as to the geological and chemical features of the Laurentian and Huronian Rocks, and as to the true nature of Eozooen. Those who are desirous of studying the later phases of the controversy with regard to Eozooen must consult the papers of Carpenter, Carter, Dawson, King & Rowney, Hahn, and others, in the 'Quart. Journ. of the Geological Society,' the 'Proceedings of the Royal Irish Academy,' the 'Annals of Natural History,' the 'Geological Magazine,' &c. Dr Carpenter's 'Introduction to the Study of the Foraminifera' should also be consulted.

[Footnote 10: In this and in all subsequently following bibliographical lists, not only is the selection of works and memoirs quoted necessarily extremely limited; but only such have, as a general rule, been chosen for mention as are easily accessible to students who are in the position of being able to refer to a good library. Exceptions, however, are occasionally made to this rule, in favour of memoirs or works of special historical interest. It is also unnecessary to add that it has not been thought requisite to insert in these lists the well-known handbooks of geological and palaeontological science; except in such instances as where they contain special information on special points.]



CHAPTER VIII.

THE CAMBRIAN PERIOD.

The traces of life in the Laurentian period, as we have seen, are but scanty; but the Cambrian Rocks—so called from their occurrence in North Wales and its borders ("Cambria ")—have yielded numerous remains of animals and some dubious plants. The Cambrian deposits have thus a special interest as being the oldest rocks in which occur any number of well-preserved and unquestionable organisms. We have here the remains of the first fauna, or assemblage of animals, of which we have at present knowledge. As regards their geographical distribution, the Cambrian Rocks have been recognised in many parts of the world, but there is some question as to the precise limits of the formation, and we may consider that their most typical area is in South Wales, where they have been carefully worked out, chiefly by Dr Henry Hicks. In this region, in the neighbourhood of the promontory of St David's, the Cambrian Rocks are largely developed, resting upon an ancient ridge of Pre-Cambrian (Laurentian?) strata, and overlaid by the lowest beds of the Lower Silurian. The subjoined sketch-section (fig. 27) exhibits in a general manner the succession of strata in this locality.

From this section it will be seen that the Cambrian Rocks in Wales are divided in the first place into a lower and an upper group. The Lower Cambrian is constituted at the base by a great series of grits, sandstones, conglomerates, and slates, which are known as the "Longmynd group," from their vast development in the Longmynd Hills in Shropshire, and which attain in North Wales a thickness of 8000 feet or more. The Longmynd beds are succeeded by the so-called "Menevian group," a series of sandstones, flags, and grits, about 600 feet in thickness, and containing a considerable number of fossils. The Upper Cambrian series consists in its lower portion of nearly 5000 feet of strata, principally shaly and slaty, which are known as the "Lingula Flags," from the great abundance in them of a shell referable to the genus Lingula. These are followed by 1000 feet of dark shales and flaggy sandstones, which are known as the "Tremadoc slates," from their occurrence near Tremadoc in North Wales; and these in turn are surmounted, apparently quite conformably, by the basement beds of the Lower Silurian.



The above may be regarded as giving a typical series of the Cambrian Rocks in a typical locality; but strata of Cambrian age are known in many other regions, of which it is only possible here to allude to a few of the most important. In Scandinavia occurs a well-developed series of Cambrian deposits, representing both the lower and upper parts of the formation. In Bohemia, the Upper Cambrian, in particular, is largely developed, and constitutes the so-called "Primordial zone" of Barrande. Lastly, in North America, whilst the Lower Cambrian is only imperfectly developed, or is represented by the Huronian, the Upper Cambrian formation has a wide extension, containing fossils similar in character to the analogous strata in Europe, and known as the "Potsdam Sandstone." The subjoined table shows the chief areas where Cambrian Rocks are developed, and their general equivalency:

TABULAR VIEW OF THE CAMBRIAN FORMATION.

Britain. Europe. America. /a. Tremadoc Slates. a. Primordial zone a. Potsdam of Bohemia. Sandstone. b. Lingula Flags. b. Paradoxides b. Acadian Upper < Schists, Olenus group of New Cambrian. Schists, and Brunswick. Dictyonema schists of Sweden. /a. Longmynd Beds. a. Fucoidal Huronian Sandstone of Sweden Formation? b. Llanberis Slates. b. Eophyton Sandstone of Sweden. Lower < c. Harlech Grits. Cambrian. d. Oldhamia Slates of Ireland. e. Conglomerates and and Sandstones of Sutherlandshire? f. Menevian Beds.

Like all the older Palaeozoic deposits, the Cambrian Rocks, though by no means necessarily what would be called actually "metamorphic," have been highly cleaved, and otherwise altered from their original condition. Owing partly to their indurated state, and partly to their great antiquity, they are usually found in the heart of mountainous districts, which have undergone great disturbance, and have been subjected to an enormous amount of denudation. In some cases, as in the Longmynd Hills in Shropshire, they form low rounded elevations, largely covered by pasture, and with few or no elements of sublimity. In other cases, however, they rise into bold and rugged mountains, girded by precipitous cliffs. Industrially, the Cambrian Rocks are of interest, if only for the reason that the celebrated Welsh slates of Llanberis are derived from highly-cleaved beds of this age. Taken as a whole, the Cambrian formation is essentially composed of arenaceous and muddy sediments, the latter being sometimes red, but more commonly nearly black in colour. It has often been supposed that the Cambrians are a deep-sea deposit, and that we may thus account for the few fossils contained in them; but the paucity of fossils is to a large extent imaginary, and some of the Lower Cambrian beds of the Longmynd Hills would appear to have been laid down in shallow water; as they exhibit rain-prints, sun-cracks, and ripple-marks—incontrovertible evidence of their having been a shore-deposit. The occurrence, of innumerable worm-tracks and burrows in many Cambrian strata is also a proof of shallow-water conditions; and the general absence of limestones, coupled with the coarse mechanical nature of many of the sediments of the Lower Cambrian, maybe taken as pointing in the same direction.

The life of the Cambrian, though not so rich as in the succeeding Silurian period, nevertheless consists of representatives of most of the great classes of invertebrate animals. The coarse sandy deposits of the formation, which abound more particularly towards its lower part, naturally are to a large extent barren of fossils; but the muddy sediments, when not too highly cleaved, and especially towards the summit of the group, are replete with organic remains. This is also the case, in many localities at any rate, with the finer beds of the Potsdam Sandstone in America. Limestones are known to occur in only a few areas (chiefly in America), and this may account for the apparent total absence of corals. It is, however, interesting to note that, with this exception, almost all the other leading groups of Invertebrates are known to have come into existence during the Cambrian period.

Fig. 28.—Fragment of Eophyton Linneanum, a supposed land-plant. Lower Cambrian, Sweden, of the natural size.

Of the land-surfaces of the Cambrian period we know nothing; and there is, therefore, nothing surprising in the fact that our acquaintance with the Cambrian vegetation is confined to some marine plants or sea-weeds, often of a very obscure and problematical nature. The "Fucoidal Sandstone" of Sweden, and the "Potsdam Sandstone" of North America, have both yielded numerous remains which have been regarded as markings left by sea-weeds or "Fucoids;" but these are highly enigmatical in their characters, and would, in many instances, seem to be rather referable to the tracks and burrows of marine worms. The first-mentioned of these formations has also yielded the curious, furrowed and striated stems which have been described as a kind of land-plant under the name of Eopkyton (fig. 28). It cannot be said, however, that the vegetable origin of these singular bodies has been satisfactorily proved. Lastly, there are found in certain green and purple beds of Lower Cambrian age at Bray Head, Wicklow, Ireland, some very remarkable fossils, which are well known under the name of Oldhamia, but the true nature of which is very doubtful. The commonest form of Oldhamia (fig. 29) consists of a thread-like stem or axis, from which spring at regular intervals bundles of short filamentous branches in a fan-like manner. In the locality where it occurs, the fronds of Oldhamia are very abundant, and are spread over the surfaces of the strata in tangled layers. That it is organic is certain, and that it is a calcareous sea-weed is probable; but it may possibly belong to the sea-mosses (Polyzoa), or to the sea-firs (Sertularians).

Amongst the lower forms of animal life (Protozoa), we find the Sponges represented by the curious bodies, composed of netted fibres, to which the name of Protospongia has been given (fig. 32, a); and the comparatively gigantic, conical, or cylindrical fossils termed Archoeocyathus by Mr Billings are certainly referable either to the Foraminifera or to the Sponges. The almost total absence of limestones in the formation may be regarded as a sufficient explanation of the fact that the Foraminifera are not more largely and unequivocally represented; though the existence of greensands in the Cambrian beds of Wisconsin and Tennessee may be taken as an indication that this class of animals was by no means wholly wanting. The same fact may explain the total absence of corals, so far as at present known.



The group of the Echinodermata (Sea-lilies, Sea-urchins, and their allies) is represented by a few forms, which are principally of interest as being the earliest-known examples of the class. It is also worthy of note that these precursors of a group which subsequently attains such geological importance, are referable to no less than three distinct orders—the Crinoids or Sea-lilies, represented by a species of Dendrocrinus; the Cystideans by Protocystites; and the Star-fishes by Palasterina and some other forms. Only the last of these groups, however, appears to occur in the Lower Cambrian.



