' fl Y OF MIA IARTH SCIENCES LIBRARY THE STUDENT'S HANDBOOK OF STEATIGEAPHICAL GEOLOGY THE STUDENT'S HANDBOOK OF STKATIGEAPHICAL GEOLOGY BY A. J. JUKES-BROWNE, B.A., F.R.S., F.G.S. % LATE OF THE GEOLOGICAL SURVEY OF ENGLAND AND WALES AUTHOR OF THE 'BUILDING OF THE BRITISH ISLES,' ETC. SECOND EDITION ILLUSTRATED WITH MAPS, DIAGRAMS, AND FIGURES OF FOSSILS LONDON: EDWARD STANFORD 12, 13, 14 LONG ACRE, W.C. 1912 All rights reserved EARTH SCIENCES LIBRARY PEEFACE IN preparing a new edition of this volume I have not only revised and partly rewritten the descriptions of the British strata, but have enlarged the scope of the work by giving more complete accounts of the continental representatives of each system or series of rocks. I have, however, strictly confined myself to the geology of Europe, because any adequate account of Asiatic or American strata, with their fossil contents, would have made the volume too bulky. For such further information the student must consult a larger treatise, such as Sir A. Geikie's Text-look of Geology or de Lapparent's TraiU de Gtologie. It has often been said that the British Isles present us with an epitome of the geology of Europe ; but though it is true that we possess representatives of nearly all the great series of strata which form the geological record, yet some of them are missing, such as the Stephanian Series, the Middle Trias (Muschelkalk, etc.), and the Miocene (except for the Bovey Beds), while others consist of freshwater deposits instead of the more universal marine facies. Consequently, a student who is only acquainted with the British rocks is ignorant of many important formations and fossils, and is apt to take limited and incomplete views of the stratigraphical systems and their problems. In thinking it desirable to supply such supplementary 258402 VI STRATIGRAPHICAL GEOLOGY information I am supported by many teachers of geology, who wish their students to take a broader view and to learn more about the geology of Europe, with perhaps less local detail of English subdivisions and zones. In order to show how such wider knowledge can be utilised and made interest- ing, I have appended to each chapter a brief sketch of the conclusions which may be drawn from the facts with regard to the physical and geographical conditions under which each great series of strata was accumulated in the European region. Further, in three cases I have ventured to express these conclusions in the form of cartographical restorations, i.e. of the Lower Devonian, the Upper Trias, and the Lower Cretaceous. Such maps form a conspicuous feature in the later editions of de Lapparent's treatise, but have not yet been inserted in an English text-book. In a future edition I may be able to include maps of other epochs ; meantime, I would suggest that students should endeavour to make their own geographical restorations from the descriptions which I have given. I need hardly seek to justify such advice, for, as Professor Watts has remarked, such maps " group together numbers of facts and inferences with which it was previously difficult to retain touch;" . . . they "systematise future research, they direct to the places where the discovery of new facts is desirable." In the work of revision I have sought the assistance of those geologists who have paid special attention to particular formations. Their writings are, of course, referred to in the text, but I have to thank the following friends and corre- spondents for reading portions of certain chapters, and for useful information thereon : C. T. Clough, Edward Greenly, W. G. Fearnsides, J. E. Marr, A. Vaughan, E. A. Newell PREFACE Vll Arber, T. G. Bonney, and H. Woods. I am specially indebted to Mr. Woods for revising all the descriptions and lists of fossils, and for amending the nomenclature, both of genera and species, in accordance with current views. Finally, I may point out that many additional illustrations have been inserted, not only of fossils but of maps and sections, so that there are now over two hundred figures and groups of fossils. Moreover, two of the older maps, those of North Whales and Devonshire, have been replaced by more correct representations, in accordance with recent publications. A. J. JUKES-BROWNE. EIGH, TORQUAY, November 1911. CONTENTS CHAPTER I THE SUCCESSION OP STRATIFIED ROCKS PAGE Scope of the Subject Succession of Stratified Rocks Nomen- clature of Rock- groups Definition and Delimitation of Systems Stratigraphical Breaks Paleeontological Evidence 1 CHAPTER II STRATIGRAPHICAL PALEONTOLOGY Origin of Species Imperfection of the Geological Record Suc- cession of Faunas Zones Synchronism and Homotaxis . 14 CHAPTER III THE LITERATURE OF HISTORICAL GEOLOGY Maps Publications on Stratigraphy Publications on Palaeon- tology Books on Palaeogeography ..... 26 CHAPTER IV THE ARCHAEAN ROCKS A. Classification and Nomenclature ; B. British Representatives : Scotland, Ireland, Wales, England ; C. Archaean Rocks in Europe : France, Austria and Switzerland, Bohemia, Scandinavia References 34 X STRATIGRAPHICAL GEOLOGY CHAPTER V THE CAMBRIAN SYSTEM PAGE A. Nomenclature and Classification B. Life of the Period C. British Cambrian Rocks : Wales, Midland Counties, Lake District and Isle of Man, Ireland, Scotland D. The Cambrian of Northern Europe : France, Iberian Peninsula, Belgium and the Ardennes, Scandinavia, Russia, Germany and Bohemia E. European Geography and Conditions of Deposi- tion References ......... 70 CHAPTER VI THE ORDOVICIAN SYSTEM A. Nomenclature and Divisions B. Life of the Period C. British Ordovician Rocks : Wales, Shropshire, Cumberland and Westmoreland, Scotland, Ireland D. Some European Areas : France, Spain and Portugal, Belgium, Bohemia. Scandinavia, Russia E. Conditions of Formation References 109 CHAPTER VII THE SILURIAN SYSTEM A. Nomenclature and Divisions B. Life of the Period C. British Silurian Rocks : Typical Silurian Area, Wales, Westmoreland, South of Scotland, Ireland D. Continental Equivalents : Scandinavia, Russia, Bohemia, Germany, Belgium, France, Pyrenees, Spain and Portugal E. Condi- tions of Deposition References . . . . . .1,53 CHAPTER VIII DEVONIAN AND OLD RED SANDSTONE SYSTEM General Classification I. The Devonian Facies A. The Marine Fauna B. British Devonian Rocks : Devon and Cornwall, The South of Ireland C. Devonian of the Continent : France, Spain, The Ardennes, Germany, Bohemia, Poland and Russia II. The Old Red Sandstone Facies A. Flora and Fauna B. Stratigraphy of the Old Red Sandstone : Wales, Scotland, North of Ireland, Norway, Arctic Regions C. Geography and History of the Period References . . . . .193 CONTENTS xi CHAPTEE IX THE CARBONIFEROUS SYSTEM /. The Avonian Series PAGE A. Life of the Period B. British Avonian Strata : Bristol and South Wales, North Wales, Stafford and Derby, South Lancashire and Yorkshire, North Lancashire and Westmore- land, Northumberland and Berwick, Scotland, Ireland, Devon and Cornwall C. Continental Representatives : France and Belgium, Germany, South of France, Pyrenees, and Spain, Russia D. Conditions of Deposition References . . 244 CHAPTEE X THE CARBONIFEROUS SYSTEM (continued] II. Westphalian Series A. Subdivisions and General Relations B. Life of the Period C. Westphalian Series in Britain : South Pennine and Midland Area, Cumberland and Northumberland, Scotland, Ireland, South Wales and Bristol Area, Devon and Cornwall D. Continental Representatives : Westphalia, Belgium and France, Sarrebruck Area, Silesia, Moravia and Poland, Russia, Mediterranean Region ...... 285 III. The Stephanian Series A. The Stephanian Fades B. The Uralian Facies C. The Formation of Coal-seams References ..... 313 CHAPTEE XI THE PERMIAN SYSTEM A. Nomenclature and Subdivisions B. Life of the Period C. British Permian Rocks : Midland Area, Devon and Somerset, North-eastern Area, North-western Area and Ireland, Scotland D. Continental Permian Strata : France, Germany Russia E. History of the Period References .... 324 Xll STRATIGRAPHICAL GEOLOGY CHAPTER XII THE TRIASSIC SYSTEM PAGE A. Nomenclature and Classification B. Life of the Period: Marine Fauna, Lacustrine Fauna C. The Trias in Britain : Bunter Series, Keuper Series D. The Trias in Europe : Germany, Alpine Region, France E. The History of the European Trias References ....... 347 CHAPTER XIII THE JURASSIC SYSTEM A. Nomenclature and Division B. Life of the Period , . 382 7. The Lias or Lower Jurassic Series A. The British Lias : Subdivisions, Fauna, Stratigraphy : South- west Province, Midland Counties, Yorkshire, Cumberland, Ireland and Scotland B. Continental Liassic Strata : France, Germany, Sweden, Alpine Region, Hungary .... 386 II. Middle Jurassic Series A. Stages and Zones B. Life of the Period C. Middle Jurassic of England : Inferior Oolite (Bajocian), Bath Oolites (Bath- onian) D. Middle Jurassic of Scotland E. Continental Equivalents : France, Germany, Alpine Fades . . . 409 777. Upper Jurassic Series A. Subdivision and Nomenclature B. Life of the Period C. Upper Jurassic of England : Oxford Clay, Corallian, Kimer- idge Clay, Portland Beds, Purbeck Beds D. Upper Jurassic of Scotland E. Continental Equivalents : France, Germany, Alpine Fades, Russian Fades History of the Jurassic Period References . 432 CHAPTER XIV THE CRETACEOUS SYSTEM A. Nomenclature and Subdivisions B. Life of the Period . 462 CONTENTS Xlll I. Lower Cretaceous Series PAGE A. Lower Cretaceous Life. B. British Representatives : South of England, Midland District, Northern District, Scotland C. Continental Representatives : Provence, The Jura, Germany, Russia ........... 466 //. Upper Cretaceous Series A. Stages and Zones B. Characteristic Fossils C. The Upper Cretaceous of Britain : England, Ireland, and Scotland D. Continental Representatives : Northern France, Denmark and Sweden, The Mediterranean Region, Germany, Bohemia and Silesia, Russia E. The History of the Period References . 491 CHAPTER XV THE PALAEOGENE SYSTEM /. The Eocene Series A. Life of the Period B. British Eocene Deposits C. Region of VoJcanic Activity D. Continental Equivalents : Belgium, The Paris Basin, Southern Europe . . . . .526 //. The Oligocene Series A. Nomenclature and Divisions B. Life of the Period C. The English Series D. Continental Representatives : Belgium, Germany, Switzerland, The Paris Basin, Central France, Italy and Dalmatia ........ 555 ///. Geography and Sedimentation . 567 CHAPTER XVI THE NEOGENE SYSTEM 7. The Miocene Series Nomenclature and Divisions B. Life of the Period C. The Miocene in England D. Continental Miocene: France, Switzerland, Austria, Italy, Belgium and Germany E. History of the Miocene Period .... 572 XIV STRATIGRAPHICAL GEOLOGY II. The Pliocene Series PAGE A. Subdivisions and Nomenclature B. Life of the Period C. British Pliocene Deposits D. Continental Pliocene Deposits : Belgium and Holland, France, Italy, Austria and the Balkans, Germany E. History of the Pliocene Period References . 586 CHAPTER XYII THE PLEISTOCENE SERIES A. General Considerations B. Glaciation and Glacial Deposits C. Glacial Deposits of Britain : Central and Northern Scotland, South of Scotland and North of England, York- shire and Lincolnshire, East Anglia, The Midland Area, Wales and the West of England, Ireland D. The Glacial Deposits of Europe E. Non-Glacial Deposits : Classification and Fauna, British Deposits, Continental Deposits F. Some Phases of Pleistocene Geography References . . .611 INDEX 653 CHAPTER I THE SUCCESSION OF STRATIFIED ROCKS Scope of the Subject. In some manuals of Geology this branch of the subject is called Stratigraphical Geology, because it chiefly consists in the description of the great stratified series of rocks which make up the mass of the earth's crust. It is some- times called historical geology because it should comprise not only a description of the various rocks which were formed during each period of the world's history, but also some description of the fossils found in each formation, and some account of the physical conditions under which the rocks themselves were deposited, with a notice of the changes in the relative position of land and sea which took place from time to time ; all this should be made as complete as the imperfection of the geological documents will allow. In brief, historical geology should gather up the knowledge acquired in all the other departments of the science, and should apply it to the elucidation of the history of the earth, or of a portion of the earth, from the earliest time of which any records exist down to the time of human history. At present there are only a few limited areas of the earth's surface about which we have anything like a full Stratigraphical knowledge, and much remains to be learnt even in the areas about which we know most. Historical geology, therefore, is yet in its infancy ; but enough is now known of the Stratigraphical geology of Europe to make it possible to give an outline of its geological history, and American geologists are rapidly making it possible to do the same for North America. The Succession of Stratified Bocks. The general principles on which Stratigraphical geology is founded are easily understood, and the manner in which the order and succession of rock-groups is established will now be briefly explained. The mere succession of strata in any district or country, however, is 1 B 2 . - / : STRATIGRAPHICAL GEOLOGY ;'oi?e\thmg, and, the grouping of such strata into larger systems, representing' definite geological periods, is another thing, and one which requires consideration from several points of view. A subsequent chapter, therefore, will be devoted to a discussion of the principles upon which divisional lines may be drawn in any established succession of rock-groups. The two principal tests by which the relative age of different strata is determined are (1), Superposition; (2), Fossil contents. The conclusions to be drawn from the superposition of strata are explained in most text-books of physical geology, and the law of vertical succession which forms the basis of the whole fabric of historical geology may be stated as follows : " In any succession of beds each one represents the conditions which prevailed over a certain area for a certain length of time, the lowest is the oldest, the uppermost is the newest, and the relative age of the others is indicated by their relative position." If the strata are inclined, the right order is ascertained by making a geological survey of the district, and constructing a section at right angles to the general strike of the beds. Again, by tracing any one set of strata horizontally along their strike from one part of a district to another, where perhaps there are other rocks of a different kind, we obtain a datum of reference by which to determine whether the rocks in the second area are newer or older than those in the first. So long as there is physical and geological continuity between the different portions of a district, i.e. so long as some one member of a conformable series of strata can be followed continuously, such a survey generally affords sufficient data for ascertaining the relative position of the rocks which occur in the district, and for constructing a table of their vertical succession. But when either geographical or geological continuity is interrupted that is, when a district or country is separated from others of like structure either by the sea or by tracts of totally different rocks then we must begin all over again, and construct an independent table of strata for the new district. For example, Wales is a district to all parts of which one system of classification and nomenclature can be applied. Similar groups of rocks occur in Ireland on the one hand, and in Cumberland on the other ; but Wales is physically separated from Ireland by the Irish Sea, and geologically separated from Cumberland by the Triassic plains of Cheshire and Lancashire ; hence the same system of nomenclature could not be used in the other districts, the vertical succession had to be determined independently, and local names given to the different subdivisions in each district before they could be in any way compared with one another. THE SUCCESSION OF STRATIFIED ROCKS 3 But, it may be asked, when two such districts have been inde- pendently investigated, how are we to correlate the two tables of strata, and ascertain which rocks or rock-groups were approxi- mately contemporaneous ? It is here that a knowledge of fossils (Palaeontology) comes to our aid, and enables us to identify rocks by their fossil contents, so that strata in different districts and of different lithological characters may be included in the same natural group or system, because they contain the same or closely similar species of fossils. This was the principle discovered and applied at the beginning of last century by Dr. William Smith, who is often called the Father of British Geology, and who was certainly the founder of that branch of the science which is the subject of the present volume. The book in which Smith recorded his discovery of the strati- graphical use of fossils was entitled Strata identified by Organised Fossils, and was published in 1816. He had previously made out the succession of the strata which occur near Bath, and had observed that each well-marked group of beds contained a special assemblage of fossils. As his acquaintance with English rocks be- came larger he noticed that there was a similar succession elsewhere, and thus (in the words of his nephew and biographer 1 ) "he inferred that each of the separate periods occupied in the formation of the strata was accompanied by a peculiar series of the forms of organic life, that these forms characterised those periods, and that the different strata could be identified in distant localities and otherwise doubtful cases by peculiar embedded organic remains." The experience of subsequent observers confirmed and established this inference, which has become a guiding principle in strati- graphical geology. Further research, moreover, has brought out the more definite conclusion that there has been a continuous succession of life-forms,, that species and genera and families, and even whole orders of beings, have come into existence, have nourished, and have then gradually died out, never to recur. It is this non- recurrence of species which gives a special value to fossils as a test of age and as a means of correlation. The same kind of rock has been formed again and again during the history of the world, but Avheri once a species has died out it has never appeared again. It must not be supposed, however, that the stratigraphical succession of rocks presents us with a complete record of the history of the earth, or that it will ever supply us with examples of all the species which have lived upon the earth's surface. On the contrary, the records preserved to us are very incomplete, and many pages are wanting in every chapter of the great volume, so 1 Memoirs of William Smith, by John Phillips, F.R.S., 1844. 4 STRATIGEAPHICAL GEOLOGY that historical geology cannot be defined as a complete History of the Earth ; it would be more correctly described as Scenes from the History of the Earth. Nomenclature of Rock-groups. As soon as it was recog- nised that the rocks which compose the earth's crust were not a mere congeries of rock-masses without order or sequence, but that when properly interpreted they arranged themselves in an orderly succession of formations, then it was seen to be necessary that names should be given to all parts of this wonderful succession of stratified deposits. The history which can be read from the "records of the rocks" is a very long one, although, as has already been stated, it is by no means a complete history. Every natural region of the earth's surface has its own geological history, but the same great periods of time can be recognised in all. Geological history, like human history, has its early or pristine times, its mediaeval times, and its later or more modern times ; it may perhaps be said to consist of several volumes, and of course each of these volumes consists of several chapters dealing with so many successive periods. Stratigraphical nomenclature is not very satisfactory, for it has been gradually developed in a synthetic manner, and is not the consistent and harmonious arrangement which it might have been had the older geologists possessed our present knowledge of the rocks. In early days the greater rock-groups were generally called formations ; thus we read of the Old Eed Sandstone formation, the New Red Sandstone formation, the Carboniferous formation, and the Chalk formation. Afterwards, as a knowledge of their fossil contents increased, the relative time- values of these formations were better understood, and some of them were united with others to form systems, but the number of such systems was not fixed, nor did any definite idea exist as to what entitled a series of rocks to rank as a system. There is still some difference of opinion on this point, but doubtless agreement will ultimately be attained. It was also early perceived that the older formations differed very greatly from the later deposits both in mineral characters and in fossil contents. It was noticed that the oldest rocks were generally crystalline schists, that the next oldest were more or less altered and indurated, while the later were often soft and clearly stratified, and the most recent were usually loose and unconsolidated. Hence Werner in Germany used names which may be translated as equivalent to primitive, transition, stratified, and alluvial rocks, and the French introduced the more convenient terms Primary, Secondary, and Tertiary divisions. It was supposed that the rocks belonging to these eras of time THE SUCCESSION OF STRATIFIED ROCKS 5 differed essentially in their original nature and mode of formation. Subsequent investigation proved this idea to be an error, and Professor Phillips consequently proposed another set of names based on palaeontological facts, namely, Palaeozoic, Mesozoic, and Ccenozoic, signifying respectively times of ancient, middle, and recent life. The two former of these have been very generally adopted, but Tertiary is still preferred to Caenozoic, and there is no reason now why its correlatives, Primary and Secondary, should not be employed as time -words, instead of the more cumbrous Palaeozoic and Mesozoic. Professor E. Forbes thought that the faunas of the Secondary and Tertiary had so great an affinity that the whole succession was more naturally divisible into two eras, which he termed Palaeozoic and Neozoic. No doubt this is strictly correct, but the triple division has been found very convenient in Europe, and is more generally adopted. Since these terms were proposed, however, a series of stratified rocks has been discovered below those which were called Primary or Palaeozoic, and for these several names have been proposed. Of these Arclicean is the best, for pre-Cambrian is too indefinite, and the others, Azoic and Eozoic, are founded on the respective assumptions (1) that they contain no organic remains, (2) that they contain the relics of the first dawn of life in the world. Such small traces of life as have been discovered in these rocks do not suggest anything very different from that of Palaeozoic time, but they may for the present be termed Eozoic. In the following table of systems the oldest is placed at the bottom and the newest at the top to indicate their natural super- position and order in the chronological scale. Bras. Systems. Tertiary or Caenozoic . l^ e gene. j t Palaeogene, g f Cretaceous, j- Secondary or Mesozoic . . . \ Jurassic. I 2. (Triassic. J ' I Permian. Carboniferous. Devonian. Silurian. Ordovician. Cambrian. f Eparchsean. Archaean or Eozoic . . . J Mesarchaean. [ Protarchaean. Nomenclature. With regard to the names of the systems in the above table some explanation is required. Such names as 6 STRATIGRAPHICAL GEOLOGY Carboniferous and Cretaceous are used in a far wider sense than their mere lithological signification. These names were originally given to certain British formations, in which the most conspicuous elements were respectively beds of coal and chalk ; but they have since acquired a chronological meaning, and are used as systematic names for the periods of time during which the rocks were formed. They have also been adopted in other countries for rock-groups, which occupy the same relative place in the geological succession. The Cretaceous System then includes, not only the Chalk, but also the underlying beds of sand and clay, which contain a similar assemblage of genera and species. Moreover, the Cretaceous System includes not only the Chalk and its associated beds in Europe, but all the rocks which were formed during the same period of the earth's history. Elsewhere, therefore, as in North America, the system may consist of sandstones, shales, and coal-seams without a bed of real chalk in it. So the Carboniferous System means that system of strata of which the British Coal-measures form an integral part, but which includes several other rock-groups, all united by a community of fossils. It is these fossils, and not the occurrence of coal-seams, which are the criterion of the so-called Carboniferous period, for, as above mentioned, in America there are coals of Cretaceous age. In the same way, when we speak of the Carboniferous Limestone Series, we do not mean a group which everywhere consists of lime- stone, but one to which the great Carboniferous limestone belongs ; when traced northward through England, this limestone is found to split up into separate beds, which finally become subordinate to thick masses of sandstone and shale. The Devonian System is now considered to include the great series of rocks which are still called the Old Eed Sandstone. This is a particularly awkward and unsatisfactory name for an important group of rocks. It was called old because it lay beneath the Carboniferous rocks, above which came another group of similar red sandstones. The latter were formerly termed New Red Sandstone, but this has been divided into two distinct systems which have received appropriate names, and it is to be hoped that "Old Red Sandstone" also will soon be replaced by a better name. As Jukes observed fifty years ago, 1 " it is avowedly a provisional designation, just exactly as the names of all the great groups of stratified rocks are provisional. They are temporary names adopted for present purposes, and have grown into use, and will continue to be used until they are superseded by more appropriate terms, which increasing knowledge only can show to be more appropriate." 1 Manual of Geology, second edition, 1862, p. 429. THE SUCCESSION OF STRATIFIED ROCKS 7 In many respects those names which, have a geographical or historical origin are the most satisfactory names taken from those of towns, rivers, mountain ranges, provinces, or the ancient races which inhabited such provinces ; as Jurassic (from the Jura range), Devonian (from the county of Devon), Cambrian (from the Roman name of North Wales), Silurian (from the tribe of the Silures). These will probably remain as permanent names for systems to which they are applied. Definition and Delimitation of Systems. The number of systems into which the geological series should be divided, and the horizons at which the planes of separation are placed, are points requiring some consideration, because they involve certain principles of classification. The divisional lines are generally drawn where natural breaks occur, either in the sequence of the strata themselves, or in the succession of fossil species which they contain. Such breaks are therefore of two kinds, (1) Stratigraphical, (2) Palseonto- logical. 1. Stratigraphical Breaks. The nature of Stratigraphical breaks or unconformities is explained in all handbooks of physical geology, and in the present connection we have to deal only with their relative importance and value in classification. The most complete and satisfactory measure of any unconformity is the thickness of strata which were deposited in other areas during the time represented by the gap between the two unconformable formations. It is not every unconformity which is of sufficient importance to form a plane of division between two systems. Thus in Wales there is a great local unconformity within the Silurian System, but this does not exist in other areas. Again, in Northampton and Lincolnshire there is unconformity between two portions of the Jurassic System, while no such break exists either in the south of England or in Yorkshire. A break which is represented elsewhere by several thousand feet of rock implies the lapse of a considerable period, and being always accompanied by a great change in the fossils, might be taken as a line of division between two systems. But if systems were to be rigidly limited by physical unconformities, and defined as a series of beds formed during a complete oscillation of downward and upward movement, the classification of the rocks in one area will break down when applied to those of other districts, because the gaps will occur at different horizons in different areas. Thus in Britain alone three different classifications at least would be required for the Primary rocks, as the following columns will show, a being taken to mean a break or unconformity. STRATIGRAPHICAL GEOLOGY South Wales. North Wales. Scotland. Carboniferous. Carboniferous. Carboniferous. Upper O.R.S. Upper O.R.S. Middle O.R.S. Lower O.R.S. Lower O.R.S. Silurian. Silurian. Silurian. Ordovician. Ordovician. Ordovician. Cambrian. Cambrian. Cambrian. Thus, if we adopted the arrangement warranted by the strati- graphical breaks which occur in South Wales, we should have two separate systems, and two only, for there is a general conformity between the Cambrian and Ordovician, and from the Silurian to the Carboniferous. In North Wales, however, the break between the Ordovician and Silurian is much less marked, while there is certainly one above the Cambrian, and there is an entire absence of Old Red Sandstone, so that we should make three separate systems. Again if we pass to the south of Scotland we find complete conformity between the Ordovician and Silurian, and no marked break between the latter and the Old Red Sandstone, but a great one between the two divisions of this Red Sandstone formation. Hence we might also make three systems in Scotland, but they would be quite different from those in Wales and would be of very unequal value. As in this case, so also in others, a classification founded on physical breaks may be conveniently used for a limited district, and until further knowledge is acquired ; but the breaks in the succession are certain to be filled up elsewhere by transitional beds, and as these may be of systematic value, we must always be prepared for the possible intercalation of another system between any two groups that are divided by a decided unconformity. 1 In brief, it must be admitted that unconformities are geological accidents, and if our classification and nomenclature are to be applicable to any large section of the earth's crust, they must be based, not on stratigraphical but on palseontological facts. The Ordovician, Devonian, and Jurassic are systems so established ; and, in process of time, as our knowledge becomes more and more complete, it is probable that the number of systems will be finally determined in the same manner. They will always be more or less arbitrary divisions, since it will be difficult to say what number of 1 The student can read with advantage Professor Lapworth's article on the " Classification of the Lower Palaeozoic Rocks " in the Geological Magazine, Dec. 2, vol. vi. p. 1 (1879). THE SUCCESSION OF STRATIFIED ROCKS peculiar genera would entitle a series of beds to the rank of a system ; but a general agreement will doubtless be reached in time. Meanwhile the student will do well to regard our present classifica- tion as provisional, and to remember that, though divisional lines must be drawn somewhere, passage beds and transition groups are of frequent occurrence in nature. 2. Palceontological Evidence. But although we must rely mainly on palseontological evidence in separating one system from another, it must not be supposed that every sudden change of fauna is to be taken as introducing a new system, or that such a change always indicates a period of non-deposition, and therefore a greater or less break in the succession ; unless this coincides with a physical break, the mere fact of such a change does not necessarily prove any great lapse of time. The change may be caused in several ways, either by rapid depression, by a sudden immigration of new forms, or by elevation with or without continuous deposition. It is conceivable, for instance, that rapid depression, causing a considerable increase in the depth of the water, and removing the area farther away from the coast line, would produce what has been termed a palseontological break in a comparatively short period of time, and without any discontinuity of deposit. The beds below a certain plane would embed the remains of a shore fauna, while those above it would contain a deep-water assemblage, which might be very different even if the deposits were of similar lithological character. Depression, by the submergence of barriers, may also cause much more extensive changes; we have only to consider what would result from the submergence of certain existent isthmuses, such as those of Panama and Suez, to realise what must have repeatedly happened in the past. The balance of life in both provinces would be altogether disturbed and upset, resulting in the extinction of some forms, the inter-migration of others, and the development of some new varieties and species. Moreover, an alteration in the physical geography of one area may indirectly, but very greatly, affect the climate and life of another distant area ; thus, if the whole isthmus of Panama were now to be submerged, the Gulf Stream would probably be diverted from its present course, the result of which would be an immediate refrigeration of the British climate and a consequent immigration of Boreal and Arctic species into the Celtic province. Again, if the Gulf Stream remained, but the opening into the Arctic ocean were closed, the British climate would be ameliorated, Boreal forms would disappear, and a more southern assemblage would occupy the area. Elevation of the sea-bottom will produce similar direct changes 10 STRATIGRAPHICAL GEOLOGY to those caused by depression ; a deep-sea area, where deposition has been very slow, may by elevation be brought nearer to a coast line, and beds containing a deep-sea fauna may be succeeded immediately by others more rapidly accumulated, and embedding a shallow-water assemblage. The change of life in such a case would be great, but the lapse of time would not be correspondingly long. If, however, elevation is continued till the area is raised above the sea-level, and remains for a time as a land surface, then a plane of erosion will be formed, and the change of fauna will be accompanied by stratigraphical discontinuity. The break then may be very considerable, but it may or may not coincide with the epoch of change from one fauna to another in an adjacent area of subsidence. Notwithstanding this, it is true that the continuity of deposition may outlast the duration of a certain fauna ; and it may become necessary to draw a line between two faunas of systematic importance in the midst of a conformable series of strata ; in such a case it will naturally be drawn where the greatest palseontological break occurs. In other words, a life-assemblage, characteristic of one period in the world's history, may continue to inhabit any area until such changes occur as to cause the introduction of an assem- blage which has been gradually developed outside that area, and these changes may take place with or without interruption of deposition in that particular area. From the preceding remarks it will be seen that it is impossible to define a system as a series of deposits formed during one complete downward and upward movement or oscillation of some one part of the earth's surface. Moreover, if we could so define a system, the systems of one region could not be coeval with those of other regions. Systems must therefore have a palseontological value, and a system may perhaps be defined as a consecutive series of strata formed during the prevalence of certain generic forms of life throughout a large part of the earth's surface. The student must be prepared to find that systems founded on the differences in successive faunas will not include anything like equal thicknesses of rock. If, as appears most probable, the differentiation of species and the development of higher forms has progressed in a constantly increasing ratio, it is clear that the time- value of systems based on such changes will become less and less as we approach modern times. It is certainly a fact that the same forms of life extend through a much greater thickness of rock in the earlier than in later geological times ; and there is no reason for supposing that the production of rocks went on more rapidly (at any rate during later Palaeozoic periods) than at the present time ; limestones must always have been of slow and THE SUCCESSION OF STRATIFIED ROCKS 11 gradual growth, and yet the persistence of species through the Carboniferous Limestone Series is truly remarkable, while the limestones of the Jurassic and Cretaceous periods, though of far less thickness, are divisible into numerous palseontological zones or belts, each having many species which do not appear in the next. The changes in the forms of life inhabiting any marine area being thus more rapid in later geological times, an equal amount of change would naturally be accomplished in less time, and consequently, if systems are to be measured by palaeontological difference, the newer systems must include less thicknesses of rock. But even when allowance is made for this, it is impossible to regard the divisions of Tertiary time introduced by Sir Ch. Lyell as being of the same relative importance as the systems of older date with which they are usually ranked. Even the Eocene, as understood by Lyell, was hardly of the same palseontological value as Cretaceous or Silurian, while the Miocene and Pliocene are still less entitled to rank with such systems. The Oligocene has been created at the expense of Lyell's Eocene, but it is now admitted that these two together form a group which may rank as a system, while the remaining three groups, Miocene, Pliocene, and Pleistocene, may be regarded as making up another system of deposits. For these two systems M. Hohnes has proposed the names Palceogene and Neogene, and though they are not altogether satisfactory the names have been largely accepted on the European continent. Systems are divided into sections or formations, which are usually denominated as Upper, Middle, and Lower, unless any special names have been proposed for them. These sections are again divided into stages or groups, to which local names are generally given, and these again are often divisible into zones. The following is a tabular view of the twelve fossiliferous systems which overlie the Archaean rocks, together with their primary divisions, according to the nomenclature which will be adopted in this volume, and is applicable to the greater part of Europe. This list of systems is merely a table from which no idea of relative thickness can be obtained, but the adjoining columnar view is drawn to scale and is intended to give some idea of the relative thicknesses of these systems. [TABLE 12 STRATIGRAPHICAL GEOLOGY Systems. NEOGENE . PALAEOGENE CRETACEOUS JURASSIC . - ." TRIASSIC . PERMIAN . CARBONIFEROUS DEVONIAN. . . SILURIAN . ORDOVICIAN CAMBRIAN Primary Divisions. f Pleistocene, -j Pliocene. I Miocene. f Oligocene. I Eocene. | Upper. I Lower. /"Upper (White Jura). - Middle (Brown Jura). I Lower (Lias). fKeuper. -! Muschelkalk. iBunter. f Magnesian Limestone (Zechstein). I Red Sandstone (Rotliliegende). ( Stephanian. -I Westphalian (Coal-Measures). lAvonian (Carboniferous Limestone). ,' Upper. \ Middle. I Lower. {Downtonian or Ludlow Beds. Salopian or Wenlock Beds. Valentian = Llandovery Beds. Bala Series. Llandilo Series. lArenig Series. ( Upper (Olenian). - Middle (Paradoxidian). I Lower (Olenellian). 1U lative Thickness. NEOOENE PAt/EOQENE. CRETACEOUS JURASSIC TRIASSIC CARBO- -NIFEROUS DEVONIAN SILURIAN ORDOVICIAN CAMBRIAN The actual thickness of any system of strata varies of course in different parts of the same country, and for the construction of the diagram the full average thickness of each stage has been taken from English districts where the whole system is well developed. The diagram does not, therefore, show the maximum thickness of any system. Moreover, in other parts of the world the Permian and Triassic Systems are represented by marine deposits which reach a thickness of 9000 to 10,000 feet. THE SUCCESSION OF STRATIFIED ROCKS 13 Further, the Neozoic division is much thicker on the European continent than it is in England, for in the Alps thick deposits come in between the Triassic and Jurassic Systems, and the Neogene Series is of much greater thickness in Southern Europe than it is in the north. CHAPTER II STRATIGRAPHICAL PALAEONTOLOGY IT was stated on p. 3 that ever since the surface of the earth became a fit habitation for plants and animal life there has been a succession of life-forms, and that remains of these faunas are found as fossils in the stratified rocks. It was also stated that this succession was incomplete owing to gaps in the series. The causes which have produced this imperfection of the record may now be indicated, so that the reader may understand what kinds of creatures are likely to occur as fossils and what are not. It will also be shown that, where the succession is continuous and fossils are fairly abundant, they can be used for the establishment of smaller sub- divisions of the series and for the comparison of such subdivisions in different areas. Finally a few words on the distribution of species at the present day will explain some limitations in the application of palaeontological evidence, and the difference between synchronism and homotaxis in correlating the rock - groups of different countries. Origin of Species. At the outset of this inquiry we are met by the question, What is a species'? Before the publication of Darwin's book on the Origin of Species most people regarded a species as the result of a special creation of a definite kind of plant or animal, yet all who studied either plants or animals were obliged to admit that some species displayed many varieties, and also that it was difficult to say what amount of difference should constitute a species. Lamarck in 1801 was the first to propound the doctrine that all existing species are descended from other species, but few naturalists gave credence to his views, and it was not till 1859, when Darwin published his book, that the world became familiar with such a doctrine. On the primary point of the appreciation of differences Darwin remarks (Origin of Species, sixth edition, p. 41), "Certainly no 14 STKATIGRAPHICAL PALAEONTOLOGY 15 clear line of demarcation has as yet been drawn between species and sub-species that is, the forms which in the opinion of some naturalists come very near to, but do not quite arrive at, the rank of species ; or again between sub-species and well-marked varieties, or between lesser varieties and individual differences ; and a series [of specimens] impresses the mind with the idea of an actual passage." Further on he says, "From these remarks it will be seen that I look at the term species as one arbitrarily given, for the sake of convenience, to a set of individuals closely resembling each other, and that it does not essentially differ from the term variety, which is given to less distinct and more fluctuat- ing forms." Darwin and Wallace developed their views about the same time, and both accounted for the existence of so many varieties, species, and genera of organisms by indicating the various causes and conditions which influence all kinds of organisms and tend to modify their habits, forms, and members. They have shown how the constant struggle for existence and the frequent changes of environment may have developed or have increased a tendency to variation, and how this variation results in the formation of varieties and species. Natural selection (or the selective action of natural processes) results also in the extinction of many forms and in the survival of a few namely, of those few which are the most capable of accommodating themselves to the more rigorous and changed conditions of existence. This is called the " survival of the fittest." It is by the infinite repetition of these processes that species which are now widely different may have been evolved from a common ancestor. Another consequence of natural selection is the preservation of those varieties and species which chance to possess peculiarities that are useful to them in the struggle for existence. Thus as Darwin says (op. cit. p. 103), "it leads to the improvement of each creature in relation to its organic and inorganic conditions of life, and consequently in most cases to what must be regarded as an advance in organisation." In other words, it leads to the evolution of higher and higher forms of life. Now if the descendants of varieties may become different species, and if the descendants of different species may in course of time become so differentiated that most naturalists would rank them as different genera, and if this process has been going on during the whole of geological time, since the world first became fit to support life ; then we might expect that the successive assemblages of fossils preserved for us in the rocks of the earth's crust would furnish us with a long succession of links between one 16 STRATIGRAPHICAL GEOLOGY species and another, and would connect the generic forms of the present with those which have long been extinct. Unfortunately, however, though many such links have been found, and though more come to light every year, so that the genealogy of many genera can be satisfactorily made out, yet there are reasons why we can never hope to find all the links in the chain. Imperfection of the Geological Record. That the truth of Darwin's theory of the origin of species cannot at once be proved, and the complete genealogy of modern genera be constructed from the records of the past, is due to the imperfection of that very geological record which would otherwise contain the obvious proof or refutation of the hypothesis. This imperfection results from two causes, (1) the conditions which limit the entombment and preservation of organic remains, (2) the frequent erosion and destruction of the deposits in which they have been buried. 1. Conditions of Entombment. If we consider the manner in which deposits are now taking place, we shall perceive that the preservation of many kinds of organic remains in these deposits is likely to be an exceptional and accidental occurrence. Deposition of sediment is not an uninterrupted process likely to ensure the conservation of specimens of the greater number of plants and animals living at any one time, but is, on the contrary, so discontinuous that only a small proportion of any fauna or flora would be entombed in such a way as to be permanently preserved, either in terrestrial or marine deposits. On land, generations of animals may die and moulder away, shells, bones, and all, without any remains being transported to places where deposits are accumulating. Mr. F. C. Selous remarks that in South Africa "no organic matter lasts very long when exposed to the weather, and even the skull and leg bones of an elephant would, I think, crumble to dust and absolutely disappear in less than fifteen years from the date of the animal's death." l In some countries where the climate is very dry they may last longer, but still ultimately decay and crumble into dust. Of the larger land animals a few are occasionally buried in bogs or drowned in lakes and rivers, and still fewer are carried down by rivers to the sea. The animals which live in fresh water and the leaves of plants stand a better chance of preservation in those parts of the country where lacustrine and fluviatile deposits are being formed ; but these purely freshwater beds occupy but small areas, and are particularly liable to subsequent destruction, so that as a rule our knowledge of the terrestrial and fluviatile fauna of any period is derived from the specimens which were carried down 1 Sport and Travel, by F. C. Selous, 1900, p. 156. STRATIGEAPHICAL PALAEONTOLOGY 17 the streams during floods, and were buried in the estuarine deposits at the mouths of large rivers. The creatures which inhabit the sea have a much better chance of being preserved as fossils, since deposits of one kind or another are constantly being formed over large areas of the sea-bottom, especially where subsidence is in progress. Yet even among marine forms of life there are many which would only rarely be entombed ; those animals, for instance, which live on rocky shores and in clear water would not be preserved except where deposits of limestone were in process of formation. The majority of the creatures embedded would be those which lived on sandy and muddy bottoms, together with some of the free-swimming forms which moved through the water above. Lastly, of the creatures so embedded only a small proportion would ultimately remain ; those, namely, which possessed hard structures capable of fossilisation after the softer parts had decayed and disappeared. The following table shows at a glance which kinds of animals, out of all those which live in the sea, would be likely to leave memorials behind them and which would perish without leaving any trace of their existence. Sub-kingdoms. Protozoa. Porifera. Ccelenterata. Echinoderma. Vermes. Arthropoda. Molluscoida. Mollusca. Vertebrata. Groups with Hard Structures. /Foraminifera. \Radiolaria. /Silicispongise. \Calcispongi0e. Sertularida. Hydrocorallina. Graptolitoida. Stromatoporoidea. Madreporaria. Some Alcyonaria. All classes. fTubicola and the horny \ jaws of Nereids. All classes. / Bryozoa. \Brachiopoda. fLamellibranchia. I Scaphopoda. | Gastropoda except (.Cephalopoda except /Teeth, bones, and \ scales of all kinds. Perishable Groups. Flagellata. Infusoria. Myxospongiae. Ceratospongiae. Hydroida (except Sertularida). Siphonophora. Discophora. Actiniaria. Alcyonaria (some). Ctenophora. None. /Hirudinea. (Oligochaeta. Nudibranchiata. some Cuttle-fish. Tunicata. But although all the classes of creatures entered in the second column possess hard parts which are capable of preservation, it does not follow that these are always preserved. Their conversion C 18 STRATIGKAPHICAL GEOLOGY into fossils depends partly on the solidity and character of the substance of which the hard parts are composed, and partly on the rapidity with which the remains are covered up and embedded in sediment. Lastly, their final preservation as fossils depends upon the conditions to which the embedding rock is subsequently ex- posed, upon its conservation or destruction. The remains of those creatures which live in deep water or over deep water, such as many Foraminifera, Radiolaria, siliceous sponges, and pteropods, have such delicate tests and spicules that they often fall to pieces and become mere debris before they can be embedded in sediment ; for in deep water the accumulation of sediment is usually a very slow process. The thin shells of the Radiolaria and the Pteropoda are especially liable to rapid disintegration, and consequently they are rare as fossils, although they are known to have existed from very early times. 2. Subsequent Removal of Organic Remains. The proportion of those remains which would otherwise be preserved is further diminished by the subsequent destruction of some of them by the action of percolating water. All water which percolates downward from terrestrial surfaces contains carbonic acid, and such carbonated water attacks and dissolves carbonate of lime, and also that form of silica which enters into organic structures. Hence all such embedded remains are liable to solution and removal under certain circumstances. So long as the deposit containing embedded organisms remains below the sea in which it was formed, it is probable that the changes which take place are slight and gradual. It is when the fossiliferous deposit is raised above the level of the sea and is added to the dry land that changes are apt to take place. Even then, if the rock has formed part of a thick series deposited during a long period of subsidence, and has remained buried beneath others for some time before the whole mass is raised to form land, most of the fossils in it are likely to be preserved, for the rock becomes consolidated and remains saturated with saline water, which probably contains as much carbonate of lime as it is capable of holding in solution under such conditions. Many fossils exhibit changes of structure which have very probably been effected during upheaval and while the rock was in a water-soaked condition ; thus the substance of shells is frequently converted into crystalline calcite, and in other cases the calcareous shell is replaced by chalcedonic silica. Those parts of a formation, however, which come to form the surface portions of newly raised land, i.e. those parts which are raised above the lowest summer level of saturation, are subjected STEATIGRAPHICAL PALAEONTOLOGY 19 to tlie action of percolating water, with the result that many shells are dissolved, leaving only the hollows in which they lay. These hollow spaces may remain empty or they may be filled subse- quently with infiltrated mineral matter, or if disturbances take place before the final consolidation of the rock, the internal cast may be squeezed on to the external impression of the shell and receive an impress of its ornamentation. In any of these cases some record of the original organism remains, but if the rock material is a loose sand or other very soft kind of sediment all trace of the organism may be destroyed. With regard to the durability of the original shell much depends on its original mineral composition. In some shells and tests the carbonate of lime consists of the variety known as calcite, while in others it has the structure of arragonite, and it is a well- ascertained fact that arragonite is much less stable than calcite. It was pointed out by Mr. Sorby 1 that shells composed of arragonite disappeared much faster than the calcite shells wlien both were exposed to carbonated waters, and he pointed out that the creatures which possess hard calcareous structures can be separated into two sets according to the nature of the calcareous substance. These two sets may be tabulated as follows : Oalcitic Structures. Arragonite Structures. Foraminifera. Calcareous sponges. Echinodermata. Actinozoa (corals, etc.). Annelida. Lamellibranchia (except Oysters, Most Bryozoa. Pectens, and the outer layer of Brachiopoda. Spondylus, Pinna, and Mytilus). Oysters and Pectens. Gastropoda (with a few exceptions). Crustacea. Cephalopoda. Professor P. F. Kendall has also made observations and experiments which confirm those of Mr. Sorby. 2 He has pointed out that in the Coralline Crag of Aldeburgh some fossils remain perfect, while of others there are only casts, and that the casts are without exception of arragonite shells, while the former are calcitic structures, such as Bryozoa, Echinoderms, Serpulae, Brachiopoda, and the bivalve Molluscan genera, Anomia, Ostrea, Pecten, Lima, and Pinna ; with the latter, however, occurred the Gastropod shell Scalaria, which he thought must be calcitic, an inference which subsequent experiment proved to be correct. 3. Destruction of Fossiliferous Deposits. There is another set of agencies which contribute to render the geological record a very imperfect one. When any mass of marine deposit is raised into 1 Quart. Journ. GeoL Soc. vol. xxxv. (1879), Pres. Address. 2 GeoL Mag. 2nd Dec. vol. x. p. 497. 20 STKATIGRAPHICAL GEOLOGY land, there is not only an absence of deposition, but there commences an actual destruction of the records previously accumulated. In the first place, as the rising deposit passes through the plane of the sea-level, it is exposed to the erosion of the waves, and if the latest deposits are not very thick, and the process of elevation is not very rapid, the probability is that only small portions will escape destruction. These portions will remain for a time as plains, marsh lands, and raised beaches ; but as elevation continues, they will be brought gradually under the influence of atmospheric agencies, and exposed to continual wear and tear so long as they remain above the sea. It follows, therefore, that only deposits of considerable thickness will survive the attacks of destructive agencies, and remain to form a permanent part of the earth's crust, but thick deposits cannot be accumulated in shallow seas, which are the most favourable to life, so long as the sea-bed remains stationary or is being elevated. Thick formations can only be accumulated in areas where constant subsidence has been in progress, or where the shore originally shelved rapidly into deep water, as off the coast of the Riviera. In the latter case the sediment will be rapidly accumulated, and will not contain so many or such varied forms of life ; but under the former conditions, if the supply of sediment was sufficient to keep the sea shallow and to embed the organic remains, a formation rich in fossils, and thick enough, when upraised, to resist erosion, would be formed. We must, therefore, conclude with Darwin, " that nearly all our ancient formations, which are throughout the greater part of their thickness rich in fossils, have thus been formed during subsidence." l This conclusion that most of our great fossil-bearing formations have been deposited during subsidence has a very important bearing on the question before us, viz. whether the rocks are likely to con- tain the remains of an uninterrupted succession of species, or whether the record is more or less incomplete ; for, as Darwin remarks : " During periods of elevation the area of the land and of the adjoining shoal parts of the sea will be increased, and new stations will often be formed all circumstances favourable, as previously explained, for the formation of new varieties and species ; but during such periods there will generally be a blank in the geological record. On the other hand, during subsidence the inhabited area and number of inhabitants will decrease (excepting on the shores of a continent when first broken up into an archi- pelago), and consequently during subsidence, though there will be much extinction, few new varieties or species will be formed ; 1 Oi'igin of Species, sixth edition, p. 273. STEATIGRAPHICAL PALAEONTOLOGY 21 and it is during these very periods of subsidence that the deposits which are richest in fossils have been accumulated." l Succession of Faunas. From the considerations above mentioned it will be seen that we cannot expect to find all the transitional forms which, if the theory of evolution be correct, must have connected one species with another, and have linked together all the past and present species of any one genus into one long chain of life. All that we can hope to find are certain cases where so large a number of closely allied species and varieties have been preserved as to furnish us with strong evidence of their having been links in such a chain. There are, moreover, two well-established generalisations from palaeontological facts which lead to a very significant inference ; these considerations are (1) that the assemblage of fossils found in any formation has a general character intermediate between the assemblages above and below : and (2) that the fossils from two consecutive formations are far more closely related to each other than the fossils from two more remote formations. Commenting on these Darwin observes : 2 " On the theory of descent, the full meaning of the fossil remains from closely consecutive formations being closely related, though ranked as distinct species, is obvious. As the accumulation of each formation has often been interrupted, we ought not to expect to find in any one, or in any two, formations all the intermediate varieties between the species which appeared at the commencement and close of these periods ; but we ought to find, after intervals very long as measured by years, but only moderately long as measured geologically, closely allied forms, or, as they have been called by some authors, representative species ; and these assuredly we do find. We find, in short, such evidence of the slow and scarcely sensible mutations of specific forms as we have the right to expect." In spite, therefore, of the imperfection of the record, the known succession of life-forms does give very great support to the theory of evolution. Every succeeding fossil assemblage contains not only a larger number and greater variety of forms, but also some which are distinctly more specialised and more advanced in organisation. It is an acknowledged fact that many of the more ancient forms of life were what are termed " comprehensive types," i.e. types which combine the structural characters of two or more different orders or classes of creatures, which are now completely differentiated from each other. In many cases also ancient animals resemble the embryos of the more recent animals belonging to the same classes. 1 Op. cit. p. 275. 2 Op. cit. sixth edition, p. 307. 22 STEATIGRAPHICAL GEOLOGY Beyond all this there can be traced a distinct and decided progress in the types of life as we ascend the geological scale. It is true that in the earliest fauna yet known we start with a number of more or less specialised invertebrate types, but the more highly organised orders of Gastropods, Cephalopods, Crustacea, and Echinoderms do not appear till later periods ; Fish are not known earlier than Silurian times ; Amphibia commence in the Carboniferous ; Keptiles in the Permian and Trias ; Birds in the Jurassic ; Mammals of low organisation in the Trias ; higher Mammals, including Lemurs, appear in the Eocene, but true monkeys are not yet known from beds of greater age than the Miocene. Zones. It only remains to indicate how the succession of life- forms may be utilised for the purposes of subdivision and correla- tion, when we have to deal with a continuous succession of deposits, and more especially with a great thickness of one kind of deposit, such as shale or limestone. During the accumulation of such a mass of sediment it would seem that small changes were continu- ally taking place in the personnel of the fauna, new forms taking the place of old ones, some species dying out and others coming in from elsewhere ; so that when the fossils are carefully collected from each successive bed it is found that they form assemblages of species, each extending through a certain thickness of beds and thus characterising a band or zone within the formation. A zone, therefore, is a certain thickness of stratified material which contains a certain assemblage of fossil species, some of these species being either restricted to the band or particularly abundant in it. Moreover, it is not merely a specially fossiliferous band in a thick mass of sediment, but is one of a succession of zones, each stage in a fossiliferous series being usually divisible into two, three, or more zones. The following remarks on zones are quoted from the author's memoir on the Cretaceous Rocks of Britain. 1 " By the fauna of a zone we generally mean all the fossils which have been or can be found in the beds which are recognised as belonging to the zone. But what may be termed the critical fauna of a zone is the much smaller assemblage of fossils which are either restricted to it or are specially abundant in it. One of these species is then chosen as the index of the restricted assemblage of species, and the name of this fossil is given to .the zone, so that we speak of it as the zone of this fossil, e.g. the zone of Ammonites varians, or the zone of Belem- nitella mucronata. " The reader must be cautioned against becoming possessed with 1 ''The Cretaceous Rocks of Britain," vol. i. p. 34, Mem. Geol. Survey. STRATIGEAPHICAL PALEONTOLOGY 23 the idea that a zone is a set of beds characterised by the occurrence of one particular species, and that every bed which yields this index-species must belong to the zone. It must be remembered that this particular species is only one of several or many, and that it is this assemblage of species which is the guide to the zone. " The value and importance of a zone depends partly on the number of restricted species and partly on the extent of country over which it can be traced. The thickness of a zone is of course a stratigraphical accident, depending merely on the amount of sediment deposited at any locality during the lifetime of the zonal fauna. Thus a zone may be in one area only a few feet in thick- ness, in another it may be 20 or 30 feet, and in a third it may expand into 200 or 300 feet. " The limits of a zone may be definite or indefinite. Where sedimentation has been continuous and fairly rapid, the limits of the zones will naturally be indefinite ; some of the characteristic species of one zone may survive into the overlying zone, and others which are characteristic of the latter may make their first appearance in the former. Thus in many cases it is impossible to say exactly where the one zone ends and the other begins, and though by careful collecting one may fix it with fair certainty in one section, one cannot be certain of taking exactly the same plane of separation in another quarry or cliff which is a few miles away from the first. Thus where a zone is complete its limits are indefinite, but the succession of zones will be the same in both places. " Where a zone has definite limits it is generally because sedimentation ceased for a time, and in many cases because an erosive current swept away some of the sediment which had previously been accumulated, thus destroying the continuity of the record and producing what is called a ' surface of contemporaneous erosion.' Even if the current was only strong enough to prevent the accumulation of sediment, it causes a break in the record, because for a time there is no embedment and preservation of organic remains. In this way, therefore, a zone or a sub-zone may die out when traced in a certain direction." In some cases still smaller zonal subdivisions may be recognised, consisting of one or two beds which are characterised by one or more peculiar species ; these may be called sub-zones or hemerce, but there is not necessarily a complete succession of such sub-zones, nor can they be expected to have so wide a geographical extension as zones ; they are merely special horizons, and equivalent to the niveaux of French geologists. Synchronism and Homotaxis. It was stated in Chapter I. that when two districts are physically separated from one another 24 STRATIGRAPHICAL GEOLOGY we have to rely entirely on community of fossils in comparing the rocks of the one with those of the other ; and if the rocks so compared were deposited originally within the limits of the same life-province, the groups which contain similar assemblages of species may safely be regarded as contemporaneous. To take an example, no one would hesitate to correlate the members of the Ordovician and Silurian Systems in North Wales and the Lake District by means of the fossils they contain, because there is every reason to suppose that the two districts formed part of the same province during those periods. Similarly the members of the Lower Cretaceous Series in the Isle of Wight and in the Weald may be correlated with some degree of exactitude ; but the same group of beds in Lincolnshire contains such a different set of fossils that it obviously belonged to a different life-province, and it is impossible to indicate the exact equivalents of the several members of the southern series. Again, when we come to compare districts at a still greater distance from one another as, for instance, Wales and Bohemia, Sussex and the South of France, or England generally with North America we find that there is certainly a general resem- blance in the vertical succession of faunas which have flourished in the respective areas, i.e. the rock- series in each country can be divided into groups, which are characterised by the presence of the same genera ; but the species are different, and it would be a mistake to conclude that portions of formations containing assemblages of the same genera in countries so far apart were contemporaneous in the same sense as the members of the Silurian Series in Wales and Cumberland are contemporaneous. A group of species originating in one locality and spreading along the coasts of a continent would take a long time to travel from Europe to America even if there were continuous coast lines. Sometimes their progress would be arrested for a time, until physical or geographical changes allowed them to advance ; at other times they would make their way by accommodating them- selves to varying conditions and by throwing out collateral varieties. The date of their first appearance in one part of the world would be very much in advance of their arrival in the other, so that the fact of two distant formations containing fossils of closely allied species can be no proof of their being really contemporaneous. It is certain, however, that in every part of the world the geological sequence discloses the same general succession of life- forms. The progress of development may have been retarded here and accelerated there, but the successive fossil assemblages are similar to one another, and the order in which they occur is the STKATIGRAPHICAL PALAEONTOLOGY 25 same all over the world. We never find a Carboniferous fauna antedating a Silurian or Devonian fauna, though we may find the first succeeding the second without the presence of the inter- mediate Devonian assemblage. This common order of succession has been termed Homotaxis, and the Silurian System of Britain is spoken of as homotaxial with and equivalent to the Silurian of America ; even the larger subdivisions of the one may be paralleled with those of the other, as homotaxial equivalents or representatives, so that we may speak of the Niagara Group as equivalent to the Wenlock Beds, understanding by this that the groups are homotaxial equivalents and not absolutely coeval formations. To sum up the preceding observations, the occurrence of the same assemblage of species at different localities may be taken as evidence that the beds containing them are of the same age, and were formed at the same time. Again, a knowledge of the general succession of generic forms enables us to fix the relative age of any set of fossils, and, consequently, the homotaxial position of the beds containing them. In other words : 1. Assemblages having many species in common may be regarded as contemporaneous. 2. Assemblages having many genera in common may be regarded as homotaxial equivalents. CHAPTER III THE LITERATURE OF HISTORICAL GEOLOGY IT is believed that some information as to the principal geological publications and as to the facilities for ascertaining what has been published about any particular system or series of beds would be useful to the student. During the first half of the last century, and indeed up to about the year 1860, publications on stratigraphical geology were comparatively few, and it was possible for any geologist to have a fair acquaintance with all that had been written on the subject ; but as time went on the number of workers increased, and more numerous additions were made every year to our knowledge of British rocks and fossils, as well as of their equivalents in other countries. Every year added a volume to the publications of the Geological Societies of London, Edinburgh, Glasgow, Manchester, and other centres, to the volumes of the Palaeontographical Society and of the Geological Magazine. The Geological Survey also have issued a long succession of maps and memoirs, most of which have appeared within the last forty years. At the present day, therefore, it is almost impossible for any one to make himself acquainted with all that has been written on the stratified rocks of Great Britain, and quite impossible to have at the same time a good knowledge of the fossils which they contain. Consequently in these branches of geology, as in the case of other sciences, the tendency is toward the specialising of knowledge. After the student has acquired a certain amount of knowledge about the successive geological systems and their characteristic fossils he probably becomes specially interested in one of them, or in the rocks which occur in some particular district, and then he wishes to know in what books and memoirs accounts of that series of rocks or of that particular district have appeared. The descriptive parts of this volume contain only the more 26 THE LITERATURE OF HISTORICAL GEOLOGY .27 important facts respecting each system of rocks. The book is designed to include only so much information about each great series of stratified deposits as the student may be expected to read and to have before him for constant reference. Descriptions of each successive series will be given with a certain amount of detail concerning the variations which they exhibit in the British Islands, and their representatives on the European continent will be indicated in a more concise manner, but yet in sufficient detail to give the student some idea of their known extent and of the conditions under which they were deposited. But the task of selecting information from the great mass of detail which has now been accumulated is not an easy one, and further, what is regarded as specially important in one part of the country may seem of less importance in another part, for naturally each teacher will wish his pupils to become specially acquainted with the rocks of the country or district in which they live. I propose, therefore, in this chapter to indicate briefly what are the most important and useful publications, how they can be obtained, and how they should be used, in order that the reader may learn how to extend his knowledge of any .particular series of rocks, or to find out what has been published about the geology of any special district. The publications may be grouped under four heads Maps, Stratigraphy, Palaeontology, Palceogeography. MAPS It is necessary that the student should have access to a geological map of the British Islands. Every teacher will doubtless see the necessity of exhibiting one in his lecture-room, but every student should as soon as possible acquire one for himself. Fortunately a good cheap map of the British Isles is no longer a desideratum, for one is now issued by the Geological Survey at the low price of 2s. There are several others on a larger scale at a higher price, as well as good separate maps of England, Scotland, and Ireland. The following general maps may be recommended : 1. Geological Survey Index Map. Scale 25 miles to an inch, 2s. 2. Geological Map of the British Isles, from Stanford's London Atlas. Price as separate map, 10s. 6d. 3. Stanford's Geological Atlas of Great Britain and Ireland, edited by H. B. Woodward, 2nd edition enlarged, with plates of characteristic fossils. Price 1 2s. 6d. net. 28 STRATIGRAPHICAL GEOLOGY 4. Geological Map of England and Wales, by Woodward and Goodchild, 1 7| miles to an inch. Price 5s., in case 7s. 6d. 5. Geological Map of England and Wales, by Sir A. Geikie. Scale 10 miles to an inch. Price 10s. 6. Geological Map of Scotland, by Sir A. Geikie (Bartholomew). Scale 10 miles to an inch. Price 6s. 7. Map of the Surface Geology of Ireland, by Sir A. Geikie (1908). Scale 10 miles to an inch. Price 4s. 6d. 8. Geological Map of Western and Central Europe, reduced from the International map. Stanford. (In preparation.) Besides the small index map above mentioned the Geological Survey publishes a larger Index Map on the scale of 4 miles to an inch, a new edition of which is in course of publication in 19 sheets, most of which are issued and priced from 2s. to 2s. 6d. If maps of a special district are required the reader should write for a catalogue of the maps issued by the Geological Survey of the United Kingdom, this catalogue being obtainable from Mr. E. Stanford, Long Acre, London ; or from W. and A. K. Johnston, Edinburgh, or from Hodges, Figgis, and Co., Dublin. Besides the index map above mentioned the Geological Survey issues two sets of maps on the scale of one inch to a mile, one being known as the " Old Series," the other as the " New Series." STRATIGRAPHY It will be convenient to mention first those books and memoirs which are of a general nature but yet give more information about certain formations than will be found in this volume. The most complete Manual of Geology in English is Sir A. Geikie's Text-book of Geology, 4th edition (1903) published by Macmillan and Co. in 2 volumes, price 32s.. The second volume deals with Stratigraphical Geology and includes accounts of the foreign equivalents of each system, not only in Europe but in other parts of the world. A similar work in French is De Lapparent's Traite' de Ge'ologie, 5th edition (1905), in 3 volumes, price 35s. A useful epitome of the geology of many counties and districts in England and Wales will be found in the Jubilee volume issued by the Geologists' Association in four parts, price 5s. each (1910). Each article in this publication has been written by a geologist who is specially well acquainted with the district he describes. The Geological Survey has published the following memoirs, which deal either with special formations or with districts of considerable size and importance : THE LITERATURE OF HISTORICAL GEOLOGY 29 The Geology of North Wales, by Sir A. Eamsay, 2nd edition (1881). Price 21s. with map and sections. The Geology of the N. W. Highlands of Scotland, by B. N. Peach, J. Home, and others (1907). Price 10s. 6d. The Geology of the Southern Uplands of Scotland, by B. N. Peach, J. Home, and others (1900). Price 15s. The Geology of London and of Part of the Thames Valley, by W. Whitaker, in two volumes (1889). Price 11s. Guide to the Geology of London and the Neighbourhood, by W. Whitaker and H. B. Woodward. Price Is. The Geology of the Bristol and Somerset Coal-field, by H. B. Woodward (1876). Price 18s. The Triassic and Permian Bocks of the Midland Counties of England, by E. Hull (1869). Price 5s. The Geology of the Weald, by W. Topley (1875). Price 17s. 6d. The Geology of the Isle of Wight, by C. Eeid and A. Strahan (1889). Price 8s. 6d. The Geology of the Isle of Purbeck and Weymouth, by A. Strahan (1898). Price 10s. 6d. The Jurassic Rocks of Britain Vol. I. Yorkshire, by C. Fox-Strangways (1892). Price 8s. 6d. Vol. II. Yorkshire, by C. Fox-Strangways. Tables of Fossils. Price 12s. Vol. III. The Lias of England and Wales, by H. B. Wood- ward (1893). Price 7s. 6d. Vol. IV. The Lower Oolitic Kocks of England, by H. B. Woodward (1894). Price 10s. Vol. V. The Middle and Upper Oolitic Eocks of England, by H. B. Woodward (1895), with Bibliography of Jurassic Literature. Price 7s. 6d. The Cretaceous Rocks of Britain VoL I. The Gault and Upper Greensand, by A. J. Jukes- Browne (1900). Price 9s. Vol. II. The Lower and Middle Chalk of England, by A. J. Jukes-Browne and W. Hill (1903). Price 10s. Vol. III. The Upper Chalk of England, by A. J. Jukes- Browne and W. Hill (1904). Price 10s. The Pliocene Deposits of Britain, by C. Eeid (1890). Price 5s. 6d. A list of the smaller or more local memoirs (explanations of various sheets of the one-inch map) will be found in the Catalogue 30 STRATIGRAPHICAL GEOLOGY of Maps and Memoirs issued by the Survey. Many of them have lists of the publications referring to the district under description. The next most important source of information is the Quarterly Journal of the Geological Society of London, the publication of which commenced in 1845. There is an excellent index to the first fifty volumes of this Journal, which can be purchased from the Geological Society, Burlington House, London (price 10s.). This index gives references to the names of authors, of places, and of subjects, so that by its means it is easy to ascertain what papers have been published in the Journal on any particular area or any set of beds. This index deals with the yearly volumes between 1845 and 1894. The Geological Society also publishes every year a list of the various items of geological literature which have been added to its library. This publication began in 1894, with a list for the latter half of that year (price Is.). Each succeeding annual part is issued in a paper cover (price 2s.), and is divided into two portions, the first being an index of authors' names with titles of papers, and the second an index of subjects, and each separate paper in all the publications purchased or received by the Geological Society is entered in these indices, as well as all the papers published in the Quarterly Journal of the Society since 1894. It will be seen, therefore, that this annual list forms a fairly good index to the geological literature of the whole world. Before this work was taken in hand by the Geological Society of London two attempts had been made in England to establish an annual record of geological literature. These were : 1. The Geological Record, consisting of six annual volumes (1874-1879), edited by W. Whitaker, giving short abstracts of all papers, British and foreign ; and two volumes (for 1880 to 1884 inclusive), edited by W. Topley, giving titles only. Published by Taylor and Francis, London. Price 16s. per volume. 2. The Annals of British Geology, by J. F. Blake (Dulau and Co., London). Four volumes (1890-1893). Price of the vol. for 1890, 5s., of others, 9s. The volumes of the Geological Magazine contain many articles on stratigraphical geology, and a general Index to all the volumes from 1864 to 1903 has been published, price 21s. The Geologists' Association publishes Proceedings which contain many useful papers, and each volume contains the proceedings and communications of two years, with accounts of excursions and a good index. The first volume is for the years 1868-69. Of the provincial societies which publish geological papers the THE LITERATURE OF HISTORICAL GEOLOGY 31 chief are the following : the Geological Societies of Edinburgh, Glasgow, Manchester, and Liverpool, the Yorkshire Geological and Polytechnic Society, and the Royal Geological Society of Cornwall. Particulars respecting the above and all other scientific societies in the United Kingdom will be found in the Official Year -Book of Scientific and Learned Societies, published by C. Griffin and Co., Strand, London. Price 7s. 6d. each annual volume. The British Islands have been the scene of repeated volcanic eruptions, and volcanic products are interbedded with the sediments of all the Palaeozoic systems, except perhaps the Permian. Brief descriptions of the contemporaneous volcanic rocks will be given in the following pages after that of the sedimentary rocks of each period, but the reader will find this part of the subject admirably treated by Sir Archibald Geikie in his Ancient Volcanoes of Great Britain (Macmillan and Co., 2 vols. 1897, price 36s.). PALEONTOLOGY For the proper study of historical geology it is essential that the student should have some knowledge of the forms and structure of the different classes of animals, and especially of the Invertebrate classes. This general acquaintance he can get from a careful perusal of Mr. H. Woods's small manual of Palaeontology in the Cambridge Natural Science Series (second edition, 1899, price 6s.), but he may sometimes find it necessary to refer to larger books on the subject. Of these the best are : A Manual of Palaeontology, by Messrs. Nicholson and Lydekker, third edition, in two volumes (1889), price 42s. Zittel's Text-book of Paleontology, the American edition of 1900, published by Macmillan and Co., vol. i., price 25s. Outlines of Vertebrate Palaeontology, by A. S. Woodward, Camb. Univ. Press, price 14s. Besides these may be mentioned a series of volumes now being issued under the title of A Treatise on Zoology, edited by Professor E. R Lankester, and published by A. and C. Black. This series will be complete in ten parts, of which nine have been published. Price from 12s. 6d. to 15s. per part. Fossil forms are dealt with in most of them. The stratigraphical geologist, however, has to deal with fossils principally as important aids in determining the relative age of rocks, and he has therefore to make himself acquainted with the fossils which are most characteristic of the different systems, and especially with those species which have a restricted range within each system and thus characterise stages and zones. 32 STRATIGRAPHICAL GEOLOGY Lists of characteristic fossils will be given in this volume, to- gether with figures of some of them. The student will, however, be glad to know where he may find figures of other species, and he may consult the following books with advantage : For Paleozoic fossils a useful book is Baily's Characteristic British Fossils, vol. i. (Palaeozoic), with 42 plates (1875). Copies of this can still be obtained from Messrs. Dulau and Co., 37 Soho Square, London. Price 10s. For Tertiary fossils Lowry's Chart of Characteristic Tertiary Fossils can be recommended. (Stanford. Price 4s.) For fossil Crustacea Woodward and Lowry's Chart of Fossil Crustacea is good, though not limited to British species. The more recent memoirs of the Geological Survey that is to say most of those published since 1890 contain figures of some of the characteristic fossils of the formations described. The memoirs on the Jurassic rocks and on the Cretaceous rocks may be speci- ally mentioned as containing many figures of fossils that will be useful to the student. For illustrations of fossil plants and for information on their structure Seward's Fossil Plants should be consulted (Cambridge Nat. Science Manuals, two volumes, price 12s. each). Studies in Fossil Botany, by D. H. Scott (A. and C. Black), is also excellent. As works of reference which ought to be found in every library of scientific books the publications of the Palaeontographical Society must be mentioned, for these volumes are the great store- house of illustrations of British fossils. The annual quarto volume of the Society costs only 21s., and contains parts of four or five monographs, with from thirty to forty plates. The Society was established in 1847, and a list of the monographs which have been completed and of those which are in progress can be obtained from Messrs. Dulau and Co. (Soho Square, London). The completed monographs can be purchased separately, and include monographs on Fossil Corals, Echinoderms, Brachiopoda, Trilobites, Devonian Fossils, Mollusca of the Crags, Cretaceous Bivalves, and Ammonites of the Lias. Another work of reference is Etheridge's Fossils of the British Isles, vol. i. (Clarendon Press, Oxford, 1888), which is a complete list of the fossils recorded from the Palaeozoic rocks of Britain up to that year, showing the range of each species. PAL^OGEOGRAPHY The geography of the earth's surface has gone through a long series of phases and mutations. Each period may be said to have THE LITERATURE OF HISTORICAL GEOLOGY 33 had a geography of its own, but the special phase with which it opened slowly changed toward the evolution of the ensuing phase by the upheaval of one region and the subsidence of others, or by a general regional movement of elevation or subsidence. Sometimes indeed great changes seem to have taken place within the duration of a single period. The present arrangement of land and sea, as well as their relative heights and depths, is merely the last of these phases, but it is at the same time a result of the geographical evolution which has been in progress ever since the earth's crust was cool enough for water to condense and settle upon it ; in other words, modern geography is the outcome of all the past geographical changes. The restoration of ancient phases of geography, the attempt to indicate the limits of the seas and the positions of the land areas of any age or period, is one of the most difficult problems that a. geologist can endeavour to solve. Shore-lines are seldom preserved, and the essay involves a careful consideration of the conditions under which various local deposits have been formed, and a studied use of the scientific imagination in estimating the original extension of deposits, of which only small portions are in many cases pre- served or exposed. Such endeavours to restore the geographical conditions of past periods must at present be tentative, and with respect to the older periods they are probably more suggestive than real ; but they exercise the faculty of inductive reasoning, and they help to show the student that there are aims and objects to be attained by the study of historical geology which are more interesting than the mere enumeration of rock-groups or the recognition of character- istic fossils. A brief account of the conditions under which each series of rocks has been accumulated and of the relative positions of land and sea in the European region during each period will be given in the following pages ; but for more detailed consideration of the geographical changes in the British area the reader is referred to the author's treatise on the Building of the British Isles (third edition, 1911, Stanford), and to Professor Hull's Contributions to the Physical History of the British Isles (1882, Stanford). He will also find maps that deal with a wider area in the last edition (fifth) of De Lapparent's Traite de Geologie (1905). CHAPTER IV THE ARCHAEAN ROCKS A. CLASSIFICATION AND NOMENCLATURE IN this chapter we shall deal with the oldest rocks which are known to enter into the surface-structure of the earth's crust, rocks which can be seen to emerge from below the oldest fossiliferous (Cambrian) strata, not only in Britain but in many other parts of the world. These primitive formations consist mainly of crystalline rocks granite, gneiss, and crystalline schists but they also include great thicknesses of stratified rocks, partly sedimentary and partly volcanic, which have not been altered beyond recognition as such rocks. Gneisses and crystalline schists belong to a category of their own and are generally termed metamorphic rocks, yet they have certainly not all been formed in the same way, and the determina- tion of their relative age and manner of formation is often a very difficult matter. Even the stratified Archaean rocks cannot be classified and correlated in the same way as those of the Palaeozoic systems can, because they are destitute of fossils, except some minute microscopic bodies, and because they vary greatly in litho- logical composition and in the extent of metamorphic change which they have undergone. Consequently, when a succession of pre-Cambrian rocks has been made out in any one area, it is very difficult to correlate the divisions which are observable in this area with those which can be made out in another. Even with regard to larger divisions, of the rank of series or systems, European geologists have not yet formulated any classification which has met with general acceptance. They differ as to the number of rock-groups which can be recognised in some of the principal areas of pre-Cambrian rocks, and also as to the relative age of some of the component rock-masses ; and in some cases it is a matter of dispute whether certain schists and 34 THE ARCH^AN ROCKS 35 phyllites are of Archaean age at all, or are metamorphosed sediments of much later date. All agree, however, in admitting that in the regions where large tracts of pre-Cambrian rocks are exposed, there is a general succession from an older to a younger series, and that in some areas there is a distinct break and unconformity between the older and younger series. Thus there is no doubt that the Archaean rocks can be divided into two distinct systems, and some continental geologists advocate the restriction of the name Archaean to the older series and apply that of pre-Cambrian to the newer and more distinctly stratified series. This, however, is not a satisfactory nomenclature, for the term pre-Cambrian is an indefinite and inconvenient name for a system or a period, though it is a useful adjective in a general sense. It is better, therefore, to call all pre- Cambrian rocks Archcean, using the term Protarchaean for the oldest and Nearchcean or Eparchcean for the younger series. Difficulties, however, still remain, because in some countries, and notably in America, the older crystalline rocks have been divided into two and even three series, while in some areas two series of distinctly stratified rocks are believed to exist. Ultimately it may be possible to divide the whole into three distinct systems, as suggested by Professor Boiiney, which may then be called Protarchaean, Mesarchaean, and Eparchaean. 1 a Meanwhile at the present time the rock-groups which have been recognised in different areas are known by different local names ; so that not only every country, but almost every separate district in which Archaean rocks are exposed, has its own special nomen- clature, and it is seldom possible to correlate the groups established in one country with those of another. Under these circumstances it is obvious that the Archaean rocks cannot be described in separate chapters on a stratigraphical basis, but must be dealt with regionally, those of each large area being described separately and successively. As regards their total thickness exaggerated ideas seem to prevail in some quarters. Attempts to estimate the thickness of crystalline schists are quite untrustworthy, and where that of the stratified series can be estimated their thickness is not more than that attained by Cambrian or Silurian strata in some districts, i.e. from 10,000 to 16,000 feet. a References will be found at the end of the chapter. 36 STKATIGRAPHICAL GEOLOGY B. BRITISH REPRESENTATIVES 1. Scotland By looking at any geological map of Scotland the reader will see that, if the Outer Hebrides are included, about two- thirds of the whole country are occupied by gneissic, schistose, and semi- crystalline rocks, which are believed to be older than the Cambrian. There is now no doubt that this great region of the Scottish Highlands includes several series of pre-Cambrian rocks, and the region is divisible into three areas : (1) a western area, including the Hebrides, the western parts of Sutherland and Ross, and tracts in Skye, Mull, and Islay ; (2) a broad medial area extending from the north coast of Sutherland to the south of Inverness ; (3) the Central Highlands from Aberdeen to Argyllshire. Parts of these three areas are shown on the accompanying map (Fig. 1), and they appear to include three different series of Archaean rocks, though each area contains representatives of more than one division, and it is only in the west that the relative positions of two of these divisions are quite clear. The western belt is marked off from the median tract by a thrust-plane of enormous displacement which is known as the Moine thrust. The western outcrop of this plane runs from Whiten Head near Loch Eriboll in a S.S.W. direction, through Sutherland and Ross and through the Sound of Sleat. Thence it probably passes outside the Isles of Mull and Islay. West of this line the Archaean rocks are comparatively unaltered and seem, indeed, to have undergone little change from the condition in which they existed when the early Cambrian sandstones were deposited, while the areas to the east of them have been subjected to enormous pressures, with resulting dynamic metamorphism produced at some post-Cambrian date. The north-west area in Sutherland and Ross has long been known for its clear presentation of three distinct rock -systems, though the precise ages of these three systems was for a long time in doubt. Each is separated from that below by a great unconformity, and the highest of them is now known to be Cambrian. The other two have been, named the Torridonian or Torridon sandstone, and lowest of all the Hebridean or Lewisian gneiss. The following description of the Hebridean gneiss is taken from the Report of the geological mapping of the area, published in 1888. 2 In the cliffs between Cape Wrath and Loch Torridon they present two types of structure, appearing in some parts as " massive rudely foliated crystalline rocks, with few divisional planes, or [in other 38 STRATIGRAPHICAL GEOLOGY parts] as well-banded gneisses in which the constituents have a distinct parallel arrangement. Both varieties are traversed by segregation veins and pegmatites. The prominent minerals are plagioclase felspar, pyroxene (augite), hornblende, quartz (frequently opalescent), and magnetite. It is worthy of note that mica is a rare constituent of these original types of gneiss. On close examination it is apparent that the bands present certain lithological varieties of variable thickness ; some consisting mainly of pyroxene or hornblende and a small quantity of plagioclase felspar; some of plagioclase, pyroxene or hornblende, and opalescent quartz ; others of opalescent quartz and felspar. These varieties frequently cross the lines of schistosity, and are evidently due to differences in the nature of the materials prior to the development of the foliation." In some places these foliated rocks include masses of non-foliated basic igneous rock (gabbros, peridotites, augite -granulites, and diorites), arranged in lenticular belts which are more or less parallel to the foliation. These patches and belts are intersected by veins of grey pegmatite, consisting mainly of felspar and quartz, and varying in thickness from a few inches to several yards. But when these non-foliated masses are carefully traced it is found that they pass laterally into the rudely foliated basic gneisses and that the pegmatites merge into quartz-felspar schists. Hence it is concluded that almost the entire mass consisted originally of igneous rocks, and that the gneisses have been developed by deformation alone ; but that there was an original differentiation of the felspathic and hornblendic elements, either by separate intrusion or by fluxion, while the rocks were still in a viscous condition. There is no doubt, however, that the members of the original complex must have been subsequently affected by intense pressure inducing plastic deformation and metamorphism. The gneisses are also traversed by other dykes of basic igneous rock diabase, epidiorite, and picrite but these are nevertheless of a date anterior to that of the overlying Torridon sandstone which they do not enter. Most of these dykes run from S.E. to N.W. or W.N.W. Moreover, detailed mapping of the area has shown that after the intrusion of the igneous dykes, and still before the deposition of the Torridonian, the whole region was subjected to intense lateral pressure which caused much plication of the gneisses and produced bands of disruption or "shear-zones" which run approximately east and west. In certain parts of the area, i.e. near Gairloch and Loch Maree, rocks of a different character make their appearance and seem to lie within a broad and deep synclinal fold of the Hebridean gneiss. 3 They consist of schists which are clearly altered sediments, but their THE ARCHAEAN ROCKS 39 relation to the gneiss is uncertain because the planes of contact seem to be shear-planes with bands of crushed rock, but they are clearly older than the Torridon sandstone. This schistose series consists for the most part of mica-schist, graphite-schist, and crystalline limestone, with bands that may represent altered quartzite and chert ; and there are also broad out- crops of hornblende-schist which are probably dykes or sills of basic igneous rock. There are several bands of limestone, varying in thickness from 3 to 30 feet ; one of the best known is the Letterewe limestone, which is a cream-coloured rock containing lumps of actinolite ; it was formerly extensively quarried. Another band at Ardlair is a white saccharoid dolomite. The existence of this tract of metamorphosed sediments to the west of the great thrust-plane is important, because it may be regarded as a remnant of a mass of such rocks which was originally much more extensive, and seems to occupy an intermediate position between the Hebridean gneiss and the Torridon sandstone. We shall presently see that the more, highly metamorphosed complex of rocks to the eastward includes representatives of a similar schistose series, and it has been reasonably supposed that the Loch Maree Series represents some of the sedimentary rocks into which the Hebridean gneisses were intruded. Hebridean gneisses are found on the western border of Inverness, opposite Skye, and in the peninsula of Sleat ; they form the islands of Coll and Tiree and occur in lona and Soa ; they are also exposed in the north-east part of Colonsay and again in the south- western part of Islay which terminates in the precipices of Khinns Point. The gneisses of Islay are similar to those of the north-west, but Dr. Teall observes 4 that under the microscope they show evidence of profound modification by pressure, " the general result of which has been to crush the original constituents and thus to produce microscopic breccias, not holocrystalline schists." The features of the belt of faulted and overthrust masses of rock which separates the western and eastern areas of the Northern Highlands are sufficiently illustrated in Figs. 2 and 3, drawn by the Officers of the Geological Survey, and the details of its structure need not be described in this place, It is only necessary to note that the rocks which occupy the country to the east of the great Moine thrust have been so crushed, crumpled, and metamorphosed that they now have special aspects of their own. Part of the great median area of schistose rocks is shown on the map (Fig. 1) ; it comprises the greater parts of Sutherland, Boss, Cromarty, and Inverness ; it extends across the " great glen " into Nairn and Elgin, and southwards through the Forest of Athol w <=> 03 "p a? 5 THE ARCHAEAN ROCKS 41 to and beyond Loch Rannoch, as well as to the eastern end of Locli Leven, but south of this line it becomes infolded with the Grampian Series. The whole of this area has now been surveyed by the staff of the Geological Survey of Scotland, and some important conclusions have been reached. These have been cautiously announced from time to time in the annual " Summaries of Progress," from which extracts and abstracts will here be given. In 1904 and 1905 it was found that over a large area embracing parts of Ross, Inverness, and Nairn the metamorphosed rocks could be separated into two divisions : (1) a complex group consisting partly of gneisses of Hebridean type and partly of altered sedimentary rocks ; (2) certain gneissic schists of both siliceous and pelitic types which, being intimately connected, were termed the Moine schists. Further, the mapping of these areas showed that the Hebridean gneisses occurred as inliers surrounded by Moine schists, and that the latter appeared to lie unconformably on all the members of group 1 ; the lowest part of the granulitic schists being a band containing fragments of felspathic gneiss and having all the aspect of a sheared conglomerate. These conclusions have been confirmed by subsequent work, and the altered sediments in the Hebridean complex are found to include crystalline marbles, calcareous silicates, and brown mica- schists which are often graphitic, associated with banded hornblendic gneisses. The reader will perceive that there is a striking resemblance between this series and that of the Loch Maree Beds above described. Dr. Flett, writing in 1905, says "such rocks are a characteristic part of the Lewisian in certain areas " (i.e. Glenelg and Loch Maree) and remarks that they differ widely from any of the Moine Schist Series. It is evident that two sedimentary series are present in the Moine schist area, the one associated with the Hebridean, and of considerable thickness, but largely eroded and removed before the deposition of the other (i.e. the Moine schists). The Moine schists present two different types of rock which can be separately mapped; these are (1) psammitic schists or flaggy granulites composed of quartz, felspar, and biotite, (2) pelitic or argillaceous schists now converted into mica-schists in which white muscovite mica predominates. The mapping of Ross and Inverness has led to the belief that these are not merely successive but that there are two bands of psammitic and two of pelitic schists ; there are also occasional seams of quartz-schist and of calcareous schist. The whole mass of the Moine schists has apparently been formed out of a series of sandstones and sandy shales. Proof of the last statement has recently been obtained by Messrs. 