The Ringed-worms (Annelida), if rightly credited with all the remains usually referred to them, appear to have swarmed in the Cambrian seas. Being soft-bodied, we do not find the actual worms themselves in the fossil condition, but we have, nevertheless, abundant traces of their existence. In some cases we find vertical burrows of greater or less depth, often expanded towards their apertures, in which the worm must have actually lived (fig. 30), as various species do at the present day. In these cases, the tube must have been rendered more or less permanent by receiving a coating of mucus, or perhaps a genuine membranous secretion, from the body of the animal; and it may be found quite empty, or occupied by a cast of sand or mud. Of this nature are the burrows which have been described under the names of Scolithus and Scolecoderma, and probably the Histioderma of the Lower Cambrian of Ireland. In other cases, as in Arenicolites (fig. 32, b), the worm seems to have inhabited a double burrow, shaped like the letter U, and having two openings placed close together on the surface of the stratum. Thousands of these twin-burrows occur in some of the strata of the Longmynd, and it is supposed that the worm used one opening to the burrow as an aperture of entrance, and the other as one of exit. In other cases, again, we find simply the meandering trails caused by the worm dragging its body over the surface of the mud. Markings of this kind are commoner in the Silurian Rocks, and it is generally more or less doubtful whether they may not have been caused by other marine animals, such as shellfish, whilst some of them have certainly nothing whatever to do with the worms. Lastly, the Cambrian beds often show twining cylindrical bodies, commonly more or less matted together, and not confined to the surfaces of the strata, but passing through them. These have often been regarded as the remains of sea-weeds, but it is more probable that they represent casts of the underground burrows of worms of similar habits to the common lob-worm (Arenicola) of the present day.

The Articulate animals are numerously represented in the Cambrian deposits, but exclusively by the class of Crustaceans. Some of these are little double-shelled creatures, resembling our living water-fleas (Ostracoda). A few are larger forms, and belong to the same group as the existing brine-shrimps and fairy-shrimps (Phyllopoda). One of the most characteristic of these is the Hymenocaris vermicauda of the Lingula Flags (fig. 32, d). By far the larger number of the Cambrian Crustacea belong, however, to the remarkable and wholly extinct group of the Trilobites. These extraordinary animals must have literally swarmed in the seas of the later portion of this and the whole of the succeeding period; and they survived in greatly diminished numbers till the earlier portion of the Carboniferous period. They died out, however, wholly before the close of the Palaeozoic epoch, and we have no Crustaceans at the present day which can be considered as their direct representatives. They have, however, relationships of a more or less intimate character with the existing groups of the Phyllopods, the King-crabs (Limulus), and the Isopods ("Slaters," Wood-lice, &c.) Indeed, one member of the last-mentioned order, namely, the Serolis of the coasts of Patagonia, has been regarded as the nearest living ally of the Trilobites. Be this as it may, the Trilobites possessed a skeleton which, though capable of undergoing almost endless variations, was wonderfully constant in its pattern of structure, and we may briefly describe here the chief features of this.



The upper surface of the body of a Trilobite was defended by a strong shell or "crust," partly horny and partly calcareous in its composition. This shell (fig. 31) generally exhibits a very distinct "trilobation" or division into three longitudinal lobes, one central and two lateral. It also exhibits a more important and more fundamental division into three transverse portions, which are so loosely connected with one another as very commonly to be found separate. The first and most anterior of these divisions is a shield or buckler which covers the head; the second or middle portion is composed of movable rings covering the trunk ("thorax "); and the third is a shield which covers the tailor "abdomen." The head-shield (fig. 31, e) is generally more or less semicircular in shape; and its central portion, covering the stomach of the animal, is usually strongly elevated, and generally marked by lateral furrows. A little on each side of the head are placed the eyes, which are generally crescentic in shape, and resemble the eyes of insects and many existing Crustaceans in being "compound," or made up of numerous simple eyes aggregated together. So excellent is the state of preservation of many specimens of Trilobites, that the numerous individual lenses of the eyes have been uninjured, and as many as four hundred have been counted in each eye of some forms. The eyes may be supported upon prominences, but they are never carried on movable stalks (as they are in the existing lobsters and crabs); and in some of the Cambrian Trilobites, such as the little Agnosti (fig. 31 g), the animal was blind. The lateral portions of the head-shield are usually separated from the central portion by a peculiar line of division (the so-called "facial suture") on each side; but this is also wanting in some of the Cambrian species. The backward angles of the head-shield, also, are often prolonged into spines, which sometimes reach a great length. Following the head-shield behind, we have a portion of the body which is composed of movable segments or "body-rings," and which is technically called the "thorax," Ordinarily, this region is strongly trilobed, and each ring consists of a central convex portion, and of two flatter side-lobes. The number of body-rings in the thorax is very variable (from two to twenty-six), but is fixed for the adult forms of each group of the Trilobites. The young forms have much fewer rings than the full-grown ones; and it is curious to find that the Cambrian Trilobites very commonly have either a great many rings (as in Paradoxides, fig. 31, a), or else very few (as in Agnostus, fig. 31, g). In some instances, the body-rings do not seem to have been so constructed as to allow of much movement, but in other cases this region of the body is so flexible that the animal possessed the power of rolling itself up completely, like a hedgehog; and many individuals have been permanently preserved as fossils in this defensive condition. Finally, the body of the Trilobite was completed by a tail-shield (technically termed the "pygidium"), which varies much in size and form, and is composed of a greater or less number of rings, similar to those which form the thorax, but immovably amalgamated with one another (fig. 31, h).

The under surface of the body in the Trilobites appears to have been more or less entirely destitute of hard structures, with the exception of a well-developed upper lip, in the form of a plate attached to the inferior side of the head-shield in front. There is no reason to doubt that the animal possessed legs; but these structures seem to have resembled those of many living Crustaceans in being quite soft and membranous. This, at any rate, seems to have been generally the case; though structures which have been regarded as legs have been detected on the under surface of one of the larger species of Trilobites. There is also, at present, no direct evidence that the Trilobites possessed the two pairs of jointed feelers ("antennae") which are so characteristic of recent Crustaceans.

The Trilobites vary much in size, and the Cambrian formation presents examples of both the largest and the smallest members of the order. Some of the young forms may be little bigger than a millet-seed, and some adult examples of the smaller species (such as Agnostus) may be only a few lines in length; whilst such giants of the order as Paradoxides and Asaphus may reach a length of from one to two feet. Judging from what we actually know as to the structure of the Trilobites, and also from analogous recent forms, it would seem that these ancient Crustaceans were mud-haunting creatures, denizens of shallow seas, and affecting the soft silt of the bottom rather than the clear water above. Whenever muddy sediments are found in the Cambrian and Silurian formations, there we are tolerably sure to find Trilobites, though they are by no means absolutely wanting in limestones. They appear to have crawled out upon the sea-bottom, or burrowed in the yielding mud, with the soft under surface directed downwards; and it is probable that they really derived their nutriment from the organic matter contained in the ooze amongst which they lived. The vital organs seem to have occupied the central lobe of the skeleton, by which they were protected; and a series of delicate leaf-like paddles, which probably served as respiratory organs, would appear to have been carried on the under surface of the thorax. That they had their enemies may be regarded as certain; but we have no evidence that they were furnished with any offensive weapons, or, indeed, with any means of defence beyond their hard crust, and the power, possessed by so many of them, of rolling themselves into a ball. An additional proof of the fact that they for the most part crawled along the sea-bottom is found in the occurrence of tracks and markings of various kinds, which can hardly be ascribed to any other creatures with any show of probability. That this is the true nature of some of the markings in question cannot be doubted at all; and in other cases no explanation so probable has yet been suggested. If, however, the tracks which have been described from the Potsdam Sandstone of North America under the name of Protichnites are really due to the peregrinations of some Trilobite, they must have been produced by one of the largest examples of the order.

As already said, the Cambrian Rocks are very rich in the remains of Trilobites. In the lowest beds of the series (Longmynd Rocks), representatives of some half-dozen genera have now been detected, including the dwarf Agnostus and the giant Paradoxides. In the higher beds, the number both of genera and species is largely increased; and from the great comparative abundance of individuals, the Trilobites have every right to be considered as the most characteristic fossils of the Cambrian period,—the more so as the Cambrian species belong to peculiar types, which, for the most part, died out before the commencement of the Silurian epoch.

All the remaining Cambrian fossils which demand any notice here are members of one or other division of the great class of the Mollusca, or "Shell-fish" properly so called. In the Lower Cambrian Rocks the Lamp-shells (Brachiopoda) are the principal or sole representatives of the class, and appear chiefly in three interesting and important types—namely, Lingulella, Discina, and Obolella. Of these the last (fig. 32, i) is highly characteristic of these ancient deposits; whilst Discina is one of those remarkable persistent types which, commencing at this early period, has continued to be represented by varying forms through all the intervening geological formations up to the present day. Lingulella (fig. 32, c), again, is closely allied to the existing "Goose-bill" Lamp-shell (Lingula anatina), and thus presents us with another example of an extremely long-lived type. The Lingulelloe and their successors; the Linguloe, are singular in possessing a shell which is of a horny texture, and contains but a small proportion of calcareous matter. In the Upper Cambrian Rocks, the Lingulelloe become much more abundant, the broad satchel-shaped species known as L. Davisii (fig. 32, e) being so abundant that one of the great divisions of the Cambrian is termed the "Lingula Flags." Here, also, we meet for the first time with examples of the genus Orthis (fig. 32, f, k, l) a characteristic Palaeozoic type of the Brachiopods, which is destined to undergo a vast extension in later ages.