42 STRATIGRAPHICAL GEOLOGY dough and Cramp ton 5 from a study of the aureole or zone of contact-metamorphism round the gneissic granite of Cam Chuin- neag in Eoss-shire. Within this zone the Moine sediment was altered and indurated before the incidence of the dynamic meta- morphism, the consequence being that the beds have to a large extent escaped the effects of that metamorphism and can be recog- nised as hornfelses formed from the alteration of ordinary sandstone and shales The hornfelses occur where a sandy shale was in contact with the granite ; it is now a hard and splintery rock, and where least affected by subsequent shearing it shows both a clastic structure and the original lamination. The more quartzose bands often contain minute pebbles of rounded quartz ; and " On the surfaces of certain layers there are markings strongly suggestive of sun-cracks and ripple-marks, and others which may possibly represent worm- tubes." Moreover, the beds lie with gentle inclinations, and they dip at right angles to the general direction of the foliation-bedding or shearing of the neighbouring schists. Lastly, intermediate stages between the typical hornfels and the foliated schists have been observed ; and it is clear that all the rocks were solid at the time when the pressure-foliation was induced, the granite being then converted into a gneiss and the shales into mica-schists, while the rocks of the indurated aureole were more resistant and are con- sequently much less altered. The Moine schists and the underlying Hebrideaii rocks have now a common foliation which has obscured that which previously existed in the older series : they have also a common flexion and plication, the rock-masses being folded into a complex system of isoclines, the prevalent strike of which is from S.S.W. to N.N.E., though they vary in direction round the inliers of Hebridean gneiss. The above description applies more particularly to the areas north and west of the Great Caledonian Glen. East of the faults which run along this glen the country is occupied by large tracts of similar quartzo-felspathic schists, which, however, differ in being more persistently flaggy, the layers being alternately more and less micaceous. Hence the name Eilde Flags has recently been used as a special designation of this portion or fades of the Moine schists. 6 We now pass to the equally large area of the Central Highlands, which extends across the centre of Scotland from parts of Aberdeen and Kincardine on the N.E. through Perthshire and Argyllshire to the coasts of Cantire and Islay. This area is occupied mainly by three different kinds of rocks, first and principally by a series of metamorphosed sediments comprising quartzites, phyllites, and limestones, secondly of small tracts of Eilde flags, and thirdly of THE ARCHAEAN ROCKS 43 large intrusive masses or bosses of granite. The stratified series has been called the Grampian Series by Dr. Hicks 7 (1883) and the Dalradian by Sir A. Geikie 8 (1891). The mapping of this extensive area by the Geological Survey has been in progress for many years, and reports on successive portions have been published in the annual "Summaries of Progress." From these we learn that the successive belts of altered stratified rocks are not difficult to follow, but that, owing to the isoclinal folding and the intense plication, it is not easy to unravel the true succession from oldest to newest. In Argyllshire, however, where the metamorphism is less and the physical structure less complicated, the succession appears to be as follows, according to E. B. Bailey, 6 though it is not yet known whether it should be read upwards or downwards. 9. Eilde flags. 8. Glen Coe quartzite (white and fine-grained). 7. Leven schists, grey phyllites, and others with quartzite bands. 6. Ballachulish limestone and black slates ( Easdale slates). 5. Striped Transition Series (not always present). 4. Appin and Loch Awe quartzite (a pebbly rock). 3. Appin limestone (cream coloured). 2. Appin phyllites, with bands of flaggy quartzite. 1. Cuil Bay slates (black). The section (Fig. 4) represents Mr. Bailey's theory of the arrange- ment of these groups in a series of recumbent folds, and from it the reader will understand the difficulty of ascertaining the real order of succession. With regard to the total thickness of these groups no exact measurements have been made, but Mr. Bailey states that "even the thinnest of them, the Appin limestone, is probably not less than 100 feet thick, while the Leven schists, Glen Coe quartzite, and Eilde flags must each of them reach about 1000 feet ; the other members of the sequence should doubtless be reckoned in hundreds of feet. These rough estimates refer to original thickness of deposition. Now over wide areas various groups are reduced to mere films, . . . elsewhere again their dimensions have been mightily increased by reduplication." We may infer, therefore, that the whole series, exclusive of the Eilde flags, must have been originally between 4000 and 5000 feet thick at least. The Appin quartzite is specially interesting because what may be its base is generally a pebble-bed or conglomerate ; moreover, the Appin limestone sometimes passes through a pebbly calcareous rock into the quartzite, so that the evidence of superposition is doubtful. The pebbles are chiefly of quartz and pink felspar. In other parts of the Highlands what seems to be the same quartzite 44 STRATIGKAPHICAL GEOLOGY has a border or base containing stones of the size of boulders as well as pebbles. The above may be an ascending or descending sequence. Hitherto it has been generally supposed that the rocks of the Central Highlands are younger than those to the north-west of them, but this is an assumption, and the intermediate position of the Moine schists between the unaltered Hebrideaii area and the Grampian range may be merely a geographical and structural accident. It does not follow that they are of intermediate age, and we shall see in the sequel that the general order of succession in the metamorphic area may have to be read from south-east to north-west. We must now return to the north-western area and give some FlG. 4. SECTION FROM LOCH LINKHE TO LOCH ETIVE. Copied from Mr. Bailey. Distance about 14 miles. The numbers refer to those in the succession given on p. 43. account of the Torridon sandstone, which in that area is the only formation occupying an intermediate position between the Hebrideaii complex and the Cambrian quartzites. There is, however, a remark- able gap and unconformity between the gneisses and the Torridon sandstones, and this gap must represent a long interval of time. In some places the surface of the "gneiss is a gently undulating plain, but elsewhere it had been carved into a series of hills and valleys, the former rising to heights of from a few hundred to more than 2000 feet above the valley-bottoms. 9 Moreover, the basement beds of the Torridonian are generally in the form of a conglomerate or breccia, the components of which have been mainly derived from the underlying gneisses, but except in this conglomerate fragments of gneiss are by no means abundan t in the overlying coarse sandstones, and most of the pebbles found in them consist of rocks which differ from any now exposed in Sutherland or Koss. They include many kinds of rocks, both igneous and sedimentary, such as spherulitic felsite, felspar-porphyry THE ARCHAEAN ROCKS 45 vein-quartz, quartzite, grit, chert, and jasper ; and Dr. Teall has found from microscopical examination that the felsites and por- phyries are "identical in all essential respects with the felsites [i.e. de vitrified rhyolites] belonging to the Uriconian Series of Shropshire." In these pebbles, therefore, we seem to have traces of a vanished formation, of later date than the Hebridean gneiss, from which the component materials of the Torridonian deposits were principally derived. It would be interesting to compare the volcanic rocks with the felsites and porphyries which occur in Sweden and seem to occupy an intermediate position between the equivalents of the Hebridean and Torridonian Series. In the extreme north between Cape Wrath and Loch Inchard, the Torridonian is not more than 1200 to 1500 feet thick, but farther south the beds not only thicken but another set of beds come in at the base, till round Loch Maree and Loch Torridon its thickness amounts to more than 8000 feet. The whole formation has been divided into three groups by Messrs. Peach and Home, as shown in the following table. Groups. Nature of Materials. Thickness. 3. Aultbea Group 2. Applecross Group 1. Diabeg Group Chocolate-coloured sandstones and flags with grey flags and partings of shale Coarse arkose with pebbles of quartz, quartzite, etc. Red sandstones and mudstones with grey grits and black shales and calcareous bands 2000 to 4500 5000 to 7000 500 in Gairloch 4400 in Kishorn Still farther south in Skye Mr. Clough finds that the lowest or Diabeg Group expands in a remarkable way, and that it there attains a thickness of about 7000 feet, so that although in that island the Aultbea Group is not represented there are 12,000 feet of Torridonian deposits, and may have been as much as 16,000 feet. 10 Mr. Clough has made the following subdivisions of the Torridonian of Skye in descending order : e. Applecross Group. Red and chocolate arkoses with pebbles of various rocks ; estimated at 5000 feet. d. Kinloch Beds. Alternating beds of dark-grey sandy shale and fine grey or buif grit ; about 3500 feet. c. Seamraig Beds. Greenish-grey grits with some beds of sandy shale ; 2600 feet &. Loch na Dal Beds. Sandy shales and fine grits ; 600 feet. a. Epidotic grits and conglomerates. About 300 feet. 46 STRATIGRAPHICAL GEOLOGY The four lower divisions represent an expansion of the Diabeg Group, and it will be noticed that they include two important developments of shaly deposits. From the facts recorded in the Geological Survey Memoir it may be inferred that the Torridoniaii Series originally presented a succession of overlapping beds, the lower members thinning out northwards beneath the higher beds as indicated in the diagram, Fig. 5. In other words we may sup- pose that deposition began in the south and extended northwards as land in that direction was gradually submerged. Small patches of Torridonian sandstone occur in the island of Lewis, and show that the series had a western extension ; outliers of it have also been detected beneath the Cambrian rocks of East Sutherland, so that it must originally have occupied a very large area, and probably extended beyond the shores of Scotland both CAP W. ROSS SKYE Pig. ;3. DIAGRAM OF THE ORIGINAL SUCCESSION OF TORRIDONIAN DEPOSITS. 3. Aultbea Group. 2. Applecross Group. 1. Diabeg Group. on the east and on the west. It was, however, to a large extent destroyed and broken up into isolated tracts during the time which elapsed before the deposition of the lowest Cambrian. The most southerly occurrence of the Torridon Series in Scotland is in the Isle of Islay, where there is an interesting and suggestive collocation of Archaean rocks. 11 Hebridean and Torridonian are only found in the western part of the island, while the eastern and larger part is occupied by two groups, quartzites and schists, one of which is a continuation of the Grampian Series. The two tracts are separated by a thrust-plane which runs from Loch Gruinard 011 the north, through Loch Skerrols and near Bridgend, to the east side of Laggan Bay 011 the south. The Torridonian sequence resembles that of Skye, having an epidotic grit and conglomerate at the base, overlain by a thick group of grey slates and grits, which are succeeded by the Bowmore grits, a series of red, green, and grey felspathic grits or arkoses like those of the Applecross Group. Owing to intense plication no estimates of thickness can be made. Moreover, the whole series has been affected by dynamic metamorphism and converted into slates, ^phyllites, and more or less schistose grits. THE ARCHAEAN ROCKS 47 The Grampian Series of Eastern Islay comprises the following members in descending order : 4. Islay limestone ) = Ballachulish limestone and slates. 3. Black slates and phyllitesj 2. Phyllites with thin quartzose bands ( = Striped Series). 1. Quartz-schists ( = Appin quartzite). This series is overlain unconforrnably by quartzites and shales which greatly resemble those of the Cambrian in Skye, but as no fossils other than worm-tracks and annelid -burrows have been found, the correlation is not proved. These newer quartzites have a con- glomerate at the base which contains pebbles of all the older rocks in the island, and includes many of the Islay limestone. In Islay, therefore, not only is the Grampian Series brought into apposition with Hebridean and Torridonian rocks, but there is evidence which amounts to a strong presumption that it is of pre-Cambrian age. It is, moreover, a noteworthy fact that nothing comparable to the Moine Series has been found in the island. The Geological Survey has not yet ventured to record any definite conclusion with respect to the relative ages of the Moine schists, the Torridonian, and the Grampian Series. In the absence of any official declaration it is only possible to indicate the trend of the evidence, so far as it has been published in the Reports and Memoirs of the Survey. Reviewing this evidence one cannot fail to be struck with the fact that on the western side of the great thrust-planes, where the succession is clear and post-Cambrian disturbance is small, only two Archaean Series are found, i.e. the Hebridean and the Torridonian. On the other hand, in the much larger area to the eastward there are at least three such series, Hebridean, Moine, and Grampian ; further, it seems very probable that both of the two last are newer than the Hebridean, and also that both are pre-Cambrian, but the relative age of the Moine schists is still a matter of doubt. Two alternatives present themselves, either (1) the Moine Series may be a separate intermediate system, or (2) they may be merely sheared and metamorphosed Torridonian deposits of the Skye facies. If they are to be regarded as a separate Mesarchsean System (either with or without the Grampian Series), two awkward questions have to be answered, (a) what has become of the Torridonian strata east of the Moine thrust ? and (&) why is there nothing in the western, except the Torridonian, that is comparable to the Moine schists ? The evidence recently disclosed enables us to assume that the Moine schists were originally a series of 48 STRATIGRAPHICAL GEOLOGY felspatliic sandstones and sandy shales ; moreover there is reason to believe that they are divisible into four zones, the first and third being purely arenaceous, the second and fourth pelitic or argillaceous. Again, there is evidence that where they are in contact with the Hebridean there is an unconformity between the two. Thus there must have been a striking resemblance between the Torridonian and the Moine schists in their original condition. The idea that the Moine schists and flags are only Torridoniaii in an altered form is by no means new, for it was suggested as long ago as 1899 by Sir A. Geikie in the following passage : 12 " Probably the Torridon sandstone was largely drawn upon in the process of the manufacture of these schists, and there may also have been Cambrian or even later sediments which underwent the same conversion into a foliated crystalline condition." The progress of investigation has tended to eliminate the latter possibility, but to confirm the first supposition. Thus Messrs. Home and Teall, writing of the eastern schists in the memoir on the N.W. Highlands, pointed out that in several cases tracts of Torridon sandstone below and near the thrust-planes show a tendency to break down into schistose grits, and again that the Moine schists which immediately overlie the Moine thrust are less completely crystalline than those lying farther to the east, similar cataclastic structures being not uncommon in them, so that "the question has arisen as to whether they represent crystalline schists more or less broken down or sedimentary rocks which are on the way, so to speak, to become Moine schist " (op. cit. p. 600). On the other hand, Dr. Home and Professor Gregory hold that the Moine schists are pre-Grampian, because in several localities the Grampian Series rests directly upon them. Some of these cases are explicable by overthrusts, but in Glen Tilt the Blair Athol lime- stone lies unconformably on an eroded surface of Moine schist. These geologists consequently believe in the existence of four Archaean Systems in the following ascending order (1) Hebridean, (2) Caledonian (i.e. Moine schists), (3) Dalradian ( = Grampian), (4) Torridonian. 2. Ireland The Archaean rocks of Ireland deserve a brief notice because they are for the most part a continuation of the Scottish Grampian Series. There are two large areas in Ireland where such rocks come to the surface from beneath the Palaeozoic formations. One of these areas includes the greater part of Donegal and the western part of Derry with a portion of Tyrone ; the other comprises the THE ARCHAEAN ROCKS 49 western parts of Mayo and Galway, and the two areas are almost connected by a narrow tract of the same rocks, which runs from south-west to north-east for a distance of 60 miles, extending from near Castlebar through the Ox Mountains to Hamilton in Leitrim. So far as yet known fundamental gneiss is only exposed in Tyrone, the gneisses of the western areas being all intrusive gneissic granites. There is no great fault or thrust-plane separating unaltered Hebridean from later schists, as in the north-west of Scotland ; or if such a plane exists it must lie outside the north- west coast of Donegal. Neither is there anything to correspond with the Torrid onian or with the Moine schists of Scotland. (1) Fundamental Gneiss. In Tyrone this forms a long narrow tract of hilly country extending north-east from near Carrickmore to Slieve Gallion, about 18 miles, and having a maximum width of 3 or 4 miles. Sir A. Geikie has described it as " an undoubted core of Archaean gneiss, resembling in all essential characters the Lewisian rocks of the north-west of Scotland." 13 More recently its position has been discussed by Professor G. A. J. Cole, who also regards it as the oldest local rock and mentions the occurrence of a much later intrusive granite, which includes fragments of the gneiss and of the schists which flank the ridge. 14 (2) Donegal Schists. The gneiss of Tyrone is surrounded by a group of coarse volcanic agglomerates tuffs and lavas which pass up into green chloritic schists and silvery mica-schists like those of the Grampian Series. The lavas are chiefly basic and are often plainly amygdaloid al and on the north-west side of the ridge are singularly fresh and unaltered, a condition which Sir A. Geikie attributes to the protection afforded them by the ridge during the time of great pressure from the south-east. The tuffs have suffered much more from deformation and are now silky green chloritic schists, but they seem originally to have been derived from the same magma as the lavas (op. cit. p. 42). Farther to the north-west, in Donegal and in Londonderry, these volcanic rocks are succeeded by quartzites and black schists with bands of crystalline limestone. A similar assemblage of sedimentary and igneous rocks is found in Galway, Mayo, and the Ox Mountains. These of Slieve Gamph and the Ox Mountains have been recently described by Mr. M'Henry, who considers the descending succession to be as follows : 15 4. Quartzite passing down into a sheared conglomerate which is often a "Boulder-bed." 3. Limestone, often altered to whitish marble or green ophicalcite. 2. Black shales and graphitic grits. 1. Pebbly grits with bands of black shale or schist. 50 STRATIGEAPHICAL GEOLOGY This series has been extensively invaded by igneous rocks, both basic and acid, the dykes and veins of the latter cutting through the former which are consequently the older, though all are compressed and sheared by still later movements of the crust. The " Boulder-bed " has been identified in varying thickness all over Galway, Mayo, and Donegal. The boulders in it are almost all of an unfoliated granite, unlike anything now visible at the surface. They are mostly angular or sub-angular, only a few being rounded, and are sparely distributed in a fine-grained matrix. They vary in size up to 3j feet across, and Mr. M'Henry has suggested that the bed is a genuine boulder-clay of glacial origin, but its constant superposition on a limestone over such a large region makes this rather unlikely. Besides the dykes and veins of igneous rock there are large tracts of more or less foliated granite which have often the appearance of being interbedded with the schists, but they sometimes pass into banded gneiss, which is thus shown to be a composite rock formed by the intrusion of tongues of granite into the schists, the banding being due partly to the incorporation of the latter and partly to a flow structure developed in the granite itself, while the whole mass has been sharply flexured, compressed, and sheared by the subsequent pressures. Describing the granitic dome of Ardara in West Donegal, Professor Cole 16 points out that the pure granite of the central mass has a border zone of banded gneiss, which is of composite origin and not an ' original gneiss nor a mylonised rock. " The phenomena," he says, " are those of intrusion along the pre-existing foliation-planes of a schist, and a good gneissose rock results, set with inclusion - flecks, in which these foliation-planes are still apparent." Whatever may be the age of these intrusive gneissic granites, that of the schistose series seems now to be practically settled by the discovery of Arenig rocks in Mayo and their superposition on the schists south of Killary Harbour. This disposes of the old view that the schists were only metamorphosed Ordovician sediments. 3. Wales Archaean rocks come to the surface both in North and South Wales ; they occupy large parts of Anglesey and smaller areas in Carnarvonshire, while in Pembroke round St. David's they are known as the Pebidian Series (see map, Fig. 16). Anglesey. Much has been written about the Archaean rocks of Anglesey, and many conflicting views have been published. A good description of the rocks themselves was given by Ramsay in THE ARCH^IAN ROCKS 51 his memoir on the "Geology of North Wales"; the plutonic complex of the central area was described by Dr. C. Callaway 17 in 1902 ; Sir A. Geikie noted the resemblance between some of the schists and those of Scotland ; the north-western part of the island was mapped by Dr. C. Matley, and the Ordovician tracts separated from the Archaean schists in 1899. 18 Finally, during the last fifteen years the whole island has been re-surveyed by Mr. E. Greenly, whose map and memoir will shortly be published by the Geological Surveys. Meantime I am indebted to Mr. Greenly for the follow- ing condensed account of the pre-Cambrian rocks. Though no fossiliferous Cambrian rocks have been found in the island the relation of the schistose complex to the Ordovician, and of the Ordovician to Cambrian on the mainland, leaves no doubt of the pre-Cambrian age of that complex. Rocks of Archaean age reach the surface in five principal areas and in several smaller inlying tracts. They consist for the most part of greenish schistose grits, alternating with gritty phyllites and fine chloritic mica-schists; there are also other mica-schists, about the origin of which some difference of opinion still exists. There are, moreover, many bands of quartzite and of limestone, some being of considerable thickness. Across the central part of the island, ranging south-west and north-east, is an axis-like tract of highly crystalline rocks, including various types of basic and acid rocks which have a deep-seated aspect, i.e. gneisses, hornfelses, diorite, and granite. The relations of this group to the sedimentary series have been much obscured by subsequent movements, but some of the granites are certainly intrusive into the adjacent rocks which are now in the condition of crystalline hornfels. Interbedded with the sedimentary rocks in many parts of the island are bands of basic igneous rock. Most of these are now chloritic and hornblendic schists, with (locally) large masses of glaucophane schist. Some of the chloritic schists were probably volcanic tuffs, but here and there, especially in the south-west near Newborough, the original structure is beautifully preserved, and the lavas are of that pillowy or ellipsoidal form and variolitic type with which radiolarian cherts are so often associated. No unaltered cherts have been found, but nodules and lumps of red jasper occur abundantly in these lavas, filling the interspaces between the ellipsoids. Similar jasper also occurs in the form of nodules and thin beds in many of the limestones, and seams of fine jaspery purple phyllite are also of frequent occurrence. From this mode of occur- rence these jaspers have been regard epl by some writers as altered radio- larian cherts, but their radiolarian origin has not yet been proved. 52 STRATIGRAPHICAL GEOLOGY The whole Archasan complex of Anglesey was affected by very powerful pre-Ordovician earth - movements. In some parts they are closely folded and minutely corrugated, in other places so sheared and torn that the harder and more resistant beds have been broken up and are now merely a series of phacoids or lenti- cular masses, floating as it were in a kind of matrix formed out of the softer beds. Further, the movements induced varying degrees of metamorphism, so that there are few of the rocks which have not been converted into more or less crystalline schists. Under these circumstances and in the absence of fossils it has not been found possible to establish anything like an order of succession ; at the same time it is nearly certain that more than one of the Archaean systems is present in the island. There is even some reason to think that the central complex of gneissic rocks is a fragment of a massif older than the great sedimentary series, but its outer members are so welded to that series by a common foliation that direct evidence can hardly be expected. The general sequence of events seems to have been as follows : Upon an ancient floor, of which hardly anything is known, but which was composed in part at any rate of plutonic rocks, a great series of grits and shales, with quartzites and limestones, was laid down. At some period during the deposition of these sediments a volcanic episode occurred, and basic lavas with ellipsoidal structure were poured out on the sea floor, after which the interstices of the lava-flows were infiltrated with silica in the form of chert. After the accumulation of all these sediments and volcanic materials, but before the beginning of Ordovician time, crust move- ments on the most powerful scale supervened, causing the whole complex to be crushed, folded, and sheared in a most complicated manner. The rocks were in fact practically reduced to the con- dition in which they are now found before the Deposition of the oldest local members of the Ordovician Series, and probably before the commencement of Cambrian time. Carnarvonshire. Rocks similar to the sedimentary series of Anglesey are found again on the western side of the Lleyn pro- montory and form a long strip from Bardsey Island to Nevin, with a width of from 1 to 2 miles. This district exhibits the same series of schists and slates, of purple and green colours, with broken bands of limestone and quartzite, and a similar association of basic lavas and tuffs with jaspery infiltrations. According to Mr. J. F. Blake there is rather a larger development of volcanic material than in Anglesey, especially in the south-west where there is much volcanic ash and some masses of coarse agglomerate. Two other small strips of pre-Cambrian rocks come to the THE ARCHAEAN ROCKS 53 surface farther north, one extending from Llanllyfin on the south to Penrhyn on the north, a distance of about 15 miles, with a maximum width of 2 miles (see map, Fig. 19). This tract consists of a rhyolitic felsite succeeded by breccias and schists. A second tract extends from Carnarvon to Bangor, a distance of about 10 miles, with a width of over a mile in the central part. This latter has been well described by Professors Bonney and Hughes, 19 who have shown that there are three sets of rocks in the ridge, (1) a granitoid rock at the southern end (Twt Hill), (2) a reddish - purple quartz - felsite, (3) a series of felspathic grits, slates, and agglomerates (the " altered Cambrian " of the Geological Survey). These last are overlain by other conglomerates of Lower Cambrian age. With respect to the relations of these rocks the following facts appear to be well established : (1) That the quartz -felsite was a rhyolite, and probably a rhyolitic lava-flow. (2) That between this rhyolite and the basal Cambrian con- glomerate there is a series of volcanic and sedimentary deposits which are not less than 3000 feet thick. (3) That the Cambrian conglomerate contains many fragments of the quartz -felsite and some of granitoid rock (consisting of quartz and felspar) like that of Twt Hill. This set of pre-Cambrian rocks has been called the Bangor Series, and it should be noted that this series differs from that found in Anglesey and the Lleyn in several particulars. The volcanic rock is acid not basic, all the rocks are less altered and less flexured, and they directly underlie the base of the Cambrian System. Pembrokeshire. The oldest rocks in South Wales crop out in the neighbourhood of St. David's (see map, Fig. 16). A pre- Cambrian age was first claimed for them by Dr. H. Hicks in 1871, and the results of his examination of the district were published in several papers. 20 His conclusions may be thus summarised : that the base of the Cambrian System is marked by a conglomerate containing pebbles of the underlying rocks ; that this conglomerate rests on a series of schistose volcanic tuffs and breccias, under which is another set of quartz-felsites and halleflintas. Below these again comes a granitoid rock which he regarded as a still older mass of metamorphic rock. To this last he gave the name of Dimetian ; the overlying volcanic series he called Pebidian (see Fig. 6). In 1883, however, these conclusions were called in question by Sir A. Geikie, who maintained the correctness of the Geological Survey mapping, holding that the granite is an intrusive mass of igneous rock, and that the volcanic series (Pebidian) is a down- I o I II On'Q - THE ARCHAEAN ROCKS 55 ward continuation of the Lower Cambrians, the apparent break between them being only a surface of contemporaneous erosion. 21 Still more recently (1908) the district has been studied and mapped by Mr. J. F. N. Green, 22 who appears to have finally established the relative positions and ages of the rocks concerned. He confirms the existence of a basal Cambrian conglomerate and of an extensive pre-Cambrian Series, which is the Pebidian of Hicks. These Pebidian rocks occupy an area of about 7 miles in length, with a maximum width of 2 miles ; and according to Mr. Green they consist almost entirely of trachytic and felsitic tuffs, more basic in the lower part, more acid and felspathic in the higher part, the total thickness being over 3000 feet. A well-marked band of conglomerate occurs in the middle of the series and lenticles of shale are seen at several horizons. There do not seem to be any interbedded contemporaneous lava-flows, but rounded pebbles of rhyolite, trachyte, and andesite occur in the tuffs. The materials of this series are evidently the products of a large volcano or group of volcanoes spread out under water, but whether on the floor of lake or sea cannot be determined. At some later epoch, but still in pre-Cambrian time, the series was invaded by a large intrusive mass of granite and by a number of quartz-porphyry dykes which probably proceed from the deeper parts of the granitic We arrive, therefore, at the conclusion that Hicks was correct in his reading of the main stratigraphical sequence, but wrong about the age of the granite ; while Sir A. Geikie rightly interpreted the relations of the granite, but was mistaken about the base of the Cambrian and in denying the existence of Archaean rocks in Pembrokeshire. 4. England Eocks of Archaean age rise to the surface in the following localities : (1) Shropshire (Longmynd, Wrekin, etc.), (2) Warwickshire (Nuneaton), (3) Leicestershire (Charnwood Forest), (4) The Malvern Hills, (5) Devon (Start Point area), (6) Cornwall (Lizard area). Shropshire. The largest and most complete area in England is that of the Longmynd, Caradoc, and Wrekin districts. These tracts are shown in Fig. 21. That of the Longmynd has a length of about 1 miles along its western border, with a further outcrop on Haughmond Hill ; its greatest breadth is about 6 miles, and it is bounded on each side by lines of fault, and inside the eastern fault another narrow strip is exposed to the south of the main outcrop. The exact age of the Longmynd rocks cannot be determined by 56 STRATIGRAPHICAL GEOLOGY superposition, for their base is not seen and the oldest rocks which overlie them are Silurian. Outside this central area are small tracts of other rocks which are demonstrably pre-Cambrian, but they are chiefly of volcanic origin while those of the Longmynd are sedimentary. Lastly, there is a small faulted inlier of micaceous schist near Rush ton which may be older than any of the others. The volcanic series is known as the Uriconian, from the hill called the Wrekin and the Roman town of Uriconium at its foot. The mass of the Wrekin ridge is made up of a succession of rhyolitic lavas with interbedded bands of ash, grit, conglomerate, and agglomerate. Midway along the ridge there is an intrusive mass of dolerite, and at each end (i.e. on Ercal Hill and Primrose Hill) are patches of granitoid rock which may be intrusive or may belong to an older series. This uncertainty is due to the fact that the junctions appear to be faults, 23 but pebbles in a conglomerate near the Wrekin prove that the rocks of the surrounding country at the time of its formation consisted principally of (1) a coarse granitoid gneiss like that of the Wrekin, (2) crystalline schists, both quartzose and micaceous, and (3) grits which have clearly been part of a sedimentary series. Rocks similar to those of the Wrekin occur in the Wrockwardine Hills to the north-west, as well as in the Caradoc and Cardington Hills near Church Stretton, and this volcanic series everywhere strikes nearly east and west, while the Longmyndian and Cambrian rocks strike north-east and south-west (see map, Fig. 21). On the western side of the Longmynd, Pontesford Hill is largely made up of rhyolitic lavas like those of the Wrekin, with intrusive dolerites. The Longmyndian or sedimentary series of the Longmynd is divisible into an eastern and western portion. The eastern part appears to be the older, and if there is no reduplication of beds the succession is as follows (see Fig. 7), in descending order : 5. Greenish slates weathering purple, with some grits. 4. Hard greywacke (indurated sandy beds). 3. Purple slates. 2. Banded slates and greywackes. 1. Soft dark shales. The apparent thickness of these rocks is over 15,000 feet, and they dip regularly to the north-west, the annelid markings and rain-spots on their upper surfaces proving this to be the true dip, and that we follow an upward succession in going from east to west. No definite fossils have been obtained except some im- pressions, which may be those of a Lingulella. Passing westward this series is succeeded by another set of beds, consisting (1) of reddish felspathic grits with a thick band of con- THE ARCHAEAN EOCKS 57 glomerate in the middle, (2) hard purple slates, and (3) purple and reddish-brown grits with lenticular beds of conglomerate. It is uncertain whether this is a true succession, for it is just possible that the double set of reddish grits and conglomerates is a reduplication. The relation of the western series to the eastern series is also a point of dispute. Professor Blake believed the former to be uncon- formable to the latter and to be of the same age as some of the grits in the Uriconian Group. In his view the eastern or grey series is the oldest rock -group in the district and the Uriconian is newer. In favour of this view it may be remarked that the basal Cambrian sandstones have only been found in contact with the Uriconian and not upon any part of the Longmyndian. On the other hand, numerous fragments of the Uriconian rhyolites occur not only in the conglomerates of the western series, but also occasionally as small fragments in the grits of the eastern series. The evidence of actual unconformity is considered very dubious by Dr. Callaway, who also points out that the strike of the two Longmyndian Series is parallel (from N.N.E. to S.S.W.), while that of the Uriconian is quite different (from east to west). 24 The balance of evidence seems to be in favour of Dr. Callaway's view, in which case the order of succession would be (1) Uriconian, (2) Eastern Longmynd Series, (3) Western Longmynd Series. In this connection it may be mentioned that this last (the red series) has a close lithological resemblance to the Torridon sandstones, and it will be remembered that the conglomerates of the latter contain pebbles of a Uriconian rhyolitic type. This tends to confirm Dr. Callaway's opinion and suggests a comparison between the Longmyndian as a whole and the Torridonian of Scotland. Malvern Hills. The oldest rocks of the Malvern Hills are also pre-Cambrian, and are interesting because the greater part consists of highly crystalline gneissic rocks, which are probably altered igneous rocks, and are comparable to the Hebridean gneisses of Scotland and Ireland. This group is known as the Malvernian. Side by side with these is a small tract of rocks comparable with the Uriconian of Shropshire, and consisting of rhyolites, andesites, basalts, and felspathic tuffs. 25 There is nothing here comparable to the Longmyndian Series, but the district has undergone an enormous amount of compression with plication, shearing, and faulting, as will be seen from the section, Fig. 8. Charnwood Forest. Small tracts of volcanic tuffs and breccias, resembling those of the Uriconian, occur at Barnt Green near Birmingham and at Caldecote near Nuneaton, and are in both cases covered unconformably by Cambrian strata. A larger tract 58 STRATIGRAPHICAL GEOLOGY of similar rocks forms the upland district of Charnwood Forest in Leicestershire. The " Charnian Series " has been described by Messrs. Bonney and E. Hill, 26 and more recently by Professor Watts, 27 from whose memoirs the following account is taken. The exposures (partly concealed by Trias) extend over an area about 8 miles long by 5 in breadth, and the strike of the rocks is from north-west to south- east, the general structure being that of a broad anticline broken by faults and flexures. The apparent thickness of the beds exposed, as estimated from a section drawn by Professor Watts, is about 3500 feet, but neither base nor top is seen. The oldest rocks (Blackbrook Beds) are hard greenish slates and hornstones, which seem originally to have consisted largely of volcanic dust; with these are some fine gritty beds containing visible quartz-grains, but they are nevertheless hard flinty rocks. The overlying (Maplewell) group consists of coarse volcanic E.SO'S. TL WLa WS wU s MS 2 Malv Fig. 8. SECTION ACROSS THE HEREFORDSHIRE BEACON. By Professor T. T. Groom (Quart. Journ. Geol. Soc. Ivi. p. 141). Tr. = Trias. Malv. = Mai vernian. F. F. = Faults. LL to MS 2 = Silurian Rocks. Ur. = Uriconian . breccias and grits with tufts and ashes, some of the latter being of such fine grain that they form hornstones and slates, either green or purple in colour. The highest group (Brand Beds) has a conglomerate at the base which contains rounded pebbles of felsite, quartz, quartzite, and slate ; this passes up into brown and purple quartzites with some interstratified sandy shales, in which what seem to be worm-casts have been found. These beds are succeeded by green and purple slates (Swithland slates). In the north-west corner of the area, forming Bardon Hill and Peldar Tor, there are irregular masses of a porphyritic andesite and dacite, associated with coarse agglomerates. These seem to indicate the site or proximity of a volcanic vent. In the southern part of the area there are large intrusive masses of augite-syenite, and on the east is the hornblendic granite of Mount Sorrel, a rock which is largely quarried for road-metal. Devon and Cornwall. The only other exposures of Archaean rocks in England are the two small areas of the Bolt and Start THE ARCHAEAN ROCKS 59 district in Devon and the Lizard promontory in Cornwall, and in neither case can the Archaean age of the rocks be proved because the adjacent strata belong to the Devonian System. The area in Devon is that part of the county which lies inside the headlands of Start Point, Prawle Point, and Bolt Head. It has a length of about 9 miles with an extreme width of 3 from north to south. The rocks of which it consists are divisible into two groups, a lower set of mica and quartz schists and an upper one of green hornblende-epidote schists, known as the Green schists, and obviously altered basic igneous rocks. The general structure of the eastern half of the area is that of an irregular anticline, bent over and inverted on the southern side. The northern border is believed to be a fault, and there is everywhere a rapid if not immediate change from foliated rocks to Devonian slates which are comparatively unaltered, except by cleavage and dislocations. Some geologists have maintained that the schists are only highly altered Devonian slates, and have appealed to the facts, that there is a progressive increase in the intensity of the cleavage in these slates south of Brixham ; that there are appearances of a passage from slates to schists in some places along the boundary line ; and that the general strike of both series is the same, i.e. from west to east. It must be remembered, however, that the dominant flexuring of this part of England dates from late Carboniferous time (Armorican folding), while the appearance of a passage may be caused by the occurrence of crushed and broken schists between two lines of fault. If this area is altered Devonian the metamorphism can only be due to the influence of some local cause, such as the existence of a subterranean mass of granite, but the balance of evidence seems to be in favour of their Archaean age. 28 C. ARCHAEAN EOCKS IN EUROPE Outcrops of Archaean rocks, or what are generally regarded as Archaean rocks, occur in France, Spain, Switzerland, Austria, Bohemia, Saxony, the north of Russia, Finland, and Scandinavia. Only the more important of these areas can here be described. 1. France In France the chief exposures are found in the north-west (Brittany and Normandy) and in the Central Plateau. Brittany. Pre- Cambrian rocks, with associated masses of 130 STRATIGRAPHICAL GEOLOGY granite, occupy large parts of Brittany, together with the western portions of Normandy, Maine, and Anjou. They have been well described by Professor C. Karrois, from whose memoirs the following account has been compiled. 29 He and other French geologists are inclined to divide them into two main series which they call (1) Archaean and (2) Brioverian, but as they again subdivide the lower series into three portions, and as no break has been observed between any of the divisions, they can be equally well arranged in three series, thus : 3. Phyllades des St. Lo ( = Brioverian). 2. The mica-schists of Groix. 1. The schists of Audierne and gneiss of Quimperle. The oldest group is a complex of which a granitoid rock some- times seems to be the oldest member, but more often the lowest rocks are mica-schists which include bands of ribboned gneiss, white leptynite, amphibolite, and eclogite. In the south these rocks form a continuous tract from 10 to 16 miles in breadth extending from the Bay of Audierne on the west to the valley of the Loire, a distance of about 140 miles. In the north there are two bands of similar rocks, one passing from Brest to Leon with probably a submarine continuation to Guernsey, the other near St. Malo. Professor Bonney informs me that he has not seen anything in Western or Northern Brittany, nor in the Cotentin, which is quite comparable with the Hebridean and Malvernian gneiss, and he is inclined to doubt whether such gneiss exists in the region. The schists of Groix seem to be an upward continuation of the series, being truly crystalline schists of micaceous, chloritic, graphitic, and sillimanitic varieties, with subordinate bands of quartzite (sericitic and graphitic), hornstone, and pyroxenic marble (cipolin). These resemble the Loch Maree Beds. The Brioverian Series consists of much less altered rocks of a typical Eparchsean character. They exhibit three different facies in different parts of the region : (a) the facies of St. Lo in Normandy and of Central Brittany, comprising the phyllades of St. Lo and Lamballe, the phyllades of Gourin, which include a band of limestone, and the green flags of Neant at the top; (6) the Tre"gorrois facies in the north of Brittany, consisting of phyllades and quartzose rocks with interstratified eruptives (porphyrites and diabases) ; (c) lastly, the southern facies, which consists mainly of shales with interbedded bands of coarse felspathic sandstone, recalling the Torridonian 6f Scotland and suggesting that the Brioverian land lay to the south of Brittany. These Brioverian deposits are unconformably overlain by the THE ARCHAEAN ROCK 61 purple conglomerate which forms the base of the Cambrian System, and no fossils have been found in them except some microscopic bodies which resemble Radiolaria and Foraminifera. The Central Plateau. This is a large region extending from La Chatre and Moulins on the north (Departments of the Indre and Allier) to the Cevennes in the south, a distance of about 220 miles, while, between Chalus and Brive on the west to the valley of the Rhone on the east, its breadth is over 160 miles. This great area consists of a complex of granites, granulites, gneisses of several kinds, micaceous schists, and phyllites, but to what extent these rocks are of pre-Cambrian age is at present uncertain, because the Palasozoic rocks by which they are doubtless surrounded are almost everywhere concealed and overstepped by Mesozoic strata. It is only in the southern border (in the Tarn and Herault Departments) that rocks of Palaeozoic age are found in contact with them. The granites are doubtless all intrusive rocks and the granitoid gneisses may be foliated portions of the same intrusions, like those of Donegal ; while the more highly crystalline portions of the complex are probably those which are most closely associated with these intrusions, being merely the more altered portions of the mica- schist series into which the igneous rocks were intruded. No break or clear plane of separation has been detected in any part of the schistose series, but in several districts along or near the borders of the region there are tracts of still less altered and obviously sedimentary rocks, including arkose sandstones, quartzites, slates, and phyllites comparable with the Brioverian. A long strip of such rocks occurs between Brive and Tulle in the Correze, where Messrs, de Launay and Mouret have described the following descending succession : 30 Dark-grey slates and argiilites. Green slates and phyllites. Greenish sericitic and micaceous quartzites. Similarly in the Cevennes on the outer borders of the gneissic rocks there is a considerable thickness of satiny schists or phyllites (schistes luisantes). In the district of Montagne Noir these schists include bands of limestone and appear to pass up into a set of shales and limestones which contain Cambrian fossils. Hence M. Bergeron regards the whole mass of the Montagne Noir as Cambrian, 31 repeated in many parallel flexures and becoming more and more metamorphosed toward the central axis of the mass till the rocks pass into mica-schists and gneisses. He states that the limestones resist alteration longer than the shales, but gradually 62 STRATIGRAPHICAL GEOLOGY become greenish from the development of augite or hornblende crystals, which increase in size till the rock passes into amphibolite or pyroxemite. By others, however, some of the phyllites are considered to be of Brioverian age. 2. Austria and Switzerland A long tract of crystalline rocks, some of which are probably Archaean, extends from the Graian and Cottian Alps on the borders of France and Italy through the southern part of Switzerland (Pennine, Lepontine, and Rhsetian Alps), and is continued eastward through the Tyrol into Carinthia and Styria. According to von Hauer's map of the Eastern Alps, the funda- mental rock of that region is a group of banded hornblendic and micaceous gneisses, which he terms " the Central Gneiss " ; outside this is a series of micaceous schists, followed by a formation called the Thonschiefer, i.e. silvery or sheeny schists. The gneisses of the " Central " group are only occasionally exposed along the central axis of the chain. Those of the Velber-Tauern Pass and the neighbouring Gross Venediger are thus described by Professor Bonney : 32 " These, whether 'micaceous or hornblendic, recalled to my mind certain members, but not the most coarsely crystalline, of the Hebridean of Scotland and the Laurentian of Canada, which are often distinctly modified by subsequent pressure. We also find slightly foliated, coarsely crystalline, granitoid rock with nodes or enclosures, probably a granite subsequently modified, the upper part of the mountain chiefly consisting of the latter rock." The Gross Venediger rises to a height of 12,000 feet. Over]ying these gneisses and apparently passing down into them is a series of fine-grained gneisses and mica-schists, often distinctly banded with white (quartzose) and dark (biotitic) layers, and this bedding-foliation is often modified by a subsequent cleavage- foliation. With these rocks are associated bands of porphyritic gneiss which Professor Bonney regards as igneous rocks intruded into them before the production of the earlier foliation. The Thonschiefer Series is a well-defined group of rocks which extends throughout the Eastern Alps from the Swiss border nearly as far as Gratz. It consists mainly of leaden-coloured micaceous schists which have glistening sheeny surfaces, but it includes bands of a more quartzose schist with some of calc-schist and crystalline limestone. Again, in the Western Alps of France, Italy, and Switzerland, Professors Bonney, Gastaldi, and others have recognised a similar succession of (1) coarse banded gneisses apparently succeeded by THE ARCILEAN ROCKS 63 (2) finer gneisses and mica-schists, often traversed by dykes of serpentine and epidiorite, and finally a series (3) of micaceous, calcareous, and chloritic schists, between which and the older schists there is a marked lithological difference and contrast. Recently, however, grave doubts have been thrown upon the view that these are all Archaean rocks, upon the age of the gneiss by Professor J. W. Gregory, 33 and upon that of the schistes lustres by Kilian and Parona. Professor Gregory admits the apparent sequence, but fails to find anything like a continuous band of " central gneiss " as represented on certain maps ; only independent isolated masses of gneiss which appear to be intrusive granites of very recent date. Monsieur Parona, again, has discovered layers of chert containing Radiolaria in the silvery schists, and the generic assemblage suggests a Carboniferous or even Mesozoic age. Professor Gregory has specially studied the gneisses of the Cottian Alps and finds that there is clear evidence of contact- metamorphism both in the gneisses and the adjoining schists ; that the gneisses invade different parts of the schist series with trans- gressive junctions ; that they include large blocks of the schists and serpentines ; that the schists are occasionally traversed by dykes of aplite which appear to be apophyses of the gneissic granite, while the latter is not itself cut by any of the later igneous rocks. He concludes that the gneiss is really a granite of Pliocene date and that the silvery schists are mainly Palaeozoic and probably of Carboniferous age. Thus if these views are correct the only Archaean rocks in the Cottian, and indeed in all the "Western and Italian Alps, are the series of mica-schists and calc-schists into which the Tertiary granite has been injected. 3. Bohemia Another large area of Archaean rocks is exposed in Bohemia and Saxony, extending from the valley of the Danube on the south through the Bohemian Highlands ^Bohmer Wald) and the Erzge- birge, and as far as Chemnitz and Dresden on the north, while eastward it ranges into Moravia. In the northern part of this region there is a basin of Palaeozoic strata extending from Pilsen to Prague, and on the border of this the highest Archaean rocks can be seen to pass below the Cambrian strata. The Archaean rocks of this region seem to resemble those of the Eastern Alps, and a similar threefold division of them has been made, those supposed to be the oldest consisting of gneisses much interpenetrated by granites. In apparent succession to these are mica-schists and phyllites, with intercalations of amphibolites, 64 STRATIGRAPHICAL GEOLOGY serpentines, eclogites, and crystalline limestones, evidently a series of metamorphosed stratified rocks. Finally on the borders of the Palaeozoic basin these schists graduate upward into dark-grey clay- slates (Thonschiefer) with some more siliceous layers of lydite or hornstone. These are the Przibram shales which formed the base of Barrande's " Stage A" but as they are overlain unconformably by Cambrian conglomerates they are now separated from the rest of that stage and classed as Archaean. As in the case of the Alpine Series much doubt has recently been thrown on the old interpretation of the stratigraphy of this area. Some of the gneisses, as well as the granites, are now regarded as intrusive rocks, and no break or definite unconformity has been detected in the series, so that the whole of it may belong to the later portion of Archaean time ; at any rate nothing demonstrably C. <& PM. X. K.M.&a. GL Gr.G&l. FIG. 9. SECTION THROUGH PART OF THE BOHMER WALD (Gumbel). C. Cambrian. K. Cryst. limestone. Gl. Mica-schist. Gyl. Gneissic granite. Ph. Phyllite. Q. Quartzitic schist. Gr. Granite dyke. Gn. Gneiss. Protarchaean and comparable to our Hebridean has yet been recognised. 4. Scandinavia The most extensive exposures of the Archaean platform in Europe are to be found in Norway, Sweden, and Finland. The whole surface area of Finland, except where covered by drift and alluvium, consists of a complex of Archaean rocks with many intrusive masses of granite and porphyry. The greater part also of the Scandinavian peninsula is occupied by a similar complex, which, however, passes beneath a broad, central, and axial band of Cambrian, Ordovician, and Silurian deposits. The natural eastern border of this band of Palaeozoic sediments is obscured by very large tracts or sheets of overthrust and meta- morphosed rocks which, according to Scandinavian geologists, include representatives both of the later Archaean and of the Palaeozoic Systems. There is consequently a close correspondence THE ARCH^AN ROCKS 65 between the general structure of Scandinavia and that of Scotland, but the overthrust sheets in the former region are on a much larger scale, and appear to include a larger number of formations. More- over there is this great difference between the two regions, that the forces producing the thrusts have acted from opposite directions T i.e. from the east in Scotland and from the west in Scandinavia ; so that in the latter case the unaltered rocks are found in. the areas- which lie to the east of the overthrust tracts, i.e. in Sweden and Finmark. The large areas occupied by Archaean rocks in the Fenno- Scandian region are partly shown in Fig. 10, which, however, does not include the more eastern districts. The Swedish and Finnish geologists 3 * now recognise no less than four series or systems of pre- Cambrian rocks : (1) a fundamental complex which is comparable with the Hebridean of Scotland (2) the Kalevian, which is a series of altered sediments and volcanic rocks ; (3) the Jatulian (princi- pally quartzites) ; (4) the Jotnian and Sparagmite sandstones, which seem to correspond with the Torridonian of Scotland. The fundamental rocks are gneisses of several kinds, some being granitic gneisses which seem to be deformed and foliated granites ; another type is the garnetiferous gneiss which is regarded as a combination, of stratified gneiss (paragneiss) and gneissic granite (orthogneiss) ; while there are also banded grey gneisses (coarse and fine) and some red granitoid gneiss. Above these gneisses, and apparently in stratigraphical sequence, is a sedimentary series which in Sweden is called the " Porphyry- Leptite " Group. This consists of mica-schists, green schists, leptites ( = altered felspathic tuffs), schistose porphyrites (lavas), crystalline limestones, and iron ores ; these rocks occur in regular bedded succession and alternation, but the bedding-planes are nearly always vertical, and the rocks themselves are crystalline. The series seems to be comparable with that of Loch Maree and Glenelg, but is on a much more extensive scale. In parts of Lapland, Finland, and Olonetz (Russia), Dr. Seder- holm has separated another series of metamorphosed sediments under the name of Kalevian (see Fig. 10), which he regards as newer than the Dalecarlian porphyrites and yet clearly older than the Jatulian. This series consists mainly of quartzites which are often micaceous, and pass into muscovite -schists ; they also include bands of conglomerate and occasionally of talc-schist and dolomite. There are also intercalated sheets of metabasite (i.e. epidiorite). This series has not yet been identified in Sweden, but Dr. Sederholm thinks it is included in the greatly folded and plicated complex of Northern Sweden, while in the south (Smaland) there F Fig. 10. -^GEOLOGICAL SKETCH-MAP OF FENNO-SCANDIA, Showing the areas of Archaean and Palaeozoic rocks. (Based on a map by A. E. Tornebohm.) THE ARCHAEAN ROCKS 67 are quartzitic schists on the border of a large area of granite which much resembles the post-Kalevian granites of Finland. Here it may be remarked that all the areas occupied by these older Archaean rocks are invaded by large tracts of granite, most of which are older than the succeeding Jatulian and Jotnian Series. The typical Jatulian is found in the east of Finland and in Olonetz. It rests with marked unconformity on the Kalevian, though itself consisting principally of white quartzites with bands of epidiorite. At or near the base are conglomerates containing rolled pebbles of granite, gneiss, and older vitreous quartzites. The Jatulian quartzites often show ripple-marks and desiccation-cracks. Above them are dolomitic limestones and shales (or slates), and in several . places beds of anthracitic coal occur. The total thickness is about 2000 feet. Whether there are any rocks of Jatulian age in Sweden is a matter of doubt. Sederholm regards the Dal formation, which occupies an area to the west of Lake Venern, as Jatulian, but his only reason is that the beds are strongly flexured and affected by pressure-metamorphism. Dr. Tornebohm dissents from this view and maintains that the Dalsland Series is Jotnian, the petrological resemblance being very close and the succession very clear, as given below : Feet. 4. Schistose grits, grading into mica-schists .... 1300 3. White quartzite, with some beds of red slate .... 1500 2. Dark slates with intercalations of earthy limestone and red sandstone, and two or three sheets of schistose greenstone ( = altered basic igneous rock) ...... 2600 1. Reddish quartzitic sandstones, with beds of arkose and con- glomerate at the base 650 About 6000 The typical Jotnian is a great red sandstone formation which has a great development in Central Sweden and Norway, extending in a nearly continuous band from the area near Lake Miosen to the extreme north-eastern end of Finmark, a distance of about 860 miles. It may be regarded as including the Sparagmite sand- stone of Norway, though, where the two come together, the latter overlies the Jotnian with a certain amount of unconformity or contemporaneous erosion, but the time value of the break does not seem to be greater than that between the English Keuper and Bunter, so that they form parts of one and the same system. Each part of the series is over 2000 feet thick. To the east of the overthrust areas the Jotnian sandstones are quite unaltered and lie in nearly horizontal sheets. At the base 68 STRATIGEAPHICAL GEOLOGY there is a conglomerate, the pebbles in which are chiefly of quartz- porphyry and orange-coloured quartzite. The mass of the formation consists of brownish-red or chocolate-coloured felspathic sandstones, in which false-bedding, ripple-marks, and desiccation-cracks are frequently observable. In the higher part thin layers of red shale occur, and there are also intercalated sheets of diabasic lava, some being intrusive sills and others extruded lava-flows which are sometimes amygdaloidal. In Scandinavia these Jotnian and Sparagmite sandstones rest nearly everywhere on the rocks of the Protarchsean complex, and the unconformity between them is as striking and profound as that between the Torridonian and Hebridean of Scotland. In the latter country, as stated on p. 39, there is evidence of an inter- mediate formation having existed. In Scandinavia remnants of this (the Jatulian) still exist, but if it ever extended far toward the north it was destroyed before the formation of the Jotnian. In the western metamorphosed tract the place of the Jotnian sandstones seems to be taken by the Are schists ; these are mainly quartzose flags and gneissic schists, including white mica-schists and brown garnetiferous schists. e Hogbom remarks that petro- grapbically and technically the Are schists very much resemble the Moine schists of Scotland. Both he and Tornebohm believe that they have originated from the Sparagmite sandstones, the latter becoming "transformed to mylonites, sericite schists, and white mica-schists within the western thrust areas and the crystal- Unity increasing from east to west." There is, however, a greater petrological difference between them than can be accounted for in this way, and there " must also have existed a primary difference between the eastern sparagmites and the western schists, a change of facies characterised partly by the entrance of basic igneous rocks into the western series, partly by the substitution of argillaceous and slaty sediments for the gritty sparagmites and conglomerates. The basic rocks just mentioned, which probably are to a great extent tufaceous (i.e. altered tuffs), repeatedly alternate with these argillaceous beds. Generally speaking, one can say that this change of facies goes on slowly from the east or south-east to the west or north-west." REFERENCES 1 Bonney in Building of the British Isles, by Jukes -Browne. Third edition, p. 21. (Stanford, 1911.) 2 " Report on Geology of North- West Highlands," Quart. Journ. Geol. Soc. vol. xliv. p. 378 (1888). 3 See "The Geological Structure of the North- West Highlands of Scotland," Mem. Geol. Survey, 1907, p. 192, etc. THE ARCHAEAN ROCKS 69 4 " Geology of the Isle of Islay," Mem. Geol. Survey (1907), p. 15. 5 "The Augen-gneiss and Moine Sediments of Ross-shire," by C. T. Clough, C. B. Crampton, and J. S. Flett, Geol. Mag. (1910), p. 337. 6 E. B. Bailey in Quart. Journ. Geol. Soc. vol. Ixvi. p. 599. 7 Hicks in Proc. Geol. Assoc. vol. viii. p. 270. 8 Sir A. Geikie, Pres. Address Geol. Soc. (1891), p. 75. 9 "Geological Structure of North-West Highlands" (supra cit.), p. 275. 10 "Geological Structure of North- West Highlands" (supra cit.), p. 348. 11 "Geology of the Isle of Islay" (supra. cit.) t p. 19. 12 Sir A. Geikie in Sum. Prog. Geol. Surcey for 1899, p. 23. 13 Sir A. Geikie in Pres. Address Geol. Soc. (1891), p. 41. 14 G. A. J. Cole in Trans. Roy. Irish Acad. vol. xxxi. p. 431 (1900). 15 M'Henry in Proc. Roy. Irish Acad. vol. xxiv. p. 371 (1903). 16 Cole, "On Composite Gneisses in Boylagh," Proc. Roy. Irish Acad. vol. xxiv. p. 203 (1903). 17 Callaway, Quart. Journ. Geol. Soc. vol. Iviii. p. 662 (1902). 18 Matley, Quart. Journ. Geol. Soc. vol. Ix. p. 635 (1899). 19 See Hughes, Quart. Journ. Geol. Soc. vol. xxxiv. p. 137 (1878) ; and Bonney, op. cit. vol. xxxv. p. 309, vol. xxxix. p. 478 (with map). 20 Hicks in Quart. Journ. Geol. Soc. vols. xxxii., xxxiv., and xxxv. 21 Sir A. Geikie, Quart. Journ. Geol. Soc. vol. xxxix. p. 261 (1883). 22 E. J. N. Green, Quart. Journ. Geol. Soc. vol. Ixiv. p. 363 (1908). 23 Callaway in Quart. Journ. Geol. Soc. vol. xxxv. p. 643 (1879). 24 Callaway in Quart. Journ. Geol. Soc. vol. xlvii. p. 109 (1891). 23 Consult Callaway, Quart. Journ. Geol. Soc. vol. xxxvi. p. 536 (1880) and vol. li. p. 1 (1895) ; Acland, vol. liv. p. 556 (1898) ; and Groom, vol. Iv. p. 129 (1899). 26 Bonney and Hill in Quart. Journ. Geol. Soc. vols. xxxiii., xxxiv., and xxxvi. (1877-1879). 27 W. W. Watts in Geol. Mag. for 1 896, p. 485, and Proc. Geol. Assoc. vol. xvii. p. 375. 28 Bonney, Quart. Journ. Geol. Soc. vol. xl. p. 1 (1884) ; Miss C. A. Raisin, op. cit. vol. xliii. p. 715 ; and W. A. E. Ussher, Mem. Geol. Survey (Kingsbridge, 1904). 29 See papers on " Central Brittany," Proc. Geol. Assoc. vol. xvi. p. 101 (1900), and "Excursion en Bretagne," Livret-Guide de Congres Geol. Internal, for 1900. t 30 See Bulletins de la Carte Geol. de France for 1889 and 1890. 31 Bergeron, "Etude Geol. du S. du Plateau Central," and Livret-Guide. du Gong. Geol. Int. for 1900, "Excursion en Montague Noir." 32 Bonney, Quart. Journ. Geol. Soc. vol. xlv. p. 89 (1889). 33 J. W. Gregory, Quart. Journ. Geol. Soc. vol. 1. p. 232 (1894). 34 Consult A. E. Tornebohm, Livret-Guide I. of Internat. Geol. Congress at Stockholm (1910) ; Hdgbom, " Pre- Cambrian Geology of Sweden," Bull. Geol. Inst. Upsala, vol. x. (1910) ; and Sederholm, "Les Roches pre-quater- naires de la Fenno-Scandia," extrait de V Atlas de Finlande, Helsingfors (1910). CHAPTEE V THE CAMBRIAN SYSTEM A. NOMENCLATURE AND CLASSIFICATION THE establishment of three separate systems among the Lower Palaeozoic rocks was not accomplished without much controversy, for in early days only two were recognised, and both were made more comprehensive than they ought to have been. The well-known name Silurian (from the British tribe of Silures who inhabited the western parts of Wales and its borders) was first introduced by Sir Roderick Murchison in 1835 to designate the series of rocks which he had studied in the land of the Silures (Shropshire, Herefordshire, and South Wales) ; he divided this system into an upper and a lower series. 1 a In the meantime, between the years 1831 and 1835, Professor Sedgwick had worked out the complete succession of the Palaeozoic rocks in North Wales, and in 1835 he proposed the name Cambrian for this series, which he divided into three portions, calling them Lower, Middle, and Upper Cambrian. For the beds which overlie this series in the Berwyn Mountains he adopted Murchison's name of Silurian. 2 Sedgwick worked out the succession from below upwards, and Murchison worked from above downwards, and for a long time neither was aware of the fact that their systems overlapped each other. Murchison supposed that Sedgwick's Cambrian lay entirely below his Silurian, but when the fossils were collected and de- scribed, it was found that the Upper Cambrian was equivalent to the Lower Silurian, and this fact led to much subsequent controversy. As Murchison became the Director-General of the Geological Survey, his nomenclature was naturally adopted by the surveyors, and as they found that his Lower Silurian had no natural base, group after group of Sedgwick's Cambrian was gradually absorbed a For references see end of the chapter. 70 THE CAMBRIAN SYSTEM 71 into it, till the Lower Silurian came to include the whole of the rocks (below the Upper Silurian) in which any fossils had been found. This was obviously unfair and unscientific, since the so- called Lower Silurian was in reality a part of Sedgwick's Cambrian Series, the delineation of which by Sedgwick was accurate and complete. Still it was done, and the name Lower Silurian was generally adopted both in England and abroad. The fact is, Sedgwick and Murchison both described more than one system of rocks, and each included two systems under one name. Sedgwick had clearly a right to call one system Cambrian and Murchison to call one Silurian, but there is really a third intermediate system, containing the second great fauna in the life- history of the globe, and to this neither geologist had a distinct claim. For this intermediate system Professor Lapworth proposed the name Ordovician (from the tribe of the Ordovices who inhabited North Wales), remarking, 3 " That so long as present systems of nomenclature survive, nothing can disturb the application of the title Cambrian to the rocks of the ' Primordial Series, 3 and that of Silurian to the strata of the * Third Fauna.' It is vastly different, however, with the intermediate series, which has been the object of incessant disputes from the day it was recognised until now." This proposal has met with very general acceptance, and has recently been adopted by the Geological Survey. The following table shows the original arrangements of Sedgwick and Murchison, that formerly used by the Geological Survey, and lastly the one which is now in general use. Sedgwick, Murchison, Geological Lapworth, 1836. 1839-47. Survey. (now used). Ludlow Beds Wenlock Beds Llandovery Beds - {s& \ fs& }-* Bala Series ] Llandilo Flags 1 Lower !- Ordovician. Llanvirn Series (Silurian Arenig Series Tremadoc Slates Ling u la Flags Menevian j Cambrian . (Primordial j Silurian Lower Silurian ^ I Cambrian. Harlech Series Cambrian Cambrian J Classification of the Cambrian System. Since the time of Sedgwick and Murchison the Cambrian rocks have been investigated much more completely, and fossils have been found at various horizons down to the very base of the system. Fresh subdivisions have been established and new names introduced into the nomenclature. A Menevian Group has been separated from the Lingula flags, and 72 STRATIGRAPHICAL GEOLOGY the Harlech Series was found to be divisible into two groups (the Caerfai and the Solva Beds). It has, moreover, been found that the whole succession is divisible into series which are characterised by special genera of trilobites, and that in many places a further sub- division can be made into zones based on the limitation of certain fossils (see p. 22). Thus the old lithological grouping has been modified to agree with the palaeontological evidence, and the best modern arrangement of the rock-groups that compose the Cambrian system in England is as follows : TJ /Tremadoc Beds =the Bryograptus fauna. \Lingula Flags (in three stages) = the Olenus fauna. Middle [ Sa V B a e ds BedS } = the Pa doxides fauna. Lower = Caerfai Beds = the Olenellus fauna. B. LIFE OF THE PERIOD Great interest attaches to the fossils which have been found in the Cambrian rocks, because, with the exception of the few organic remains mentioned on p. 61, they include the remains of the oldest animals hitherto discovered. At the same time the number of different classes of creatures which exist in the oldest Cambrian fauna make it certain that ancestors of these creatures must have existed in pre-Cambrian times, and there can be little doubt that some of them will sooner or later be discovered in the higher pre-Cambrian rocks. The fossils which have been found in the Lower and Middle Cambrian of Wales and Shropshire are referable to the following groups Porifera, Cystidea, Annelida, Brachiopoda, Gastropoda, and Crustacea ; while representatives of Actinozoa, Crinoidea, Graptolitoida, and Lamellibranchia have also been obtained from rocks of the same age in North America. The very existence of so many different kinds of animals at this early period is a remarkable fact, and is a proof of the great gap which exists between the Cambrian and the pre-Cambrian rocks of Western Europe. ' But although referred to many different classes, examination of the fossils themselves reveals the fact that most of them are repre- sented by lowly and more or less generalised forms that is, forms which combine in themselves characters that are now only found in two or more distinct families or orders of beings. Thus Archceo- cyathus and its allies have been classed by some with the Sponges and by others with Corals, and they are now regarded as a primitive family intermediate between Porifera and Actinozoa, and possibly representing the stock from which both these classes were derived. THE CAMBRIAN SYSTEM 73 The Cystidea again are placed by systematists at the bottom of the Echinoderma, and appear to represent an early stage in the evolution of the Echinoderm race. Two of the Lower Cambrian Brachiopods (Lingulella and Orbiculoidea] are low forms of that class, and the others, Obolella and Kutorgina, were not much more highly organised. No Lamelli- branchia have yet been found in the Lower Cambrian of Europe, but some small fossils found in North America are referred to this class. The Gastropoda are represented by the genus Hyolithes and its allies, which are usually referred to the Pteropods, though their real affinities are very doubtful. More typical Gastropods appear in the Lower Cambrian of America and in the Middle Cambrian of Scotland. Crustacea are represented by the Entomostraca and Trilobita. The former are true Crustacea of low organisation ; the latter exhibit a curious mingling of characters, which have led some authors to exclude them from the Crustacea, but Professor Beecher 4 remarks that Trilobites show the clearest evidence of primitive Crustacean affinities. In the Middle and Upper Cambrian Hydrozoa (Graptolites), Crinoid and Asteroid Echinoderms, Phyllocarid Crustacea, a few Lamellibranchs and Gastropods, with some Nautiloid Cephalo- poda make their appearance. The occurrence of Cephalopoda, the highest of the Molluscan classes, suggests that the evolution and development of marine Mollusca must have been especially rapid during this early period of the earth's history, or else that their ancestors existed somewhere long before Cambrian times. The following are the principal characteristic species of the several faunas which have been recognised in the Cambrian System. Lower Cambrian (Olenellus Fauna] Ostracoda. Leperditia cambrensis. Annelida. Serpulites fistula, Scolithus linearis. Trilobita. Olenellus Callavei, 0. Lapworthi, 0. intermedius, Olenel- loides armatus. Brachiopoda. Lingulella primseva, Orbiculoidea caerfaiensis, Kutorgina cingulata, Obolella Salted, Acrotreta Sabrinse. Gastropoda. Hyolithes princeps, Orthotheca Geeri, Stenotheca rugosa, Coleoloides typicalis, Salterella pulchella, S. Maccullochi. Middle Cambrian (Paradoxides Fauna] Porifera. Protospongia fenestrata (S. and M.) 5 P. major (S.). rt Cystidea. Protocystis menevensis (M.). a S. indicates occurrence in Solva Beds, M. in Menevian. 74 STRATIGRAPHICAL GEOLOGY 4 Fig. 11. GROUP OF LOWER CAMBRIAN FOSSILS. 1. a, b. Obolella Groomi (enlarged). 2. The same (shell highly enlarged). o. Olenellus Callavei (reduced). (I. Kutorgina cingulata (enlarged). Fig. 12. GROUP OF IRISH CAMBRIAN FOSSILS. c. Histioderma hibernicum. e. Arenicolites didyma. a. Oldhamia antiqua. b. Oldhamia radiata. THE CAMBRIAN SYSTEM 75 Trilobita. Paradoxides Harknessi (S.), P. aurora (S.), P. Davidis (M. ), Erinnys venulosa (M.), Conocoryphe Lyelli (S.), C. solvensis (S.), Plutonia Sedgwicki (S.), Solenopleuro variolaris (M.)> Agnostus cambrensis (S.), Microdiscus sculptus (S.), M. punctatus (M.). Brachiopoda. Lingulella ferruginea (S. and M.), Obolella sagittalis, (S. and M.). Orbiculoides pileolus (S. and M.), Orthis Hicksi (M.). Fig. 13. GROUP OF LOWER CAMBRIAN FOSSILS. (. Conocoryphe Lyelli. b. Microdiscus sculptus. c. Paradoxides Davidis. d. Hyolithes corrugatus. e. Lingulella ferruginea. /. Protospongia fenestella. m. Agnostus Cambrensis (nat. size and enlarged). n. Orbiculoidea pileolus (much, enlarged). Gastropoda. Hyolithes antiqua (S. ), H. corrugata (M.), Cyrtotheca hamula (M.). Upper Cambrian (Lingula Flags) Annelida. Cruziana semiplicata, Oldhamia antiqua, 0. radiata, Histioderma hibernica. Trilobita. Olenus cataractes, 0. micrurus, Peltura scarabseoides, Parabolina spinulosa, Conocoryphe bucephala, Agnostus pisiformis. A. trisectus. Phyllocarida. Hymenocaris vermicauda. 76 STRATIGRAPHICAL GEOLOGY Brachiopoda. Lingulella Davisii, Orthis lenticularis (and in Tremadoc). Gastropoda. Bellerophon cambrensis. Upper Cambrian (Tremadoc} Hydrozoa. Dictyograptus socialis ( = Dictyonema), Bryograptus Callavei, B. Kjerulfi. Echinoderma. Dendrocrinus cambrensis, Palseasterina ramseyensis. Trilobita. Asaphellus Homfrayi, Ogygia scutatrix, Niobe Hom- frayi, Angelina Sedgwicki, Sphserophthalmus alatus, Peltura punctata. Brachiopoda. Lingulella lepis, Obolella Belti. 0. sabrinse, Orthis lenticularis, and 0. Carausi, Orthisina festinata. Lamellibranchia. Palsearca Hopkinsoni, Glyptarca primseva, Ctenodonta menapiensis. Gastropoda. Hyolithes cuspidatus, Conularia Homfrayi, Bellero- phon arfonensis. Cephalopoda. Orthoceras sericeum, 0. mendax, Cyrtoceras prsecox. C. BRITISH CAMBRIAN ROCKS Range and Relation to Rocks below. Cambrian rocks rise to the surface in several parts of England and Wales, as well as in Ireland and Scotland, but they nowhere occupy any large tract of country and they exhibit several very different facies, so that it is with some difficulty that the component members of these isolated exposures can be compared with one another. In Wales, the typical area, there are four districts where Cambrian strata emerge from beneath the overlying Ordovician, namely in Pembrokeshire, Merionethshire, Carnarvonshire, and Anglesey. In England there are six outcrops, but all are small ; these are (1) The Malvern Hills, (2) The Wrekin area in Shropshire, (3) The Lickey (N. Worcestershire), (4) near Nuneaton in Warwick- shire, (5) The Lake District, and (6) The Isle of Man. In Ireland rocks which are believed to be of Upper Cambrian age occur in Wicklow and Wexford, but they contain very few organic remains, and none of the Welsh Cambrian fossils have yet been found in them. In Scotland they have only yet been recognised in the extreme north-west (Sutherland, Ross, and Skye), where they consist mainly of limestones, and thus exhibit a very different facies from those occurring in the more southern areas. Wherever the base of the Lower Cambrian Series is exposed, it is seen to rest unconformably upon an eroded and planed down surface of the Archaean rocks. This is conspicuously the case in Scotland and in most of the English localities, but the base is not exposed in Cumberland, the Isle of Man, nor in Ireland. THE CAMBRIAN SYSTEM 77 Fig. 14. GROUP OF LINGULA FLAG FOSSILS. a. Cruziana semiplicate. b. Dictyonema sociale. c. Lingulella Davisi. d. Olenus micrurus. e. Agnostus pisiformis. /. Hymenocaris vermicauda. gf. Sphserophthalmus alatus. h. Orthis lenticularis. m. Conocoryphe bucephala. n. Parabolina spinulosa. 78 STRATIGHAPHICAL GEOLOGY Fig. 15. GROUP OF TREMADOC FOSSILS. a. Asaphus Homfrayi. b. Angelina Sedgwicki. c. Conularia Homfrayi. d. Niobe Homfrayi. e. Bellerophori arfonensis. /. Palaearca Hopkinsoni. g. Orthis Carausi. h. Orthoceras mendax. m. Bryograptus Kjerulfi. /I. Part of same magnified. THE CAMBRIAN SYSTEM 79 1. The Cambrian of Wales Though the exposed tracts of Cambrian rocks in Wales are rather far apart it will be convenient to treat the whole as one area, because the several tracts are doubtless portions of one continuous subterranean mass, and because in process of time the same stages and subdivisions will probably be recognised in all the separate districts. Both in this and subsequent chapters I shall assume that the student possesses a copy of the cheap general Geological Map of the British Islands, published by the Geological Survey (see p. 27), so that he may see at a glance the relative positions of the areas described. The descriptions here given will in some cases be illustrated by special maps, and for other parts of the country the reader can obtain the Index maps of England and Southern Scotland, which are published in sheets (see p. 28). The Cambrian rocks of North Wales were first described and classified by Professor Sedgwick, and the lowest series of beds, in which he found no fossils, were termed by him the Harlech grits. The beds in Pembrokeshire were first properly identified and described by Dr. H. Hicks, 5 who obtained fossils from the equi- valents of the Harlech grits, and was thus led to divide the series into two groups which he named respectively the Caerfai Beds and Solva Beds. He was also the first to separate the Menevian from the Lingula flags, and to make a detailed map of the St. David's area which is reproduced in Fig. 16. We shall now indicate the representatives of the three divisions of the Cambrian (Lower, Middle, and Upper) in each of the separate areas, so far as it is possible to do so. Olenellian or Lower Cambrian. In Pembrokeshire this series, named Caerfai Beds by Dr. Hicks, is estimated to have a thickness of about 1600 feet, and to consist of the following sets of beds in descending order : Feet. Purple sandstones with Annelid tracks ..... 1000 Ked shales with Lingulella primceva, Discina, and fragments of Olenellus ......... 50 Flaggy sandstones with Annelid markings .... 460 Conglomerates . . . . . . . . . 60 to 150 About 1600 The basal bed of this series is the Cambrian conglomerate so frequently mentioned in the preceding chapter. Near St. David's it is about 60 feet thick, and consists of well-rounded pebbles set in a matrix which is a mixture of sandy and ashy materials ; 80 STRATIGRAPHICAL GEOLOGY sometimes the ashy element appears to predominate, sometimes it is a true arkose, consisting of quartz -sand and decomposed felspar. The pebbles vary in size, ranging up to a foot in diameter ; they consist principally of a reddish quartzite and of quartz, but include some of quartz -felsite, porcellanite, schist, and granitoid rock, all derived from the underlying pre-Cambrian rocks. In Merioneth the equivalents of the Caerfai Series are to be found in the central parts of the Harlech dome (see Fig. 19). The GEOLOGICAL MAP OP THE COUNTRY NEAR ST DAVID'S Fig. 16. MAP OF PART OK SOUTH WALES (after Hicks). great Harlech Series, believed to be from 5000 to 6000 feet thick, was not subdivided by Sir A. Ramsay and the officers of the Geological Survey, but quite recently Professor Lapworth and Dr. T. S. Wilson have mapped a large part of the district and have determined the sequence of the beds. Their detailed descriptions have not yet been published, but they have allowed a summary of the succession to be printed in a paper by Mr. A. R. Andrews, 6 and from this the following brief account is gleaned. The lowest beds occupy the coast between Egryn Abbey and Llanfair and extend eastward to the Rhinog Mountains. The red felspathic grits of Dolwen are probably not far from the base of the THE CAMBRIAN SYSTEM 81 series, which, however, is not exposed. Above them are blue and purple slates and grits (the Cefu Slate Group) of which there may be 1000 feet. These are succeeded by the Rhinog grits, which have a thickness of about 2500 feet. These two groups may be regarded as representing the Lower Cambrian. Passing under the great mountain mass of the Snowdoii range, the Harlech Series again emerges and occupies a considerable area in Carnarvonshire from the coast near Clynnog to Penrhyn and Llanllechid on the north-west, lying on both sides of the axial ridge of Archaean rocks. The basal beds are grits and conglomerates, which contain pebbles of granite, quartz-felsite, and porcellanite derived from the pre- Cambrian rocks described in the preceding chapter. These are succeeded by a thick series of purple and green slates, some of which are worked at the celebrated slate quarries of Penrhyn and Llanberis ; above these again are coarse grits and sand- stones. The total thickness on the eastern side of the anticline near Llanberis is estimated at about 3000 feet. The beds exposed in and near the Penrhyn slate quarries were described by Professor T. M'K. Hughes in 1889, from whose paper the following condensed account of the succession is taken : Lingula flags with L. Davisii. Feet. | 6. Coarse grits and sandstones of Bro nil wyd ? 500 -| 5. Purple and bluish-green slates, yielding Conocoryphe viola [ in the upper part 1500 4. Band of fine red grit (15 feet). 3. Purple and veined slates (quarried) ..... 540 2. Green sandy slates 300 1. Grits, slates, and basal conglomerate ? 300 3140 Of these beds 1 to 4 may represent the Caerfai Series, but as no fossils have yet been found in them correlation is un- certain. Paradoxidian or Middle Cambrian. This series is distinguished by the presence of several species of the Trilobite genus Paradoxides. In Pembrokeshire it was divided by Dr. Hicks into two stages, the Solva Group and the Menevian Group, Menevia being the ancient name of St. David's. He also recognised the following subdivisions or zones : 82 STRATIGRAPHICAL GEOLOGY Feet. g ( Sandstones and shales, with Orthis Hicksi .... 100 "> -I Flags and shales, with Paradoxides Davidis .... 350 (Flags and shales, with Paradoxides Hicksi .... 300 jjd f Grey flagstones and grits with Paradoxides aurora . . 150 =8 I Red, purple, and grey sandstones and slates with Paradoxides solvensis 1500 Yellowish sandstones and grey flags with Par adoxides Harknessi 150 2550 I The lowest beds are yellowish sandstones with some pebble beds, and above these are grey flags which have yielded a number of fossils, including Paradoxides Harknessi, Plutonia Sedgwicki, Micro- discus sculptus, Conocoryphe Lyelli, Agnostus cambrensis, and Hyolithes antiqua. The central group is of great thickness and yields fragments of fossils at different horizons, but it was only near the top that Dr. Hicks found Paradoxides solvensis and Conocoryphe solvensis with some other fossils. The upper beds contain other species of Paradoxides and Conocoryphe, while Agnostus cambrensis and Lingulella ferruginea occur in all three groups. The Menevian Group consists mainly of dark-grey flagstones and black shales, and from the middle zone Dr. Hicks obtained 28 species of fossils belonging to the Trilobite genera Paradoxides, Conocoryphe, Microdiscus, Anopolenus, Arionellus, Erinnys, Agnostus, and Holocephalina with Brachiopods of the genera Lingulella, Discina, Obolella, and Orthis ; and some Gasteropod Mollnsca of the genera Hyolithes, Stenotheca, and Cyrtotheca. In Merioneth, according to Messrs. Lap worth and Wilson, the series is represented by the following succession of beds : Feet. CloganBeds ( = Menevian), black shales and slates . . . 250 Gamlan Shales, a series of grey and greenish shales and flags, with beds of grit which thicken eastward 1000 Barmouth Grits, thick felspathic grits with some bauds of pebbles 600 Hafotty or Manganese Shales, grey and green shales and flags with grits which thicken eastward ....... 1000 About 2850 No fossils have yet been found in the great series of greenish- grey shales and grits which seem to be the equivalents of the Solva stage, but the Menevian has yielded 19 species, includ- ing 5 species of Agnostus, 2 of Anopolenus, 3 of Paradoxides with Microdiscus punctatus, Lingulella ferruginea, and Hyolithes corrugata. In Carnarvonshire the Solva Group is evidently represented by the upper green and purple slates of Llanberis and of Penrhyn THE CAMBRIAN SYSTEM 83 <0 ft. a i 3 -2 3. a 5 >> 5 feQC 1 1 03 c I '00 a -- & a 1^51 I frill 84 STRATIGRAPHICAL GEOLOGY in which. Gonocoryphe viola was found (see p. 81), but nothing like the black Menevian slates and mudstones appears to exist in that area. "Whether originally present and now cut out by faults, or whether they were eroded before the deposition of the Lingula flags, is a question for future investigation. Mr. Fearnsides is of opinion that there are local unconformities at the base of the Upper Cambrian, and that this absence of the Menevian in Carnarvonshire is probably due to erosion. Olenian or Upper Cambrian. As already stated this consists of the groups known as the Lingula flags and the Tremadoc Beds, and as in the case of the underlying Middle and Lower Cambrians their thickness is greatest in Merioneth, which seems to have been near the centre of the Welsh basin of deposi- tion in Cambrian time. In Pembrokeshire the total thickness of this division is not half that which it attains in Merioneth, though some of the same zones or subdivisions seem to be represented. The area has not yet been resurveyed by the Geological Survey and Dr. Hicks' account of them is still the best available, but Mr. J. F. K Green considers it probable that there is a break or non-sequence be- tween the Lingula flags and the Tremadoc. 7 The succession is as follows : m f Feet. la ^ I Black flagstones and slates with Neseuretus Ramsey ensis, S Jg | Niobe Homfrayi, and Lamellibranchs .... 1000 f Upper Fine grey slates without fossils .... 300 3 SjJ Middle Hard grey micaceous flags with Lingulella, Davisi in abundance 1000 Lower Grey flagstones and black slates without fossils . 700 These beds appear to form a continuous band along the northern side of the Cambrian area from Whitesand Bay to the country south of Llanvirn and beyond Crugglas (see Fig. 16) ; east of St. David's they are seen in Solva Harbour, and inland they spread eastward round another tract of Lower and Middle Cambrian rocks of which little is yet known. Another small tract of Lower Tremadoc Beds is found round Tremanhire, a place about 4 miles east of St. David's ; they are brought in above the Lingula flags by a synclinal flexure, and have yielded many of the same fossils as in Ramsey Island. Tremadoc slates have also been found to the south of Caermarthen by the Misses Crossfield and Skeat, and their presence there is interesting as it was unsuspected. They yielded Orthoceras sericeum and two new species of Trilobites, Peltura punctata and Ogygia THE CAMBRIAN SYSTEM 85 marginata. The outcrop forms a narrow band between that of the Ordovician and the area of Old Red Sandstone to the south. GEOLOGICAL MAP OF A PART OP NORTH WALES- Carboniferous Silurian Ordovician bvSoi Up. Cambrian !''' " -I Mid & lower-Cambrian . f . r Granite &.Felsite LLUU Archaean Schists Stanford's Gfyyrafih l EstatfJ<>rnim Fig. 19. MAP OF NORTH WALES (from recent sources). Tn Merioneth the Lingtila flags alone attain a thickness of over 6000 feet, while the overlying Tremadoc Beds have a maximum thickness of about 2000 feet. The Lingula flags form a continuous belt all round the 86 STRATIGRAPHICAL GEOLOGY Harlech dome (see Fig. 19). They are well exposed near Dolgelly and Ffestiniog and were studied in 1867 by Mr. Belt, who sub- divided them into three groups, each characterised by a certain assemblage of Trilobites and other fossils. More recently they have been examined by Professor Lapworth and Dr. Wilson (op. cit.\ who appear to accept and confirm Belt's main divisions, which are as follows with thicknesses corrected by the later estimates. Feet. Dolgelly / Soft black shales with Peltura scarabceoides . . 300 Beds ^Hard blue slates with Parabolina spinulosa . . 300 Ffestiniog /Bluish-grey flags with Olenus micrurus ... 50 Beds \Grey micaceous flags with Hymenocaris . . . 3000 Maentwrog /Black slates and flags with Olenus gibbosus . . 1600 Beds ^Dark-grey flags and shales with 0. cataractes . . 1000 About 6250 The lower portion of the Maentwrog Beds has been termed the Vigra Beds by Messrs. Lapworth and Wilson, and the higher 14 18 12 11 fir 10 9 8765 4 3 2 la 1 FIG. 20. SECTION ACROSS THE PENRHYN PROMONTORY (after Salter). 12, 13. Llanvirnian. 2 to 4. Lower Tremadoc. 11. Arenig Grit. 1 and la. Lingula flags. 5 to 10. Upper Treiradoc. g. Intrusive dyke of igneous rock. There is probably a fault below the letter F. portion the Pen Rhos Beds. The former contain beds of fine siliceous grit from 2 to 3 feet thick, while in the latter such grits are only from ^ to 2 inches thick. Two species of Olenus and Agnostus pisiformis are the characteristic fossils. The Ffestiniog Group is a thick mass of flagstones with only a few bands of black shale ; they are usually hard and consequently form higher ground than the underlying shales. The most abund- ant fossil throughout is Lingulella Davisii. The Dolgelly Group consists of soft blue and black shales, which appear to have been formed in much deeper water and though only 600 feet probably represent as long a period of time as the whole of the Ffestiniog Beds. They resemble the Alum shales of Sweden. The best section of the Tremadoc slates in North Wales is in the cliffs of the Penrhyn promontory south-east of Tremadoc in Merioneth. According to Mr. Salter 8 the succession which can here be seen is as follows : THE CAMBRIAN SYSTEM 87 f 12. Black slates with iron stains. Aremg ^ n> Garth Hm grit Upper Tremadoc" 10. Hard bluish flags with many fossils, Angelina SedgivicJci, Asaphellus ffomfrayi, Ogygia scutatrix, etc. 9. Soft sandy shales with Shumardia. 7. Alternating beds of hard bluish slate and black pencil slate. 6, 5. Thick beds of pencil slate, with some hard sandy layers, H Asaphellus Homfrayi. ( 4. Hard bluish slaty rock in massive beds. Lower J 3 ^ 2 . Dark iron-stained slates, with Psilocephalus innotatus, emadoc ^ Mobe Hom f ray ^ Dictyograptus socialis. 1. Lingula flags Black slates with Olenus. Mr. Fearnsides has recently described a similar succession near Tremadoc on the east side of the Lleyn peninsula, where the Upper Tremadoc Beds (440 feet thick) yield species of Asaphellus, Angelina Shumardia, Symphysurus, and Dikellocephalus? The Tremadoc Beds do not form a continuous band round the Lingula flags of Merioneth, for in certain places they seem to disappear beneath the base of the Arenig Series (Ordoviciaii) ; it is now believed that there is an actual unconformity between the two series and that the Arenig often completely oversteps the whole of the Tremadoc Group so as to rest on Lingula flags. In Carnarvonshire the thickness of the Lingula flags is much less, but near Llanberis and Nant Ffrancon they are apparently from 2000 to 2500 feet thick. On each side of Llyn Peris the nearly vertical Harlech grits are succeeded by dark slates which represent the Maentwrog Beds, with a thickness of about 1000 feet, and these are followed at the eastern end of the lake by hard quartzose grits (? Ffestiniog beds), which are estimated by Kamsay to be about 1300 feet thick. 10 These beds are traceable north- ward through Elidyr-fawr, and a good section is exposed in the southern cliffs of Marchlyn-mawr, where Eamsay found Olenus micrurus and Lingulella Davisii in certain grey and brown grits. It is noticeable that some of these grits are of coarse grain, and occasionally conglomeratic, thus differing from the finer-grained flagstones of Merioneth. The Upper Lingula flags or Dolgelly Beds have not yet been identified in Carnarvon owing to the rarity of fossils, but are probably represented by a part of the overlying slates. On the western side of the Llanberis anticline the thickness of the Cambrian Series seems to be still further reduced, and when they again emerge on the eastern flank of the Bangor-Carnarvon ridge there are only slates overlying the basal grits and conglomerates. Professor Hughes n has traced these basal beds along the eastern flank of the pre-Cambrian ridge all the way from Twt Hill near 88 STRATIGEAPHICAL GEOLOGY Carnarvon to Garth Point near Banger. The pebbles vary according to the rocks on which the conglomerate locally rests ; near Carnarvon they are chiefly quartz, near Llanddeiniolen and near Bangor they include pebbles of quartzite, andesite, felsite, and quartz. Sir A. Kamsay was of opinion that the whole of the Lingula flags had thinned out near Bangor, that the grits were of Harlech age and were overlain unconformably by Arenig shales. This, however, is not the generally accepted view, for there is no evidence of such unconformity, and if the diminution of thickness is due to thinning out toward a shore line we should expect the lower beds to thin out and disappear before the upper beds. The facts may be explained as a simple case of overlap, and we may suppose that the whole of the Harlech Series (i.e. Lower and Middle Cambrian) has thinned out with probably some of the Lingula slates (Maentwrog Beds), and that the lowest Cambrians between Llanddeiniolen and Bangor represent the Lingula grits of Elidyr-fawr and Marchlyn- mawr. They may, however, be of still later date. Whether Tremadoc Beds exist to the north-west of Snowdon has not yet been ascertained because of the scarcity of fossils in the mass of dark slates which intervenes between the Lingula grits and the Arenig grits. 2. Midland Counties Cambrian rocks are brought to the surface in four distinct and isolated areas in the Midlands, viz. (1) Malvern Hills, (2) The Wrekin (Shropshire), (3) the Lower Lickey Hill (Worcestershire), and (4) Nuneaton (Warwickshire). In lithological characters, and to some extent also in their fossil contents, these Midland rocks differ very much from the typical Cambrian of Wales, and accord- ing to Professor Lapworth 12 they approximate more closely to the Cambrian of Scotland and of the continent of Europe. The total thickness of the system is very much less in the Midlands than in Wales, and though it is believed that in the western areas the succession is incomplete, it appears to be unbroken near Nuneaton, where the thickness of the exposed beds is only about 2200 feet, and if 800 feet are added for the concealed upper beds, the total is then only 3000. The whole system is divisible into a lower arenaceous division, comparable to the Caerfai and Solva Beds of South Wales, and an upper shaly division, representing the Lingula flags and the Tremadoc Beds. The Arenaceous Division. At the south-west end of the Malvern Hills the Archaean schists are bordered by beds of quartzite THE CAMBKIAN SYSTEM 89 and of greenish flaggy micaceous sandstone which have been called the Hollybush sandstone. In these beds Dr. T. T. Groom has recognised the following descending succession : 13 2. Hollybush Sandstone, consisting of : 6. Massive glauconitic sandstones, from 550 to 700 feet thick, with several species of Hyolithes (including H. fistula), Kutorgina cingulata var Phillipsi, etc. a. Shaly sandstone, with calcareous bands and seams of quartzite, not less than 220 feet, with Kut. Phillipsi, Obolella sp., Linnarssonia sagittalis, Hyolithes sp. 1. Hollybush Quartzite. Quartzites and conglomerates of un- known thickness, with Kutorgina, Obolella, and Hyolithes. The conglomerates in the quartzite contain pebbles of rneta- morphic quartzite, rhyolite, andesite, and red granophyre, and the mass was supposed to rest unconformably on the pre-Cambrian rocks. It is probable that such was their relation, but Professor Groom believes the present junction planes to be faults, and thinks that the quartzite may have been several hundred feet thick, though only 50 or 60 feet of it can now be seen. In Shropshire the Wrekin ridge, as well as the Uriconian rocks of the Caradoc and Cardington Hills, are flanked on their eastern sides by a series of Cambrian rocks, the lowest being a hard quartzite like that of Malvern (see map, Fig. 21, here reproduced by permission of Professor C. Lap worth) ; but its base is seen to rest directly and unconformably on the Archaean rocks, and it contains pebbles of rhyolite, tuff, etc., derived from them. This quartzite is about 300 feet thick, and is succeeded by a series of sandstones, limestones, grits, and shales which are gener- ally known as the Comley sandstone, though they include repre- sentatives of two faunal divisions. This series is supposed to be about 500 feet thick. Its true age was first recognised by Dr. C. Callaway in 1877, and the Olenellus fauna was first discovered in it by Professor C. Lapworth in 1888. Still later exploration by Mr. G. S. Cobbold has resulted in the discovery of more fossils and has shown that the succession of fossiliferous beds at Comley is as follows : 14 Paradoxides flaggy shales and grits. Beds | Coarse calcareous grit with fragments of sandstone and [ 'limestone derived from beds below, f Black limestone with phosphatic nodules (1 foot). Olenellus I Grey limestones with black phosphatic nodules (4 feet). Beds 1 Purplish-red sandy limestone with Olenellus. ^ Greenish felspathic sandstones. ! , 5 ...k ^ sa ^ ^ THE CAMBRIAN SYSTEM 91 The red limestone contains Olenellus Callavei, Microdiscus helena with new species of Ptychoparia and Micmaccia, and Kutorgina cingulata. The grey limestones have yielded new species of Protolemis, Microdiscus, Anomocare, and Mohicana, with some Brachiopoda. The black limestone contains Hyolites, Satterella, and Brachiopoda. The overlying conglomeratic grit with its derived fragments indicates a local break with erosion of the beds below so that it may be inferred that the latter are only a part of the series originally deposited. This upper grit has yielded three species of Paradoxides (P. Groomi, P. rugulosus, and P. cf. Davidis). In the Lickey district there is an exposure of quartzite which is almost certainly of Lower Cambrian age, though no fossils except worm-burrows have yet been found in it. It consists of flaggy Fig. 22. SECTION FROM NUNEATON TO CALDECOTE HILL (LapWOrtll). />. Keuper marls. d 5 . Coal-measures. a*. Oldbury shales. a 3 . Purley shales. 