Of the higher groups of the Mollusca the record is as yet but scanty. In the Lower Cambrian, we have but the thin, fragile, dagger-shaped shells of the free-swimming oceanic Molluscs or "Winged-snails" (Pteropoda), of which the most characteristic is the genus Theca (fig. 32, g). In the Upper Cambrian, in addition to these, we have a few Univalves (Gasteropoda), and, thanks to the researches of Dr Hicks, quite a small assemblage of Bivalves (Lamellibranchiata), though these are mostly of no great dimensions (fig. 32, h). Of the chambered Cephalopoda (Cuttle-fishes and their allies), we have but few traces; and these wholly confined to the higher beds of the formation. We meet, however, with examples of the wonderful genus Orthoceras, with its straight, partitioned shell, which we shall find in an immense variety of forms in the Silurian rocks. Lastly, it is worthy of note that the lowest of all the groups of the Mollusca—namely, that of the Sea-mats, Sea-mosses, and Lace-corals (Polyzoa)—is only doubtfully known to have any representatives in the Cambrian, though undergoing a large and varied development in the Silurian deposits.



An exception, however, may with much probability be made to this statement in favour of the singular genus Dictyonema (fig. 33), which is highly characteristic of the highest Cambrian beds (Tremadoc Slates). This curious fossil occurs in the form of fan-like or funnel-shaped expansions, composed of slightly-diverging horny branches, which are united in a net-like manner by numerous delicate cross-bars, and exhibit a row of little cups or cells, in which the animals were contained, on each side. Dictyonema has generally been referred to the Graptolites; but it has a much greater affinity with the plant-like Sea-firs (Sertularians) or the Sea-mosses (Polyzoa), and the balance of evidence is perhaps in favour of placing it with the latter.

LITERATURE.

The following are the more important and accessible works and memoirs which may be consulted in studying the stratigraphical and palaeontological relations of the Cambrian Rocks:—

(1) 'Siluria.' Sir Roderick Murchison. 5th ed., pp. 21-46. (2) 'Synopsis of the Classification of the British Palaeozoic Rocks.' Sedgwick. Introduction to the 3d Fasciculus of the 'Descriptions of British Palaeozoic Fossils in the Woodwardian Museum,' by F. M'Coy, pp. i-xcviii, 1855. (3) 'Catalogue of the Cambrian and Silurian Fossils in the Geological Museum of the University of Cambridge.' Salter. With a Preface by Prof. Sedgwick. 1873. (4) 'Thesaurus Siluricus.' Bigsby. 1868. (5) "History of the Names Cambrian and Silurian." Sterry Hunt.—'Geological Magazine.' 1873. (6) 'Systeme Silurien du Centre de la Boheme.' Barrande. Vol. I. (7) 'Report of Progress of the Geological Survey of Canada, from its Commencement to 1863,' pp. 87-109. (8) 'Acadian Geology.' Dawson. Pp. 641-657. (9) "Guide to the Geology of New York," Lincklaen; and "Contributions to the Palaeontology of New York," James Hall.—'Fourteenth Report on the State Cabinet.' 1861. (10) 'Palaeozoic Fossils of Canada.' Billings. 1865. (11) 'Manual of Geology.' Dana. Pp. 166-182. 2d ed. 1875. (12) "Geology of North Wales," Ramsay; with Appendix on the Fossils, Salter.—'Memoirs of the Geological Survey of Great Britain,' vol. iii. 1866. (13) "On the Ancient Rocks of the St David's Promontory, South Wales, and their Fossil Contents." Harkness and Hicks.—' Quart. Journ. Geol. Soc.,' xxvii. 384-402. 1871. (14) "On the Tremadoc Rocks in the Neighbourhood of St David's, South Wales, and their Fossil Contents." Hicks.—'Quart. Journ. Geol. Soc.,' xxix. 39-52. 1873.

In the above list, allusion has necessarily been omitted to numerous works and memoirs on the Cambrian deposits of Sweden and Norway, Central Europe, Russia, Spain, and various parts of North America, as well as to a number of important papers on the British Cambrian strata by various well-known observers. Amongst these latter may be mentioned memoirs by Prof. Phillips, and Messrs Salter, Hicks, Belt, Plant, Homfray, Ash, Holl, &c.



CHAPTER IX.

THE LOWER SILURIAN PERIOD.

The great system of deposits to which Sir Roderick Murchison applied the name of "Silurian Rocks" reposes directly upon the highest Cambrian beds, apparently without any marked unconformity, though with a considerable change in the nature of the fossils. The name "Silurian" was originally proposed by the eminent geologist just alluded to for a great series of strata lying below the Old Red Sandstone, and occupying districts in Wales and its borders which were at one time inhabited by the "Silures," a tribe of ancient Britons. Deposits of a corresponding age are now known to be largely developed in other parts of England, in Scotland, and in Ireland, in North America, in Australia, in India, in Bohemia, Saxony, Bavaria, Russia, Sweden and Norway, Spain, and in various other regions of less note. In some regions, as in the neighbourhood of St Petersburg, the Silurian strata are found not only to have preserved their original horizontality, but also to have retained almost unaltered their primitive soft and incoherent nature. In other regions, as in Scandinavia and many parts of North America, similar strata, now consolidated into shales, sandstones, and limestones, may be found resting with a very slight inclination on still older sediments. In a great many regions, however, the Silurian deposits are found to have undergone more or less folding, crumpling, and dislocation, accompanied by induration and "cleavage" of the finer and softer sediments; whilst in some regions, as in the Highlands of Scotland, actual "metamorphism" has taken place. In consequence of the above, Silurian districts usually present the bold, rugged, and picturesque outlines which are characteristic of the older "Primitive" rocks of the earth's crust in general. In many instances, we find Silurian strata rising into mountain-chains of great grandeur and sublimity, exhibiting the utmost diversity of which rock-scenery is capable, and delighting the artist with endless changes of valley, lake, and cliff. Such districts are little suitable for agriculture, though this is often compensated for by the valuable mineral products contained in the rocks. On the other hand, when the rocks are tolerably soft and uniform in their nature, or when few disturbances of the crust of the earth have taken place, we may find Silurian areas to be covered with an abundant pasturage or to be heavily timbered.

Under the head of "Silurian Rocks," Sir Roderick Murchison included all the strata between the summit of the "Longmynd." beds and the Old Red Sandstone, and he divided these into the two great groups of the Lower Silurian and Upper Silurian. It is, however, now generally admitted that a considerable portion of the basement beds of Murchison's Silurian series must be transferred—-if only upon palaeontological grounds—to the Upper Cambrian, as has here been done; and much controversy has been carried on as to the proper nomenclature of the Upper Silurian and of the remaining portion of Murchison's Lower Silurian. Thus, some would confine the name "Silurian" exclusively to the Upper Silurian, and would apply the name of "Cambro-Silurian" to the Lower Silurian, or would include all beds of the latter age in the "Cambrian" series of Sedgwick. It is not necessary to enter into the merits of these conflicting views. For our present purpose, it is sufficient to recognise that there exist two great groups of rocks between the highest Cambrian beds, as here defined, and the base of the Devonian or Old Red Sandstone. These two great groups are so closely allied to one another, both physically and palaeontologically, that many authorities have established a third or intermediate group (the "Middle Silurian"), by which a passage is made from one into the other. This method of procedure involves disadvantages which appear to outweigh its advantages; and the two groups in question are not only generally capable of very distinct stratigraphical separation, but at the same time exhibit, together with the alliances above spoken of, so many and such important palaeontological differences, that it is best to consider them separately. We shall therefore follow this course in the present instance; and pending the final solution of the controversy as to Cambrian and Silurian nomenclature, we shall distinguish these two groups of strata as the "Lower Silurian" and the "Upper Silurian."

The Lower Silurian Rocks are known already to be developed in various regions; and though their general succession in these areas is approximately the same, each area exhibits peculiarities of its own, whilst the subdivisions of each are known by special names. All, therefore, that can be attempted here, is to select two typical areas—such as Wales and North America and to briefly consider the grouping and divisions of the Lower Silurian in each.

In Wales, the line between the Cambrian and Lower Silurian is somewhat ill-defined, and is certainly not marked by any strong unconformity. There are, however; grounds for accepting the line proposed, for palaeontological reasons, by Dr Hicks, in accordance with which the Tremadoc Slates ("Lower Tremadoc" of Salter) become the highest of the Cambrian deposits of Britain. If we take this view, the Lower Silurian rocks of Wales and adjoining districts are found to have the following general succession from below upwards (fig. 34):—

1. The Arenig Group.—This group derives its name from the Arenig mountains, where it is extensively developed. It consists of about 4000 feet of slates, shales, and flags, and is divisible into a lower, middle, and upper division, of which the former is often regarded as Cambrian under the name of "Upper Tremadoc Slates."

2. The Llandeilo Group.—The thickness of this group varies from about 4000 to as much as 10,000 feet; but in this latter case a great amount of the thickness is made up of volcanic ashes and interbedded traps. The sedimentary beds of this group are principally slates and flags, the latter occasionally with calcareous bands; and the whole series can be divided into a lower, middle, and upper Llandeilo division, of which the last is the most important. The name of "Llandeilo" is derived from the town of the same name in Wales, where strata of this age were described by Murchison.

3. The Caradoc or Bala Group.—The alternative names of this group are also of local origin, and are derived, the one from Caer Caradoc in Shropshire, the other from Bala in Wales, strata of this age occurring in both localities. The series is divided into a lower and upper group, the latter chiefly composed of shales and flags, and the former of sandstones and shales, together with the important and interesting calcareous band known as the "Bala Limestone." The thickness of the entire series varies from 4000 to as much as 12,000 feet, according as it contains more or less of interstratified igneous rocks.

4. The Llandovery Group (Lower Llandovery of Murchison).—This series, as developed near the town of Llandovery, in Caermarthenshire, consists of less than 1000 feet of conglomerates, sandstones, and shales. It is probable, however, that the little calcareous band known as the "Hirnant Limestone," together with certain pale-coloured slates which lie above the Bala Limestone, though usually referred to the Caradoc series, should in reality be regarded as belonging to the Llandovery group.

The general succession of the Lower Silurian strata of Wales and its borders, attaining a maximum thickness (along with contemporaneous igneous matter) of nearly 30,000 feet, is diagramatically represented in the annexed sketch-section (fig. 34):—



In North America, both in the United States and in Canada, the Silurian rocks are very largely developed, and may be regarded as constituting an exceedingly full and typical series of the deposits of this period. The chief groups of the Silurian rocks of North America are as follows, beginning, as before, with the lowest strata, and proceeding upwards (fig. 35):—

1. Quebec Group.—This group is typically developed in the vicinity of Quebec, where it consists of about 5000 feet of strata, chiefly variously-coloured shales, together with some sandstones and a few calcareous bands. It contains a number of peculiar Graptolites, by which it can be identified without question with the Arenig group of Wales and the corresponding Skiddaw Slates of the North of England. It is also to be noted that numerous Trilobites of a distinct Cambrian facies have been obtained in the limestones of the Quebec group, near Quebec. These fossils, however, have been exclusively obtained from the limestones of the group; and as these limestones are principally calcareous breccias or conglomerates, there is room for believing that these primordial fossils are really derived, in part at any rate, from fragments of an upper Cambrian limestone. In the State of New York, the Graptolitic shales of Quebec are wanting; and the base of the Silurian is constituted by the so-called "Calciferous Sand-rock" and "Chazy Limestone."[11] The first of these is essentially and typically calcareous, and the second is a genuine limestone.

[Footnote 11: The precise relations of the Quebec shales with Graptolites (Levis Formation) to the Calciferous and Chazy beds are still obscure, though there seems little doubt but that the Quebec Shales are superior to the Calciferous Sand-rock.]

2. The Trenton Group.—This is an essentially calcareous group, the various limestones of which it is composed being known as the "Bird's-eye," "Black River," and "Trenton" Limestones, of which the last is the thickest and most important. The thickness of this group is variable, and the bands of limestone in it are often separated by beds of shale.

3. The Cincinnati Group (Hudson River Formation[12]).—This group consists essentially of a lower series of shales, often black in colour and highly charged with bituminous matter (the "Utica Slates "), and of an upper series of shales, sandstones, and limestones (the "Cincinnati" rocks proper). The exact parallelism of the Trenton and Cincinnati groups with the subdivisions of the Welsh Silurian series can hardly be stated positively. Probably no precise equivalency exists; but there can be no doubt but that the Trenton and Cincinnati groups correspond, as a whole, with the Llandeilo and Caradoc groups of Britain. The subjoined diagrammatic section (fig. 35) gives a general idea of the succession of the Lower Silurian rocks of North America:—



[Footnote 12: There is some difficulty about the precise nomenclature of this group. It was originally called the "Hudson River Formation;" but this name is inappropriate, as rocks of this age hardly touch anywhere the actual Hudson River itself, the rocks so called formerly being now known to be of more ancient date. There is also some want of propriety in the name of "Cincinnati Group," since the rocks which are known under this name in the vicinity of Cincinnati itself are the representatives of the Trenton Limestone, Utica Slates, and the old Hudson River group, inseparably united in what used to be called the "Blue Limestone Series."].

Of the life of the Lower Silurian period we have record in a vast number of fossils, showing that the seas of this period were abundantly furnished with living denizens. We have, however, in the meanwhile, no knowledge of the land-surfaces of the period. We have therefore no means of speculating as to the nature of the terrestrial animals of this ancient age, nor is anything known with certainty of any land-plants which may have existed. The only relics of vegetation upon which a positive opinion can be expressed belong to the obscure group of the "Fucoids," and are supposed to be the remains of sea-weeds. Some of the fossils usually placed under this head are probably not of a vegetable nature at all, but others (fig. 36) appear to be unquestionable plants. The true affinities of these, however, are extremely dubious. All that can be said is, that remains which appear to be certainly vegetable, and which are most probably due to marine plants, have been recognised nearly at the base of the Lower Silurian (Arenig), and that they are found throughout the series whenever suitable conditions recur.

The Protozoans appear to have flourished extensively in the Lower Silurian seas, though to a large extent under forms which are still little understood. We have here for the first time the appearance of Foraminifera of the ordinary type—one of the most interesting observations in this collection being that made by Ehrenberg, who showed that the Lower Silurian sandstones of the neighbourhood of St Petersburg contained casts in glauconite of Foraminiferous shells, some of which are referable to the existing genera Rotalia and Texularia. True Sponges, belonging to that section of the group in which the skeleton is calcareous, are also not unknown, one of the most characteristic genera being Astylospongia (fig. 37). In this genus are included more or less globular, often lobed sponges, which are believed not to have been attached to foreign bodies. In the form here figured there is a funnel-shaped cavity at the summit; and the entire mass of the sponge is perforated, as in living examples, by a system of canals which convey the sea-water to all parts of the organism. The canals by which the sea-water gains entrance open on the exterior of the sphere, and those by which it again escapes from the sponge open into the cup-shaped depression at the summit.



The most abundant, and at the same time the least understood, of Lower Silurian Protozoans belong, however, to the genera Stromatopora and Receptaculites, the structure of which can merely be alluded to here. The specimens of Stromatopora (fig. 38) occur as hemispherical, pear-shaped, globular, or irregular masses, often of very considerable size, and sometimes demonstrably attached to foreign bodies. In their structure these masses consist of numerous thin calcareous laminae, usually arranged concentrically, and separated by narrow interspaces. These interspaces are generally crossed by numerous vertical calcareous pillars, giving the vertical section of the fossil a lattice-like appearance. There are also usually minute pores in the concentric laminae, by which the successive interspaces are placed in communication; and sometimes the surface presents large rounded openings, which appear to correspond with the water-canals of the Sponges. Upon the whole, though presenting some curious affinities to the calcareous Sponges, Stromatopora is perhaps more properly regarded as a gigantic Foraminifer. If this view be correct, it is of special interest as being probably the nearest ally of Eozooen, the general appearance of the two being strikingly similar, though their minute structure is not at all the same. Lastly, in the fossils known as Receptaculites and Ischadites we are also presented with certain singular Lower Silurian Protozoans, which may with great probability be regarded as gigantic Foraminifera. Their structure is very complex; but fragments are easily recognised by the fact that the exterior is covered with numerous rhomboidal calcareous plates, closely fitting together, and arranged in peculiar intersecting curves, presenting very much the appearance of the engine-turned case of a watch.



Passing next to the sub-kingdom of Coelenterate animals (Zoophytes, Corals, &c.), we find that this great group, almost or wholly absent in the Cambrian, is represented in Lower Silurian deposits by a great number of forms belonging on the one hand to the true Corals, and en the other hand to the singular family of the Graptolites. If we except certain plant-like fossils which probably belong rather to the Sertularians or the Polyzoans (e.g., Dictyonema, Dendrograptus, &c.), the family of the Graptolites may be regarded as exclusively Silurian in its distribution. Not only is this the case, but it attained its maximum development almost upon its first appearance, in the Arenig Rocks; and whilst represented by a great variety of types in the Lower Silurian; it only exists in the Upper Silurian in a much diminished form. The Graptolites (Gr. grapho, I write; lithos, stone) were so named by Linnaeus, from the resemblance of some of them to written or pencilled marks upon the stone, though the great naturalist himself did not believe them to be true fossils at all. They occur as linear or leaf-like bodies, sometimes simple, sometimes compound and branched; and no doubt whatever can be entertained as to their being the skeletons of composite organisms, or colonies of semi-independent animals united together by a common fleshy trunk, similar to what is observed in the colonies of the existing Sea-firs (Sertularians). This fleshy trunk or common stem of the colony was protected by a delicate horny sheath, and it gave origin to the little flower-like "polypites," which constituted the active element of the whole assemblage. These semi-independent beings were, in turn, protected each by a little horny cup or cell, directly connected with the common sheath below, and terminating above in an opening through which the polypite could protrude its tentacled head or could again withdraw itself for safety. The entire skeleton, again, was usually, if not universally, supported by a delicate horny rod or "axis," which appears to have been hollow, and which often protrudes to a greater or less extent beyond one or both of the extremities of the actual colony.