2. Hyolithes Beds. a 1 . Cambrian quartzite. A. Caldecote volcanic rocks. Diorite dykes shown in black. quartzites, with intercalated purple shales like those next de- scribed. Near Nuneaton there is a remarkable tract of Cambrian rocks which before the discovery of fossils was supposed to be of Car- boniferous age. It is about 9 miles in length, with a width of a mile where broadest, and the structure of the district is shown in the section drawn by Professor Lapworth (Fig. 22). The arenaceous division has been called the Hartshill quartzite, and is about 600 feet thick. The lower part consists of alternating beds of siliceous grit and purple or grey shale, the grits consisting of sand set in a chalcedonic cement, and forming a very hard rock which has usually a pale pink colour. No fossils have yet been found in them. The middle portion consists of thick beds of quartzitic grit with very little shale, and has not yielded any fossils. The Upper or Camp Hill Group is a much more varied set of beds, consisting at the base of thin-bedded glauconitic grits, followed by sandy micaceous shales about 50 feet thick, in which are two notable beds, one a pebble bed or conglomerate, and the other a bed of red limestone about 2 feet thick. The highest member is a mass of 92 STKATIGRAPHICAL GEOLOGY purple false- bedded quartzitic sandstone containing many grains of glauconite and about 50 feet thick. The limestone and associated red shales have yielded several species of Hyolithes and of Orthotheca, with Stenotheca rugosa, Coleoloides typicalis, Kutorgina cingulata and Iphidea labradorica, and Orthisina transversa. The facies of the fauna is that of the Lower Cambrian or Olenellus zone, though no trilobites have yet been found. The Shaly Division (Upper Cambrian). In the Malvern area this is over 2000 feet thick, and is divisible into two parts. The lower 800 feet are black shales with bands of basaltic lava which have generally been regarded as contemporaneous lava-flows, but which Dr. Groom finds to be intrusive sills. These shales have yielded Peltura scarabceoides, Gtenopyge bisulcata, Agnostus trisectus, Sphcerophthalmus alatus, etc. species which are character- istic of the Lower Dolgelly Beds of North Wales. The upper part of the Malvern shales is composed of bluish or yellowish -grey shales, with many intrusive sills of basalt and diabase. Dr. Groom 15 found them to be of much greater thickness than was previously supposed, and from his section, reproduced in Fig. 18, it will be seen that they are at least 1300 feet thick. Their highest beds contain Dictyograptus socialis, Agnostus dux, Cheirurus Frederici, Acrotreta Sabrince, with species of Asaphus and Olenus ; most probably they are of Lower Tremadoc age. No representatives of the Menevian, Maentwrog, or Ffestiniog beds have yet been recognised, but the boundary between the Hollybush sandstone and the Malvern shales seems to be a faulted one, so that beds of intermediate age may be faulted out. In Shropshire the Shineton shales occupy a tract two miles broad in the valley of the Severn by Shineton and Cressage, and they extend both to the north-east and south-west (see map, Fig. 21). They also occur at Pedwardine, and again in a long narrow strip on the west side of the Longmyiid, where they underlie the Arenig Beds of the Stiper stones. Their true age was first made known by Dr. Ch. Callaway, 16 who discovered in them fossils of Upper Cambrian types. They are soft, fissile, micaceous shales of a dark - blue colour, weathering to olive - green and yellow. Their base is not seen, and their junction with the Comley sandstone appears to be a fault, so that here, as at Malvern, the central part of the Cambrian System seems to be missing. The lowest beds contain Dictyonema. The highest beds are more fossiliferous and have yielded many trilobites, including Eulonia monile, Olenus Salteri, Asaphellus Homfrayi, A. Crofti, and Shumardia salopiensis, with the Cystidean Macrocystella Marice. THE CAMBRIAN SYSTEM 93 The total thickness is estimated at about 1500 feet and the beds are clearly of Tremadoc age. Passing again to Nuneaton, where the Hartshiil quartzite is succeeded by the Stockingford shales, we have what appears to be a complete succession without break or faulting. Professor Lapworth has divided the shaly series into three groups, of which the following is a brief description, taking them in descending order :- Upper or Merevale Shales. Only seen near Merevale, and consisting of greenish-grey shales in which Dictyograptus is abundant. The thickness exposed may be 200 feet. Middle or Oldbury Shales. Black shales, with some bands of grey shale ; in highest beds are Sphcerophthalmus alatus and Ctenopyge pecten; in the lower Agnostus pisiformis, Lcptoplastus Salteri, Bcyrichia Angelini, and a Lingulella. Thickness probably 800 feet. Loicer or Purley Shales. Reddish-purple mudstones and shales, with some green and grey bands, containing minute Lingulella, Obolella sagittalis, Acrothelc granulata, and a Conocoryphe. Thickness about 600 feet. The fauna of the Purley shales is comparable with that of the Conocoryphe exulans zone of Sweden, which belongs to the Paradoxides division and is homotaxial with our Menevian Series. The black Oldbury shales are comparable with the black shales of Malvern and the Lower Dolgelly Beds. The Merevale shales appear to represent the lowest part of the Shineton shales and of the Malvern grey shale. No horizon comparable with the mass of Lingula flags has been recognised, but this may be due to the scarcity of fossils in the lower part of the Oldbury shales. 3. Lake District and the Isle of Man In the northern part of the Lake District around Skiddaw Mountain, Derwent, and Crummock Waters there is a great thick- ness of dark -coloured slates, with bands of hard grit and occasional beds of conglomerate. To these beds Sedgwick gave the name of Skiddaw slates ; the higher portion of them is of Ordovician (Arenig) age, but the lower part is Cambrian (Tremadoc and Lingula flags). The Cambrian age of this portion was suspected by Sedgwick and maintained by Clifton Ward, but not until 1894 was this view confirmed by good palseontological evidence. 17 Mr. Ward took a certain band of grit occurring north of Skiddaw as the base of the Arenig Series and identified with it certain flaggy grits near Buttermere, but this has not been confirmed and the boundary line between Cambrian and Ordovician has not yet been 94 STKATIGHAPHICAL GEOLOGY determined, the difficulty being that the rocks are so much folded, contorted, and faulted that it is almost impossible to trace particular beds in the mass of slates. Few fossils except graptolites have been found, and these occur in definite bands of a more earthy kind of slate than the rest, so that, as Dr. Marr observes, " whether these beds are of the same age as the black glossy unfossiliferous slates, or whether they are newer beds folded as synclines among these slates, yet remains to be ascertained." The only locality indeed where graptolites of Tremadoc species have yet been found is Earth near Keswick, where the following species were obtained by Miss G. L. Elles 18 in 1898 Bryograptus Kjerulfi (see Fig. 15m), B. ramosus, B. Callavei, Chonograptus tenellus, and Ch. flexilis. The greater part of the Isle of Man is occupied by a similar series of slates and grits, now known as the Manx slates. These have been divided by Mr. Lamplugh into three portions which seem to be consecutive, though no recognisable fossils have been found in any of them, and the crush-phenomena produced by lateral compression are even more strongly developed in that island than in Cumberland. 19 The subdivisions traced by Mr. Lamplugh strike from N.E. to S.W., and the Lonan flags on the eastern side are believed to be the lowest ; these seem originally to have been flagstones, mudstones, and shales, but are now hard, cleaved, crumpled, and frilled rocks. North-west of them is a band of quartz-veined, thin-bedded grits (the Agneash grits), and west of them are the Barrule slates, a mass of dark-blue slaty rocks, which closely resemble the Skiddaw slates. Of the Lonan flags Mr. Lamplugh observes that lithologically they have a close resemblance to the Bray Head Series described below. 4. Ireland So far as present knowledge extends Cambrian rocks only occur in the south-east of Ireland in the counties of Dublin, Wicklow, and Wexford. The facies is one of grits, flagstones, and shales like that of Wales, but the base is not exposed and none of the Welsh fossils have yet been discovered in the Irish area. The rocks are indeed so cleaved and altered by metamorphic agencies that until recently it has been an open question whether they were of Eparchsean or Cambrian age. The determination of their age rests mainly on the discovery of their characteristic fossil Oldhamia (see Fig. 12) in rocks of Cambrian age elsewhere. It is certainly curious that Oldhamia should not have been found in the Cambrian of Wales, but in the THE CAMBRIAN SYSTEM 95 v f) * 84 la Ardennes of North-west France both the Irish species of Oldhamia occur in the slates of Fumay which underlie beds of Tremadoc age. Oldhamia radiata has also been found in. Belgium, and another species (0. occidens) has been obtained from the Upper Cambrian of North America. In further confirmation of this determina- tion of age it has been ascertained that the supposed unconformity between the Arenig Beds and the underlying rocks in Ireland is of an illusory character. The beds formerly re- garded as conglomerates in Wexford have proved to be crush-breccias, and elsewhere the junction beds of the two series are so plicated into one another that it is impossible to find a dividing line, and so far as Messrs. Egan and M'Henry | could ascertain in their revision of 1898, 20 ~ there is no evidence of any stratigraphical break between them. H One of the best sections of these Cambrian rocks is that exhibited in the cliffs of Bray Head near Dublin, whence they are sometimes called the Bray Head Series. At the northern end of this cliff section are red, purple, and green slates and grits, succeeded on the south by alternating grits and shales. The beds are bent into numerous sharp folds, so that the real thickness is much less than appears at first sight. Jukes and Du Noyer estimated the exposed succession at between 3000 and 4000 feet, but neither the base nor summit is actually visible. These rocks are also well exposed on the coast of Howth, east of Dublin, where they g include some remarkable breccias. No fossils have been found in these beds g except the two or three species of Oldhamia, Q the trumpet -shaped worm - tubes known as Histioderma, and some spherical bodies with radiating spines which may be Radiolaria. The nature of the markings called Oldhamia has not been definitely ascertained ; Professor ^ Sollas has pointed out that those of 0. radiata are grooves and may have been made by the anterior end of a worm, but those of 0. antiqua are ridges on the upper sur- 96 STRATIGRAPHICAL GEOLOGY faces of bedding planes, and no explanation of them lias yet been 5. Scotland In the north-west of Scotland Cambrian rocks flank the eastern side of the tract occupied by the Hebridean and Torridonian, all the way from Durness and Eriboll on the north coast down to the island of Skye. Before the discovery of Olenellus in these rocks they were supposed to be of Ordovician age. In the north of Sutherland there are two tracts of Cambrian, the relative positions of which are shown in Fig. 23. The beds of the Durness area are arranged in the form of a basin truncated on the eastern side by a fault which brings them against the Archaean gneiss ; on the west they are seen to rest unconformably on this gneiss and on the red Torridon sandstone. The Durness basin is separated from that of Loch Eriboll by a prominent ridge of Archaean gneiss, which, however, bears an outlier of the Cambrian basement beds. The Eriboll area exhibits the same series as that of Durness, but they are crushed and faulted in a remarkable manner, and are cut off on the east by a thrust-plane, which carries some of the Archaean gneiss over the Cambrian Series, so that here the ordinary test of age by superposition becomes useless. The total thickness of the beds composing the Cambrian System of Scotland is estimated as at least 2000 feet, but the upper limit is not reached, being everywhere concealed beneath the masses of crushed and sheared rock which have been carried westward onto and over the Cambrian strata. The exposed portion is divisible into two series, an arenaceous below and a calcareous series above. These can also be subdivided as follows : 21 '3. Grey limestones, some dark and some light grey, with Calcareous Series or Durness limestone, 1400 feet many layers of chert ; many fossils throughout. 2. Dolomitic limestones, grey, white, mottled, and pink, with layers of chert nodules. The Sangomore and Sailmhor Groups. Few fossils. 1. Light-coloured dolomites and limestones, with bands of chert. 5. Dolomites and limestones witli SaltcreUa. 4. Dolomitic grits with Salterella (Serpulite grit) \ 200 Arenaceous 3. Calcareous mudstones (Fucoid Beds) / feet. Series, - 2. Fine-grained quartzites perforated by worm burrows 720 feet (pipe rock). 300 feet. 1. Coarse flaggy grits and quartzites, with a basement bed of brecciated conglomerate. 200 feet. The most abundant fossils in the Durness limestone are Cephalopoda Orthoceras, Piloceras, and Lituites. Next in number THE CAMBKIAN SYSTEM 97 are Gastropoda, but bivalves (Lamellibranchia and Brachiopoda) are not rare. Archceocyathus (see p. 72) is of frequent occurrence, as are also sponges of the genus Calathium or Archceoscyphia. In 1891 fragments of Olenellus (0. Lapworthi) and specimens of Hyolithes were found in the Serpulite grits and Fucoid Beds of Ross, proving them to be of Lower Cambrian age. Hence it is probable that the lower part of the Calcareous Series is of Middle Cambrian age (Paradoxidian), but the higher and more fossiliferous part is of Upper Cambrian age, corresponding to the Upper Lingula flags and Tremadoc slates of Wales. The Cambrian succession in Skye has been worked out by Mr. C. T. Clough, who found that the same subdivision could be made as in Durness. Moreover, the Olenellus fauna was again discovered in the Fucoid Beds at Tokavaig and Ord. No fossils, however, could be found in the lower part of the Durness limestone, the thickness of which to the top of the Sailmhor Group Mr. Clough estimates at 830 feet. But higher beds are found in Strath, and in 1898 these yielded many fossils, including Calathium and Archceoscyphia ; a trilobite of the genus Solenopleura ; Orthisina festinata and 0. striatula ; a bivalve, Euchasma Blumeribachice ; species of Pleurotomaria, Murchisonia, Maclurea, and Holopea, with Piloceras, Orthoceras, and Trocholites. The fauna of the Durness limestone, as regards both genera and species, has much more affinity with that of the Cambrian of North America than with that of Wales or England. It may in fact be inferred that the Scotch Cambrian belongs to a North Atlantic or North American area of life and deposition, in which the physical conditions differed considerably from those of the South British and Central European area. The only other part of Scotland where rocks occur which seem to be of Cambrian age is along the south-eastern border of the Central Highlands in Perth, Forfar, and Kincardine. They occur in narrow faulted strips along a line from Blairgowrie in Perth to Stonehaven on the coast, and consist of grey shales and grits, jasper and jaspery phyllite, with lenticular sills and possibly interbedded flows of basic igneous rock (dolerite). They were described by Mr. G. Barrow as the " Green Rock and Jasper Series," 22 and from the resemblance between them and rocks of Arenig age in the Southern Uplands he regarded them as probably Ordovician ; but a more recent discovery of fossils in the same series as exposed on the coast near Stonehaven makes it probable that they are of Cambrian age. This discovery was made by Mr. R. Campbell 23 in a set of black shales with layers of jasper and chert intercalated with bands or sills H 98 STRATIGRAPHICAL GEOLOGY 1 &sJi D So p . ^ - 6 J 1 o '.B -^ 1 1 9 S I I CC o5 of of 3 F-H . 0) l t ^. . r "S rC "S 03 "S ^ -2 "S Warwick! ^<2 M o __ 81 ^s & 31 11 IS &8 '5 ^ lii 11 52 ! l t Ehinog Grit Cefn Slates Dolwen Gri 3000 feet. fr |S 02 9 IPPIIM THE CAMBRIAN SYSTEM 99 of green igneous rock. Dr. Peach reports that the fossils include species of Lingulella, Obolella, Acrotreta, Linnarssonia, and Siphono- treta, and a few specimens of a Phyllocarid Crustacean allied to Lingulocaris. These genera, he remarks, are found in both Upper Cambrian and Lower Ordovician, but the absence of graptolites is indicative of the former rather than the latter. D. THE CAMBRIAN OF NORTHERN EUROPE 1. France (North- West) Cambrian strata are found in many parts of Brittany and Normandy troughed in between the tracts of Archsean rocks, the main flexures of the district striking roughly from west to east. The system is most fully developed in the southern and eastern parts of the region, some portions of it being apparently absent in the west and north. In the north-west (around Brest and Treguier) the Cambrian Series consists of two groups differing much in lithological characters. These are (1) a basal purple conglomerate passing up into sandstones, 300 to 500 feet, (2) purple and green slates, about 400 feet. 24 No fossils have been found in these rocks, and they are succeeded by coarse grey felspathic sandstones (Grds felspathique), which are either of Tremadoc or Arenig age. In the south and south-east of Brittany the Cambrian is repre- sented by much thicker masses of deposit. Thus to the south and west of Rennes the purple conglomerate and sandstone is 1700 feet thick, and the overlying slates are estimated to be 7000 feet. Farther east in Mayenne there is a more varied series comprising a basal conglomerate, purple slates with beds of grit and limestone, coarse sandstones, and andesitic lavas, with finally green slates containing Lingulella Criei. In the lower beds only a few specimens of Lingulella and Dinobolus have yet been found, but Mr. ^Ehlert believes them to represent the whole of the Lower and Middle Cambrian Series (Olenellian and Farad oxidian). 25 The correlation of the southern and northern series is not easy because of the rarity of fossils. Professor de Lapparent gave no definite correlation, but seemed to assume that the basal con- glomerates are everywhere of the same age. If this is so then there is probably a gap in the northern series between the slates and the Gr&s felspathique. On the other hand, the northern area may have been land in Lower Cambrian time, and the conglomerates may be shore deposits which would be of later and later date as we passed from south to north and 100 STRATIGRAPHICAL GEOLOGY north-west, till only the Upper Cambrian is represented. On this view the correlation would be as follows, and the great difference in thickness more easily accounted for : Southern area. Northern area. ^ f Green and purple slates. Green and purple slates \ggQ f ee ^ g I Felspathic sandstones and lavas. Purple sandstones, etc. / -j Purple slates, g I Purple sandstones, co ^ Basal conglomerate. In Normandy a somewhat different facies of the Cambrian is found, inasmuch as it includes an important limestone formation which is largely quarried for marble. Near Caen, May, and Clecy the succession is : Feet. Green slates and limestone, without fossils 130 Massive limestones (red and grey marbles) ..... 250 Red slates and red marble 20 Purple sandstone and conglomerate ...... 200 600 All these beds thicken southward till, near Montbard and Fourneaux, they have a combined thickness of nearly 2000 feet, while the overlying felspathic sandstones thin out entirely, and finally, south of St. Lo, the whole series disappears beneath an overstep of the Ores armoricain (Arenig) on to the Archaean rocks, and the Cambrian only comes in again to the south of Ernee and Mayenne. 2. Iberian Peninsula Cambrian rocks occur in several parts of Spain and Portugal, in Asturias and Galicia, in Leon, in Seville near Ciudad-Real, and near Busaco in Portugal ; the Paradoxides fauna having been found in several of these districts. The area which has been most completely explored and described is that of Asturias and Galicia in the north-west, and in 1882 Dr. Barrois referred the following groups to the Cambrian : 26 T , r f Greenish sandy shales with beds of green sandstone 100 to tt^M 300 feet. ( Limestones and shales 60 to 200 feet. Eivadeo /Green quartzites and slates. Beds \Green slates and bluish phyllites. The upper group yielded species of Paradoxides, Conocoryphe, Arionellus, and Trochocystites bohemicus, so that it is clearly of Middle Cambrian age. It is overlain by conglomerates and sand- stones with Bilobites which are comparable with the Gres fels- THE CAMBRIAN SYSTEM ,/ 101 pathique of Brittany, and there is consequently 'sin 'albseMde of the ' Upper Cambrian (Olenus fauna). Below the La Vega Beds and apparently in conformable succession are the unfossiliferous Bivadeo slates, which may in part represent the Lower Cambrian, but Dr. Barrois compared them with the Phyllades de St. Lo which were at that time classed as Cambrian. Further investigation is there- fore required in this area where no base has yet been recognised. 3. Belgium and the Ardennes In the Ardennes district there are four separate areas where Cambrian rocks come to the surface, but only those of Kocroi and of Stavelot are of any considerable size. 27 The one lies mostly in France, forming a hilly plateau traversed by the deep ravine of the Meuse between Deville and Fumay ; the other is on the borders of Belgium and Germany around Spa and Stavelot. The rocks have the same facies in both areas they are much flexured and faulted, but a definite succession has been made and the Belgian geologists group them in three stages as below : 3. Salmian, consisting of slates and quartzo-phyllites in which Dictyo- graptus has been found. 2. fievinian, black slates and dark-grey quartzites. 1. Devillian, greenish and grey quartzites with bands of purple and green slate containing Oldhamia. The base is not exposed, and no fossils except Oldhamia radiata and 0. antiqua have been found in the Devillian, which may be of Middle or Upper Cambrian age ; while the Salmian may certainly be correlated with the British Tremadoc Beds. In Belgium there is another area where a succession of Cambrian rocks are found, i.e. in the valley of the Thyle south of Brussels. Here the lowest beds resemble the Devillian and also contain Oldhamia ; above them, near Mousty, M. Malaise has described a set of black slaty shales with black limestones and bands of chert, and in the limestones he has found small Ostracoda which he refers to Leperditia punctatissima and Primitia solvensis (Lingula flag species) ; these black beds are succeeded by quartzo-phy Hades which he also refers to the Lingula flags, and he correlates the Cambrian of the Ardennes and of Brabant as follows : 28 Ardennes. Brabant. a -, f Upper. . . Not recognised. Salmlan \Lower. . . Assize de Villers. Revinian . . . _. Assize de Mousty. n -IT f Upper . . Assize de Tubize. Deyllllan [Lower . . Assize de Blaumont. 102 ct STKATTGKAPHICAL GEOLOGY In connectibn v with the Cambrian Series of Belgium the recent recognition of similar rocks beneath the eastern counties of England is of much interest, for at Stutton near Ipswich the boring made in 1894-5 entered hard quartzitic rocks at a depth of 994 feet, and traversed them to a total depth of 1525 feet. These rocks are really dark-grey quartzo-phyllades, which M. Mourlon has recognised as resembling those of Villers, near Ath. At Harwich and Weeley in Essex the rocks found below the Gault are dark-grey and purple slates, in which an Orthoceras was found. These must be either of Tremadoc age or a part of the overlying Ordovician Series as developed in Belgium. 4. Scandinavia It is interesting to note that in Norway, Sweden, and Northern Kussia the Cambrian strata occur in broad, flat-lying sheets as originally deposited, and are so little altered that they have more the aspect of our Mesozoic clays, shales, and sandstones than of Palaeozoic rocks. Moreover, those of Scandinavia present a complete succession of highly fossiliferous beds, and yet their thickness is only a few hundred feet. The Cambrian Series of this region consists everywhere of a lower arenaceous division and an upper argillaceous division, the lower being from 350 to 550 feet thick and the upper from 200 to 400 feet. The system is most fully developed in Sweden, and especially in the districts of Scania and Westrogothia, where there is the following succession in descending order (slightly altered from Linnarsson and Tullberg). 29 IV. Passage Beds (thickness about 50 feet = Tremadoc). 3. Ceratopyge limestone, with G. forficula. 2. Shumardia shales. 1. Dictyograptus shales, with Dictyograptus and Obolella Salteri. III. Olenus shales in five zones, with three bands of limestone, thick- ness from 70 to 200 feet ( = Lingula flags). 5. Zone of Parabolina heres. 4. Zone of Peltura scarabceoides and Sphcerophthalmus alatus. 3. Zone of Eurycare and Leptoplastus. 2. Zone of Parabolina spinulosa. 1. Zone of Olenus truncatus, 0. gibbosus, and Ag. pisiformis. II. Paradoxides shales, divisible into four principal zones, and from 80 to 160 feet thick ( = Menevian and Solva Beds). 4. The Andrarum limestone with 5 feet of shale above ; Agnostus Icevigatus, Paradoxides Forchhammeri, and Orthis Hicksi. 3. Zone of Par. Davidis and Conocoryphe cequalis. 2. Zone of Par. Tessini with the exsulans limestone at base containing Conocoryphe exsulans and Par. palpebrosus. 1. Zone of Agnostus atavus, Paradoxides olandicus, and the black Alum shale at the base. THE CAMBRIAN SYSTEM 103 I. Olenellus Beds, chiefly sandstones homotaxial with the Hartshill quartzite and Caerfai Beds, 350 to 550 feet. 4. Phosphatic limestone (few fossils). 3. Sandy shales with Olenellus Kjerulfi and Arionellus primcevus (5 to 10 feet). 2. Fucoid sandstone with Obolella favosa, 200 to 350 feet. 1. Eophyton sandstone with Mickwitzia monilifera and the markings called Eophyton and Cruziana, 150 to 200 feet. The basal sandstones rest unconformably upon, the Archaean 5. Russia In the Baltic provinces of Russia there is a still more remarkable development of the Cambrian System. The arenaceous division is partly replaced by blue clay, and the black trilobite shales are absent. The succession is as follows : Feet. f6. Glauconitic sand with Obolus and Siphonotreta ... 10 5. Shales with Dictyograptus and Bryograptus ... 20 4. The Ungulite sandstone full of Obolus Apollinis and other minute Brachiopods ....... 4 Alternating clays and sandstones with Olenellus Michwitzi, Mickwitzia monilifera, and a Scenella .... 50 2. The blue clay, without fossils 300 Ferruginous sandstones, resting on granitoid rock . . 300 if 3 ' 684 Professor F. Schmidt 30 has pointed out that the base of the Ungulite sandstone is sharply divided from the beds below, and some- times contains pebbles derived from them. Hence there appears to be a gap at this point, which will account for the complete absence of any representatives of the Paradoxides and Olenus shales, unless the thin band of Ungulite sandstone is a partial equivalent of the latter. The underlying clays and sandstones are evidently the equivalents of the Fucoid and Eophyton sandstones, while the two highest members are comparable with the beds which occupy the same position in Sweden. 6. Germany and Bohemia It is very probable that Cambrian rocks underlie the greater part of Germany, but they are only exposed in a few places. One of these is the Hohe Venn, south of Aix-la-Chapelle, where there are slates, quartzites, and phyllites, resembling those of the Ardennes and containing both Oldhamia and Dictyograptus. Another and larger area is found in the Fichtel Mountains and the 104 STRATIGRAPHICAL GEOLOGY Thuringia Wald, north-east of Bavaria, but it is only at Leimitz, near Hof, that well-preserved fossils have been obtained ; the fauna, consisting chiefly of Trilobites, Brachiopoda, and a few Cystidea, appears to belong to the highest Cambrian (Upper Tremadoc and Ceratopyge limestone). In Bohemia the existence of Cambrian rocks was first made known by Barrande in 1852, when he recognised his Stage C as containing a "primordial fauna." Subsequent writers have modified Barrande's classification, and have found a still older fauna in his Stage B. To agree with the British Cambrian System it must also include part of D, and will then consist of the following groups : Stage D 1. Pisolitic ironstones, sandstones, and sandy conglomerates with diabasic lavas and tuft's. Stage C. Greenish slates with a Paradoxides fauna (P. bohemicus and P. spinulosus), Conocoryphe, Arionellus, Ellipsocephalus, etc. Stage B. Sandstones and flags with an Olenellus fauna passing down into coarse conglomerate which rests unconformably on Stage A (Przibram schists). A more detailed study of this Cambrian Series has been made by Pompeckj, 31 who describes the lowest part (Stage B) as consisting of a conglomerate passing up into grits, sandstones, and flags which contain species of the Trilobites Sao, Solenopleura, Ellipsocephalus, and Ptychoparia, with Stenotheca and Orthis ; this is an Olenellus fauna, though that genus has not yet been found. Dr. Marr 32 remarks that these rocks strongly resemble our Harlech Series and are of great thickness. The Paradoxides Beds are slaty shales with a thickness of about 1000 feet. Above them is a conglomerate supporting greenish grits which contain Lingula Feistmanteli and probably represent part of our Lingula flags. Above these are pisolitic ironstones of black, green, and red colours which contain a few fossils Erinnys Grimmi, Amphion Lindnaueri, and Orthis desiderata ; this small assemblage having the aspect of a passage fauna and being therefore probably of Tremadoc age. From this account it will be seen that the Bohemian sequence is not quite complete ; in fact, there seems to be a break and un- conformity between Stages C and D. The beds of Stages B and C are only found over a small portion of the area near the town of Beroun, and are overlapped by the conglomerates of D 1 which then rest directly on the Archaean rocks. Moreover, 110 species of Olenus has yet been found in Bohemia, from which it may be inferred that there is little if anything to represent the lower and middle parts of our Lingula flags, the Lingula grits of Stage D corresponding probably with the Dolgelly Beds of Wales. The THE CAMBRIAN SYSTEM 105 total thickness of the Cambrian in Bohemia may be about 3000 feet. The only other exposure of Cambrian rocks in Central Europe is in Russian Poland, round Sandomir on the Vistula, where some shales, associated with quartzites and conglomerates, have yielded species of Paradoxides (? P. bohemicus), Agnostus (fallax and gibbus), and Liostracus Linnarsoni ; evidently a Middle Cambrian fauna. E. EUROPEAN GEOGRAPHY AND CONDITIONS OF DEPOSITION Any attempt to restore the geographical conditions of the early Palaeozoic periods must be a task of great difficulty. There are large areas where the rocks of these early systems are deeply buried below those of newer times, and the exposures which we can study are only so many isolated portions of some ancient sea-floor ; while the traces of contemporaneous lands are only to be found where some series of strata are absent or where coarse detrital deposits are unusually thick. Nevertheless, wherever Palaeozoic rocks can be followed over a considerable area some indications of the direction from which the sediment was transported can usually be found, and of the conditions under which it was deposited. It will therefore be useful to indicate such evidence as we possess for answering the question under what geographical and physical conditions were the Cambrian strata of Northern and Western Europe accumulated 1 In the first place, some general considerations are suggested by a comparison of the deposits in Wales, England, Sweden, and Russia. The enormous thickness of the Welsh Cambrians, which have a maximum thickness of about 11,000 feet, as compared with a possible 3000 feet in Warwickshire, less than 1000 feet in Sweden, and only 680 feet in Russia, is a point of great significance ; but the inference to be drawn from these figures depends on other stratigraphical facts. If the lower beds were not represented in the Swedish and Russian succession, it might be inferred that the north of Europe was a land-area during the earlier part of the period, and was gradually submerged by the waters of a sea which spread over it from the south-west, and this view was actually suggested in 1876 before the Olenellus fauna was recognised in Europe. The facts as we now know them, however, will not allow of such an inference ; the Swedish succession is as complete as that of Warwick- shire or South Wales, and the difference in thickness is caused mainly by the great increase in the amount of sandy material in the Cambrian sediments as we pass from east to west, or rather from north-east to south-west. 106 STIIATIGEAPHICAL GEOLOGY In the western areas arenaceous material prevails not only in the Lower Cambrian, but also in the Middle and Upper divisions, for in North Wales a large part of the Lingula Flag Series consists of flagstones which become coarse quartzose grits as they are followed through Carnarvonshire. On the other hand, in Scotland, Sweden, and Russia sandstone is practically confined to the lowest stage, all the higher portions of the system consisting of limestone and shale. As the Swedish geologist Linnarsson wrote in 1876, "The facts rather tend to show that most of the Swedish Cambrian rocks were deposited in a deeper sea and farther from land than the British." He points out that the small thickness of the Olenus Beds over such a large area in Sweden can only be explained on this view, for they are so thin compared with the Lingula flags of Wales that the rate at which sediment was accumulated in Wales must have been fifty or sixty times as rapid as it was in Sweden. This must have been due to the much greater proximity of land, while all the facts connected with the Upper Cambrians of Sweden lead us to infer that they were deposited in a deep sea and far from land. We may therefore conclude that the greater part of the materials forming the Cambrian deposits in France, Wales, and England were derived from lands which lay to the west of France and to the north-west of Wales. That which lay west of France and south-west of Wales was probably part of a large land-area or continent, occupying part of what is now the Atlantic Ocean, and terminating eastward in a promontory which included the northern part of Brittany during early Cambrian time, but was submerged in the later part of the period. The land to the north-west of Wales may have been an island or another promontory of the Atlantic continent. It included Anglesey and probably stretched across the Irish Sea into Ireland, but until more is known about the structure of the north of Ireland it is impossible to say how far it reached in that direction. Open sea seems to have extended across the North Atlantic region from America to the north of Scotland with circumpolar or Arctic land again to the north of it, for it is only in this way that the American character of the Scottish Cambrian fauna can be satisfactorily accounted for (see The Building of the British Isles, 3rd edition, p. 63). This northern sea may not have been very deep, but its waters must have become deep enough and clear enough in Middle Cambrian time for the formation of limestone, and in the Scottish limestones there is evidence that the accumula- tion of shell-debris and calcareous matter went on very slowly ; for THE CAMBRIAN SYSTEM 107 the condition of the larger fossils in these beds is such that they must have lain for a long time on the sea-floor and have suffered greatly from decay and partial solution before they were protected by a covering of other calcareous materials.. Mr. Peach 33 has observed that the shells are generally imperfect and seldom show the external markings ; the Lamellibranchs, though often found with attached valves and in their natural position, are yet often so imperfect that they have clearly suffered much from solution before complete embedment. Again the outer walls and the septa of the chambered Cephalopoda have in most cases been largely or wholly destroyed, so as to leave only the thick column of the central siphuncular portion of the shell. Moreover, these limestones contain many nodules of chert arranged in layers like the flints and cherts in Cretaceous deposits, and sometimes in continuous layers or sheets. These cherts contain sponge-spicules, and though no Radiolaria can now be recognised in them it is very probable that the silica was derived partly from Radiolaria and partly from siliceous sponges. The occurrence of glauconite grains in certain beds, not only in Scotland but in the Upper Cambrian of Sweden and Russia, is also indicative of clear water, slow deposition, and some distance from land ; for it is probable that the conditions favourable for the production of glauconite on sea-floors have not materially changed from Cambrian time to the present day. Those conditions are found where the water is clear and the bottom currents are fairly strong but carry little sediment, and where the depth of water is from 200 to 500 fathoms ; the actual depth, however, seems to be less important than the other conditions. Passing now to the more southern parts of the European region we find proofs of the existence of a Cambrian Sea in Portugal, Spain, Provence, and Sardinia ; so that it seems probable that the whole of Western and Southern Europe was covered by this sea, though there may have been islands within its limits. It is only in Bohemia that we seem to approach one of the borders of this sea, for there the older Cambrian deposits are limited to one part of the area, the Upper Cambrian overlapping on to Archajan rocks. Even this land may have been an island, but if it was part of a large tract that must have lain to the east or south of Bohemia. Another point to be noted with regard to the physical conditions of Cambrian time is that very little volcanic activity showed itself in the European region till toward the close of the period. After the violent and long -continued outbursts of volcanic energy which prevailed in pre-Cambrian times, there seems to have been a period of quiescence and subsidence before the subterranean forces acquired 108 STRATIGRAPHICAL GEOLOGY power enough to burst forth again into surface eruptions. During the later part of the period volcanic activity became rife in several parts of the region, notably in Wales, Brittany, and Bohemia. EEFERENCES 1 On the "Silurian System of Rocks," by R. I. Murchison, Phil. Mag. third series, vol. vii. p. 46 (1835). 2 On the "Silurian and Cambrian Systems," by Professor Sedgwick and R. I. Murchison, Rep. Brit. Assoc. 1835, part ii. pp. 59-61. 3 On the "Tripartite Classification of the Lower Palaeozoic Rocks," Geol. Mag. Dec. 2, vol. vi. p. 12 (1879). 4 In Zittel's Palaeontology, English edition (1900), vol. i. p. 622. 5 Hicks in Quart. Journ. Geol. Soc. vol. xxvii. p. 384, and vol. xxi. p. 39 ; also Popular Science Rev. 1881. 6 A. R. Andrew, Geol. Mag. for 1910, pp. 159, 201, 261. 7 J. F. N. Green, Proc. Geol. Assoc. vol. xxii. p. 135 (1911). 8 Salter in Ramsay's Geology of North Wales, ed. 2, p. 62. 9 Fearnsides, Quart. Journ. Geol. Soc. vol. Ixvi. p. 142 (1910). 10 Ramsay, "Geology of North Wales," Mem. Geol. Surv. 2nd edition, p. 166. 11 T. M'K. Hughes, Quart. Journ. Geol. Soc. vol. xxxiv. p. 137, vol. xxxv. p. 628 ; and Proc. Geol. Assoc. vol. viii. p. 199. 12 C. Lapworth, " Geology of the Birmingham District," Proc. Geol. Assoc. vol. xv. p. 313 (for 1898). 13 T. T. Groom, Quart. Journ. Geol. Soc. vol. Iviii. p. 89. 14 Cobbold, Quart. Journ. Geol. Soc. vol. Ixvi. p. 19 (1910). 15 T. T. Groom, Quart. Journ. Geol. Soc. vol. Iv. p. 157 et seq (1899). 16 C. Callaway, Quart. Journ. Geol. Soc. vol. xxxiii. p. 652 (1877). 17 J. E. Marr, Geol. Mag. for 1894, p. 122. 18 Miss G. L. Elles, Quart. Journ. Geol. Soc. vol. liv. p. 463. 19 G. W. Lamplugh, "Geology of the Isle of Man," Mem. Geol. Surv. 20 Egan and M 'Henry, Sum. Prog. Geol. Surv. for 1898, p. 57. 21 See "Geological Structure of the N.W. Highlands," Mem. Geol. Surv. 22 G. Barrow, Quart. Journ. Geol. Soc. vol. Ivii. p. 328 (1901). 23 R. Campbell, Geol. Mag. for 1911, p. 63. 24 C. Barrois, Proc. Geol. Assoc. vol. xvi. p. 101, and Congres Geol. Internal, for 1900 (Excursions). 25 D. P. ^Ehlert, Congres Geol. Internal, for 1900, Livret Guide (Mayenne). 26 Barrois, "Terrains anciens des Asturies," etc., Mem. Soc. Geol. Nord (de France), Tome 2 (1882). 27 J. Gosselet, Congres Geol. Internal, for 1900, Livret Guide (Ardenne). 28 C. Malaise, Bull. Acad. Royal de Belg. for 1910. 29 See Mobery in Geol. For en. Stockholm. Forandl. for 1910. 30 Schmidt, Quart. Journ. Geol. Soc. vol. xxxviii. p. 514. 31 Pompeckj, Jahrb. K. K. Geol. Reichsaust. xlv. p. 495. 32 Marr, Palaeozoic Rocks, Sedgwick Prize Essay for. 33 Peach in "Geol. Structure of the N.W. Highlands." CHAPTER VI THE ORDOVICIAN SYSTEM A. NOMENCLATURE AND DIVISIONS THE grounds on which this system has been established have already been explained (see p. 70). It is the Upper Cambrian of Sedgwick and the Lower Silurian of Murchison and the Geological Survey. The system has usually been divided into three series with the following names : 3. Upper Ordovician or Bala Series. 2. Middle Ordovician or Llandilo Series. 1. Lower Ordovician or Arenig Series. The names are taken from places in Wales where the several series are well developed, i.e. the Arenig Mountains in Merioneth, the town of Llandilo in Carmarthenshire, and the town of Bala in Merioneth. This succession was first established in North Wales by Sedgwick. Part of it was studied at the same time in Shropshire and Carmarthen by Murchison, who described the local equivalents of the Bala Series under the name of Caradoc Sandstone, and after their identity had been demonstrated the beds were for a time called the Bala and Caradoc Series. These divisions, however, were not separately mapped either by Murchison or by the Geological Survey, so that they were never clearly marked off from one another, nor was any definite horizon recognised as the base of the Llandilo Series or as that of the Bala Series. The first to establish divisions on a strictly palEeonto- logical basis was Dr. H. Hicks 1 who in 1875 recognised both the Arenig and Llandilo Beds in Pembrokeshire, and subsequently found that they were separated by a set of black shales which contained a new and distinct fauna. 2 These he called the Llanvirn Beds, but for a long time this group was regarded as a superfluous division, only embodying what should or could be classified as ^the Upper Arenig and the Lower Llandilo stages. 109 110 STKATIGRAPHICAL GEOLOGY Gradually, however, it has been found that the fauna of the Llanvirn Beds has more than a local value and is really distinct from those above and below. A portion of the Llanvirnian outcrop was mapped by Messrs. Marr and Roberts in 1885, 3 and other portions subsequently by the Geological Survey, with the result that in a recent memoir Dr. Strahan announces that the only possible course is to restrict both the Arenig and Llandilo Series, and to establish an intermediate Llanvirn Series. 4 Dr. Marr has also advocated the separation of the Upper Bala from the rest of that series because it contains many species which do not occur or are rare in the lower beds, but the true value of this Ashgillian fauna has yet to be proved, and in this volume I shall not recognise more than four main divisions in the Ordovician System, namely : 4. Bala Series (Ashgillian and Caradocian). 3. Llandilo Series or Llandeilian. 2. Llanvirn Series or Llanvirnian. 1. Arenig Series or Arenician. B. LIFE OF THE PERIOD Taken as a whole, the special features of the Ordovician fauna are (1) the abundance of branched and biserial graptolites belonging to the families Dichograptidce, Dicranograptidce, Glossograptidce, Leptograptidce, and Diplograptidce. Of these the first includes the following genera Dichograptus, Tetragraptus, Loganognaptus, Didy- mograptus, and Phyllograptus ; in the Dicranograptidae are Dicello- graptus and Dicranograptus. The Glossograptidae include Glosso- graptus and Lasiograptus. The Leptograptidse include Leptograptus and Nemagraptus ( = Ccenograptus), while Diplograptus, Crytograptus, and Climacograptus belong to the Diplograptidse. (2) The appearance of many new trilobite genera *^Eglina, Acidaspis, Ampyx, *Amphion, Asaphus, Calymene, Encrinurus, Harpes, Homalonotus, Illcenus, Lichas, Phacops, *Placoparia y Proetus, Remopleurides, Sphcerexochus, Staurocephalus, *Stygina, and *Trinucleus, those with an asterisk not surviving this period. (3) The abundance of Cystideans, eight genera, and twenty-three species, occurring in the Bala Group. (4) The abundance of Orthidae, and the appearance of the following genera of Brachiopoda Atrypa, Leptcena, *0rthisina, Rhynchonella, Siphonotreta, Strophomena, Trematis, and Triplesia. The occurrence of graptolites throughout the Ordovician System in almost all areas where shales were deposited has afforded a means of subdivision into stages or zones and of correlating the THE ORDOVICIAN SYSTEM 111 successive members of one local series with those of another and with those of other countries. The following is a table of the zones recognised in the British Islands : Bala Series Llandilian | Llanvirnian- Arenician -j /Zone of Dicellograptus anceps. I , , complanatus. ies J Pleurograptus liriearis. Dicranograptus Clingani. Climacograptus Wilsoni. Didymograptus superstes. Dicellograptus divaricatus. Didymograptus Murchisoni. , , bifidus. , , hirundo. extensus. Fig. 24. GROUP OF AEENIG AND SKIDDAW GBAPTOLITES. 1. Didymograptus hirundo. 3. Cryptograptus antennarius. 2. Didymograptus gibberulus. 4. Phyllograptus typus. 5. Tetragraptus quadribrachiatus. The following are the principal species which characterise the several divisions of the Ordovician System : Fossils of the Arenig Series Hydrozoa. Tetragraptus quadribrachiatus, T. serra, Didymograptus extensus, D. hirundo, D. gibberulus, Phyllograptus typus, Dendrograptus arbuscula, Diplograptus dentatus, Azygograptus suecicus. 112 STRATIGEAPHICAL GEOLOGY Crustacea. Ogygia Selwyni, O. marginata, JEglina binodosa, JE. caliginosa, Calymene parvifrons, Caryocaris Wrighti. Brachiopoda. Lingula petalon, Monobolina plumbea, Orthis alata. Lamellibranchia. Kedonia anglica, Palsearca amygdalus. Gastropoda. Hyolithes (Theca) vaginula. Cephalopoda. Orthoceras sericeum. Fossils of the Llanvirn Series Hydrozoa. Didymograptus bifidus, Did. Murchisoni, Did. nanus, Did. acutidens, Diplograptus (Amplexograptus) confertus, Diplograptus calcaratus, Glossograptus ciliatus. Trilobita. Ampyx nudus, Illsenus Hughesi, Placoparia cani- brensis, Phacops llanvirnensis, Barrandea Homfrayi. Brachiopoda. Lingula attenuata. Fig. 25. GROUP OF ARENIG FOSSILS. a. Ogygia Selwyni. b. (Eglina binodosa. c. Palsearca amygdalus. d. Euomphalus corndensis. Gastropoda. Cephalopoda. Radiolaria. Foraminifera. Hydrozoa. Crustacea. Brachiopoda. Gastropoda. Cephalopoda. Euomphalus corndensis, Pleurotomaria llanvirnensis, Conularia llanvirnensis, Bellerophon perturbatus. Conoceras llanvirnense, Orthoceras caeresiense. Fossils of the Llandilo Serie.s Species of Styptosphsera, Spongoplegma, Diploplegma, Stauroplegma, Haliomma, Dorysphaera, Doryplegma, Triplosphsera. Saccammiua Carteri. Didymograptus superstes, Diplograptus foliaceus, Dicel- lograptus divaricatus, and D. sextans, Nemagraptus (Coenograptus) gracilis, Dicranograptus ramosus. Asaphus tyrannus, Ogygia Buchi, Ampyx nudus, Tri- nucleus Jamesi. nucleus fimbriatus, Agnostus M pyx 'Coyi, Acidaspis Lingula brevis, Orthis alata, 0. confinis, Siphonotreta micula (ranges into Bala). Maclurea Logani, M. magna, Ophileta compacta, Bellero- phon perturbatus (passes to Bala). Orthoceras Avelinei. THE ORDOVICIAX SYSTEM 113 Fig. -26. OROUP OF ARENJO AND LLANDILO FOSSILS. a.. Didymograptus Murchisoni. /. Acidaspis Jamesi (Irish), ft Orthisalata(Areuig).' it. Trinucleus fimbriatus. e. Asaphus tyrannus, f. OgygiaBuchi. Fig. 27. GROUP OF LLANDILO AND BALA GRAPTOLITES. 1. Climacograptus bicornis. "2. Dicellograptus sextans. 3. Dicellograptus ziczac. 4. Glossograptus Hincksi. 5. Dicranograptus ramosus. 0. Diplograptus foliaceus. 114 STRATIGRAPHICAL GEOLOGY Fig. 28. GROUP OF BALA FOSSILS. ff. Orthis flabellulum. h. Orthis elegantula. c. Ctenodonta semitruncata. (I Modiolopsis expansa. i\ Holopea concinna. /. Lituites hibernicus. Fig. 29. GROUP OF BALA FOSSILS. a. Echinospluera aurantium. b. Sphseronites munitus. c. Illamus Davisii. (/. Liclias hibernicus. e. Phacops apiculatus. /. Agiiostus trinodus. THE ORDOVICIAN SYSTEM 115 Actinozoa. Hydrozoa. Echinoderma. Crustacea. Bryozoa. Fossils of the Bala Series Favosites crassa, Monticulipora iibrosa, Nebulipora lens, Lyopora favosa, Halysites catenularius. Climacograptus Wilsoni, Dicranograptus Clingani, Pleurograptus linearis, Dicellograptus anceps, D. complanatus, D. truncatus, Leptograptus flaccidus. Rhaphanocrinus basalis, Echinosphsera aurantium, Sphse- ronites munitus, S. Litchi, Hemicosmites rugatus. Agnostus trinodus, Trinucleus concentricus, T. seticornis, Illaenus Davisi, I. Bowrnanni, Homalonotus bisul- catus, Phacops apiculatus, P. Brongniarti, Lichas laxatus, L. hibernicus, Staurocephalus globiceps. Monticulipora favulosa, M. petropolitana, Phyllopora Hisingeri, Ramipora hochstellaris. lOrthis vespertitio Brachiopoda. Orthis calliqramma/ Fig. 30. THREE BALA FOSSILS. Orthis actonise, 0. flabellulum, 0. calligramma, 0. biforata, 0. hirnantensis, 0. vespertilio, Leptsena (Plectambonites) sericea, Strophomena expansa, S. siluriana, Trematis corona. Lamillibranehia. Ctenodonta varicosa, Palfearca edmondiiformis, Modo- lopsis obliqua, M. expansa, M. modiolaris. Gastropoda. Holopea concinna, Cyclonema crebristria, Murchisouia simplex, Ecculiomphalus Bucklandi, Tentaculites anglicus, Hyolithes (Theca) triangularis, Bellerophon nodosus. .Cephalopoda. Cyrtoceras sonax, Orthoceras vagans, Trochoceras cor- nuarietis. C. BRITISH ORDOVICIAN ROCKS Ordovician rocks occupy much larger areas of the surface in Great Britain than the Cambrian rocks, and are also found in several parts of Ireland. In Wales they occupy nearly all the western and central portions of the country, outside the Cambrian outcrops. They pass eastward beneath the overlying Silurian .strata, and portions of them emerge again in Shropshire and Radnorshire. 