The above gives the elementary constitution of any Graptolite, but there are considerable differences as to the manner in which these elements are arranged and combined. In some forms the common stem of the colony gives origin to but a single row of cells on one side. If the common stem is a simple, straight, or slightly-curved linear body, then we have the simplest form of Graptolite known (the genus Monograptus); and it is worthy of note that these simple types do not come into existence till comparatively late (Llandeilo), and last nearly to the very close of the Upper Silurian. In other cases, whilst there is still but a single row of cells, the colony may consist of two of these simple stems springing from a common point, as in the so-called "twin Graptolites" (Didymograptus, fig. 40). This type is entirely confined to the earlier portion of the Lower Silurian period (Arenig and Llandeilo). In other cases, again, there may be four of such stems springing from a central point (Tetragraptus). Lastly, there are numerous complex forms (such as Dichograptus, Loganograptus, &c.) in which there are eight or more of these simple branches, all arising from a common centre (fig. 39), which is sometimes furnished with a singular horny disc. These complicated branching forms, as well as the Tetragrapti, are characteristic of the horizon of the Arenig group. Similar forms, often specifically identical, are found at this horizon in Wales, in the great series of the Skiddaw Slates of the north of England, in the Quebec group in Canada, in equivalent beds in Sweden, and in certain gold-bearing slates of the same age in Victoria in Australia.



In another great group of Graptolites (including the genera Diplograptus, Dicranograptus, Climacograptus, &c.) the common stem of the colony gives origin, over part or the whole or its length, to two rows of cells, one on each side (fig. 41). These "double-celled" Graptolites are highly characteristic of the Lower Silurian deposits; and, with an exception more apparent than real in Bohemia, they are exclusively confined to strata of Lower Silurian age, and are not known to occur in the Upper Silurian. Lastly, there is a group of Graptolites (Phyllograptus, fig. 42) in which the colony is leaf-like in form, and is composed of four rows of cells springing in a cross-like manner from the common stem. These forms are highly characteristic of the Arenig group.



The Graptolites are usually found in dark-coloured, often black shales, which sometimes contain so much carbon as to become "anthracitic." They may be simply carbonaceous; but they are more commonly converted into iron-pyrites, when they glitter with the brilliant lustre of silver as they lie scattered on the surface of the rock, fully deserving in their metallic tracery the name of "written stones." They constitute one of the most important groups of Silurian fossils, and are of the greatest value in determining the precise stratigraphical position of the beds in which they occur. They present, however, special difficulties in their study; and it is still a moot point as to their precise position in the zoological scale. The balance of evidence is in favour of regarding them as an ancient and peculiar group of the Sea-firs (Hydroid Zoophytes), but some regard them as belonging rather to the Sea-mosses (Polyzoa). Under any circumstances, they cannot be directly compared either with the ordinary Sea-firs or the ordinary Sea-mosses; for these two groups consist of fixed organisms, whereas the Graptolites were certainly free-floating creatures, living at large in the open sea. The only Hydroid Zoophytes or Polyzoans which have a similar free mode of existence, have either no skeleton at all, or have hard structures quite unlike the horny sheaths of the Graptolites.

The second great group of Coelenterate animals (Actinozoa) is represented in the Lower Silurian rocks by numerous Corals. These, for obvious reasons, are much more abundant in regions where the Lower Silurian series is largely calcareous (as in North America) than in districts like Wales, where limestones are very feebly developed. The Lower Silurian Corals, though the first of their class, and presenting certain peculiarities, may be regarded as essentially similar in nature to existing Corals. These, as is well known, are the calcareous skeletons of animals—the so-called "Coral-Zoophytes"—closely allied to the common Sea-anemones in structure and habit. A simple coral (fig. 43) consists of a calcareous cup embedded in the soft tissues of the flower-like polype, and having at its summit a more or less deep depression (the "calice") in which the digestive organs are contained. The space within the coral is divided into compartments by numerous vertical calcareous plates (the "septa"), which spring from the inside of the wall of the cup, and of which some generally reach the centre. Compound corals, again (fig. 44), consist of a greater or less number of structures similar in structure to the above, but united together in different ways into a common mass. Simple corals, therefore, are the skeletons of single and independent polypes; whilst compound corals are the skeletons of assemblages or colonies of similar polypes, living united with one another another as an organic community.



In the general details of their structure, the Lower Silurian Corals do not differ from the ordinary Corals of the present day. The latter, however, have the vertical calcareous plates of the coral ("septa") arranged in multiples of six or five; whereas the former have these structures arranged in multiples of four, and often showing a cross-like disposition. For this reason, the common Lower Silurian Corals are separated to form a distinct group under the name of Rugose Corals or Rugosa. They are further distinguished by the fact that the cavity of the coral ("visceral chamber") is usually subdivided by more or less numerous horizontal calcareous plates or partitions, which divide the coral into so many tiers or storeys, and which are known as the "tabulae" (fig. 45).



In addition to the Rugose Corals, the Lower Silurian rocks contain a number of curious compound corals, the tubes of which have either no septa at all or merely rudimentary ones, but which have the transverse partitions or "tabulae" very highly developed. These are known as the Tabulate Corals; and recent researches on some of their existing allies (such as Heliopora) have shown that they are really allied to the modern Sea-pens, Organ-pipe Corals, and Red Coral, rather than to the typical stony Corals. Amongst the characteristic Rugose Corals of the Lower Silurian may be mentioned species belonging to the genera Columnaria, Favistella, Streptelasma, and Zaphrentis; whilst amongst the "Tabulate" Corals, the principal forms belong to the genera Choetetes, Halysites (the Chain-coral), Constellaria, and Heliolites. These groups of the Corals, however, attain a greater development at a later period, and they will be noticed more particularly hereafter.

Upper Silurian, America; B, Pleurocystites squamosus, showing two short "arms," Lower Silurian, Canada; C, Pseudocrinus bifasciatus, Upper Silurian, England; D, Lepadocrinus Gebhartii, Upper Silurian, America. (After Hall, Billings, and Salter.)]

[Footnote 13: The genus Caryocrinus is sometimes regarded as properly belonging to the Crinoids, but there seem to be good reasons for rather considering it as an abnormal form of Cystidean.]

Passing onto higher animals, we find that the class of the Echinodermata is represented by examples of the Star-fishes (Asteroidea), the Sea-lilies (Crinoidea), and the peculiar extinct group of the Cystideans (Cystoidea), with one or two of the Brittle-stars (Ophiuroidea)—the Sea-urchins (Echinoidea) being still wanting. The Crinoids, though in some places extremely numerous, have not the varied development that they possess in the Upper Silurian, in connection with which their structure will be more fully spoken of. In the meanwhile, it is sufficient to note that many of the calcareous deposits of the Lower Silurian are strictly entitled to the name of "Crinoidal limestones," being composed in great part of the detached joints, and plates, and broken stems, of these beautiful but fragile organisms (see fig. 12). Allied to the Crinoids are the singular creatures which are known as Cystideans (fig. 46). These are generally composed of a globular or ovate body (the "calyx"), supported upon a short stalk (the "column"), by which the organism was usually attached to some foreign body. The body was enclosed by closely-fitting calcareous plates, accurately jointed together; and the stem was made up of numerous distinct pieces or joints, flexibly united to each other by membrane. The chief distinction which strikes one in comparing the Cystideans with the Crinoids is, that the latter are always furnished, as will be subsequently seen, with a beautiful crown of branched and feathery appendages, springing from the summit of the calyx, and which are composed of innumerable calcareous plates or joints, and are known as the "arms." In the Cystideans, on the other hand, there are either no "arms" at all, or merely short, unbranched, rudimentary arms. The Cystideans are principally, and indeed nearly exclusively, Silurian fossils; and though occurring in the Upper Silurian in no small numbers, they are pre-eminently characteristic of the Llandeilo-Caradoc period of Lower Silurian time. They commenced their existence, so far as known, in the Upper Cambrian; and though examples are not absolutely unknown in later periods, they are pre-eminently characteristic of the earlier portion of the Palaeozoic epoch.



The Ringed Worms (Annelides) are abundantly represented in the Lower Silurian, but principally by tracks and burrows similar in essential respects to those which occur so commonly in the Cambrian formation, and calling for no special comment. Much more important are the Articulate animals, represented as heretofore, wholly by the remains of the aquatic group of the Crustaceans. Amongst these are numerous little bivalved forms—such as species of Primitia (fig. 47, f), Beyrichia (fig. 47, e), and Leperditia (fig. 47, i and j). Most of these are very small, varying from the size of a pin's head up to that of a hemp seed; but they are sometimes as large as a small bean (fig. 47, i), and they are commonly found in myriads together in the rock. As before said, they belong to the same great group as the living Water-fleas (Ostracoda). Besides these, we find the pod-shaped head-shields of the shrimp-like Phyllopods—such as Caryocaris (fig. 47, d) and Ceratiocaris. More important, however, than any of these are the Trilobites, which may be considered as attaining their maximum development in the Lower Silurian. The huge Paradoxides of the Cambrian have now disappeared, and with them almost all the principal and characteristic "primordial" genera, save Olenus and Agnostus. In their place we have a great number of new forms—some of them, like the great Asaphus tyrannus of the Upper Llandeilo (fig. 47, a), attaining a length of a foot or more, and thus hardly yielding in the matter of size to their ancient rivals. Almost every subdivision of the Lower Silurian series has its own special and characteristic species of Trilobites; and the study of these is therefore of great importance to the geologist. A few widely-dispersed and characteristic species have been here figured (fig. 47); and the following may be considered as the principal Lower Silurian genera—Asaphus, Ogygia, Cheirurus, Ampyx, Caiymene, Trinucleus, Lichas, Illoenus, AEglina, Harpes, Remopleurides, Phacops, Acidaspis, and Homalonotus, a few of them passing upwards under new forms into the Upper Silurian.

Coming next to the Mollusca, we find the group of the Sea-mosses and Sea-mats (Polyzoa) represented now by quite a number of forms. Amongst these are examples of the true Lace-corals (Retepora and Fenestella), with their netted fan-like or funnel-shaped fronds; and along with these are numerous delicate encrusting forms, which grew parasitically attached to shells and corals (Hippothoa, Alecto, &c.); but perhaps the most characteristic forms belong to the genus Ptilodictya (figs. 48 and 49). In this group the frond is flattened, with thin striated edges, sometimes sword-like or scimitar-shaped, but often more or less branched; and it consists of two layers of cells, separated by a delicate membrane, and opening upon opposite sides. Each of these little chambers or "cells" was originally tenanted by a minute animal, and the whole thus constituted a compound organism or colony.



The Lamp-shells or Brachiopods are so numerous, and present such varied types, both in this and the succeeding period of the Upper Silurian, that the name of "Age of Brachiopods" has with justice been applied to the Silurian period as a whole. It would be impossible here to enter into details as to the many different forms of Brachiopods which present themselves in the Lower Silurian deposits; but we may select the three genera Orthis, Strophomena, and Leptoena for illustration, as being specially characteristic of this period, though not exclusively confined to it. The numerous shells which belong to the extensive and cosmopolitan genus Orthis (fig. 50, a, b, c, and fig. 51, c and d), are usually more or less transversely-oblong or subquadrate, the two valves (as more or less in all the Brachiopods) of unequal sizes, generally more or less convex, and marked with radiating ribs or lines. The valves of the shell are united to one another by teeth and sockets, and there is a straight hinge-line. The beaks are also separated by a distinct space ("hinge-area"), formed in part by each valve, which is perforated by a triangular opening, through which, in the living condition, passed a muscular cord attaching the shell to some foreign object. The genus Strophomena (fig. 50, d, and 51, a and b) is very like Orthis in general character; but the shell is usually much flatter, one or other valve often being concave, the hinge-line is longer, and the aperture for the emission of the stalk of attachment is partially closed by a calcareous plate. In Leptoena, again (fig. 51, e), the shell is like Strophomena in many respects, but generally comparatively longer, often completely semicircular, and having one valve convex and the other valve concave. Amongst other genera of Brachiopods which are largely represented in the Lower Silurian rocks may be mentioned Lingula, Crania, Discina, Trematis, Siphonotreta, Acrotreta, Rhynchonella, and Athyris; but none of these can claim the importance to which the three previously-mentioned groups are entitled.



The remaining Lower Silurian groups of Mollusca can be but briefly glanced at here. The Bivalves (Lamellibranchiata) find numerous representatives, belonging to such genera as Modiolopsis, Ctenodonta, Orthonota, Paloearca, Lyrodesma, Ambonychia, and Cleidophorus. The Univalves (Gasteropoda) are also very numerous, the two most important genera being Murchisonia (fig. 52) and Pleurotomaria. In both these groups the outer lip of the shell is notched; but the shell in the former is elongated and turreted, whilst in the latter it is depressed. The curious oceanic Univalves known as the Heteropods are also very abundant, the principal forms belonging to Bellerophon and Maclurea. In the former (fig. 53) there is a symmetrical convoluted shell, like that of the Pearly Nautilus in shape, but without any internal partitions, and having the aperture often expanded and notched behind. The species of Maclurea (fig. 54) are found both in North America and in Scotland, and are exclusively confined to the Lower Silurian period, so far as known. They have the shell coiled into a flat spiral, the mouth being furnished with a very curious, thick, and solid lid or "operculum." The Lower Silurian Pteropods, or "Winged snails," are numerous, and belong principally to the genera Theca, Conularia, and Tentaculites, the last-mentioned of these often being extremely abundant in certain strata.



Restoration of Orthoceras, the shell being supposed to be divided vertically, and only its upper part being shown. a, Arms; f, Muscular tube ("funnel") by which water is expelled from the mantle-chamber; c, Air-chambers; s, Siphuncle.]

[Footnote 14: This illustration is taken from a rough sketch made by the author many years ago, but he is unable to say from what original source it was copied.]

Lastly, the Lower Silurian Rocks have yielded a vast number of chambered shells, referable to animals which belong to the same great division as the Cuttle-fishes (the Cephalopoda), and of which the Pearly Nautilus is the only living representative at the present day. In this group of Cephalopods the animal possesses a well-developed external shell, which is divided into chambers by shelly partitions ("septa"). The animal lives in the last-formed and largest chamber of the shell, to which it is organically connected by muscular attachments. The head is furnished with long muscular processes or "arms," and can be protruded from the mouth of the shell at will, or again withdrawn within it. We learn, also, from the Pearly Nautilus, that these animals must have possessed two pairs of breathing organs or "gills;" hence all these forms are grouped together under the name of the "Tetrabranchiate" Cephalopods (Gr. tetra, four; bragchia, gill). On the other hand, the ordinary Cuttle-fishes and Calamaries either possess an internal skeleton, or if they have an external shell, it is not chambered; their "arms" are furnished with powerful organs of adhesion in the form of suckers; and they possess only a single pair of gills. For this last reason they are termed the "Dibranchiate" Cephalopods (Gr. dis, twice; bragchia, gill). No trace of the true Cuttle-fishes has yet been found in Lower Silurian deposits; but the Tetrabranchiate group is represented by a great number of forms, sometimes of great size. The principal Lower Silurian genus is the well-known and widely-distributed Orthoceras (fig. 55). The shell in this genus agrees with that of the existing Pearly Nautilus, in consisting of numerous chambers separated by shelly partitions (or septa), the latter being perforated by a tube which runs the whole length of the shell after the last chamber, and is known as the "siphuncle" (fig. 56, s). The last chamber formed is the largest, and in it the animal lives. The chambers behind this are apparently filled with some gas secreted by the animal itself; and these are supposed to act as a kind of float, enabling the creature to move with ease under the weight of its shell. The various air-chambers, though the siphuncle passes through them, have no direct connection with one another; and it is believed that the animal has the power of slightly altering its specific gravity, and thus of rising or sinking in the water by driving additional fluid into the siphuncle or partially emptying it. The Orthoceras further agrees with the Pearly Nautilus in the fact that the partitions or septa separating the different air-chambers are simple and smooth, concave in front and convex behind, and devoid of the elaborate lobation which they exhibit in the Ammonites; whilst the siphuncle pierces the septa either in the centre or near it. In the Nautilus, however, the shell is coiled into a flat spiral; whereas in Orthoceras the shell is a straight, longer or shorter cone, tapering behind, and gradually expanding towards its mouth in front. The chief objections to the belief that the animal of the Orthoceras was essentially like that of the Pearly Nautilus are—the comparatively small size of the body-chamber, the often contracted aperture of the mouth, and the enormous size of some specimens of the shell. Thus, some Orthocerata have been discovered measuring ten or twelve feet in length, with a diameter of a foot at the larger extremity. These colossal dimensions certainly make it difficult to imagine that the comparatively small body-chamber could have held an animal large enough to move a load so ponderous as its own shell. To some, this difficulty has appeared so great that they prefer to believe that the Orthoceras did not live in its shell at all, but that its shell was an internal skeleton similar to what we shall find to exist in many of the true Cuttle-fishes. There is something to be said in favour of this view, but it would compel us to believe in the existence in Lower Silurian times of Cuttle-fishes fully equal in size to the giant "Kraken" of fable. It need only be added in this connection that the Lower Silurian rocks have yielded the remains of many other Tetrabranchiate Cephalopods besides Orthoceras. Some of these belong to Cyrtoceras, which only differs from Orthoceras in the bow-shaped form of the shell; others belong to Phragmoceras, Lituites, &c.; and, lastly; we have true Nautili, with their spiral shells, closely resembling the existing Pearly Nautilus.

Whilst all the sub-kingdoms of the Invertebrate animals are represented in the Lower Silurian rocks, no traces of Vertebrate animals have ever been discovered in these ancient deposits, unless the so-called "Conodonts" found by Pander in vast numbers in strata of this age [15] in Russia should prove to be really of this nature. These problematical bodies are of microscopic size, and have the form of minute, conical, tooth-shaped spines, with sharp edges, and hollow at the base. Their original discoverer regarded them as the horny teeth of fishes allied to the Lampreys; but Owen came to the conclusion that they probably belonged to Invertebrates. The recent investigation of a vast number of similar but slightly larger bodies, of very various forms, in the Carboniferous rocks of Ohio, has led Professor Newberry to the conclusion that these singular fossils really are, as Pander thought, the teeth of Cyclostomatous fishes. The whole of this difficult question has thus been reopened, and we may yet have to record the first advent of Vertebrate animals in the Lower Silurian.

[Footnote 15: According to Pander, the "Conodonts" are found not only in the Lower Silurian beds, but also in the "Ungulite Grit" (Upper Cambrian), as well as in the Devonian and Carboniferous deposits of Russia. Should the Conodonts prove to be truly the remains of fishes, we should thus have to transfer the first appearance of vertebrates to, at any rate, as early a period as the Upper Cambrian.]