116 STRATIGKAPHICAL GEOLOGY Hocks of tliis age occupy a large part of the Lake District, where they were first studied by Sedgwick. The Isle of Man is an isolated fragment of this Cumbrian area, while to the eastward, in Yorkshire, some small exposures occur, proving the extension of Ordovician rocks beneath the newer strata in that direction. In the south of Scotland, Ordovician and Silurian rocks range across from Wigtown to Berwickshire, and form the mass of the Southern Highlands, their northern boundary being generally the great line of fault which runs from Girvan in Ayr to the Lammermuir Hills. Ordovician rocks occur in many parts of Ireland and exhibit several different facies, one of these facies being found in the north- west (Galway, Mayo, Donegal, and Londonderry), another in the north-east (Down, Cavan, and Meath), another in the south-east (Wicklow, Wexford, and Waterford, with extensions into Clare and Tipperary). In most places the Ordovician rocks seem to succeed the Cambrian in conformable sequence, but in North Wales there appears to be a decided break of continuity, the basal grit of the Arenig Series lying unconformably on the Cambrian and passing tran sgressively across the members of the Tremadoc Series on to- those of the Ffestiniog Group. From an examination and survey of the country around the Arenig and Aran Mountains Mr. Fearnsides- has expressed his belief that the Cambrian rocks of North Wales were ridged up into a series of low flexures striking north and south before the deposition of the Arenig sediments. 5 This physical disturbance was probably due to the intense volcanic action which shortly afterwards led to the establishment of surface volcanoes, and to the outpouring of the great masses of lava and ashes which are such conspicuous features in the Ordovician Series of Merioneth and Carnarvonshire. 1. Wales Arenig Series. The typical district of this group is supposed to be that of the Arenig Mountains in Merioneth ; but the series as now restricted is not there very thick, and it will be more convenient to begin with South Wales, Pembroke, and Carmarthen, where it has a fuller development. The outcrop of the Arenig Series on Ramsey Island and round St. David's Head is shown on Fig. 16, p. 80. From the latter point its outcrop runs eastward and curves round the St. David's anticline to near Haverfordwest, and thence eastward to Carmarthen and Llangarthney. THE OEDOYICIAN SYSTEM 117 Dr. H. Hicks described the Lower Arenig, near St. David's, as consisting of fine black slates and shales, which are best exposed in Kamsey Island, where they have yielded Didymograptus extensus, Phyllograptus stella, Trigonograptus ensiformis, and T. truncatus. They also contain Ogygia scutatrix and Asaphellus homfrayi. Some of the same beds are seen in Whitesand Bay on the main- land, but are faulted against the Tremadoc slates. He believed this group to be about 1000 feet thick. What must be called the Upper Arenig (Middle Arenig of Hicks) are a series of black slates and flagstones with some beds of gritty sandstone, and they have yielded many of the characteristic Arenig trilobites, Ogygia peltata, &glina grandis, Trinucleus Gibbsi, and Ampyx Salteri, and the graptolites Tetragraptus serra, T. crucialis, and Didymograptus patulus. These beds are about 1500 feet thick. The equivalents of this series have recently been mapped in Carmarthenshire for the Geological Survey by Messrs. Cantrill and H. Thomas, who have termed them the Tetragraptus Beds and have divided them into two stages or zones, (1) that of Didymo- graptus extensus, (2) that of Did. hirundo. The former includes the greater part of the series, and is over 1000 feet thick ; litho- logically it consists of blue-black shales and mudstones with grits and conglomerates at the base which mark it off from the under- lying Tremadoc slates. The upper zone appears to be only about 200 feet thick. North of Carmarthen the Arenig Series passes under the flexured mass of higher Ordovician rocks which occupy the central area of Wales (Cardigan, Brecknock, and Montgomery), not again coming to the surface till brought up by the great Harlech or Merioneth- shire anticline. Round this it forms a nearly continuous out- crop from Towyn on the coast through the Cader Idris, Aran, and Arenig ranges, and thence through the Migneint district to Manod Mawr, north of Ffestiniog (see Fig. 19), where it seems to be faulted against higher beds. Throughout this tract of country from Cader Idris to the Arenig Mountains, the place of the Arenig Series appears to be entirely below the beds of volcanic ash and the intrusive masses of andesite which were formerly included in this series. In the district of Arenig Fawr and Moel Llyfnant the succession of beds has been mapped and described by Mr. Fearnsides, 6 and the position of the Arenig Beds is shown in the section drawn by him and reproduced by his permission in Fig. 31. The series here is only about 600 feet thick, but both zones can be recognised, and the sequence is as follows : >0 1) II I CO s 5!li I ||1 111 * t til? 1111 rj O i-^ oo THE OEDOVICIAN SYSTEM 119 Feet. 5. Dark calcareous shales with Didymograptus hirundo and Tetragraptus serra ........ 200 4. Erwent or Ogygia, limestone, yielding Ogygia Selwyni, Orthoceras sericeum, Monobolina plumbea, etc. . . 20 3. Heullan ashes hard, grey, compact beds with Calymene parvifrons ......... 150 2. Llyfnant flags with Did. extensus and D. deflexus . . 200 1. Basal grit, varying in thickness from . . . . to 100 About 620 On the north side of the Harlech anticline the series is found in the Moelwyn Mountains, north-west of Ffestiniog, the basal grit being there overlain by 800 feet of hard spotted flagstones in which only a species of Tetragraptus has yet been found. The limestone is absent, and the Did. hirundo shales seem to be represented by bands of slate intercalated with thick beds of volcanic ash and agglomerate. Westward the basal grit has been found above the Tremadoc Series, but the overlying beds are so broken by faults that no succession can be made out. They come in again in the southern part of the Lleyn peninsula where grits and^flags with extensiform graptolites pass across the edges of the Lingula flags and rest on the Archaean rocks near Aberdaron. On the eastern side these beds are overlain by a band of grit and breccia 60 to 70 feet thick containing fragments of schist, granulite, etc., derived from the Archaean rocks, and this is succeeded by dark shales which have yielded Didymograptus hirundo. Above them are beds of ironstone and manganese-ore with blue mudstones containing Azygograptus suecicus. Arenig slates also form part of the great slate country in Carnarvonshire north-west of the Snowdon range, but the details of this area have not yet been worked out. The complete series has, however, been indicated by Miss G. L. Elles 7 along the banks of the river Seiont at Carnarvon, where the zone of D. extensus consists of calcareous flags and shales containing that species with Azygograptus suecicus and OEglina binodosa. To the south of these beds are micaceous flags and shales yielding D. hirundo^ D. nitidus, and numerous phyllopods, and these are succeeded by black shales with a Llanvirn fauna. The outcrops extend for a distance of about 2000 yards with a dip of from 40 to 50 to the south-east, and the series seems therefore to be about 1400 feet thick. Arenig Beds have also been recognised on both, sides of the Menai Strait, but in Central Anglesey they are believed to have thinned out and to be overlapped by the Llanvirn Beds. 120 STRATIGRAPHICAL GEOLOGY 33 ti S i 5 "g * 3 ^^ I 11! 6 O G O Cfi < (_ rQ p9 15 vi < o 1 !! III THE OEDOVICIAN SYSTEM 1^ Llanvirn Series. This group was separated from the beds below and above it by Dr. Hicks in 1881. It takes its name from that of a village near Llanrain in Pembrokeshire (see map, Fig. 16), where the slates have yielded many of the characteristic trilobites and graptolites of the Lower Llanvirnian or zone of Didymograptus bifidus. This species 'and its congeners D. artus, D. nanu8 t and D. Murchisoni are often termed the tuning-fork graptolites, as distinguished from the extensiform species which prevail in the underlying Arenig Beds. The lower stage is probably from 1200 to 1500 feet thick in Pembrokeshire, and consists of dark-grey flags and slates. The upper stage or zone of D. Murchisoni consists of fine black slates interbedded with felspathic tuft's and has a thickness of about 500 feet. These beds have now been mapped by the Geological Survey from Haverfordwest eastward, through Narberth and Carmarthen. Along this tract the lower stage maintains its thickness, but the Murchisoni Beds are much reduced, being only 230 feet near St. Clears, while east of Carmarthen they appear to thin out below the base of the overlying Ffairfach Grit (Llandilo). Passing now to Merioneth there is no doubt that the Llanvirnian can be followed all round the great Harlech geanticline, but they have not yet been separated from the Arenig and Llandilo Series. Their position on Arenig Favvr is indicated in Mr. Fearnside's section (Fig. 31, p. 118), the base being marked by a band of shale with D. bifidus which is overlain by a volcanic series consisting of ash and agglomerate with thick intrusive masses or laccolites of hypers then e-andesite. Above these, however, is an outcrop of shale containing Diplograptas foliaceus and other species which suggest a high horizon in the Murchisoni zone. The position of the Llanvirnian in the Moelwyn area has not yet been ascertained, but farther west in the promontory of Penrhyn (see Fig. 20), it appears to be faulted down against the basal Arenig grit, for the black slates of Ty Obrey which adjoin that grit have yielded fossils which prove them to be of Llanvirn age. Above them are ashy slates and ash beds passing under a thick volcanic series. In Carnarvon the Llanvirn Series has not yet been separated as a stratigraphical division from the beds above and below, but as mentioned on p. 119, the bifidus zone has been recognised as succeeding the hirundo -beds on the river Seiont near Carnarvon. Its existence in Anglesey has a special interest, because in the central part of that island it forms the base of the Ordovician System, and has a basal conglomerate which rests directly on the 122 STEATIGRAPHICAL GEOLOGY Archaean rocks. The lowest horizon from which fossils have been obtained are sandstones overlying the basal conglomerate near Llanfaelog (see Fig. 34). These fossils are an Orthis resembling 0. Carausi and fragments of a trilobite regarded by Professor Hughes as a Neseuretus, but the specimen is probably Calymene Tristani, while the slates which succeed the sandstones contain graptolites of Upper Llanvirn species. Moreover, as Dr. Matley has pointed out, there is a complete absence both in Central and Northern Anglesey of the characteristic Arenig graptolites. 8 He therefore infers that the lowest fossiliferous rocks in Central Anglesey are equivalents of the beds now classed as Llanvirnian, while in the north of the island even these do not appear, and must have been overlapped by the Bala Series. Tan-y-bry Fig. 34. SECTION WEST OF LLANFAELOG (Dr. H. Hicks). Scale 6 inches to a mile. 6. Blown sand. 5. Black slates and dark flags. 4. Flagstones and sandstones. 3. Basal conglomerate (? Ordovician). 1. Granitoid Series. Llandilo Series. The typical area of these beds and the place from which they take their name is Llandilo in Carmarthen- shire, where the series originally described by Murchison as the " Llandilo flags " forms a natural group, well marked off, both in regard to its lithological characters and its fossil contents, from the underlying Llanvirn Series. Hence the Geological Survey have recently found it necessary (as already stated) to adopt the latter series as a primary division and to restrict the Llandilo Series within its original limits. 9 Near Llandilo and north-eastwards to Llangadock the group consists mainly of grey calcareous flagstones weathering to a yellowish brown. These are about 2500 feet thick, and in their lower part they contain many beds of hard bluish -grey limestone, most of them being fossiliferous and yielding both Ogygia Buchi and Asaphus tyrannus. Under these beds is a hard massive calcareous grit, the Ffairfach grit, which has a thickness of about 150 feet and also contains Asaphus tyrannus. THE ORDOVICIAN SYSTEM 123 When traced westward, however, this grit thins out and its place is taken by a band of volcanic ash, while the upper part of the flagstone series passes into black shale, the whole at the same time greatly diminishing in thickness, so that near Carmarthen the group is less than 900 feet thick and the succession is : Feet. Hendre shales with Dicellograptus divaricatus . . . 500 Llandilo flags and limestones ...... about 250 Volcanic ash with Asaphus tyrannus ..... about 70 Still farther west the " Asaphus ash " also thins out, and the flaggy limestones are condensed into one band of dark grey or black limestone which rests directly on the Llanvirnian shales. On the north side of the St. David's anticline the series has again become much thicker and is divisible into two stages as below : Feet. Black slates and flags with Ogygia Buchi and Calymene duplicate*, ......... 1000 Calcareous flags and limestone with Asaphus tyrannus and Calymene cambrensis about 800 The restricted Llandilo Series has not yet been mapped either in Merioneth or Carnarvonshire. The upper andesitic and rhyolitic ashes of the Arenig Mountains are regarded by Mr. Fearnsides as belonging to this series, and they are unconformably overlain by a fossiliferous limestone which seems to belong to the Bala Series. In the anticline of the Berwyn Mountains to the eastward Llandilo Beds appear again with a more normal facies, having a band of limestones at or near the base which are about 400 feet thick and are full of such fossils as Asaphus tyrannus, Calymene cambrensis, Orihis turgida, etc. In the Snowdonian district there is a great development of volcanic rocks which were included by Ramsay in the Bala Series, but are now thought to be largely of Llandilo age ; for upon the highest volcanic rocks of Snowdon rest black shales crowded with species of Diplograptus which seem to indicate a low horizon in the Bala Series. The lavas and ashes of the Snowdon range seem to have issued from a group of volcanic vents which probably formed volcanic islands in the Ordovician sea. Near the actual centres of eruption both ashes and lavas are found, the latter being usually rhyolites, but in other localities, presumably more distant from the volcanic vents, ashes are the prevalent deposit, and these are inters tratified with some bands of slaty shale. This volcanic series extends to the very northern extremity of Carnarvonshire. 124 STRATIGRAPHICAL GEOLOGY Bala Series. In Pembrokeshire this series was described for the first time in any detail by Messrs. Marr and Roberts in 1885. 10 They examined the succession of beds near Haverfordwest and Nar- berth, and established divisions which have been adopted by the Geological Survey officers, who have since mapped the beds through Pembroke and Carmarthen, and have at the same time fixed the horizon of passage between the Llandilo and Bala Series in this area. The limestones are seldom all developed in one locality, being lenticular beds which thin out entirely in some directions, but when each occurs it is on the same horizon and the complete succession is given as follows : Feet. ., ,. f Slade and Red hill Beds .... 2000 ' n \ Sholeshook limestone . . . - . . up to 200 | Robeston Wathen limestone . . ,, 100 Caradocian -j Mydrim shales with Dicranoyraptus . . 500 \ Mydrim limestone . . . . to 150 About 2900 The Mydrim limestone is quite local, and in its absence the horizon can only be distinguished by the change in the species of graptolites, the lower part of the Mydrim shales being characterised by Dicranograptus Nicholsoni, the middle part by Mesograptu s multidens, Dicranograptus Clingani, and Diplograptus calcaratus ; while the uppermost 100 feet yield Diplograptus truncatus and Glimacograptus minimus. The overlying limestones are most fully developed near Narberth, the lower beds consisting of fairly pure limestone, black in tint and crowded with corals of the genera Heliolites, Halysites, and Favosites, with many Brachiopods, but few trilobites. The Sholeshook limestone has quite a different assemblage, being grey or greenish in colour, impure and sandy in texture, and char- acterised by an abundant trilobite fauna including Staurocephalus globiceps, Trinucleus seticornis, Phacops Brongniarti, and Cheirurus juvenis, with many Cystideans, Gastropoda, and Cephalopoda. The overlying beds form a monotonous series of bluish-grey mudstones which are only fossiliferous on certain horizons ; they sometimes include bands of limestone containing Favosites fibrosus and sometimes more sandy beds which contain many species of Orthis, while the shales contain Leptcena sericea. Northward this Bala Series extends over a considerable area, and probably occupies large parts of Central Wales, along the anticlinal axes of the numerous flexures into which the rocks of that region have been folded, but little information about them is at present available. THE OEDOVICIAN SYSTEM 125 In North Wales they are typically developed near Bala, but the base of the series has not yet been determined by graptolitic evidence. There is a great thickness of sandy shales below the Bala limestone, and some portion of them is probably of Bala age, but the greater part of the Lower Bala Group of Sedgwick is now regarded as Llandilian. The beds which have been grouped as Bala limestone include two distinct horizons or /ones ; and the name Bala limestone is now restricted to the lower of these zones, the Rhiwlas limestone, which was formerly exposed at Ehiwlas, north of Bala, being the equivalent of the Sholeshook limestone in Pembrokeshire. Even the main mass of the Bala limestone is often split up into several beds by the intercalation of calcareous sandstone or shale, Fig. 35. SECTION THROUGH THE BALA LIMESTONE NEAR OELLIORIN, SOUTH OF BALA LAKE (Ruddy). S. Shales with Le.ptcena sericea. 7. Coarse sand, many fossils, l). Crystalline limestone. 5. Coarse sand like No. 7. 4. Sandy shales, trilobites. 8. Sandy limestone, brachiopods. L>. Volcanic ash. 1. Shales with Orthis. as at Gellygrin, to the south of Bala Lake, as described by Mr. T. Ruddy. 11 Here there is a lower limestone, 12 feet thick and containing many fossils, succeeded by sandy shales with many of the characteristic Bala , trilobites, and some feet higher a hard, massive, crystalline limestone 18 or 20 feet thick, the whole group of beds (5 to 8 in Fig. 35) being about 50 feet thick. The characteristic fossils of the Bala limestone are Trinueleus concentricus, Asaphus Poivisi, Phacops apiculatus, Homalonotus bisulcatus, Orthis vespertilio, 0. spiriferoides, and 0. biforata. Those of the Rhiwlas limestone are Cyrtoceras sonax, Orthoceras vagans, Trinucleus seticornis, Ampyx tumidus, Encrinurus sexcostatus, Staurocephalus clavifrons, Phacops Brongniarti, Leptcena tenuicincta, Sphceronites Litchi y Echinosphcera aurantium, Hemicosmites rugatus and other cystid Echinoderms. Above this limestone there is some 1500 feet of sandy shale, the upper beds of which contain graptolites of Silurian (i.e. Llandovery) species. The exact horizon at which the divisional line should be drawn has not yet been ascertained, but the thick- ness of Upper Bala Beds is probably not less than 1000 feet. 126 STRATIGRAPHICAL GEOLOGY From the above account it will be seen that the succession in the typical Bala district may be summarised as below : Feet. Upper /Grey shales and shaly sandstones .... about 1000 Bala \ Rhiwlas limestone . . . . . . 10 to 20 Lower f Bala limestone ....... 10 to 50 Bala X Grey sandy shales . . . . . perhaps 1600 About 2600 The succession in the Berwyn Mountains to the eastward has been described by Mr. D. C. Davies, and more recently by Messrs. Groom and Lake. 12 From the account given by the latter of the series exposed near Glyn Ceiriog, it appears that a great thickness of beds (the Paudy Series) below the Bala limestones must be included in the Bala Series. These beds consist of volcanic ashes, sandstones, and slates, and some of them contain many of the characteristic Bala fossils ; their total thickness being about 1600 feet. In the Ceiriog district the true Bala limestone appears to be faulted out, but elsewhere there is from 200 to 300 feet of such limestone, containing the usual trilobites, brachiopoda, and corals. Above this are calcareous shales and bands of limestone containing the fauna of the Khiwlas beds. These are succeeded by shales of a considerable thickness. To the north and north-west of the Bala district the limestones appear to thin out or to pass into a nodular calcareous tuff. In the Lleyn peninsula the place of the Bala Series is occupied by a thick set of sandy shales and mudstones of an olive-green colour, and including thin beds of felspathic grit which are largely made up of volcanic detritus. The Bala Series can be traced northward to Conway, and the succession near that place has been described by Miss G. L. Elles, 13 who has divided the beds into the following groups : Feet. Deganwy mudstones ......... 80 Bodreida mudstones or Trinucleus Beds ..... 350 p , , ( Upper, shales with Orthograptus truncatus \ uaanani i Lowerj flags witll Mesograptus multidens, Climaco- \ . 310 ( graptus, and Dicranograptus Conway Mountain Volcanic Series ..... over 2000 Miss Elles regards the Lower Cadnant shales as Llandilian, but almost all the species of graptolites recorded by her from these beds occur also in the Mydrim shales of Carmarthen, and as the total thickness of the sedimentary beds at Conway is only 740 feet, it is probable that the base of the Bala Series is to be found in the volcanic rocks. THE ORDOVICIAN SYSTEM 127 sa c >rH Jo . s 1 lsl suo 128 STRATIGRAPHICAL GEOLOGY 2. Shropshire As the Ordovician Series of South Shropshire is a continuation of the Welsh facies of the formation, it is unnecessary to give a detailed account of the local subdivisions. For a description of these the student is referred to that given by Professors Lapworth and Watts in Proc. Geol. Assoc. vol. xiii. p. 312. Only a brief summary will here be given, with a separation of the Llanvirnian by means of the graptolite zones which Miss Elles has more recently been able to establish. The Ordovician Series is brought up to the surface in Shrop- shire by the broad compound geanticline, of which the Longmynd may be regarded as the central axis (see map, Fig. 21, and section, Fig. 7). The principal area of Ordovician rocks is on the western side of the Longmynd, around Shelve and Corndon, where the general dip is westward, and the whole succession is found from Arenig to Upper Bala. On the eastern side (Caradoc area) the out- crop is narrow and only equivalents of the Bala Series occur, resting directly either on Archaean or on Cambrian rocks. From this disposition it is evident that all the lower members of the Ordovician System thinned out against the shores of a land-area consisting of the older rocks, and that this land did not sink below the Ordovician Sea until the beginning of Caradocian time. Arenig Series. The base of this division consists of a hard siliceous grit or quartzite forming the ridge called the Stiper stones ; above this are grey and green flagstones with interbedded shales, in which Professor Lapworth has found Ogygia Sehvyni, Tetragrap- tus quadribrachiatus with species of Phyllograptus and Trigono- graptus which mark them as Lower Arenig. In the higher part of the series both Didymograptus extensus and D. hirundo have been found, so that probably the whole of the Welsh Arenigjs represented and the total thickness may be 1500 feet. Llanvirn Series. The Hope shales, which have yielded Did. bifidus, form the lower part of this series and are succeeded by hard andesitic tuffs with intercalations of shale which prove the whole to have been deposited in water. These beds are overlain by flags and shales with few fossils, but at the top are the Betton shales with Didymograptus Murchisoni. The thickness of this series is probably about 2000 feet. Llandilo Series. This is represented by the Meadowtown Beds, a series of flags and flaggy limestones containing Asaphus tyrannus and Ogygia Buchi, overlain by black mudstones and shales with Leptograptus flaccidus and Nemograptus gracilis. Their thickness has not been estimated, but is probably about 1000 feet. THE ORDOVICIAN SYSTEM 129 Bala and Caradoc Series. Continuing the above succession in the Shelve district the Bala Beds are well exposed in the Spy Wood section. At the base are the Spy Wood grits and flags passing up into shales ; these beds yield Trinucleus concentricus, Orthis calligramma, Diplograptus truncatus, Climacograptus bicornis, and other fossils. Above comes the Harrington Group, a set of grey and black shales with interstratified beds of andesitic ashes and breccias or agglomerates ; the highest shales are overlain uncon- formably by the Silurian (Upper Llandovery Beds), the Hirnant Beds not being exposed. Hope Bawdier T/cfrprton Harton Wenlock S. Fig. 38. SECTION NEAR HOPE BOWDLER IN SHROPSHIRE (after E. S. Cobbold). z. Wenlock limestone. 1. Longville flags. li. Harnage shale. ?/. Wenlock shale. s. Chatwall sandstone. g. Hoar Edge grits. w. Llandovery Beds. On the eastern side of the Longmynd the succession in the Caradoc district was first worked out by Murchison and formed the type of his " Caradoc sandstone." This was subsequently proved to be only a local facies of the Bala Series. The succession as corrected by Callaway and Lapworth is as follows : Llandovery Beds (unconformable) / Trinucleus shales. Caradoc I Acton shales and limestone, sandstone, < Cheney Longville flags. 2000 feet Harnage shales. V Hoar Edge grits and sandy limestone. All these beds contain fossils, and the Acton Beds have yielded Lichas laxatus, Ampyx, Orthis Actonice, 0. flabellulum, and many corals, so it is probable that they represent the Bala limestone. The highest or Hirnant Beds are concealed by the overstep of the Silurian. 3. Cumberland and Westmoreland Ordovician rocks occupy a large area in the Lake District, which was the scene of Professor Sedgwick's earliest work on the K Fig. 39. MAP OF CUMBERLAND AND WESTMORELAND. THE ORDOVICIAN SYSTEM 131 rocks of this system (1832-36). All the mountainous region lying north-west of a line drawn from the estuary of the Duddon by Hawkshead and Ambleside to the valley of the Lowther Beck consists essentially of Ordovician rocks (see the map, Fig. 39, and section, Fig. 37). They are also found in a long narrow inlier, known as the Cross Fell inlier, on the eastern side of the Vale of Eden in the north-east of Westmoreland, bounded on the one side by Carboni- ferous limestone and on the other by New Eed Sandstone. The Ordovician rocks of this area are divisible into four series, which are roughly equivalent to those of Wales. Skidd avian Series. It was stated on p. 93 that the greater part of the great mass of slates which have generally been known as the Skiddaw slates was of Ordovician age, but as the highest portions belong to the Llanvirnian Series it is only the central portion which is strictly the equivalent of the Welsh Arenig Series, and for this part the name Skiddavian may be used. Its basement beds are certain beds of grit which are called the Watch Hill grits in the memoir of the Geological Survey on the country between Appleby, Ulles water, and Hawes water (1897). Above this horizon the slates have yielded both trilobites and graptolites, which have enabled Dr. J. E. Marr and Miss G. L. Elles 14 to recognise three zonal divisions which, however, cannot at present be correlated exactly with those of Wales. They are : 3. Upper Tetragraptus Beds, yielding Tetragraptus serra, Didymograptus indentus (var. nanus), and Phyllograptus typus. 2. Dicliograptus Beds, Dichograptus 8-branchiatus, and Did. patulus. 1. Lower Tetragraptus Beds, with Tetr. Bigsbyi. The whole series is so greatly flexured and faulted that it is impossible to form any accurate estimate of its thickness. Llanvirn Series. This is typically developed in the Cross Fell area, where the two well-known components of the series have been distinguished under the names of Ellergill and Milburn Beds respectively ; the former being black slates containing Didymograptus bifidus and Glossograptus armatus, the latter being mainly volcanic rocks (ashes and lavas) interstratified with slates which have yielded Did. Murchisoni and Diplograptus dentatus. In the Lake District beds belonging to the lower stage occur near Threlkald and Troutbeck, east of Keswick, and those of the upper stage near Ulleswater and elsewhere. Llandilo Series. This consists of an immense thickness of volcanic rocks rhyolitic and andesitic lavas with thick beds of ash and agglomerate (see map, Fig. 39), which are generally of 132 STRATIGKAPHICAL GEOLOGY green or purple colour. They are the " Green Slate and Porphyry Group " of Sedgwick, afterwards called the Borrowdale Series. Their total thickness is estimated to be 12,000 feet, but not all of this is of Llandilo age ; as above stated the lower part is evidently of Llanvirnian date, the central part must be Llandilian, and the highest part may be of Bala age ; for in the Lake District the volcanic series is directly succeeded by the Coniston limestone, but in the Cross Fell inlier there are stratified beds containing Bala fossils between the limestone and the highest rhyolite of that area. Bala Series. The thickness of the stratified portion of this series, even including those of Roman Fell, is much less than the corresponding beds in Wales (i.e. about 300 feet), but a regular zonal succession can be made out. The series has been subdivided by Dr. Marr 15 as follows : Coniston area. Cross Fell area. A ^n-'ir /Shales. Ashgill shales. Sleddale Beds -! Conglomerate. ) -p. f , -, , 1 Stile 8 End Beds. ) Dufton shales. Roman Fell Group Rhyolite. Corona Beds. The " Corona Beds " are a set of calcareous shales, limestones, and ash beds, about 100 feet in thickness, and they contain many brachiopods, such as Trematis corona, Lingula tennigranulata, Orthis testudinaria. The limestones consist largely of the tests of a small Crustacean called Beyrichia. The Applethwaite limestone consists of calcareous shales with bands of limestone and a peculiar bed of white horny limestone at the top. Overlying this is a bed of grey limestone about 5 feet thick which contains the fauna of the Rhiwlas and Sholeshook limestones, i.e. many Cystidean Echinoderms with Staurocephalus clavifrons, Phacops apiculatus, and Orthoceras vagans. The limestones are succeeded by dark blue and grey shales which yield many of the same fossils that occur at similar horizons in North Wales. 4. Scotland The southern uplands of Scotland are largely composed of Ordovician and Silurian rocks folded together into a number of anticlinal and synclinal folds, each fold having its subsidiary plications, so that the country has a complicated structure, and can only be successfully mapped by careful attention to fossils. The older views about this region were mistaken because the fossils were not sufficiently studied, and because it was supposed that ! * l || B IS- tprt 1'S ll * M 5 Is II OQ J ! 11 // w 8 ill PS fe M O'g P ^T3 o T; ...... /. 134 STRATIGEAPHICAL GEOLOGY graptolites were characteristic of Llandilo Beds and did not occur outside their limits. The true structure and succession of strata in this region was first established by Professor Lapworth, whose researches on the Girvan and Moffat districts are embodied in two detailed papers, and who afterwards summed up the structure of the whole region in another one, to which the student should refer. 16 More detailed descriptions are given in the " Geology of the Southern Uplands," Mem. Geol. Surv. 1899, The sections, Figs. 40 and 41, are taken from Lap worth's papers. The special interest of this region is that it presents us with two very different facies of sedimentation, which, however, can be traced through changing intermediate facies from the one area to the other. In Ayrshire (Girvan, etc.) the Ordovician has the ordinary facies of a formation accumulated at no great distance from a continental coast-line, consisting as it does of a considerable thickness of conglomerates, sandstones, shales, with at least one bed of limestone. The other type, found in Dumfries and Wigtownshire, is one that must have been formed in much deeper water, for there the whole system is condensed into a small thickness of dark- coloured mudstones, shales, and chert -beds. The base of it is not exposed, but the total thickness from the summit of the Bala Series down to the base of the Arenig chert-beds is only about 185 feet, as compared with a thickness of 3500 feet in Girvan. Professor Lapworth divided the Ordovician of this region into three groups or series, which are roughly equivalent to Arenig, Llandilo, and Bala, and these divisions have been adopted by the Geological Survey of Scotland. The graptolitic facies of the formation lends itself to easy zonal subdivision and to correlation with other areas, so that the names given to the middle and upper parts of this succession, i.e. Glenkiln shales and Hartfelt shales, have been widely used as if they were synonymous with Llandilo and Bala. The progress of discovery in Wales, however, has shown that the divisional line between the Glenkiln and Hartfell shale is not quite on the same horizon as that taken in South Wales between the Llandilo and Bala. Moreover, nothing comparable to the Llanvirn Series has yet been identified in Scotland. No shales yielding Didymograptus bifidus or Did. Murchisoni have been found. In the one area there is a distinct unconformity at the base of the Llandilo with a conglomerate above the break ; in the other there is merely an abrupt change from one kind of sediment to another. It is evident, however, that a chapter of the geological record is missing in this region, and the following tabular view gives the established divi- sions without any attempt to correlate them with those of Wales. THE OBDOVICIAN SYSTEM 135 Girvan District. Moflat District. ( DRUMMUCK BEDS, 400 feet b. Shales with Staurocephalus. a. Shales with Trinucleus seticornis. SHALLOCH FLAGS, 800 feet Flagstones and shales with Dicellograptus truncatus. WHITEHOUSE BEDS, 300 feet b. Beds with Dicellograptus complanatus. a. Beds with Leptogr.flacddus. ARDWELL FLAGS, 1000 feet b. Beds with DicTano. ramosus and Diplograptus rugosus. a. Beds with Crypto, tricornis smdDiplograptusfoliaceus. BALCLATCIIIE BEDS, 100 feet Grits and sandstones with Climacograptus bicornis and Oryptogr. tricornis. Shales with Glossograptus Hincksi. HARTFELL SHALES, 100 feet 6. Zone of Dicellograptus an- ceps and D. truncatus. 5. Barren nnidstones. 4. Zone of Dicello. complanatus. 3. Zone of Pleurograptus line- aris and Lepto. flaccidus. 2. Zone of Dicranograptus Clingani and D. ramosus. 1. Zone of Climacogr. Wilsoni and Cryptogr. tricornis. BENAN CONGLOMERATE, 500 feet. STINCHAR GROUP, 100 feet b. Shales with Didymograptus superstes and Dicello- graptus sextans, a. Limestones with Maclurea Logani. KIRKLAND BEDS, 200 feet Sandstones and conglomerates with Orthis confinis. Unconformity here. GLENKILN SHALES, 50 feet- Black shales with Didymo- graptus superstes, Dicel. sextans, and Diplograptus dentatus. Yellow mudstones with chert- bands. Black shales with Ccenograptus gracilis and Didymo- graptus superstes. RADIOLARIAN CHERTS, red and grey cherts, mudstones, and volcanic tuffs, 70 feet. Black shales of Bennane Head with Tetragraptus quadribrachiatus, 4 feet. VOLCANIC ROCKS, lavas and tuffs with thin shales containing Tetr. quadribrachiatus, 1500 feet seen. RADIOLARIAN CHERT BEDS Mudstones and shales with layers and lenticular nodules of chert, 150 to 200 feet. Volcanic tuffs at Trowdale, 150 feet. Base not seen. 136 STRATTGRAPHICAL GEOLOGY Arenig 1 Series. From this table it will be seen that so far as the beds are exposed the greater part consists of volcanic rocks, but that the highest member is a group of mudstones and shales inter- stratified with lenticular layers of red and grey chert which are full of the remains of Radiolaria. From this fact it has been inferred that these beds were oceanic deposits formed under a great depth of water and at a great distance from land. Elsewhere, however, I have shown that such an inference cannot be drawn from the mere occurrence of Eadiolaria (see Building of the British Isles, 3rd edition (1911), pp. 12 and 72). The narrow compass, within which the beds representing the Ordovician System are compressed in the Moffat district, is shown in Fig. 42, which is reduced from a section in the Survey Memoir. Fig. 42. SECTION THROUGH TEOWDALE IN THE URR VALLEY. (Scale about 750 feet to an inch.) Silurian/ 6 ' Gala Beds - nrrinviniim f 4 - Hartfell shales. 2, 1. Chert Beds and in \ 5. Birkhill shales. ician ) 3. Glenkiln shales. Volcanic tuffs. In the Barr Series the Stinchar limestone is an interesting horizon, for many of its fossils are not found anywhere in England, but occur in the Trenton limestone of North America ; such are Maclurea Logani, M. magia, Murchisonia angustata, the sponge- like coral Tetradium, and Saccamina Carteri ; there are also many corals (Lyopora favosa, etc.) and brachiopods, with Lichas sexcostatus ? Illcenus latus, I. Bowmanni, and Calymene Blumenbachi, most of which range from Upper Llandilo to Bala Beds. The Balclatchie Beds are included in the Barr Series by Professor Lapworth, but he speaks of them as forming a transitional band or zone, and many of the graptolites range into the Hartfell shales. Messrs. Peach and Home place these beds in the Ardmillan or Bala Series, but leave the Benan conglomerate in the Barr Series. It is possible that the plane of divisions should be drawn between the Balclatchie shales and the overlying grits. The Ardmillan Series, which is the equivalent of the Welsh Bala Beds, consists of a great thickness of alternating flagstones, THE ORDOVICIAN SYSTEM 137 sandstones, mudstones, and shales. It is interesting to note that near the top of the highest (Drummuck) group there is a thin bed of grit which has yielded Staurocephalus gloUceps, Trinucleus Buck- landi, a species of Palwaster and other fossils, associated with shales containing Dicellograptus anceps and Diplograptus truncatus. The graptolites show that these beds are equivalent to the highest part of the Hartfell shale, but there are no limestones comparable with those of Bala and Coniston. Beds of the Moffat type, dark shales and mudstones, have been traced all across Southern Scotland, from the Lammermuir Hills through the northern parts of Selkirk and Dumfries and the central parts of Kirkcudbright and Wigtown. Beds of an inter- mediate type form a broad belt to the north of this, varying from 5 to 15 miles in width, from the Moorfoot Hills to the northern part of Wigtownshire. No deposits of Ordovician age have yet been proved to exist in the Central or Northern parts of Scotland. The only rocks which may possibly be of that age are the green igneous rocks, green shales, and jasper beds which occur between lines of fault along the southern border of the Central Highlands ; but these are now believed to be Upper Cambrian (see p. 97). 5. Ireland Ordovician rocks occur in several parts of Ireland, as will be seen by a reference to the map in Chapter IX. ; and they must originally have been deposited over nearly the whole of the Irish region. The largest area is that which spreads through Waterford, Wexford, Wicklow, and Kildare, and is penetrated by the large granitic massif which occupies so much of Carlow and Wicklow. The sediments of this area have much resemblance to those of Wales, and may be regarded as a continuation of the latter, but they are much more altered by subsequent igneous intrusions and by concomitant earth pressures, so that they are greatly plicated, crushed, and cleaved. The tract in the north, extending from Down to County Cavan, presents a facies similar to that of the condensed series of Southern Scotland, and the same type of graptolitic shale recurs in the south-west round Lough Derg on the borders of Clare and Tipperary. In the north-west, however (Galway and Mayo), there is a very different set of beds of a more littoral character. Eastern Pacies. The eastern parts of this area were revised by Messrs. Egan and M 'Henry of the Geological Survey during the years 1898 and 1899. l ~ They found that there was apparently 138 STRATIGRAPHICAL GEOLOGY a conformable passage from the Cambrian into the Ordovician, though the beds along the line of junction are everywhere so mingled together by subsequent plication and crushing that no definite boundary could be drawn between them. The lowest Ordovician Beds are a set of banded slates and fine grits which are generally known as the Kibband Series. They have yielded fossils near Arklow, Courtown, and Kilrea, which included species of Tetragraptus, Phyllograptus, and Didymograptus, and a Bryo- graptus like Kjerulfi. Hence they appear to represent a Tremadoc horizon as well as the Arenig and Llanvirn Series. Above the Eibband Series are black slates which contain a graptolite fauna of Llandilo age, i.e. Dicranograptus Nicholsoni, D. ramosus, Ccenograptus gracilis, Diplograptus mucronatus, and Lane. Chair of Kildare. Shales and \ Andesites and ftasalts Shales \ rrboniferonsGrits \ werlainby Limestone] Litnest imest.o-verlain * sft Dnft Limestone and Grits and Grits ^ Drift y Drift Fig. 43. SECTION ACROSS THE CHAIR OF KILDARE (Reynolds and Gardiner). Horizontal scale, 6 inches to a mile. Leptograptus flaccidus. These beds are succeeded by shales and limestones of Bala age which have yielded many fossils. At Portraine, on the coast north of Dublin Bay, there is a small inlier of Ordovician, and the cliffs show an interesting section of Bala deposits. The succession has been described by Messrs. Gardiner and Roberts, 18 and consists of the following ; (1) andesitic lavas, (2) coarse ashy conglomerate, (3) calcareous ash beds with layers of limestone and shale, (4) limestones full of corals, (5) compact grey limestone with many trilobites and brachiopods, including Illcenus Bowmanni, Staurocephalus sp., Trinucleus seticornis, and Sphcerexochus mirus. Thus it appears that after a series of volcanic eruptions in the near vicinity, the water became clear and suitable for the growth of corals, which produced layers and beds of coralliferous limestone, but nothing at all resembling a coral reef. Some change then occurred, probably subsidence, causing a deepening of the water, in consequence of which the corals died out, and their place was taken by trilobites and brachiopods, whose remains are the chief components of the upper limestone. A more extensive exposure of the series is found in another inlier THE OKDOVICIAN SYSTEM 139 among the Carboniferous rocks of Kildare. The succession there is shown in Fig. 43, reproduced by permission of Messrs. Eeynolds and Gardiner 19 who have described the rocks seen. All the beds belong to the Bala Series, and have a thickness of about 2200 feet, the main limestone being about 550 feet thick, and containing many trilobites and brachiopods ; the fauna is that of the Bala limestone, nor do the overlying shales yield Ashgillian species, though beds of a higher horizon may be present. Northern Facies. In the counties Down, Armagh, Monaghan, and Cavan the thickness of the Ordovician Beds is reduced to a band of black shales and mudstones exactly resembling those of Moffat and the southern Scottish belt. Many years ago, from exposures south of Belfast Lough, Mr. Swanston obtained all the characteristic graptolites of the Glenkiln and Hartfell shales. There are also some narrow inlying anticlinal exposures among the neighbouring Silurians, and in one of these near Slane (County Meath) Mr. M'Henry has found some layers of banded Eadiolarian chert and dark shales yielding Lower Llandilo or Arenig graptolites. 20 The main band of Ordovician runs from north-east to south- west, and in Monaghan and Cavan is from 3 to 9 miles broad. The strata pass north-westward below the Carboniferous rocks, and a portion of them emerges on the southern side of the Archaean ridge near Pomeroy in Tyrone. The Arenig and Llandilo Series appear to have thinned out and to be overlapped by the Bala deposits, which rest directly on the old rocks. The beds are about 350 feet thick and have a conglomerate at the base, overlain by sandstones, flagstones, and mudstones which have yielded a fauna comparable to that of the Drummuck Beds of Ayrshire (Ashgillian) ; moreover they are conformably succeeded by Silurian strata. 21 Western Facies. Ordovician rocks occupy a considerable area on the borders of Mayo and Galway between Killary Harbour and Lough Mask, and there is another tract between Clew Bay and Castlebar. The most complete succession is found in the cliffs and escarpments of the tracts along Killary Harbour. On the south side the lowest beds seen are black shales and cherts about 60 feet thick, containing four species of Tetragr'aptus with Dichograptus 8-brachiatus. These beds are succeeded by coarse conglomeratic grits with bands of shale which have yielded Didymograptus extensus and Diplograptus dentatus ; this group is 2500 feet thick, and the grits contain pebbles of red granite and quartz-felsite like those which occur in the Archaean rocks to the north and south. 22 On the south side of the Harbour these grits seem to be 140 STRATIGRAPHICAL GEOLOGY replaced by the Doolough green slates and grits which contain the same graptolites. They pass up into the Mweelrea grits red and green felspathic grits which in the lower 3000 feet include bands of shale yielding Ogygia Buchi and brachiopods, but no graptolites ; they are evidently of Llandiliaii age. In conformable succession is an enormous and monotonous series of reddish felspathic grits, without a single parting of shale and quite destitute of fossils, but they are probably of Bala age. Thus we have the following correlation : Feet. Bala / Mweelrea /Upper 9,000 Llandilo\ Grits \Lovver 3,000 T Innvirn f Leenane Grits and l 9 500 Llan I Doolough Slates / ' ' ' ' 2 ' 50( A . 1 Bencratf shales 60 Arenig I Lower bedg not geeu e t m at i east 500 15,000 To the eastward in the Glensaul and Tourmakeady districts a much greater thickness of Arenig Beds is exposed, and they consist partly of coarse grits and conglomerates, partly of a thick mass of volcanic materials, which lower include some fossiliferous beds of limestone and shale. Omitting the felsitic lavas, most of which are intrusive, the succession described by Messrs. Gardiner and Eeynolds in this area may be summarised as follows : 23 Feet. Coarse conglomerate and sandstone (very thick) Calcareous tuffs, with bands of limestone . . . . . 1 300 Massive agglomerate of felsite- fragments . . . . .750 Volcanic tuffs with a shale yielding Didymograptus hirundo and D. gibberulus ......... 150 Coarse felspathic grits and tuffs . . . . . . .700 Coarse quartzose and some felspathic grits . . . . .150 Fine siliceous grits including a black shale with Tetragrapti and cherts containing Radiolaria . . . . . .150 Coarse grits and conglomerates .... seen for over 800 3000 It will be seen that the base of the Arenig Series is not exposed, but as the lowest conglomerates contain pebbles of gneiss, mica- schist, and hard grit they probably rest upon Archaean rocks. The lowest fossiliferous band has yielded five species of Tetra- graptus, associated with Didymograptus extensus, D, nanus, and small forms of D. bifidus. The only other species found at the higher horizon is Diplograptus dentatus, and the fauna might be Arenig or Llanvirn. The limestones contain a peculiar trilobite fauna, which leads Mr. Cowper Keed to correlate them with the THE OKDOVICIAN SYSTEM 141 Cystidean and Asaphus limestones (Llanvirnian) of Sweden and Eussia, species common to both being Nileus armadillo and Chasmops cf. Odini ; while others occur in Canada, the genera Bathyurus and Bathyurellus being essentially American. The overlying conglomerate has the appearance of lying unconformably on these upper beds, but it is lithologically similar to the Leenane grits, and is probably of the same age. D. SOME EUROPEAN AREAS 1. France To British geologists the most important region in France where Ordovician rocks are exposed is that of Brittany, Normandy, and Anjou. In this region they lie in the long synclinal troughs which run roughly from west to east and are most fully developed in the central part of the region, from Grandchamp north of Vannes by Redon and Chateaubriand to Angers in Maine-et- Loire. 24 Along this tract the following general succession has been established (see Fig. 44), though some of the subdivisions are not continuous throughout its extent. Feet. -D 1 /Limestone of Rozan with Orthis actonce . . \ t Sandstones of Redon and St. Germain (Diplograpti] j 00 T1 ,.-, ( The Riadan slates Trinucleus Ponqerardi Llandll \ Chatelier sandstone ... \-5QO Llanvirn The Sion slates with Placoparia 500 200 . /The Armorican sandstone & \Felspathic sandstone and conglomerate The felspathic sandstones (Gres felspathique) are poorly developed in the south of Brittany, but are very thick in the north, from 500 to 1000 feet, and have clearly been formed from the erosion of Archaean and granitic rocks during the interval between the Cambrian and Ordovician sedimentation in this part of France. They may indeed be partly of Tremadoc age (see p. 99). The Armorican sandstone (Gres armoricain) consists of hard white sandstones with few fossils, chiefly Lamellibranchs of the genera Eedonia, Gtenodonta, and Actinodonta and annelid tracks (Scolithes, Bilobites). It also is much thicker along the northern outcrops than along the southern, where it is sometimes less than 100 feet. Eastward, in Anjou, it is largely represented by shales. The Sion slates generally have a bed of oolitic ironstone at their base, and the slates contain graptolites of the geminus and Murchisoni types with the trilobites Placoparia Zippei, Asaphus Guettardi, and Galymene Tristani. 142 STRATIGRAPHICAL GEOLOGY In the south of Brittany the Llandilo Series begins with the sandstone of Chatelier (see Fig. 44), which is overlain near Poligny by the black slates of Eiadan, the latter containing Trinucleus Pongerardi, Acidaspis Buchi, Illcenus Beaumonti, and Calymene arago. In Northern Brittany the series is largely composed of sandstone. The sandstones of Redon and St. Germain-sur-Ille represent part of the Bala Series, but as they contain few fossils and pass up into Silurian sandstones it is difficult to separate the one from the other, except in the west (Crozon, near Brest), where the two sand- stones are separated by a band of limestone yielding Orthis Actonice and Illcenus Munieri. In Normandy the facies is somewhat different and the general succession is as follows : ' 25 Feet. Bala Green slates with Trinucleus Grenieri .... 