CHAPTER X.

THE UPPER SILURIAN PERIOD.

Having now treated of the Lower Silurian period at considerable length, it will not be necessary to discuss the succeeding group of the Upper Silurian in the same detail—the more so, as with a general change of species the Upper Silurian animals belong for the most part to the same great types as those which distinguish the Lower Silurian. As compared, also, as regards the total bulk of strata concerned, the thickness of the Upper Silurian is generally very much below that of the Lower Silurian, indicating that they represent a proportionately shorter period of time. In considering the general succession of the Upper Silurian beds, we shall, as before, select Wales and America as being two regions where these deposits are typically developed.

In Wales and its borders the general succession of the Upper Silurian rocks may be taken to be as follows, in ascending order (fig. 57):—

(1) The base of the Upper Silurian series is constituted by a series of arenaceous beds, to which the name of "May Hill Sandstone" was applied by Sedgwick. These are succeeded by a series of greenish-grey or pale-grey slates ("Tarannon Shales"), sometimes of great thickness; and these two groups of beds together form what may be termed the "May Hill Group" (Upper Llandovery of Murchison). Though not very extensively developed in Britain, this zone is one very well marked by its fossils; and it corresponds with the "Clinton Group" of North America, in which similar fossils occur. In South Wales this group is clearly unconformable to the highest member of the subjacent Lower Silurian (the Llandovery group); and there is reason to believe that a similar, though less conspicuous, physical break occurs very generally between the base of the Upper and the summit of the Lower Silurian.

(2) The Wenlock Group succeeds the May Hill group, and constitutes the middle member of the Upper Silurian. At its base it may have an irregular limestone ("Woolhope Limestone"), and its summit may be formed by a similar but thicker calcareous deposit ("Wenlock Limestone"); but the bulk of the group is made up of the argillaceous and shaly strata known as the "Wenlock Shale." In North Wales the Wenlock group is, represented by a great accumulation of flaggy and gritty strata (the "Denbighshire Flags and Grits"), and similar beds (the "Coniston Flags" and "Coniston Grits") take the same place in the north of England.

(3) The Ludlow Group is the highest member of the Upper Silurian, and consists typically of a lower arenaceous and shaly series (the "Lower Ludlow Rock") a middle calcareous member (the "Aymestry Limestone"), and an upper shaly and sandy series (the "Upper Ludlow Rock" and "Downton Sandstone"). At the summit, or close to the summit, of the Upper Ludlow, is a singular stratum only a few inches thick (varying from an inch to a foot), which contains numerous remains of crustaceans and fishes, and is well known under the name of the "bone-bed." Finally, the Upper Ludlow rock graduates invariably into a series of red sandy deposits, which, when of a flaggy character, are known locally as the "Tile-stones." These beds are probably to be regarded as the highest member of the Upper Silurian; but they are sometimes looked upon as passage-beds into the Old Red Sandstone, or as the base of this formation. It is, in fact, apparently impossible to draw any actual line of demarcation between the Upper Silurian and the overlying deposits of the Devonian or Old Red Sandstone series. Both in Britain and in America the Lower Devonian beds repose with perfect conformity upon the highest Silurian beds, and the two formations appear to pass into one another by a gradual and imperceptible transition.

The Upper Silurian strata of Britain vary from perhaps 3000 or 4000 feet in thickness up to 8000 or 10,000 feet. In North America the corresponding series, though also variable, is generally of much smaller thickness, and may be under 1000 feet. The general succession of the Upper Silurian deposits of North America is as follows:—

(1) Medina Sandstone.—This constitutes the base of the Upper Silurian, and consists of sandy strata, singularly devoid of life, and passing below in some localities into a conglomerate ("Oneida Conglomerate"), which is stated to contain pebbles derived from the older beds, and which would thus indicate an unconformity between the Upper and Lower Silurian.

(2) Clinton Group.—Above the Medina sandstone are beds of sandstone and shale, sometimes with calcareous bands, which constitute what is known as the "Clinton Group." The Medina and Clinton groups are undoubtedly the equivalent of the "May Hill Group" of Britain, as shown by the identity of their fossils.



(3) Niagara Group.—This group consists typically of a series of argillaceous beds ("Niagara Shale") capped by limestones ("Niagara Limestone"); and the name of the group is derived from the fact that it is over limestones of this age that the Niagara river is precipitated to form the great Falls. In places the Niagara group is wholly calcareous, and it is continued upwards into a series of marls and sandstones, with beds of salt and masses of gypsum (the "Salina Group"), or into a series of magnesian limestones ("Guelph Limestones"). The Niagara group, as a whole, corresponds unequivocally with the Wenlock group of Britain.

(4) Lower Helderberg Group.—The Upper Silurian period in North America was terminated by the deposition of a series of calcareous beds, which derive the name of "Lower Helderberg" from the Helderberg mountains, south of Albany, and which are divided into several zones, capable of recognition by their fossils, and known by local names (Tentaculite Limestone, Water-lime, Lower Pentamerus Limestone, Delthyris Shaly Limestone, and Upper Pentamerus Limestone). As a whole, this series may be regarded as the equivalent of the Ludlow group of Britain, though it is difficult to establish any precise parallelism. The summit of the Lower Heiderberg group is constituted by a coarse-grained sandstone (the "Oriskany Sandstone"), replete with organic remains, which have to a large extent a Silurian facies. Opinions differ as to whether this sandstone is to be regarded as the highest bed of the Upper Silurian or the base of the Devonian. We thus see that in America, as in Britain, no other line than an artificial one can be drawn between the Upper Silurian and the overlying Devonian.

As regards the life of the Upper Silurian period, we have, as before, a number of so-called "Fucoids," the true vegetable nature of which is in many instances beyond doubt. In addition to these, however, we meet for the first time, in deposits of this age, with the remains of genuine land-plants, though our knowledge of these is still too scanty to enable us to construct any detailed picture of the terrestrial vegetation of the period. Some of these remains indicate the existence of the remarkable genus Lepidodendron—a genus which played a part of great importance in the forests of the Devonian and Carboniferous periods, and which may be regarded as a gigantic and extinct type of the Club-mosses (Lycopodiaceoe). Near the summit of the Ludlow formation in Britain there have also been found beds charged with numerous small globular bodies, which Dr Hooker has shown to be the seed-vessels or "sporangia" of Club-mosses. Principal Dawson further states that he has seen in the same formation fragments of wood with the structure of the singular Devonian Conifer known as Prototaxites. Lastly, the same distinguished observer has described from the Upper Silurian of North America the remains of the singular land-plants belonging to the genus Psilophyton, which will be referred to at greater length hereafter.

The marine life of the Upper Silurian is in the main constituted by types of animals similar to those characterising the Lower Silurian, though for the most part belonging to different species. The Protozoans are represented principally by Stromatopora and Ischadites, along with a number of undoubted sponges (such as Amphispongia, Astroeospongia, Astylospongia, and Paloeomanon).

Amongst the Coelenterates, we find the old group of Graptolites now verging on extinction. Individuals still remain numerous, but the variety of generic and specific types has now become greatly reduced. All the branching and complex forms of the Arenig, the twin-Graptolites and Dicranograpti of the Llandeilo, and the double-celled Diplograpti and Climacograpti of the Bala group, have now disappeared. In their place we have the singular Retiolites, with its curiously-reticulated skeleton; and several species of the single-celled genus Monograptus, of which a characteristic species (M. Priodon) is here figured. If we remove from this group the plant-like Dictyonemoe, which are still present, and which survive into the Devonian, no known species of Graptolite has hitherto been detected in strata higher in geological position than the Ludlow. This, therefore, presents us with the first instance we have as yet met with of the total disappearance and extinction of a great and important series of organic forms.



The Corals are very numerously represented in the Upper Silurian rocks some of the limestones (such as the Wenlock Limestone) being often largely composed of the skeletons of these animals. Almost all the known forms of this period belong to the two great divisions of the Rugose and Tabulate corals, the former being represented by species of Zaphrentis, Omphyma, Cystiphyllum, Strombodes, Acervularia, Cyathophyllum, &c.; whilst the latter belong principally to the genera Favosites, Choetetes, Halysites, Syringopora, Heliolites, and Plasmopora. Amongst the Rugosa, the first appearance of the great and important genus Cyathophyllum, so characteristic of the Palaeozoic period, is to be noted; and amongst the Tabulata we have similarly the first appearance, in force at any rate, of the widely-spread genus Favosites—the "Honeycomb-corals." The "Chain-corals" (Halysites), figured below (fig. 59), are also very common examples of the Tabulate corals during this period, though they occur likewise in the Lower Silurian.