200 Llandilo Gres de May with Homalonotus, etc. .... 500 Llanvirn Blue slates with Calymene Tristani .... 300 A . /Gres armoricain ........ 400 Arenig |Q res felspathique, thinning out S. and E. . . variable It will be seen that in most parts of this area the total thick- ness is not more than 1500 feet, for the Gres felspathique is only found in the west of the Cotentin and is overlapped to south and west by the Gres armoricain. The overlying Calymene slates correspond with those of Sion and Angers and generally have a bed of ironstone at the base. The Gres de May has recently been subdivided into 4 zones characterised by certain species ; these in descending order are : 4. Sandstones with Conularia pyramidata. 3. Shales and grits with Trinucleus Bureaui and Cal. Tristani. 2. Sandstone with Homalonotus ( Vicaryi and 3 other sp. ). 1. Sandstones with Dalmanites armoricanus and Homalonotus. Orthis budleighensis is another common fossil in the lower part of the Gres de May. The highest member of the Ordovician, the shale with Trinucleus Grenieri, is not always present ; it is found near Cap la Hague and has a bed of sandstone above it, and thus it is possible that elsewhere it is represented by some of the highest beds of the Gres de May. In connection with the Ordovician of Normandy brief mention may be made of the area of Ordovician rocks which has long been known to exist in Cornwall. The area occupied by the older Palaeozoic rocks in the south-west of Cornwall has recently been resurveyed, but only a small portion of it has been proved to be THE OKDOVICIAN SYSTEM 143 'a o: N rn .2 | | 5 1 a OD rf I! .P CCO2 ^" a' a' bb G w t I S3 -8 B a ?l I ^5 5 S , 11 g 5 2 cop . 1 111 144 STRATIGRAPHICAL GEOLOGY Ordovician by the occurrence of fossils ; this is the Gorran quartzite near Mevagissey, in which Calymene Tristani has been found and which may therefore represent a part of the Ores de May. Following this quartzite in what is believed to be descending order are the following divisions : 26 Veryan Beds, bluish shales with thin beds of limestone and of chert with Radiolaria. Portscatho Beds, bluish slates and fine gritty flags. Falmouth Beds, green and grey slates with layers of sandstone. Mylor Series, banded slates with alternating layers of fine siliceous grit. The base is not exposed and no fossils have been found in any of these beds, so that their age is quite uncertain, and their general facies is different from any series in Normandy or Brittany. There is, however, a certain amount of resemblance between them and the Cambro-Ordovician sequence of Belgium (as described below). 2. Spain and Portugal Kocks of Ordovician age are found in many parts of Spain and Portugal, and the facies which they present in the northern and * central parts of these countries is similar to that of Brittany. This facies is well exposed in the cliffs of Asturia on the north coast of Spain. The succession there seen has been described by Professor C. Barrois 27 and compared with that of Brittany Asturia. Brittany. Calcareous shale of el Homo with Endoceras. \ Angers s i a t e s. Luarca slates with Calymene Tristani. ) White and green grits and dark slates (Cabo Busto). Ores armoricain. Variegated grits with Lingulella Heberti. Gres felspathique. The strata are much flexured and faulted, so that accurate estimates of the thickness could not be made, but Dr. Barrois con- sidered the two lower sets of arenaceous deposits to have a thick- ness of from 1200 to 1500 feet. The Luarca slates are probably about 300 feet, and the highest beds about 50 feet. The similar succession found at Almaden in Ciudad Keal has been described by de Prado (Bull Soc. Geol, France, ser. 2, vol. xii. p. 91), and that of Portugal by Delgado in Terrenes Palceozoicos do Portugal (Lisboa, 1876). 3. Belgium Although the exposures of Ordovician rocks in Belgium are very small, they are interesting because they and another small THE ORDOVICIAN SYSTEM 145 tract in Thuringia are the only exposed areas of this system between the extreme west of Europe and Bohemia. In Belgium it is probable that Ordovician rocks form a continuous sub- terranean outcrop, running for some distance from west to east beneath the Cretaceous deposits south of Ostende and Brussels, but they are only exposed in two or three valleys where the rivers have cut down to them through the superincumbent strata. 28 One of these tracts is round the highest tributaries of the river Senne, south of Brussels ; another is at the head of the Dyle valley, east of Nivelles ; and a third forms a long band north of Namur, from Gembloux on the west to near Huy on the Meuse. The rocks are much compressed, cleaved into slates, and folded into steep-sided anticlines and synclines so that only a general succession has been made out by the discovery of fossils at certain localities. Thus in Brabant the Llanvirn and Arenig faunas have not yet been detected ; in Namur a Llanvirn horizon has been found, but the Arenig does not seem to be exposed. The lowest beds near Huy are black satiny slates yielding dEglina binodosa and Didymograptus Murchisoni. Near Oxhe to the west are black micaceous grits and slates containing Homa- lonotus bisulcatus, Trinucleus favus, and Orthis redux (probably Llandilo), while at Fosse a Bala fauna has been found in gritty slates with bands of felspathic sandstone ; the fossils including Trinucleus seticornis and Calymene incerta. The same fossils with Orthis Actonice have been found near Gembloux. 4. Bohemia In the Bohemian basin, the position of which is shown on Stanford's map of Central Europe, the Ordovician succession is fairly complete and has a total thickness of about 3000 feet. It is wholly comprised in the Stage D of Barrande, who indicated his systems by letters without giving them special names (see Fig. 45). His sequence is as follows : D 5. Grey and green shales with Ainpyx Portlocki, Agnostus tardus, Sphcerexochus latens, and Dicellograptus anceps. D 4. Sandy micaceous shales, with species of Trinucleus, ^glina, Calymene, and Cheirurus. D 3. Black shales with Asaphus nobilis and Trinucleus Gfoldfussi. D 2. Hard grits and shales with Asaphus ingens, Placoparia grandis, Acidaspis Buchi, etc. D 1, g. Black shales with Placoparia Zippei, Barrandea crassa, Ogygia desiderata, Eedonia, and Didymograpti. Below the last beds and dividing them from the Upper Cambrian L 146 STRATIGRAPHICAL GEOLOGY are a series of volcanic rocks porphyritic lavas with massive beds of ash and agglomerate, which seem to occupy the place of the Arenig Series of Wales. The groups D 1 and 2 represent the Llaiivim Series, and might be called the Placoparia Beds. D 3 occupies the place of the Llandilo Series, though it is only a thin band, and part if not the whole of D 4 is probably of the same age, but D 5 is undoubtedly the equivalent of our Bala Series ; its highest portion, however, seems to have been removed by erosion before the deposition of the Silurian. 5. Scandinavia When we pass from Central to Northern Europe we come to a region which was evidently at a considerable distance from the contemporaneous lands and continents, and we find that sedimenta- tion was confined to deposits of limestone and shale. Some geologists have too hastily inferred that these deposits are formed under a great depth of water, but the facts do not warrant such a conclusion, only pointing to the existence of an open sea far away from continental land and favourable for the formation of shelly and crinoidal limestones. It was, however, clearly traversed by currents which carried a certain amount of fine argillaceous sediment, and this in some areas completely prevented the con- tinuous formation of limestones. In Norway Ordovician rocks are believed to occupy con- siderable areas, but in most places they are more or less meta- morphosed into schists and are plicated with similarly altered Cambrian and Silurian rocks, so that their stratigraphy is difficult to unravel. It is only in the south around Christiania (see the map, Fig. 10) that an undisturbed sequence is found. In the south of Sweden also there are several tracts (marked on Fig. 10) where a similar succession can be studied ; the chief of these are (1) Dalarne, north-west of Stockholm ; (2) Nerike, west of Stockholm ; (3) in the Gothlands, on each side of Lake Vettern ; (4) in Skima, the most southerly part of Sweden ; and (5) in the islands of Oland and Gottland (Visby), in the Baltic. Within this area also the succession varies, large portions which consist of graptolitic shales in the west and south being replaced by limestones towards the east, and also toward the north in the upper part of the series. The general succession and correlation are as follow : 29 THE ORDOVICIAN SYSTEM 147 Argillaceous facies. Calcareous facies. IBrachiopod shales with Stauro- \ cephalus j- Leptsena limestone of Dalarne. Trinucleus shales (T. anceps) J Zone of Climacograptus rugosus ,, Dicranograptus Clin- Chasmops limestone (C. mac- gani rums). J f * Beyrichia limestone. Diplograptus mucrona- | tus V Limestone with Ancistroceras. ^ ^ ,, Glossograptus Hincksi } ^ , > Lst. with Illsenus ccntaurus. II ^^o^^^m^ I l ^ Sphnites and trilo- bites. Isograptus gibberulus } ,, Megalaspis limbata Phyllograptus angusti- \ and many Cephalo- folius pods. Symphysurus. There are several important points to be noticed in regard to this sequence. The first is that the whole succession, though quite complete and capable of minute subdivision, is only from 300 to 400 feet thick. The second point is that the beds are horizontal and undisturbed. The third is that there is an upward passage from the Cambrian (see p. 102) to the lowest bed which can be classed as Arenig, so that the Swedish geologists prefer to include the Tremadoc in the Ordovician System. Fourthly there is the replacement of shales by limestones. In the Christiania and Westgothland districts, and in the greater part of Scania, the Arenig and Llanvirn Series are represented by the Lower Graptolite shales about 150 feet thick, but parts of these are replaced by limestone eastward till in Oland the whole consists of limestone. This limestone formation is generally known as the Orthoceras limestone from the abundance of Orthoceratida3, and it is only in recent years that its complete equivalence to the graptolitic shales has been established by Moberg, Tornebohm, Hedstrom, and Wiman. In the same way the graptolitic shales with a Glenkiln species are replaced eastward by an Ancistroceras limestone and a higher band principally composed of the tests of Beyrichia (a small Crustacean), while the Chasmops limestone is a still higher horizon occurring in the Christiania district. In Scania the Bala Series is represented by shales with 148 STRATIGRAPHICAL GEOLOGY Hartfell species of graptolites, but in Dalarne it is replaced by a limestone full of Leptcena sericea. The brachiopod shales appear to represent the Upper Bala or Ashgillian division, their lower part in Scania consisting of a greenish argillaceous limestone and the upper of blue flaggy shales. 6. Russia Ordovician and Silurian rocks occupy a broad area along the south side of the Gulf of Finland, both in the islands of Dago and Oesel, and on the mainland as far east as the southern shore of Lake Ladoga. Part of this area is shown in Fig. 10. The deposits are chiefly limestones with soft marls and shales, quite unaltered but slightly inclined to the south, and their total thickness is about 300 feet. 30 The succession is as follows : Feet. (8. J7. Borkholm Beds (marls and limestone) BalaX 7. Lyckholm limestones (corals and trilobites) . [6. Wesenberg limestone and marls Llandilo f 5. Jewe limestones and shale, Ohasmops bucculenla and -j 4. Limestones with Chasmops Odini . Llanvirn [3. Echinosphserites limestone .... ^' Vaginatus limestone (Orthoceras vaginatus) 1. Glauconitic limestone with Megalaspis . 40 50 30 100 40 30 10 30 Of these beds Nos. 1 and 2 correspond with the lower part of the Orthoceras limestone of Sweden, and the next three with its upper part. The age of No. 6 is uncertain, but it contains species of Chasmops and Encrinurus. The higher beds represent the Trinucleus shales and Leptsena limestone. It will be seen that the whole series is essentially calcareous and just as condensed as the Swedish facies, having evidently been formed very slowly in clear water and far from land. E. CONDITIONS OF FORMATION From the facts mentioned in the preceding pages it may be inferred that so far as the European region is concerned the geographical conditions of Ordovician time were on the whole similar to those of Cambrian time. Scandinavia and Eussia were still farther removed from any large tract of land, and the continued slow subsidence of that area favoured the continual deposition of fine mud or the formation of shelly limestones in water of considerable but not oceanic depth. The much greater thickness of deposit in Bohemia shows that in passing southwards we approach one of the larger Ordovician THE ORDOVICIAN SYSTEM 149 land-areas, and the position of this land seems to be fairly well indicated by the fact that the fauna found in Bohemia differs greatly not only from that of Northern Europe but also from those of the French and British Ordovician. Thus the number of species common to the Ordovician of Britain and Sweden is much larger than the number common to Sweden and Bohemia. It seems clear, therefore, that the Bohemian area belonged to a different life province, and that the land from which the Bohemian sediments were derived lay to the south and south-east of that country, over the region now occupied by Hungary, Roumania, Turkey, the Black Sea, and Asia Minor. It is probable that nearly the whole of Central and Western Europe was covered by the Ordovician Sea, though there may have been some islands of a fair size. We have indeed evidence of one such island in the centre of the English area, for the absence of Arenig, Llanvirn, and Llandilo rocks to the west of the Longmynd proves that part of Shropshire was land during the greater portion of the period, while the presence of the Caradoc sandstone shows that it was submerged (in part at least) toward the close of Ordovician time. Of the size and extent of this island we have little to guide us in forming an opinion, but it cannot have reached very far north since Ordovician strata occur at Settle in Yorkshire, and its chief extension was probably eastward. Where the Cambrian is exposed in the Midland counties, as in the Lickey Hills (Worcestershire) and near Nuneaton (Warwickshire), we find it succeeded either by Silurian or Carboniferous rocks, so that probably the whole of the Midland area from Shropshire to Northamptonshire was land during the Ordovician period. Another island may have existed in the area of the Ardennes where Devonian rocks rest directly on Cambrian. With regard to the Atlantic region the existence of continental land to the west and north-west of Ireland at this time is proved not only by the enormous thickness of Ordovician deposits in Mayo, but also by the presence of American forms of trilobites in the Arenig limestones of that area (see p. 141). It will be re- membered that the Cambrian limestones of Scotland contain a fauna of American affinities, and this fauna is supposed to have migrated along the shores of a North Atlantic land. In Arenig time this land seems to have been greatly enlarged so as to touch the north-west of Ireland, and thus to afford similar facilities for the migration of American species to the Irish area. How far this Atlantic continent extended over Scotland we do not know at present, but though the old supposition that parts of 150 STRATIGRAPHICAL GEOLOGY the metamorphic complex of the Central Highlands are Ordovician rocks is still held tenaciously by certain authorities the progress of geological research has made this theory more and more improbable. There is therefore good reason to suppose that no such rocks are there, and if this is so the Atlantic land probably extended over the greater part of Scotland, as indeed I have ventured to indicate elsewhere (Building of the British Isles'). Lastly there must also have been a land-area of considerable size in a -more southern part of the Atlantic to the west of France and Spain, for it is only from that direction that the materials constituting the Ordovician deposits of those countries can have been derived. The felspathic sandstones and conglomerates of Brittany show that part of this land extended very near to France, and also that it consisted very largely of granitic and Archaean rocks. The greater part of the Ordovician sediments are of the normal detrital character conglomerates, sandstones, flagstones, grits, mudstones, and shales, the materials of which must have been derived from the land tracts which existed in the positions above indicated. Limestones were also formed wherever the water was clear enough to permit shell-bearing creatures to live and flourish in great numbers, so that their shells and tests accumulated to form layers on the sea-floor. The chief components of such limestones are Mollusca, Brachiopoda, Crustacea, Bryozoa, and Cystidean Echinoderms. A few genera of corals also contributed to their formation, but do not seem to have been very abundant. Besides these sediments some of the Arenig shales include layers and nodules of chert which contain remains of Kadiolaria, and concerning these some erroneous ideas have arisen as to the conditions indicated by the presence of Radiolaria. This matter has been discussed in the third edition of my Building of the British Isles (1911), but it may be well to mention here that those, however, who claim that the radiolarian cherts of this region are " oceanic deposits " jump to a very important conclusion from a very small amount of evidence. Only two reasons have been given for such an inference : (1) the existence of the Radiolaria, which at the present day form deposits on oceanic floors, (2) the small thickness of the Arenig-Llandilo Beds ; but neither of these can be taken as evidence of oceanic conditions. As a matter of fact Radiolaria occur in waters of all depths, and have been found in deposits of many ages, notably in the Oxfordian and in the Lower Eocene of Northern France, in each case combined with sponge spicules to form a peculiar siliceous rock termed gaize. THE ORDOVICIAN SYSTEM 151 Thus the cherts and black shales of the Arenig Series may be compared with the gaize and black clays of the Oxfordian, but have no resemblance to modern oceanic deposits, which are red, yellow, grey, and white, but never black. There are no fine red clays nor foraminiferal limestones among the Scottish Arenig and Llandilo Series, like those which are associated with Tertiary oceanic deposits in Barbados and elsewhere, and the utmost that can safely be inferred from the character and small thickness of the Scottish deposits is that they were formed in clearer water and at a greater distance from land than the Arenigs of Wales. Finally, it should be noted that in the British area this was a period of great volcanic activity, volcanic vents being opened from time to time on the sea-floor, and probably in most cases they were built up into volcanic islands from which subaerial eruptions took place. This happened in one part of the region or another throughout the whole period. Thus there are volcanic rocks of Arenig age in Pembroke, Merioneth, Scotland, and Ireland (Mayo) ; of Llanvirn age in Merioneth and Cumberland ; of Llandilo age in Merioneth, Carnarvon, and Cumberland ; of Bala age in Waterford and Dublin. EEFERENCES 1 H. Hicks, Quart. Journ. Geol. Soc. vol. xxxi. p. 167 (1875). 2 H. Hicks, Pop. Sci. Revieiv for 1881, p. 289 ; see also Proc. Geol. Assoc. vol. vii. p. 291 (1882). 3 Marr and Roberts, Quart. Journ. Geol. Soc. vol. xli. p. 476 (1885). 4 Australian, "Geology of the Country round Carmarthen," Mem. Geol. 5 W. G. Fearnsicles, " North and Central Wales," in Geology in the Field, published by Geol. Assoc. (Stanford 1910), p. 786. 6 W. G. Fearnsides, Quart. Journ. Geol. Soc. vol. Ixi. p. 608 (1905). 7 Miss G. L. Elles in Geol. Mag. for 1904, p. 199. 8 C. Matley, Quart. Journ. Geol. Soc. 9 See Memoirs of Geol. Survey on Llandilo, Ammanford, and Carmarthen. 10 Marr and Roberts (op. cit. above). 11 T. Ruddy, Quart. Journ. Geol. Soc. vol. xxxv. p. 200. 12 Groom and Lake, Quart. Journ. Geol. Soc. vol. xlix. p. 426. 13 G. L. Elles, Quart. Journ. Geol. Soc. vol. Ivi. p. 169 (1909). 14 J. E. Marr, Geol. Mag. for 1894, p. 122. 15 J. E. Marr, Geol. Mag. for 1892, p. 97. 16 C. Lapworth in Quart. Journ. Geol. Soc. vol. xxxiv. p. 25, and vol. xxxviii. p. 538 ; and in Geol. Mag. for 1889, pp. 20 and 69. 17 Messrs. Egan and M 'Henry, Sum. Prog. Geol. Survey for 1899 and 1900. 18 Gardiner and Roberts, Quart. Journ. Geol. Soc. vol. liii. p. 520. 19 Reynolds and Gardiner, Quart. Journ. Geol. Soc. vol. lii. p. 523. 20 Summary Prog. Geol. Survey for 1896, p. 48. 21 Carruthers and Muff, Irish Naturalist for 1909, p. 7. 152 STRATIGRAPHICAL GEOLOGY 22 J. R. Kilroe, Proc. Roy. Irish Acad. vol. xxvi. p. 129 (1907). 23 Gardiner and Reynolds, Quart. Journ. Oeol. Soc. vol. Ixv. p. 104, and vol. Ixvi. p. 253. 24 Barrois in Proc. Geol. Assoc. vol. xvi. p. 108. 25 A. Bigot, Bull. Soc. Geol. France, s. 4, tome iv. p. 861. 26 "Geology of Mevagissey," Mem. Geol. S?mv(1907). 27 Barrois, "Terrains anciens des Asturies," Mem. Soc. Geol. Nord, 1882. 28 C. Malaise, Bull. Acad. Roy. de Belg. for 1910. 29 See W. G. Fearnsides in Geol. Mag. for 1907, pp. 186 and 257 ; and the Livret-guides of the International Geol. Congress for 1910. 30 0. Schmidt, Quart. Journ. Geol. Soc. vol. xxxviii. p. 514 (1882). CHAPTER VII SILURIAN SYSTEM A. NOMENCLATURE AND DIVISIONS Nomenclature. The reasons for restricting the name Silurian to the system called Upper Silurian by Murchison have been given on p. 71, but no mention was there made of a subordinate question, namely, the line of separation between the Ordovician and Silurian. Originally the whole of the strata subsequently called Llandovery Beds were included by Murchison in his Upper Caradoc Group, but in 1854 Sedgwick and M'Coy proved that the uppermost sandstones (to which they gave the name of May Hill sandstone) had no connection physically or palaeontologically with the Bala rocks, but formed the natural base of Murchison's (Upper) Silurian Series in the typical Silurian areas. Recognising the justice of this correction, Murchison then separated the beds from the Caradoc sandstone (Bala), and created a new group under the name of " Llandovery rocks," subdividing it into two stages a lower (linked by some species to the Bala Series) and an upper (closely connected with the Wenlock rocks), but clearly united to one another by a community of Pentameri and other fossils. This Llandovery Group he regarded as forming a transition series between his Lower and Upper Silurian. Some geologists, however, objected to this arrangement on the ground that there is frequently a physical break (overlap leading to unconformity) between the Lower and Upper Llandovery, so that the latter in such localities is dissociated altogether from the former and from the Bala Beds. Consequently, Sedgwick, Jukes, and others drew the line of separation at the base of the Upper Llandovery or May Hill sandstone. But elsewhere there is little or no want of conformity between the members of the Llandovery, and it was pointed out in Chapter II. that small regard must be paid to such local transgressions and unconformities, unless they coincide with a palaeontological break. 153 154 STRATIGKAPHICAL GEOLOGY Recent researches have shown that the Upper Llandovery is usually succeeded by a series of pale grey and green shales (Tarannon shales), and that these three subdivisions (Lower Llandovery, Upper Llandovery, and Tarannon shales) " are far more closely allied to each other than they are to the beds above or below, and that they should be considered as the three con- secutive members of a single formation." l The Llandoveries were formed in fact during a period of local upheaval and oscillation which intervened between the quiet periods of the Bala and Wenlock Beds. For their equivalents in South Scotland Professor Lapworth has proposed the name Valentian, from the Roman name of that province, and this is a convenient designation for the lowermost group or series of the Silurian rocks in Britain. The second natural division of the Silurian System is typically developed at Wenlock in Shropshire and at Woolhope in Hereford- shire. This series was described by Murchison under the name of the " Wenlock formation," and at the localities above-mentioned it consists mainly of shales with a band of limestone at the base and another at the top ; but in North Wales and other parts of Britain there are no such limestones, and the only means of delimitation is by the graptolite fauna. The third division is the " Ludlow formation " of Murchison, which in Shropshire and Herefordshire consists of shales and mud- stones with a band of limestone in the middle. Although the shales below the Aymestry limestone were placed in this group by Murchison, he admitted that they were merely a continuation of the Wenlock Series, and for some time it seemed doubtful if they could be separated from it when the Wenlock limestone was absent. Recently, however, it has been shown that they can be dis- tinguished by their graptolites, 2 and, further, that the Aymestry limestone comes within the highest graptolite zone. Since, however, these beds are linked to the Wenlock Group by the presence of graptolites and by a common assemblage of trilobites and brachiopoda, Professor Lapworth has proposed to unite them with the Wenlock Beds and to regard them as a primary division under the name of Salopian. The highest beds referable to the Silurian System are a set of shales, flagstones, and sandstones in which no graptolites are found, but which contain remains of a remarkable group of Arachnida the Eurypterida and also the bones of certain fish which are the earliest vertebrates found in the British Isles. For these beds the name Downtonian has been proposed, but Clunian from Clun Forest, proposed by me in 1885 (Geol. Mag.), would in some respects be a better name. SILURIAN SYSTEM 155 This arrangement of the Silurian succession into Valentian, Salopian, and Downtonian has been adopted by many English writers, but it may be doubted whether it is any real improvement on that hitherto employed and still used by the Geological Survey. It is quite as important to recognise that the central mass of beds can be divided into two groups by the different types of graptolites as it is to admit that they are united by a certain community of species. Thus the Wenlock Beds are characterised by species of Oyrtograptus and by the Monograptus priodon type, and the higher beds by species of the M. colonus type. Thus the whole system can be divided into four much more equal groups or series on a palaeontological basis, and it is probable that this will form the classification of the future. Meantime the several arrangements are shown in the following table : Downtonian-Eurypterid Series {^^^ Beda I LudlOTV f M. Colonus Series {^ Sd^S j Salopian J M. priodon and ( Wenlock limestone 1 -, | Cyrtograptus -I Wenlock shales V G I Series (Woolhop'e limestone J fc Valentian-Rastritea and f UppTLlandovery Ulandovery In dealing with the stratigraphy of the system, I shall use the new grouping as affording convenient divisions in the description of the beds, but shall indicate the correlation of the priodon and colonus series in different areas, so that the student can, if he chooses, adopt the alternative arrangement of Llandovery, Wenlock, and Ludlow Series. B. LIFE OF THE PERIOD The following is a brief synopsis of the principal genera which are found in the Silurian System : Plantse. It is in this system that the earliest remains of plants have been found. Some of these remains are doubtful, but there is general agreement with regard to Nematophycus and Pachytheca. The former occurs in the form of silicified stems and have been found in Wenlock Beds. Pachytheca is a name given to small spherical bodies which may be seeds. Hydrozoa. Graptolites are abundant in the lower beds, but gradually decrease in numbers and die out in the upper beds. The family Monograptidse, including the genera Monograptus, Cyrtograptus, and Rastrites, appears to be confined to the Silurian, 156 STRATIGRAPHICAL GEOLOGY and the only other families represented are the Diplograptidse and Retiolitidse. Stromatoporoids become abundant in the Silurian, and are associated with corals as important rock-builders. The principal genera are Stromatopora, Labechia, and Clathrodictyon. Actinozoa. Corals are very abundant in the limestones and calcareous beds ; most of the Ordovician genera occur, Favosites, Heliolites, Halysites, and Omphyma are very common, and the following commence their existence in the British area : Acervularia, Alveolites, Gystiphyllum, Arachnophyllum, and Stauria : while the following have only been found in the rocks of this system, Goniophyllum, RhizophyUum, Thecia, and Palceocyclus. Echinoderma. Of this phylum the crinoids are by far the most abundant, more than fifty species having been found in British Silurian strata. The following genera are specially characteristic of the Silurian : Gallicrinus, Glonocrinus, Grotalocrinus, Dimerocrinus, Herpetocrinus, Marsipiocrinus, Pisocrinus, and Periechocrinus ; and besides these the genera Cyathocrinus, Ichthyocrinus, Eucalyptocrinus, Hapalocrinus, and Taxocrinus make their first appearance. Cysti- deans continue to occur, and the genera Lepadocrinus, Placocystis, Pseudocrinus, Prunocystis, and Schizocystis are specially characteristic of the Silurian period. Echinoids, which are first found in the Ordovician of Russia, make their appearance in the Silurian of Britain, where they are represented by the genera Echinocystis, Palaaodiscus, and Palcechinus. Crustacea. Trilobites, though not quite so numerous as in the Ordovician, continue to be common, most of the species belong- ing to the genera Calymene, Phacops, Encrinurus, Harpes, Acidaspis, Illcenus, Homalonotus, Lichas y Cheirurus, Proetus, and Deiphon, only the last being exclusively Silurian. Arachnida. Besides trilobites, another remarkable and extinct order of Arthropods is abundant the Eurypterida, repre- sented by the genera Eurypterus, Pterygotus, Stylonurus, and Slimonia. The Xiphosura, which are allied to the Eurypterida, also occur, but are much less numerous, the principal genera being Hemiaspis, and Neolimulus. Brachiopoda are remarkably abundant. Most of the Ordo- vician genera recur, such as Lingula, Orthis, Leptcena, Camarotcechia, and other Rhynchonellidse, Strophomena, Glassia, and Triplecia ; and the following genera make their first appearance : Atrypa, Chonetes, Dayia, Gyrtia, Cyrtina, Meristina, 1 Meristella, Pentamerus (Conchidium), Stricklandia, Rhynchotreta, Wilsonia, Nucleospira, and Spirifer. Lamellibranchia. Members of this class also become more SILUEIAN SYSTEM 157 Fig. 46. GROUP OF LLANDOVERY FOSSILS. a. Atrypa hemispherica. a. Stricklandia lens. t>. Menstella angustifrons. e . Pentamerus oblongus. c. Ortlns reversa. /. Chiton Griffith!. g. Cyrtoceras approximatum. Fig. 47. SILURIAN GRAPTOLITES. a. Monograptus spiralis. ft. Rastrites peregrinus. c. Monograptus gregarius. d. Monograptus exiguus. e. Monograptus Sedgwicki. /. Monograptus priodon. 158 STRATIGRAPHICAL GEOLOGY abundant in the Silurian period, about 100 species being known in England belonging chiefly to the following genera, Cardiola, Cardiomorpha, Ctenodonta, Orthonota, Cleidophorus, Grammysia, Am- bonychia, Pterincea, Conocardium, Lunulicardium, and Modiolopsis. G-astropoda are fairly numerous, about sixty species being known from the English Silurian. Most of them are referable to the following genera : Pleurotomaria, Murchisonia, Bellerophon, Platyschisma, Omphalotrochus (Horiostoma\ Cyclonema, Eunema, Holopea, Holopella, Platyceras ( Acroculia), and Subulites. The last genus is noteworthy as possessing a channelled mouth, and Murchisonia has a very short anterior canal, all the rest being holo- stomous. Cephalopoda. These are also numerous ; but all the genera belong to the Nautiloidea ; they include species of Orthoceras, Phragmoceras, Gomphoceras, Trochoceras, Tretoceras, Ophidioceras, Cyrtoceras, Actinoceras, and Ascoceras, the last being only known as a Silurian form. Fishes. The earliest remains of vertebrate animals hitherto found in Britain occur in the Ludlow Series, and are the remains of fish belonging to the sub-class or order Ostracodermi. The chief genera are Pteraspis, Cephalaspis, Auchenaspis, Scaphaspis, and Eukeraspis. The sub-class Elasmobranchii is represented by Onchus (spines) and shagreen-like plates referred to Gcelolepis and Thelodus. The greater part of the Silurian System is capable of subdivision into zones by means of the graptolite faunas, and though these are not equally developed in all localities they always occur in the same order of succession. No species have yet been recognised in the Upper Ludlow Beds, but from the Aymestry limestone downward the following zones have been recognised : Lower Ludlow Wenlock Beds Valentian (Llandovery) 'Zone of Monograptns Leintwardinensis. tumescens. scanicus. Nilssoni. vulgaris. Cyrtograptus Lundgreni. rigidus. Linnarssorii. symmetricus. Monograptus Riccartonensis. Cyrtograptus Murchisoni. ,, Grayse. Monograptus crispus and M. exiguus. Rastrites maximus. Monograptus spinigerus and Cryptograptus coineta. ,, gregarius. Diplograptus acuminatus and vesiculosus. SILURIAN SYSTEM 159 Fig. 48. GROUP OF WENLOCK FOSSILS. a. Acervularia luxurians. 1). Omphyma subturbinatum. c. Atrypa reticularis. (1. Leptsena rhombpidalis. h. Cyathophyllum truncatum. ,i. Arachnophyllum Murchisoni. . Strophonella euglypha. /. Spirifer plicatellus. g. Horiostoma discors. The following are some of the other fossils which are specially characteristic of each of the three Silurian series or divisions : Hydrozoa. Actinozoa. Fossils of the Llandovery and Valentian Series (See table of graptolite zones.) Lindstrosmia bina, Lind. subduplicata, Pinacopora Grayi. 160 STKATIGEAPHICAL GEOLOGY Echinoderma. Palaechinus Phillips! (but none arc common). Crustacea. Illsenus semulus, Phacops Weaveri, Ph. elegans. Brachiopoda. Lingula crumena, Atrypa hemispherica, Meristella crassa, M. angustifrons, Stricklandia lens, Penta- merus oblongus, P. undatus, Orthis reversa. Lamellibranchia. Ctenodonta Eastnori, Orthonota amygdalina, and 0. inornata (both range into Wenlock). Fig. 49. GROUP or WENLOCK FOSSILS. a. Lepadocrinus quadrifasciatus. &. Periechocrinus moniliformis. c. Calymene Blumenbachi. d. Phacops caudatus. e. Bellerophon dilatatus. /. Orthoceras annulatum. Gastropoda. Turbo tritorquatus, Holopella cancellata, Omphalo- trochus prenuntius. Cephalopoda. Tretoceras bisiphonatum, Cyrtoceras approximatum, Orthoceras conicum. Fossils of the Wenlock Series Hydrozoa. (See table of graptolite zones. ) Actinozoa. Acervularia luxurians, Arachnophyllum Murchisoni, A. typus, Strombodes diffluens, Cyathophyllum truncatum, Omyphyma turbinatum, Heliolites in- terstinctus (from Bala), Favosites gottlandicus. SILURIAN SYSTEM 161 Fig. 50. OROCP or LUDLOW FOSSILS. a.. Orthoceras tenuicinctum. l>. Orthis lunata. c. Pentamerus Knighti. (1. Rhynchonella nucula. e. Pterinsea retroflexa. /. Avicula Dan by i. ft. Dayia navicula. h. Homalonotus clelphinocephalus. M 162 STRATIGRAPHICAL GEOLOGY Echinoderma. Actinocrinus (Glyptocrinus ?) pulclier (iromLlandov.), Periechocrinus moniliformis, Crotalocrinus rugosus, Marsipiocrinus cselatus, Lepadocrinus quadrifas- ciatus. Crustacea. Illaenus barriensis, Phacops caudatus (Val. to Lud.), P. Downingise, Calymeue Blumenbachii (Bala to Lud.), Encrinurus punctatus, Lichas anglicus (to Lud.). Brachiopoda. Atrypa reticularis, Pentamerus galeatus, Strophonella euglypha, Leptpena rhoraboidalis (Llan. to Lud.), Orthis rustica, 0. elegantula (Bala to Lud.), Spiriier plicatellus, Retzia? Barrandi, Wilsonia Wilsoni, Camarotcechiaborealis, Cyrtia exporrecta. Lamellibranchia. Cardiola interrupta, Grammysia cingulata. Gastropoda. Horiostoma discors, H. rugosus, Eunema cirrhosis, Platyceras (Acroculia) haliotis, Bellerophon dila- tatus. Cephalopoda. Orthoceras annulatum, Phragraoceras ventricosuni, Gomphoceras pyriforme, Trochoceras asperum, Bar- randeoceras bohemicum. Fossils of the Ludlow Series Porifera. Ischadites Koenigi, Favospongia Ruthveni. Echinoderma. Ichthyocrinus pyriformis (from Wenlock), Lap - worthura Miltoni, Palseaster Ruthveni, Palseocoma Colvini, Palaeodiscus ferox, Echinocystis pomum. Crustacea. Homalonotus delphinocephalus (from Wenl.), H. Knighti, Acidaspis coronata, A. Hughesi. Arachnida. Pterygotus arcuatus, Eurypterus abbreviatus, Hemi- aspis limuloides. Brachiopoda. Pentamerus Knighti, Dayia navicula, Camarotoechia nucula (from Val. ), Orthis lunata, Chonetes striatella (and var. lata), Orbiculoidea rugata, Lingula cornea, L. Lewisi (from Wenl.). Lamellibranchia. Avicula Danbyi (from Val.), Pterinea retroflexa, Car- diola striata, and C. interrupta (both from Wenl.), Orthonota semisulcata. Gastropoda. Cyclonema coralli, Murchisonia coralli, Holopella gregaria, Platyschisma helicites. Cephalopoda. Orthoceras bullatum, 0. tenuicinctum, 0. ludense, Trochoceras giganteum, Ascoceras Barrandi. Pisces. Scaphaspis ludensis, Cephalaspis ornatus, Auchenaspis Salteri, Thelodus parvidens. C. BRITISH SILURIAN ROCKS Bange and Belation to Older Rocks. Silurian rocks occupy surface - tracts in the following districts. They form a continuous outcrop of varying width through Wales from Car- marthenshire in the south to the Denbigh coast on the north. They spread westward through Radnorshire and parts of Shropshire, and they reappear from beneath newer rocks as inliers in Stafford- SILURIAN SYSTEM 163 shire, Hereford, and Monmouth. In all probability they have a considerable subterranean extension under Central and Southern England, for they appear as far south as Cardiff and the Mendip Hills, and have been reached in borings at Ware in Hertfordshire and at Eochester in Kent. In North Lancashire and Westmoreland they again rise to the surface in a considerable tract south of that occupied by the Ordovician, extending from the Duddoii estuary on the west to Sedburgh and Rathay Bridge on the east, a distance of about 35 miles with an average width of 14 miles. In the south of Scotland they cover a much larger area, ranging completely across the Southern Uplands from the Mull of Galloway to the coast of Berwick. There is also an outlying tract in Lanark, and it is probable that they underlie parts of the Central Lowlands, for the same Lanarkian Beds have recently been recognised 011 the coast of Kincardine. In Ireland there are many isolated exposures in the south and west and a large area in the north-east, occupying parts of Down, Armagh, Monaghan, Cavan, and Louth. In most parts of the British Isles there is no great break or unconformity between the Silurian and Ordovician, but it so happens that in the typical districts of Shropshire and Radnorshire a strong local discordance and unconformity is found, due evidently to local uplifts which produced more than one break in the sequence. In other areas the sea-floor may have been raised and the depth of water lessened without any part of it being lifted above the surface of the sea and brought into the sphere of erosive agencies. Under such circumstances uplift is indicated only by changes in the character of the sediment and in the component members of the marine fauna. 1. The Typical Silurian Area We shall consider this as including all the more eastern exposures, and especially those which occur between the Welsh borders and the valley of the Severn. It will embrace not only the Weiilock and Ludlow districts of Shropshire, but that of Presteign in Radnor, as well as those of Malvern, Woolhope, and May Hill in Hereford^ Usk in Monmouth, and the outcrops in Staffordshire. These were the places where the beds were first studied by Murchison, and where they present a more calcareous and more richly fossiliferous facies than in any other part of Britain. Llandovery Beds. In this region the Lower Llandovery is absent, and the Tarannon shales are very thin, so that the 164 STEATIGRAPHICAL GEOLOGY total thickness of the Valentian Series is much less than it is in Wales. In the southern districts (Usk, May Hill, Woolhope, and Malvern) the group consists of the May Hill sandstone and the Woolhope shales. The former is about 1000 feet thick, and consists of (1) a lower set of sandstones and conglomerates of grey and purple colours about 600 feet thick, and yielding the usual fossils, together with Ctenodonta Eastnori and Linyula crumena ; (2) an upper set of grey laminated sandstones and shales from 320 to 500 feet thick with Pentamerns lens, P. undatus, etc., the highest beds being shales which are doubtless equivalent to the WbyS High Wood z\ Nort^ Hill EbyN> \VArch\-\ Arch. Tr \ X U\ WSWL,TS MS 2 Fig. 51. SECTION THROUGH HIGH WOOD NEAR WEST MALVERN (Professor GrOOm). (Scale 6 inches to 1 mile.) WL. Woolhope limestone. WS. Woolhope shale. BS. Black shales (Cambrian). Arch. Pre-Cambrian. FF. Faults. Tarannon shales. The conglomerates rest unconformably upon the Cambrian shales where these are exposed, but elsewhere they are faulted against the Archaean rocks. The Woolhope shales are not more than 200 feet thick, consisting of green and purple shales with thin layers of limestone. In Shropshire the group consists of the following members : Purple shales ( = Taraunon shale) . Pentamerus limestone . . . Basement grits and conglomerates . 200 to 400 feet. 30 to 40 ,, to 120 , The basement beds rest unconformably 011 Ordovician, Cambrian r and Longmyndian in different parts of the area, varying greatly in thickness, and in places thinning out altogether so as to leave the overlying limestone in contact with the older rocks. Thus the Pentamerus limestone is really a sandstone rendered calcareous by the quantity of included fossil shells. The commonest fossils are SILURIAN SYSTEM 165 Pentamerus oUongus, P. undatus, Stricklandia lens, Atrypa hemi- spherica, Strophomena pecten, Encrinurus punctatus. The overlying shales consist in the Wenlock district of purple, red, and green shales with few fossils, and are distinguishable by their colours from the overlying Wenlock shales. Salopian Series. The succession of limestones and shales which form this series at Woolhope and Malvern is shown in Figs. 51 and 52. The Woolhope limestone is from 50 to 250 feet thick, the Wenlock from 250 to 400, and the Aymestry about 250 feet, while the intervening shales vary from 500 to 900 in different places, and the average total thickness of the series is over 2000 feet. The Woolhope limestone is also present near Presteign in Kadnor, but thins out northward and is only represented in Shropshire by a band of calcareous and fossiliferous shale. The whole series forms a broad band from the southern border of this county to the river Severn at Coalbrookdale, the Wenlock and Aymestry limestones forming well-marked escarpment ridges, rising above the outcrops of the shale as the beds are here dipping steadily to the east (see map, Fig. 21). The succession found near Wenlock was well described by Mr. W. Maw in 188 1, 3 from whose paper Fig. 53 has been copied. The Buildwas Beds, which here represent the Woolhope lime- stone, are exposed in a small cliff opposite Buildwas Abbey, and are shales which yield an immense number of small brachio- poda, Orthis biloba, 0. deyantula, 0. Lewisi, 0. hybrida, Leptcena segmentum, Nudeospira pisum, and Atrypa (? Retzia) Barrandi. Mr. Maw considered these beds to represent the Woolhope limestone of more southern localities. The Tick wood Beds are shales with layers of nodular limestone, the shales predominating in the lower part and the limestones in the higher part below the massive limestone of Wenlock edge, so that it is not easy to say where the one ends and the other begins. Fossils are abundant in these beds. The Wenlock limestone is well exposed at Benthall Edge and in quarries near Wenlock ; it is a grey earthy limestone, generally lying in thin beds with shaly partings, and is highly fossiliferous. Some of the beds appear to be largely made up of corals which weather out finely in the spoil-heaps, while others contain quantities of Brachiopoda and Mollusca. The Lower Ludlow shales consist of soft grey sandy shales, often showing a tendency to spheroidal structure, and their total thick- ness is about 900 feet in the Wenlock and Ludlow district. In the Wenlock area they do not contain graptolites, but have yielded many other fossils, of which the following are some of the B / I '- / / ' / > \ \ I c I 5 s c -2*5 i I fe .5 O -O' B w (f|| *- if "S 6 83 f '* .11 I'll If as * life's tasttl I Illll i -.3 Masai * 1 ^ dd ! B 1 ^ o 8ggS8 ^-o!! SILURIAN SYSTEM 167 commonest : Lingula lata, Wilsonia Wilsoni, Camarotcechia nucula, Leptcena Icevigata, Chonetes striatella. At the top of these shales there are some beds of earthy flagstone which are quarried and are known as the Leintwardine flags ; they contain remains of starfish (Lapworthura and Palceocoma), large Orthoceras, and Arthropods of the genera Pterygotus and Eurypterus. The Aymestry limestone is a bluish-grey earthy limestone of varying thickness, and often split into several beds by intercalated bands of shale. At Aymestry it is well exposed in the gorge of the river Lugg and is said to be 175 feet thick. Its commonest fossils are Pentamerus Knighti, Lingula Lewisi, Wilsonia Wil- soni, and Dayia navicula. The outcrops of the Ludlow Beds in the typical district of Ludlow and Aymestry are shown in Fig. 56. The total thickness of the series in Shropshire is about 2600 feet. In Staffordshire the Woolhope limestone reappears at Great Barr, east of Walsall, the Wenlock Beds are well exposed at Dudley and the higher beds at Sedgeley, the Silurian at all these places being brought up in anticlinal flexures from which the over- lying Carboniferous rocks have been removed by denudation. Downtonian. In Shropshire the Upper Ludlow shales consist in the lower part of soft grey shales with thin bands of limestone, and in the upper part of grey and green calcareous flags at the top of which is a bed of greenish -grey argillaceous sandstone surmounted by greenish laminated sandstones, including a thin layer consisting of the bones and spines of fish and large Crustacea ; this layer is known as the " bone bed," and has been met with at many localities. The bone bed is succeeded by the Downton sandstone, a fine- grained and thin-bedded yellowish sandstone which is quarried for building stone at Downton Castle near Ludlow, and contains Lingula cornea, Pteraspis Banksi, Eurypterus lineatus, and other Merostomata. This is about 50 feet thick, and is succeeded by reddish flagstones and olive-green shales which yield Lingula cornea with remains of fish and Crustacea. Above these are purple sandstones without fossils. In a recent paper the Misses Elles and Slater 4 recognise the following subdivisions, with a total thickness of 540 feet : Feet. Eurypterus shales . . . 120 Downton sandstone ... 50 Chonetes flags . . . .150 Rhynchonella flags . . .120 Dayia shales . . . .100 A similar series reappears in the Woolhope and Malvern districts 168 STKATIGRAPHICAL GEOLOGY and are well exposed in the cutting near Ledbury station, west of Malvern, where the sequence is as follows : Feet. Old Red marls and sandstones ..... {Ledbury shales ....... 400 Down to n sandstone ...... 90 Upper Ludlow shales ... .140 Aymestry limestone (seen in the tunnel) From the Ledbury shales Linyula cornea with remains of Cephalaspis and Pterygotus have been obtained. The most southerly occurrences of Silurian rocks of Salopian type are at Fortworth, in Gloucestershire, and in the Mendip Hills, where a small inlier has recently been discovered and described. It is noteworthy that where the base of the Devonian comes to the surface in Cornwall it does not rest on Silurian but on Ordovician and older rocks, having probably overlapped the Silurian. 2. Wales The Valentian attains its greatest thickness in Central Wales where it occupies large parts of Cardigan and Kadnor, and has a maximum thickness of 5000 feet. The most complete development of the Llandovery Beds is found within the western part of the area near Rhyader in Radnor, and has recently been worked out by Mr. H. Lap worth, from whose account the following is taken. 5 He describes the Lower Llandovery under the name of the Gwastaden Group, the Upper Llandovery as the Caban Group, and the Tarannon shale as the Rhyader pale slates. The Gwastaden Beds seem to succeed the Bala Beds without any break, but there is a strong unconformity between the Gwastaden and Caban Groups, and the latter is overlapped by the Rhyader slates, so that the structure of the district is somewhat complicated. Fig. 54 is a diagrammatic view constructed by Mr. Lapworth of the succession and of the relations of the rock-groups when faults and folds are eliminated. He estimates the thickness of the Gwastaden Series at over 1000 feet, that of the Caban Group as 1500 (maximum); that of the Rhyader (Tarannon) slates is not given, but they extend over a large area of ground between Rhyader and Llanidloes, where they are probably 1500 feet thick, so that in this district the Llandovery or Valentian Series has a total thickness of nearly 5000 feet. When traced eastward or south-eastward the Gwastaden Beds are found to thin out entirely and the thickness of the Caban Beds and Tarannon shales is greatly reduced. North-east of Builth they form quite a narrow band and rest unconformably on Llandilo 170 STRATIGRAPHICAL GEOLOGY slates. Southward they extend into Carmarthenshire, both Upper and Lower divisions being well developed at Llandovery, but south of that place they are cut out by faults. Northward the Rhyader facies is continued through Mont- gomery, but the several members of the series become much thinner till, near Bala and Corwen on the Dee, the whole is reduced to less than 1000 feet. At Corwen it consists of the following beds : Tarannon shales .... 300 feet. Graptolitic mudstones . . 300 , , Corwen grits .... 200 ,, The position of these beds is shown in Fig. 57. The Corwen grits are grey and white sandstones with bands of coarse and pebbly sandstone, and are believed to vary from 50 to 300 feet in thickness. They contain few fossils, but Meristella crassa and Pentamerus oblongus have been found, which are Lower Llandovery species. Above them are soft bluish-grey or black mudstones which contain Monograptus gregarius, M. tennis, and Climacograptus normalis ; and these pass up into the pale Tarannon slates. In North Denbigh the whole series is less than 500 feet thick, but the zonal succession of beds has been more fully made out by the Misses Elles and Wood, 6 these being : Beds. Zones. If Monograptus crenulatus. Light-coloured shales J Monograptus crispus. with black bands. 