Amongst the Echinodermata, all those orders which have hard parts capable of ready preservation are more or less largely represented. We have no trace of the Holothurians or Sea-cucumbers; but this is not surprising, as the record of the past is throughout almost silent as to the former existence of these soft-bodied creatures, the scattered plates and spicules in their skin offering a very uncertain chance of preservation in the fossil condition. The Sea-urchins (Echinoids) are said to be represented by examples of the old genus Paloechinus. The Star-fishes (Asteroids) and the Brittle-stars (Ophiuroids) are, comparatively speaking, largely represented; the former by species of Palasterina (fig. 60), Paloeaster (fig. 60), Paloeocoma (fig. 60), Petraster, Glyptaster, and Lepidaster—and the latter by species of Protaster (fig. 61), Paloeodiscus, Acroura, and Eucladia. The singular Cystideans, or "Globe Crinoids," with their globular or ovate, tesselated bodies (fig. 46, A, C, D,), are also not uncommon in the Upper Silurian; and if they do not become finally extinct here, they certainly survive the close of this period by but a very brief time. By far the most important, however, of the Upper Silurian Echinodenns, are the Sea-lilies or Crinoids. The limestones of this period are often largely composed of the fragmentary columns and detached plates of these creatures, and some of them (such as the Wenlock Limestone of Dudley) have yielded perhaps the most exquisitely-preserved examples of this group with which we are as yet acquainted. However varied in their forms, these beautiful organisms consist of a globular, ovate, or pear-shaped body (the "calyx"), supported upon a longer or shorter jointed stem (or "column"). The body is covered externally with an armour of closely-fitting calcareous plates (fig. 62), and its upper surface is protected by similar but smaller plates more loosely connected by a leathery integument. From the upper surface of the body, round its margin, springs a series of longer or shorter flexible processes, composed of innumerable calcareous joints or pieces, movably united with one another. The arms are typically five in number; but they generally subdivide at least once, sometimes twice, and they are furnished with similar but more slender lateral branches or "pinnules," thus giving rise to a crown of delicate feathery plumes. The "column" is the stem by which the animal is attached permanently to the bottom of the sea; and it is composed of numerous separate plates, so jointed together that whilst the amount of movement between any two pieces must be very limited, the entire column acquires more or less flexibility, allowing the organism as a whole to wave backwards and forwards on its stalk. Into the exquisite minutioe of structure by which the innumerable parts entering into the composition of a single Crinoid are adapted for their proper purposes in the economy of the animal, it is impossible to enter here. No period, as before said, has yielded examples of greater beauty than the Upper Silurian, the principal genera represented being Cyathocrinus, Platycrinus, Marsupiocrinus, Taxocrinus, Eucalyptocrinus, Ichthyocrinus, Mariacrinus, Periechocrinus, Glyptocrinus, Crotalocrinus, and Edriocrinus.



The tracks and burrows of Annelides are as abundant in the Upper Silurian strata as in older deposits, and have just as commonly been regarded as plants. The most abundant forms are the cylindrical, twisted bodies (Planolites), which are so frequently found on the surfaces of sandy beds, and which have been described as the stems of sea-weeds. These fossils (fig. 63), however, can be nothing more, in most cases, than the filled-up burrows of marine worms resembling the living Lob-worms. There are also various remains which belong to the group of the tube-inhabiting Annelides (Tubicola). Of this nature are the tubes of Serpulites and Cornultites, and the little spiral discs of Spirorbis Lewisii.



Amongst the Articulates, we still meet only with the remains of Crustaceans. Besides the little bivalved Ostracoda—which here are occasionally found of the size of beans—and various Phyllopods of different kinds, we have an abundance of Trilobites. These last-mentioned ancient types, however, are now beginning to show signs of decadence; and though still individually numerous, there is a great diminution in the number of generic types. Many of the old genera, which flourished so abundantly in Lower Silurian seas, have now died out; and the group is represented chiefly by species of Cheirurus, Encrinurus, Harpes, Proetus, Lichas, Acidaspis, Illoenus, Calymene, Homalonotus, and Phacops—the last of these, one of the highest and most beautiful of the groups of Trilobites, attaining here its maximum of development. In the annexed illustration (fig. 64) some of the characteristic Upper Silurian Trilobites are represented—all, however, belonging to genera which have their commencement in the Lower Silurian period. In addition to the above, the Ludlow rocks of Britain and the Lower Helderberg beds of North America have yielded the remains of certain singular Crustaceans belonging to the extinct order of the Eurypterida. Some of these wonderful forms are not remarkable for their size; but others, such as Pterygotus Anglicus (fig. 65), attain a length of six feet or more, and may fairly be considered as the giants of their class. The Eurypterids are most nearly allied to the existing King-crabs (Limuli), and have the anterior end of the body covered with a great head-shield, carrying two pairs of eyes, the one simple and the other compound. The feelers are converted into pincers, whilst the last pair of limbs have their bases covered with spiny teeth so as to act as jaws, and are flattened and widened out towards their extremities so as to officiate as swimming-paddles. The hinder extremity of the body is composed of thirteen rings, which have no legs attached to them; and the last segment of the tail is either a flattened plate or a narrow, sword-shaped spine. Fragments of the skeleton are easily recognised by the peculiar scale-like markings with which the surface is adorned, and which look not at all unlike the scales of a fish. The most famous locality for these great Crustaceans is Lesmahagow, in Lanarkshire, where many different species have been found. The true King-crabs (Limuli) of existing seas also appear to have been represented by at least one form (Neolimulus) in the Upper Silurian.



Coming to the Mollusca, we note the occurrence of the same great groups as in the Lower Silurian. Amongst the Sea-mosses (Polyzoa), we have the ancient Lace-corals (Fenestella and Retepora), with the nearly-allied Glauconome, and species of Ptilodictya (fig. 66); whilst many forms often referred here may probably have to be transferred to the Corals, just as some so-called Corals will ultimately be removed to the present group.



The Brachiopods continued to flourish during the Upper Silurian Period in immense numbers and under a greatly increased variety of forms. The three prominent Lower Silurian genera Orthis, Strophomena, and Leptoena are still well represented, though they have lost their former preeminence. Amongst the numerous types which have now come upon the scene for the first time, or which have now a special development, are Spirifera and Pentamerus. In the first of these (fig. 69. b, c), one of the valves of the shell (the dorsal) is furnished in its interior with a pair of great calcareous spires, which served for the support of the long and fringed fleshy processes or "arms" which were attached to the sides of the mouth.[16] In the genus Pentamerus (fig. 70) the shell is curiously subdivided in its interior by calcareous plates. The Pentameri commenced their existence at the very close of the Lower Silurian (Llandovery), and survived to the close of the Upper Silurian; but they are specially characteristic of the May Hill and Wenlock groups, both in Britain and in other regions. One species, Pentamerus galeatus, is common to Sweden, Britain, and America. Amongst the remaining Upper Silurian Brachiopods are the extraordinary Trimerellids; the old and at the same time modern Linguloe, Discinoe, and Cranioe; together with many species of Atrypa (fig. 68, e), Leptocoelia (fig. 68, a), Rhynchonella (fig. 68, b, c), Meristella (fig. 69, a, e, f), Athyris, Retzia, Chonetes, &c.

[Footnote 16: In all the Lamp-shells the mouth is provided with two long fleshy organs, which carry delicate filaments on their sides, and which are usually coiled into a spiral. These organs are known as the "arms," and it is from their presence that the name of "Brachiopoda" is derived (Gr. brachion, arm; podes, feet). In some cases the arms are merely coiled away within the shell, without any support; but in other cases they are carried upon a more or less elaborate shelly loop, often spoken of as the "carriage-spring apparatus." In the Spirifers, and in other ancient genera, this apparatus is coiled up into a complicated spiral (fig. 67). It is these "arms," with or without the supporting loops or spires, which serve as one of the special characters distinguishing the Brachiopods from the true Bivalves (Lamellibranchiata).]



The higher groups of the Mollusca are also largely represented in the Upper Silurian. Apart from some singular types, such as the huge and thick-shelled Megalomi of the American Wenlock formation, the Bivalves (Lamellibranchiata) present little of special interest; for though sufficiently numerous, they are rarely well preserved, and their true affinities are often uncertain. Amongst the most characteristic genera of this period may be mentioned Cardiola (fig. 71, A and C) and Pterinea (fig. 71, B), though the latter survives to a much later date. The Univalves (Gasteropoda) are very numerous, and a few characteristic forms are here figured (fig. 72). Of these, no genus is perhaps more characteristic than Euomphalus (fig. 72, b), with its flat discoidal shell, coiled up into an oblique spiral, and deeply hollowed out on one side; but examples of this group are both of older and of more modern date. Another very extensive genus, especially in America, is Platyceras (fig. 72, a and f), with its thin fragile shell—often hardly coiled up at all—its minute spire, and its widely-expanded, often sinuated mouth. The British Acroculioe should probably be placed here, and the group has with reason been regarded as allied to the Violet-snails (Ianthina) of the open Atlantic. The species of Platyostoma (fig. 72, h) also belong to the same family; and the entire group is continued throughout the Devonian into the Carboniferous. Amongst other well-known Upper Silurian Gasteropods are species of the genera Holopea (fig. 72, g), Holopella (fig. 72. e), Platyschisma (fig. 72, d), Cyclonema, Pleurotomaria, Murchisonia, Trochonema, &c. The oceanic Univalves (Heteropods) are represented mainly by species of Bellerophon; and the Winged Snails, or Pteropods, can still boast of the gigantic Thecoe and Conularioe, which characterise yet older deposits. The commonest genus of Pteropoda, however, is Tentaculites (fig. 73), which clearly belongs here, though it has commonly been regarded as the tube of an Annelide. The shell in this group is a conical tube, usually adorned with prominent transverse rings, and often with finer transverse or longitudinal striae as well; and many beds of the Upper Silurian exhibit myriads of such tubes scattered promiscuously over their surfaces.