1 Rastrites maximus. I Monograptus Sedgwicki. Grey shales, flags, and /Monograptus gregarius. mudstones. \Mesograptus modestus. Con way Castle grits and Bryn Dowsi flags, 150 feet. The lowest beds contain Diplograptus tamariscus, Climacograptus normalis, and other species of Lower Llandovery graptolites. Salopian. When the Wenlock Beds are traced westward from Presteign through Radnorshire both the limestones are found to thin out, leaving a continuous series of mudstones and shales. The Lower Ludlow shales also persist, but the Aymestry limestone at their summit disappears. The subdivision of this mudstone series was practically impos- sible so long as lithological differences only were relied upon as a means of classification, but in 1880 Professor Lapworth drew attention to the graptolite fauna of these beds, and expressed the belief that " by the aid of the lowly graptolite the geologist of the future will be able to read off the natural succession " of the monotonous Silurian mudstones. This has been accomplished SILURIAN SYSTEM 171 in Kadnorshire and West Shropshire by Miss G. L. Elles 7 and Miss E. M. B. Wood, 8 the former dealing with the Wenlock succession and the latter with the Lower Ludlow. The country round Builth may be taken as the type of this graptolitic facies of the Wenlock Series, and Miss Elles has shown Fig. 50. MAP OF THE LUDLOW DISTRICT. that it contains a fauna of twenty-three species and varieties of graptolites, by means of which it can be divided into six zones. The succession of the beds is shown in Fig. 55, which is taken by permission from her paper. The total thickness appears to be about 2500 feet, but there are two partial breaks in the series, for the flagstones of zone 4 overlap the beds below, so that in one place they rest on the lower part of zone 2, and again, there is 172 STRATIGRAPHICAL GEOLOGY reason to think that the upper beds of zone 5 are sometimes concealed by an overlap of zone 6 (Cyrtograptus Lundgreni), which represents the Wenlock limestone. The shales overlying this zone are similarly divisible, and Mi^s Wood has distinguished five zones in the following descending order : Feet. 5. Zone of Monograptus Lcinticardinensis, including the Aymestry limestone 400 t*o 500 4. ,, Monograptus tumescens . . . . 100 to 220 3. , , Monograptus scanicus . . . . . 250 to 1 00 2. ,, Monograptus Nilssoni ^ qoo tn 400 1. Monograptus vulgaris . . . . . J 8 Thus the total thickness of the Salopian Series in the Builth district is about 2600 feet. When these beds are followed northward they become more arenaceous, thick felspathic grits and micaceous flags appearing in the Wenlock shales, while the higher beds are gradually replaced by a great mass of sandstones, grits, and flagstones in which fossils are much less abundant. This is the aspect presented by the Salopian Series in Denbighshire, where parts of them have been studied by Mr. Philip Lake, 9 Professor T. M'K. Hughes, 10 Dr. H. Hicks, 11 and by the Misses Elles and Wood 12 (see Figs. 57 and 58). Combining the observations and correlations of these authors, we find that the succession may be generalised as follows : 6. Gritty sandstones and wavy banded concretionary mudstones of Moel Ganol. "5. Grey banded mudstones and sandstones with Acidaspis Hughesi. 4. Nantglyn flags with Actinocrinus pulcher, CamarotcecMa nucula, Dayia navicula, Orthoceras SedgwicJci, 0. primceviim, Cardiola inter rupta, and Monograptus Nilssoni. 3. Moel Ferna slates with Monograptus priodon and M. Flemingi. 2. Pen-y-glog grit with Meristella tumida, Strophomena deprcssa, Phacops sp., and Encrinites. 1. Pen-y-glog slates with Cyrtograptus Murchisoni, Gladiograptus geinitzianuS) Monograptus priodon, and M. vomerinus. The slates at the base are the equivalents of the lower part of the Wenlock shales, and the same fossils have been found in beds overlying the Tarannon shales near Con way. The Moel Ferna Beds must include the representative of the Wenlock limestone, but they yield very few fossils. The Nantglyn flags represent the Lower Ludlow Beds, and Miss Elles has recognised two zones in them, a lower zone of Monograptus vulgaris and a higher one of M. Nilssoni. In the valley of the Dee Mr. Lake found some micaceous flagstones which contain Monograptus Leintwardinensis in abundance, SILTJKIAN SYSTEM 173 X Si 174 STRATIGKAPHICAL GEOLOGY and these cannot be far below the horizon of the Aymestry lime- stone. This zone is probably to be found in the banded beds (No. 5) of the Clyvvd valley section, but has not yet been detected there. In South Denbigh the highest beds occur at Dinas Bran near Llan- gollen, and have yielded Orthoceras SedgwicJci, Cucullela coarctata, Dayia navicula, Chonetes minima, and other fossils. The series is believed to be over 5000 feet thick, and is nevertheless incomplete owing to the unconformable superposition of the Carboniferous rocks. Downtonian. In the country round Kington, dim, and Radnor Forest these graptolitic shales are succeeded by a great thickness of brownish sandy shale with beds of brown flagstone and sandstone, which are shown in the sections of the Geological Survey to be from 1700 to 2000 feet thick. The highest beds in Denbighshire, i.e. the sandstones and mudstones of Moel Ganol, may represent the Upper Ludlow shales, but proof of such equivalence has not yet been found. That beds of Ludlow age did originally extend over Denbigh is proved by the occurrence of fossiliferous pebbles in the conglomerate at the base of the Carboniferous Series near Abergele. 13 These are pebbles of green and red sandstone containing Orthis lunata, 0. filosa, Chonetes striatella, Spirifer elevatus, Pterincea retroflexa, and a few other species. 3. Westmoreland The Silurian sequence in the north of England does not differ greatly from that of North Wales, but it is much more complete, has been more fully investigated, and the beds are also rather more fossiliferous, so that they are more easily subdivided. For a long time the correlation of the Westmoreland Groups with those of Shropshire and Radnor was very uncertain, but the recent studies of graptolites by Messrs. Marr and Nicholson, 14 by Miss G. Elles, 15 and by Misses Watney and Welch 16 have placed the comparison on a sounder basis. Valentian. Where the sections are most complete, as in Stockdale and at Spengill, there are no arenaceous beds of the Llandovery type, but a series of shaly mudstones resting directly on the Ashgill shales with apparent conformity (see Fig. 37). In some parts of the area, however, the lowest shales are replaced by a few feet of gritty conglomerate. The ordinary facies is known as the Stockdale Shales ; their total thickness is not more than 250 feet, but divisible into two sub-groups: (1) the Skelgill Beds, (2) the Browgill Beds. By the Geological Survey SILUKIAN SYSTEM 175 the groups are called (1) the graptolitic mudstones and (2) the Pale slates. The Skelgill Beds have a thickness of 70 or 75 feet, and have at the base a thin limestone containing Atrypa flexuosa, succeeded by shale with Dimorphograptus confertus. The middle beds contain Monograptus gregarius and Rastrites peregrinus, with Encrinurus punctatus and Phacops glaber. The higher beds yield Monograptus spinigerus and M. Clingani, with many trilobites, and in the Spengill section there is a still higher band containing Rastrites maximus in abundance. The Browgill Beds are thicker (about 140 feet) but less fossili- ferous. They are divisible into three parts : (1) a zone of Mono- graptus turriculatus ; (2) a zone of Monograptus crispus ; (3) barren shales and grits. Salopian. This series attains a very great development in Westmoreland and consists almost entirely of flagstones, sandstones, and grits, with only subordinate bands of shale. The equivalents of the Wenlock Beds are really of less thickness than they are in Shropshire, but those above the horizon of the Wenlock limestone (zone of Cyrt. Lundgreni) are very much thicker. The following is a tabular view of the lithological divisions and the graptolitic zones : Divisions. Zones. Bannisdale slates (4000 feet). Sandy mudstones | with beds of hard sandstone. Orthoceras j- M. Leintwardinensis. primcevus and 0. subundulatum. Coniston grits (4000 feet). Red grits and flags ] M with Cardiolainterrupta,Pterineatenuistriata. 1-JJ* tlim sceils - Dayia navieula, and Orthis lunata. j M " scamc " s - Coldwell flags (2000 feet). Grey flags and grits \ ^ , with Acidaspis Hughcsi, Actinocrinus pulcher, \ , ^ m S aris ' Monograptus colonus, and M. Rcemeri. } lP rot)abl y)- Coldwell grits (800 feet). Coarse grey grits. C. Lundgreni. Brathay flags (1000 feet). Fine-grained bluish "j C. rigidus. laminated flagstones, comparable with the [M. Riccartonensis. Wenlock shales and divisible into 3 zones. J C. Murchisoni. Downtonian. This group seems to be represented by certain beds which succeed the Bannisdale slates and consist in the lower part of bluish thin-bedded sandstones with subordinate beds of shale, containing Protaster Miltoni, Dayia navicula, Cucullella Cawdori, Phacops Downingiw, and other fossils. There is also a calcareous band which contains Palceaster, Proetus latifrons, and Dayia navieula, and has been identified by Mr. Aveline with the Aymestry limestone. These beds are about 1200 feet thick. The highest beds, or The Kirkby Moor flags, are massive greenish and grey sandstones, with calcareous (fossiliferous) bands, passing 176 STKATIGKAPHICAL GEOLOGY up into thin-bedded flags. This group is more than 2000 feet in thickness. Fossils are abundant, the following being some of the most characteristic : Lingida cornea, Discina rugata, Spin/era elevata, Grammysia cingulata, Orthoceras ludense, and Trochoceras ibex. 4. South of Scotland The Silurian succession in this region was first deciphered by Professor Lapworth, 17 who described the two special facies exhibited by the rocks of Valentian age near MofYat in Dumfries, and near Girvan in Ayrshire. More recently the whole region of the southern uplands has been described in detail by Messrs. Peach and Home. 18 The Silurian rocks occupy the southern half of the upland region in a broad belt from 20 to 30 miles in width, extending from the Berwick coast on the east to that of Wigtown on the west. On the northern side they occur as isolated exposure in Ayrshire, in Lanark, and in the Pentland Hills. Valentian. In Dumfries and the central belt the lowest beds are soft black shales, known as the Birkhill shales ; they lie conformably upon the Hartfell shales (Bala), are equivalent to the Skelgill shales of Westmoreland, and are of about the same thickness (140 feet) (see Figs. 40 and 41, p. 133). They are rich in grapto- lites, and are divisible into the following zones : Upper ] Zon Birkhilll Lower 1 ' Birkhffll e of Kastrites maxim us. Monograptus spinigerus. Petalograptus cometa. Monograptus gregarius. Diplograptus venculosus. Diplograptus acuminatus. These shales are succeeded by a thick series of sediments, from 3000 to 4000 feet, which are the equivalents of the Browgill Beds and the Tarannon shales. They are called the Gala Group by Professor Lapworth, and have been divided by the Survey into three stages, but fossils are rare except in the lowest or Abbotsford flags, which contain Monograptus turriculatus, M. crispus, M. exiguus, etc. In the Girvan area the Birkhill shale is represented by a much thicker series of a Llandovery type, and about 1000 feet thick. The lower beds contain Lower Llandovery fossils, while the suc- ceeding beds include a limestone with many Upper Llandovery fossils, such as Pentamerus oblongus, Stricklandia lens, Illcvnus cemulus, and Phacops Stokesi. Above this, and representing the Gala Group, are the Penkill Beds (also 1000 feet thick), the topmost of these being mudstones containing Cyrtograptus Grayce. SILURIAN SYSTEM 177 Salopian. In the main belt the Gala Group is succeeded by the Eiccarton Beds, a series of conglomerates, grits, mudstones, and shales from 1000 to 1500 feet thick, and containing Cyrtograptus Murchisoni, Monograptus vomerinus, and M. priodon. At Kaeberry near Kirkcudbright there are still higher beds, consisting of green marls with limestone nodules, which are either highest Wenlock or possibly Lower Ludlow. These are 700 feet thick. In the Girvan area the representatives of the Wenlock Beds consist largely of uiifossiliferous flagstones, and only the lower part of the series is exposed. This is divided into two groups by Professor Lapworth. The lower or Bargany Group has a thickness of about 700 feet, the only fossils being the graptolites Monograptus acus and M. priodon. The higher group he called the Straiten Beds, at the base of which are olive-green flags well seen at Drumyork, succeeded by shales containing Monograptus vomerinus, Cardiola fibrosa, Orthonota truncata, and other fossils. The highest beds seen are purplish grits and conglomerates, the exposed thick- ness of the Straiton Group being about 900 feet. The highest beds referable to the Wenlock Series are well exposed in Lanark, where they are no less than 2200 feet thick (see table below), and yet do not include any beds comparable to those near Straiton ; hence there seems to be a gap in the succession, and the full thickness of this series in Scotland is not yet known. Ludlow Series. These beds are not found in the main belt nor in the Girvan district, but only in Lanark and the Pentland Hills, where the following succession has been recognised : Feet. fChocolate-coloured sandstones . . . ~\ Downtonian | Conglomerate with quartz pebbles . . . Beds "I Mudstones and shales with fish and eurypterids I Red and yellow sandstones and mudstones . Ludlow . Red and green shales with Platyceras simulans 1480 f Brown sandstones with Orthoceras . w i i Green and grey shales and mudstones . Beds I Massive sandstones and grits .... I r 2200 j Green, red, and purple shales with eurypterids ^Starfish, crinoids, and corals .... The beds which are bracketed as having a thickness of 2700 feet were formerly grouped with the Old Red Sandstone, but are regarded by the Geological Survey as the equivalents of the Downton and Ledbury Beds. They include a band of shales which has yielded many remains of fish and Merostomata. The fish fauna is of great interest, as two of the genera belong to a family which has hitherto been known only from scales ; these genera are Thelodus and N 178 STRATIGRAPHICAL GEOLOGY Lanarkia, belonging to the family Cwlolepidce, which, according to Dr. Traquair, are nearly related to the Pteraspidce, although they have superficial resemblances to the sharks. Other new genera found in these beds are Ateleaspis, Birkenia, and Lasanius. The Merostomata include species of Stylonurus and Eurypterus, and the Crustacean Ceratiocaris is also present. The underlying Ludlow shales have yielded the earliest British arachnid in the form of a scorpion (Palceophonus), a genus first described from remains in the Silurian of Gottland (Sweden). 5. Ireland Kocks of Silurian age succeed the Ordovician Series in many parts of Ireland ; they may be described as exhibiting two different facies, the one comparable with the Scotch type and the other with the Llandovery-Wenlock type. Eastern Pacies. Valentian rocks of the Birkhill-Gala type occur in the north-eastern and central counties, and probably they extend under the whole of Central Ireland down to the coasts of Clare and Waterford. Kocks having the same general character as the Birkhill shales and Gala Series of Southern Scotland range through the counties of Down, Armagh, Monaghan, and Cavan ; it has, in fact, been found that the greater part of the tract which was coloured as "Lower Silurian" on the older geological maps of Ireland consists of Valentian rocks. 19 The older beds (Birkhill shales) occur chiefly along the north-west border of the area, and have yielded the characteristic graptolites, Diplograptus folium, Monograptus gregarius, M. Sedgwicki, and Rastrites maximus. South of this zone are rocks of Gala type containing Monograptus priodon, M. turriculatus, and M. riccartonensis, while in County Louth there are rocks comparable to the Kiccarton Group with Wenlock fossils. Graptolitic Valentian Beds have also been found near Pomeroy in Tyrone, and again far to the south-west in Queen's County, Clare, Tipperary, and Limerick, where large inliers of Silurian rocks are brought up by anticlinal domes from beneath the Old Red Sandstones. "Western Facies. In. the district lying between Lough Mask and Killary Harbour (County Mayo), and again in the Dingle pro- montory of Kerry, there are thick developments of the whole Silurian System, and here the deposits give evidence of the close proximity of land. In Mayo, north of Killary Harbour, Llandovery sandstones rest on Arenig Beds, but south of that inlet they overlap the Ordovician SILURIAN SYSTEM 179 a "s *l II II m * a .SP oO ^ II CO *,- S=W 31-- s s III III *O a3 08 J- Ofefe 2 O' f I 180 STRATIGRAPHICAL GEOLOGY and rest directly on the Archaean rocks, their base being a breccia consisting of angular lumps of quartz and schist. Above this are red mudstones and a band of calcareous grit with Upper Llandovery fossils succeeded by a massive conglomerate consisting mainly of rounded pebbles and boulders of quartzites like those of Connemara. Then comes a thick series of coarse grits alternating with green sandy slates which have yielded a few fossils of Wenlock species, Monograptus riccartonensis and M. vomerinus. These beds may be about 2000 feet thick. Above them are the Salrock Beds, dull red and green shales, with thin bands of grit, and having a thickness of about 3000 feet. They contain a small species of Lingula (L. Symondsi] with Pterincea retrqftexa and Modiolopsis complanata. They probably represent the greater part of the Ludlow and Downton Beds. 20 In the Dingle promontory the ascending succession is as follows 21 (see Fig. 60): (1) Smerwick Beds, a thick series of red, green, and yellow sandstones with bands of conglomerate ; fossils rare, but Atrypa hemispherica and others have been found. (2) Ferriters Cove Beds, greenish sands and shales with bands of red sandstone ; these beds are 2500 feet thick and contain fossils of Wenlock species. (3) Croaghmarhin Beds, hard brown calcareous grits with Pen- tamerus Knighti and other Ludlow fossils. These dip southward at a high angle and pass below a great series of green and purple grits and shales which are known as the Glengarriff grits, and believed to be of Devonian age. Here, also, there are evidences of contemporaneous volcanoes. The best sections are in the cliffs near Clogher Head, where the interstratification of the lavas and tuff's with the fossiliferous Wenlock Beds is beautifully exposed. Sir A. Geikie describes the lavas as nodular felsites, the nodules being from the size of a pea to that of a hen's egg, and the associated pyroclastic rocks vary from fine pumiceous tuffs to coarse agglomerates. One of these beds is a coarse agglomerate about 1 5 feet thick, consisting of large blocks of felsite embedded in calcareous sandstone which is full of the casts of brachiopods, crinoids, and corals. The district has been recently examined by Messrs. S. H. Eeynolds and C. T. Gardiner, who found that some of the lavas are rhyolites, and that all are of acidic character. 22 From these facts we learn that a volcano existed in this part of the Silurian Sea, and that from time to time it burst into eruption, pouring forth lava streams and ejecting showers of stones and ashes, SILURIAN SYSTEM 181 which fell into the surrounding sea and were spread out on its floor, while between each epoch of eruption marine sediments of the usual kind were greatly accumulated. The part of the sea in which these deposits were formed was probably not very far from the shores of a continent occupying a large area in what is now the Atlantic Ocean. D. CONTINENTAL EQUIVALENTS Rocks of Silurian age occur in Scandinavia, Russia, Poland, Thuringia, Bohemia, Carinthia, the Alps, Belgium, France, and Spain. We can only refer to the more important areas. 1. Scandinavia Undisturbed and unaltered Silurian strata not only occur in the same parts of Norway and Sweden, and succeed those of the Ordovician System (see p. 146), but they also form the island of Gottland in the Baltic, .which thus becomes a link between the Swedish and Russian areas (see map, p. 66). The Silurian Series exhibits several different facies in different parts of the Scandinavian region, and shows great variation in thickness. It is most fully developed in Scania, where sedimenta- tion in the form of graptolitic shales continued throughout the whole of Valentian and Wenlock times, a bed of limestone only occurring at the top of the system, the full thickness of which is there about 4000 feet. This great series is divisible into zones which can be correlated with those of Great Britain. In the Christiania district the deposits are very different, con- sisting of an alternating series of limestones and shales, and the total thickness is not much over 1200 feet, but is not complete at top. In Oland and Gottland, again, to the east the facies is essentially calcareous, shales prevailing only in the lowest part, and the whole system is condensed into a thickness of something less than 200 feet in Gottland. The base, however, is not seen, though the sequence extends down as low as the equivalent of our May Hill sandstone (Upper Llandovery), neither do the highest beds come into these islands. The Gottland limestones are shallow- water and not deep-water deposits. 23 They consist largely of crinoidal limestones, but these are replaced locally by beds which are so largely composed either of Stromatoporoids or Corals (or of both) that they have clearly been formed by the growth of these organisms in situ. There 182 STEATIGRAPHICAL GEOLOGY SILURIAN SYSTEM 183 are also beds of oolite and of marly limestone which are rich in Spongiostroma, a kind of calcareous alga, and thin bands made up of Sphcerocodium, which is another alga of solid globular form. The succession of beds found in these three districts is given in an abstract form on p. 182, the figures 1, 2, 3 indicating the equi- valents of the Valentian, Salopian, and Downtonian Series. From the above account it will be seen that the Silurian of Gottland is quite an abnormal development of the system, and as such is not at all suitable to be taken as its typical facies. Hence the fashion which has come into vogue among continental geologists of using the term Gothlandian instead of Silurian is unfortunate and unnecessary. It will be noticed that the equivalents of our Llandovery and Wenlock Beds are of comparatively small thickness, and that throughout the Scandinavian region the larger part of the sediments belong to the time of our Ludlow Beds with representatives of the Downtonian in Scania. The highest beds are most fossiliferous in the Kamsasa district of Scania, where they are often full of Leperditia inwqualis and Tentaculites. Homalonotus Knighti, Phacops Downingice, Spirifera elevata, Pterincea retroflexa, and a Pterygotus also occur. The Gottland type of the system is found again on the Kussian side of the Baltic in Esthonia and Livonia, the beds having clearly been once continuous across the intervening space. From the following table it will be seen that they are very little thicker than in Gottland, when allowance is made for their forming a more complete succession both at top and base. 24 Feet. ~/Grey and yellow limestones with Eurypterus Fischeri, Hemiaspis \ rufgosa, Platyschisma helicites ...... 60 ( Dolomite and marls, with a "Wenlock fauna .... 60 2- Limestone with Pentcwnerus estonus and Illcenus barriensis, [ passing into a coralliferous limestone ..... 50 f Raikull Beds with few fossils (Phacops elegans] .... 100 n J Limestone largely composed of Pentamerus borealis ... 30 I Joerden Beds shales and marls with Encrinus ptmctatus, V. Rhynchonella affinis, and corals ...... 25 325 These beds pass southward beneath the Devonian Series,, and when the Silurian reappears in Eussian Poland near Kielce it presents a very different facies. The sequence has not yet been completely worked out, but consists in the lower part of shales 184 STRATIGRAPHICAL GEOLOGY which seem from their graptolites to include representatives of zones from the horizon of the Tarannon shales to the Lower Ludlow (Climacograplus, Mon. priodon, Mon. colonus), succeeded by arenaceous beds (grauwacke) containing Beyrichia Klcedeni and Spirifer elevatus. Above these are sandstones, some of the beds con- taining remains of Placoderm fishes and others Spirifers. 3. Bohemia Passing now to Central Europe we still find in Bohemia a con- centrated and largely calcareous facies of the Silurian with a very rich fauna, the limestone beds being crowded with species of Cephalopods, Trilobites, Corals, Brachiopoda, and Mollusca of various kinds. Some of the species are identical with those of Britain and Western Europe, but many are different, though for the most part belonging to the same genera. The beds which represent the Silurian System are all comprised within the stages E and F of Barrande, the succeeding stages G, H, which he included in the system, having since been relegated to the Devonian (see Fig. 45 and p. 221). Barrande divided his stage E into two parts, E x comprising all the shales and E 2 the overlying limestone, but these do not corre- spond with palaeontological subdivisions. Moreover, the highest beds of shale are very calcareous and contain large spheroidal limestone - balls or concretions, while layers of shale containing Monograptus colonus are interbedded with the overlying limestones. The succession with Barrande's grouping is : Feet. FI. Black nodular limestone, with graptolites and other fossils . 150 E 2 . Grey limestones in thick beds with occasional layers of shale (many fossils) ......... 400 f Calcareous shales with Mon. colonus and Cyrtograptus Lundgreni] -c, I Black shales with Cyrtograptus Murchisoni .... [ ^ 1 Shales with Mon. turriculatus i M() ^Shales with Rastrites peregrinu s ........ J 750 The two lower zones are Valentian, the black shales correspond with our Wenlock shale, and the calcareous shale may be equivalent to the Wenlock limestone. The overlying limestones must repre- sent the whole of the Ludlow Series, for they contain a mixture of Wenlock species of corals and brachiopods with such essentially Ludlow forms as Pentamerus Knighti, Dayia navicula, and Mono- graptus colonus, as well as a host of peculiar trilobites, lamelli- branchs, gasteropods, and cephalopods. SILURIAN SYSTEM 185 F l has until recently been regarded as Devonian by German geologists, but the most recent study of the beds and their contents, by F. Seemann, 25 has shown that it is more intimately connected, both stratigraphically and faunally, with the top of Stage E than with the beds above. Thus it contains graptolites, Mon. colonus, M. priodon, Callograptus dichotomus, and others, with Orthoceras subannulare and other species, and lastly, no Goniatites have been found in it. Seemann, therefore, regards this black limestone (Fj) as the top of the Silurian Series, and as forming to some extent a passage from the Silurian to the Devonian. It is, how- ever, clearly marked off from the overlying white and red lime- stones which, in his opinion, form the base of the Devonian System (see p. 222). The fauna of the thick mass of limestone forming Stage E is a very large one, including no fewer than 183 species of trilobites belonging to the genera Acidaspis, Arethusina, Phacops, Proe'tus, Illcenus, LichaSj Cyphaspis, Cheirurus, Sphcerexochus, etc. Of Orthoceras there are 250 species, and of Cyrtoceras, 150. Lamelli- branchs are very numerous, and include the genera Dualina (100 sp.), Panenka (84 sp.), Gardiola (65 sp.), Lunulicardium (85 sp.), Paracardium, Prcecardium, Kralovna, Pterincea, Vlasta, Redonia, and Prcelucina. 4. Germany In Thuringia there is again a normal succession up to a zone of Mon. colonus, above which there appears to be a break in the sequence ; and in the Kellerwald the case is similar, but there shales with Gardiola interrupt^ Monograptus, and Cyrtograptus are succeeded by beds with Phacops, Harpes, and Hercynella before they are cut off by the unconformity. In the Hartz Mountains, however, the stratigraphy is very different. There the oldest rocks exposed are a thick series of grits, mudstones, and shales, with some limestones which seemed to contain a mixed fauna of Silurian and Devonian forms with graptolites. By Beyrich and Lessen they were called the Hercynian and were regarded as an intermediate series between the Silurian and the Devonian Systems, but recent researches have shown that the central core of the range is a dislocated and faulted mass or com- plex, and that large parts of it are really Silurian while others are Devonian. To the Silurian must be referred the greywacke of Tanne, and the black limestones of the Tannenthal, which contain Gardiola interrupta, as well as the Monograptus slates of Wieda ; though these seem to overlie beds with Devonian fossils, and were formerly regarded as the highest member of the series. 186 STRATIGRAPHICAL GEOLOGY 5. Belgium Silurian rocks of a normal character are only exposed in the same valleys as disclose those of Ordovician age (see p. 145). The succession according to the most recent researches of Professor Malaise is as given below : ~ 6 Dark slates with Monograptus colonus and M. Nillssoni. Sandy slates with Mon. vomerinus, Cyrtograptus Murchisoni, and Cardiola interrupts. Flagstones and shales with Mon. bohemicus. Black slates and quartzites with Climacograptus normalis. Grey slates with Phacops StoTcesi and Halysites. It will be noticed that nothing above the horizon of the Lower Ludlow is exposed, but higher beds probably exist under the cover of newer deposits. The Silurian, however, must thin out south- wards, for no such beds occur in the Ardennes, where a series known as the Gedinnien, and generally regarded as Lower Devonian, rests unconformably on Cambrian rocks. 6. France Silurian sediments occur in several parts of France, but generally present a shaly facies with small developments of lime- stone and are not rich in fossils. The most complete sequence of zones yet recorded is that of the west of Brittany (Crozon peninsula), where M. Kerforne has been able to recognise the succession of graptolite zones given in the table below. 27 West. East. Shales with Bolbozoe. " tnd OrtEL tU i S n ! ' * "-tone of Erbray. ,, with Mon. Salweyi. \ Black limestone of Briasse with ,. with Mon. colonus. / Mon. colonus and M. vomerinus. ~\ Shales with layers of calcareo- , , with Mon. Flemingi. siliceous nodules, Cardiola inter- ,, with M. riccartonensis. j rupta, and species of PanenJca, } Dualina, etc. with Cyrtograptus. S1 ^ 8 with Cyrtograptus gein- with M convolutns Cherts with Radiolaria and shales withJterfr&fandJfon. lobiferus. ,, with Rastrites Linnaei. Sandstones without fossils. In the central and eastern parts of Brittany, as well as in Mayenne and Anjou, a similar set of beds is found, though there are some lithological differences, and the graptolite zones have not yet been worked out. The two areas are compared above. SILURIAN SYSTEM 187 The thickness of this series of beds has not been estimated because of the frequent plications. There is also some doubt about the position of the Erbray limestone which contains some Devonian species, such as Spirifer Decheni, Cryphceus pectinatus, and at the top Harpes venulosus. In Normandy (Cotentin) the Silurian similarly consists of sand- stone at the base succeeded by black carbonaceous shales (ampelites), the zonal details being at present unknown. In Southern France (Languedoc) the Silurian presents a series of black shales and limestones about 650 feet thick. The lower part yields Mon. priodon, M. Flemingi, and higher up M. colonus and M. bohemicus. Black limestones at the top contain Gardiola interrupta, Slava bohemica, MaminJca comata, and other Bohemian species (of E 1 and 2) with numerous fragments of Orthoceras, the fauna being of essentially southern type. 7. Pyrenees, Spain, and Portugal In these countries the Silurian has for the most part an argillaceous facies like that of Brittany. In the Pyrenees it con- sists of black carbonaceous shales containing graptolites of Valentian and Wenlock species, but there are some calcareous shales, and in the Haute Garonne a black limestone contains several species of Orthoceras with Monograptus priodon. In Catalonia there is a more complete succession, as below : Black calcareous shales, Mon. colonus, M. Nillssoni. Soft black shales with M. priodon. Black nodular shales with M. turriculatus. Pale shales with M. proteus (Valentian). A similar succession is found in Portugal (Alemtejo and Barrancos), where the top bed is a thin band of limestone com- parable to that of Kozan in Brittany. E. CONDITIONS OF DEPOSITION The general geographical conditions over Northern and Western Europe during the earlier part of Silurian time seem to have been similar to those prevailing in the Ordovician period. There was the same mass of continental land in the Atlantic area, along the borders of which thick deposits of sand and mud were accumulated, but there seems to have been more land over Central England and in the area of the North Sea, for it was probably from such land that the great mass of arenaceous material forming the Wenlock 188 STRATIGRAPHICAL GEOLOGY Series of North Wales and Westmoreland was derived (see Building of the British Isles, 3rd edition). The much greater thickness of Silurian deposits in Norway and Scania (as compared with the underlying Ordovician) also testifies to the nearer presence of land. There was, however, the same open sea over the Baltic area wherein little but limestone was deposited. In the North British area, too, the same strip of comparatively deep water persisted for a time with small sedimentation, though after the epoch of the Birkhill shales the whole British area received a large share of the detritus carried off the land which lay to the north and north-east of it. The Silurian Sea probably extended over the same parts of the European region as were covered by that of the Ordovician period, but the land of Ardennes and the Ehine provinces, which may have existed in Ordovician time, becomes much more of a certainty in Silurian time. Thus in Belgium the Silurian extends south- ward to the Crete du Condroz, which separates the basins of Namur and Dinant, but must be overlapped farther south by the Devonian, for it does not emerge on the Cambrian areas of Rocroi and Stavelot (see Fig. 70, p. 216). Passing eastward we find the Devonian rocks forming a continuous band through the Eifel, Hunsruch, and Taunus districts, and resting near Frankfort on Cambrian exactly as they do in the Ardennes. Southward the land probably included the Odenwald and the Schwarz Wald of Baden, together with the Vosges district on the western side of the Ehine, where a patch of Devonian lies directly on the crystalline Archaean rocks, and no Silurian or Ordovician occur. It may have extended still farther south into Switzerland, but of this there is no actual evidence. The region above indicated measures about 230 miles each way, i.e. from west to east and from north to south, so that it probably included an area of over 70,000 square miles, which is nearly as large as England. Silurian strata are not found again till we go as far east as the Carnic and Carinthian Alps in Austria, and as far south as Sardinia, but this is probably due to their concealment by newer rocks, and it is very likely that the Silurian Sea was continuous from the south of France (Pyrenees and Languedoc) through Northern Italy to Carinthia, Styria, and Bohemia. It is noticeable that in Southern Europe, i.e. in Languedoc, Pyrenees, Spain, the Carinthian Alps, and Bohemia, the fauna found in the Silurian strata is very different from that in the more northern areas ; so that two distinct life provinces seem to have existed at this time in the European region. The conditions of deposition which prevailed in this Silurian SILURIAN SYSTEM 189 Sea of Western and Central Europe seem to have remained uniform for a considerable time and with little change except that of a slow subsidence ; so that little but shale was deposited, while here and there beds of limestone were accumulated, especially in the north-east. During the later part of the period crust movement took place, causing elevation and raising much of the northern part of the region above the level of the sea. With regard to the composition and construction of the Silurian deposits some interesting questions present themselves, especially in connection with the limestones. These often consist of a varied assemblage of organic remains, all the creatures then living in the sea which possessed calcareous shells or structural parts contributing to their formation, but occasionally some particular class of animals, or even one particular species, flourished in such numbers at certain places that their remains constitute almost the entire mass of the resulting bed of limestone. Thus in some parts of the sea-floor there must have been fields of Crinoids, growing in myriads as they do now in parts of the Caribbean Sea. By the growth and death of successive generations of these fixed " stone lilies," and by the subsequent disintegration of the steins, cups, and arms, beds of crinoidal limestone were slowly accumulated. In other places certain species of Brachiopoda flourished in such numbers that they must have occupied the sea-floor almost to the exclusion of other creatures, and their remains consequently form a Brachiopod limestone. Occasionally, too, one species so pre- dominates over the others as to be the main constituent of the limestone, as in the case of the Pentamerus borealis limestone of Russia (see p. 183). Other important components of Silurian limestones are Stroma- toporoids and Corals which often occur together, but sometimes separately form the chief constituent of a bed or mass of limestone. Some coralliferous limestones are of such a nature that it has been imagined they were coral-reefs of the same kind as those built up by corals at the present day. This, however, is very doubtful for the following reasons. In the first place, modern coral-reefs are either shore-formations resting unconformably on other rocks (fringing reefs), or they are steep-sided masses of limestone built up in the open sea with deep water on one or both sides of them ; and in areas of subsidence such reefs often reach to a great vertical height from the level from which they started. Now the only ancient coral-reefs that are likely to have been preserved are those which had been formed in areas of subsidence, where their lower parts would be flanked and 190 STRATIGBAPHICAL GEOLOGY protected by deposits of marly limestone composed of coral mud and the remains of other organisms. If such an ancient reef were subsequently raised and truncated by marine erosion and finally preserved as part of the earth's crust, we should have a mass of unstratified reef-rock sharply denned on all sides from the sur- rounding beds, and probably of great vertical height in proportion to its horizontal extension. It would not form a lenticular mass nor would it pass into other kinds of limestone. Again the corals which build up such reefs at the present day belong to families which did not exist in Palaeozoic seas, and only made their appearance in those of Triassic time. Corals of the single cup-shajted form are not likely to have been reef- builders, and though there were compound corals in the Palaeozoic seas, such as Heliolites, Favosites, Pachypora, and Alveolites, they belong to the Alcyonaria, and though the living genus Heliopora lives on coral-reefs, no reliable inference can be drawn from this fact with respect to the habit and growth of Heliolites and others ; we can only judge from what can be seen of their mode of occurrence in stratified rocks. The other group of animals which has been credited with reef- forming capabilities are the Stromatoporidce, which are not corals nor Actinozoa of any kind, but are believed to have been Hydrozoa allied to the modern Millepora. They formed massive hemi- spherical or irregular globular growths of concentric laminae, and varied in size from a few inches to masses of 3 or 4 feet in diameter. They were sometimes so abundant as to be the principal con- stituents of thick beds of limestone, which on account of the size of the component masses are often unstratified ; thus they certainly formed submarine banks, but such banks are not " reefs." The beds of coralliferous limestone which occur in the Silurian and Devonian strata of Britain, France, and Germany do not present the aspect of coral-reefs. They are merely lenticular beds or masses, of greater or less horizontal extent, and are often inter- stratified with shales or occur as bands in a mass of crinoidal limestone. Such a mode of occurrence proves that they were formed on a sea-floor in the ordinary way under a certain depth of water. The corals seem in some places to have formed fields or groves, like those of some modern species of Crinoidea, and the resulting limestone is unstratified, but such limestones must not be called reefs because there is no evidence that they were ever built up to the surface of the sea. Limestones which have a greater resemblance to coral-reefs occur in Sweden, and some are beautifully exposed in the cliffs of the island of Gottland, where they can be seen to terminate in some SILURIAN SYSTEM 191 cases abruptly against another kind of limestone. The Swedish geologists call these beds reefs, and apply the term not only to such beds as are largely made up of corals, but also to those of which the principal constituents are Stromatoporoids or Bryozoa. Thus, writing of the Ascoceras limestone, Munthe says, " It may be considered as a reef-limestone formed chiefly of Stromatopora and of large fragments of Crinoids with marly sediment to fill the interspaces," and it should be noted that the Cephalopod Ascoceras is common in this limestone. Good descriptions and photographs of these beds are given in the livret-guides provided for the Geological Congress at Stockholm in 1910 and from them it is plain that by the term " reef -limestone " the writers merely mean limestones composed of organic remains which have been accumulated in such a manner as to be unstratified. One such bed is shown to be overlain by stratified Sphaerocodium limestone, another by crinoid limestone, and massive colonies of Stromatoporoids sometimes replace crinoid limestone. It is evident that all the beds were formed in shallow water where the sea-floor was swept by strong currents, to the action of which the frequent abrupt termination of some beds and their replacement by others may be ascribed, but it is very doubtful whether any of them ever formed a reef in the proper sense of that term. KEFERENCES I C. Lapworth, "On the Geological Distribution of the Rhabdophora, " Ann. and Mag. Nat. Hist. ser. 5, vol. v. and vi. (1880). a Miss E. M. Wood, Quart. Journ. Geol. Soc. vol. Ivi. p. 415 (1900). 3 G. Maw in Geol. Mag. for 1881, p. 100 ; and C. Lapworth in Proc. Geol. Assoc. vol. xiii. p. 324. 4 The Misses Elles and Slater, Quart. Journ. Geol. Soc. vol. Ixii. p. 195 (1906). 5 H. Lapworth, Quart. Journ. Geol. Soc. vol. Ivi. p. 67. 6 The Misses Elles and Wood, Quart. Journ. Geol. Soc. vol. Hi. p. 273 (1897). 7 Miss G. L. Elles, Quart. Journ. Geol. Soc. vol. Ivi. p. 370. 8 Miss E. M. Wood, Quart. Journ. Geol. Soc. vol. Ivi. p. 415. 9 P. Lake, Quart. Journ. Geol. Soc. vol. li. p. 9. 10 T. M'K. Hughes, Quart. Journ. Geol. Soc. vol. xxxv. p. 694. II H. Hicks, Quart. Journ. Geol. Soc. vol. xxxvii. p. 482. 12 Miss G. Elles, Quart. Journ. Geol. Soc. vol. Ivi. p. 169. iy A. Strahan, Quart. Journ. Geol. Soc. vol. xxxv. p. 268. 14 Messrs. Marr and Nicholson, Quart. Journ. Geol. Soc. vol. xliv. p. 658. 15 Miss G. L. Elles, Quart. Journ. Geol. Soc. vol. liv. p. 463 (1898). 16 Misses Watney and Welch, Quart. Journ. Geol. Soc. vol. Ixvii. p. 215 (1910). 17 C. Lapworth, Quart. Journ. Geol. Soc. vol. xxxiv. p. 240 and vol. xxxviii. p. 587. 192 STRATIGRAPHICAL GEOLOGY 18 "Silurian Rocks of Britain: vol. i. Scotland," Mem. Geol. Survey. 19 See Summaries of Progress (Geol. Survey) for 1897 and 1898. 20 See Carruthers and Muffin Irish Naturalist, Jan. 1909. 21 Geol. Survey, Ireland, Expl. of sheets 160, 161, and 171. 22 Messrs. Reynolds and Gardiner, Quart. Journ. Geol. 'Soc. vol. Ivii p. 267. 23 See Munthe, Geol. Foren. Stockholm Forhand., Bd. xxxii (1900). 24 0. Schmidt, Quart. Journ. Geol. Soc. vol. xxxviii. p. 514. 25 F. Seemann, Beitr. Palaont. Oesterr.-Ung. xx. p. 69 (1907). 26 C. Malaise, Bull. Acad. Roy. Beige (1910). 27 Kerforne in Bull. Soc. Sci. and Med. Quest. (Rennes 1 ), vol. x. pp. 1-234 (1901). CHAPTER VIII DEVONIAN AND OLD RED SANDSTONE SYSTEM GENERAL CLASSIFICATION AND SUBDIVISIONS WE have seen in the last chapter that, where complete sections are found, the Silurian rocks pass upwards into a thick series of red sandstones (England and Scotland), or into an equally thick series of grey and purple grits (Ireland). It will be shown in a future chapter that the marine Carboniferous rocks always rest upon a set of red and yellow sandstones, which often repose unconformably upon some of the more ancient Palaeozoic strata. The marine fauna of the Silurian dies out at the base of the red sandstones, and when we reach the next marine fauna that of the Carboniferous System it is totally different from the preceding, so that a long period of time must have elapsed between the close of the Silurian era and the commencement of the Carboniferous. The groups known as the Old Ked Sandstone in England and Scotland, and the Glengariff grits or Dingle Beds in Ireland, are clearly some of the rocks which were formed in this interval ; and if these were the only records of this portion of geological time, we might find it convenient to regard the whole of the beds which succeed the Silurian Series in conformable sequence as belonging to the Silurian System. But these are not the only strata which are interposed between the Silurian and Carboniferous Systems. In Devonshire, and in many localities in Europe, there is a thick series of rocks contain- ing a marine fauna which is evidently of intermediate age ; further, in Germany the stratigraphical proof of their interposition between Silurian and Carboniferous rocks is complete. It is this marine series which must be taken as the normal type of the intermediate system, and any freshwater deposits which can be shown to be of the same age must be treated as local contemporaneous or homotaxial strata. In America also both types of the Devonian System are present. 193 194 STRATIGRAPHICAL GEOLOGY The Devonian System was established by Sedgwick and Murchison in 1839 because they found in Devon a great series of rocks containing a special assemblage of fossils, and they divided this Devonian System into Lower, Middle, and Upper Groups; the convenience of this division has been confirmed by the researches of continental geologists. Sedgwick. and Murchison also took a leading part in the investi- gation of the European equivalents of these Devonian rocks, visiting the Rhenish mountains and indicating the existence of rocks con- taining a Devonian fauna in the Eifel district (1842). The more complete delimitation of the Devonian System was accomplished by F. Romer (1844), Von Dechen (1855), and the Sandbergers (1850-56). A little later came the important researches of Gosselet on the rocks of the Ardennes (borders of France and Belgium), where he established the same order of succession. The Old Red Sandstone, which elsewhere occupies the position of the Devonian System, is believed to be of lacustrine or estuarine origin. It has been divided into a lower and upper series, with a middle series in Wales, but in Scotland there is a marked uncon- formity between the upper and lower series, and some are doubtful whether the upper series is really of Devonian age. It is probable that the Lower Old Red Sandstone may be correlated with the Lower Devonian, that the Middle Devonian was formed during the time represented by the unconformity in Scotland and Ireland, but how much of the Upper Old Red Sandstone is really of Devonian age will depend very largely on the results of further researches in North Devon and Ireland. In dealing with the records of the Devonian period we shall describe the marine facies first, both in England and on the conti- nent, and secondly the rocks known as the Old Red Sandstone. I. THE DEVONIAN FACIES A. THE MARINE FAUNA As already mentioned, the Devonian fauna is intermediate in character between those of the Silurian and Carboniferous. Most of the characteristic Silurian genera become extinct in the Lower and Middle Devonian, while in the Middle and the Upper Devonian many genera which are prominent members of the Carboniferous fauna make their appearance. This is especially the case with the Brachiopoda, for all the Lower Devonian genera are found in the Silurian, while all those in the Upper Devonian range into the Carboniferous. The following is a brief account of the principal genera which DEVONIAN AND OLD RED SANDSTONE SYSTEM 195 survive from the Silurian period, and of the more important genera which make their first appearance in Devonian rocks. Among Hydrozoa the Stromatoporoids, which had some representatives in Silurian time, become specially abundant, and are characteristic of Devonian limestones. The principal genera are Stromatopora, Stromatoporella, Actinostroma, Parallelopora, Amphipora, and Stachyodes. Among Actinozoa, Acervularia, Alveolites, Arachnophyllum, Cystiphyllum, Pachypora, Favosites, Heliolites, Cyathophyllum, Plasmopora, and Syringopora continue from the Silurian ; and Zaphrentoid corals, which are rare in the Silurian, now become abundant. Calceola, Heliopliyllum, and Pleurodictyum occur only in Devonian rocks. Michelinia, Zaphrentis, Mesophyllum, and Phillipsastrea appear and range into the Carboniferous. Among Echinoderma the crinoids Cyathocrinus, Ichthyocrinus, and Taxocrinus continue. Ctenocrinus, Cupressocrinus, Hexacrinus, Melocrinus, Sphcerocrinus, and Rhipidocrinus are specially Devonian genera. Poteriocrinus and Haplocrinus appear and survive. Of Echinoids certain archaic forms occur, such as Lepidocentrus, which has from five to eleven rows of interambulacral plates. Among Crustacea trilobites are on the decline, though indi- viduals are sometimes common. Eepresentatives of twelve genera Tiave been found in British Devonian rocks, Acidaspis, Bronteus, Cheirurus, Cyphaspis, Lichas, Harpes, Homalonotus, Dalmanites, Phacops, Trimerocephalus, Proetus, and Dechenella. Of these Phacops, Proetus, Bronteus, and Cyphaspis are the commonest. The species of Dalmanites belong to the section or sub-genus Gryphceus. The Ostracods Beyrichia and Entomis are also common. Of Brachiopoda many of the Silurian genera survive and many new forms make their appearance ; of the latter Centronella, Davidsonia, Rensselceria, Megalanteris, Stringocephalus, and Uncites are known in Britain as Devonian fossils only, while Camarophoria, Productus, Strophalosia, and Terebratula (Dielasma) range into the Carboniferous. Of Lamellibranchia the following Silurian genera occur, but those with an asterisk die out within the period : Ambonychia,* Myalina, Grammysia,* Leptodomus, Ctenodonta, Cardiola* Cleido- phorus* Conocardium, Cyrtodonta,* Nucula. Pterinea is very abundant, and the sub-genus with elongate wings (Actinodesma) is specially characteristic. The following genera appear and range into higher strata : Aviculopecten, Ptyclwpteria, Cucullcea, Paral- lelodon, Pleurophorus, Edmondia, Prothyris, Megalodon, Curtonotus, Myophoria (?), and Solenopsis. Among Gastropoda the following genera survive from the 196 STRATIGRAPHICAL GEOLOGY Silurian and have many Devonian species : Murchisonia, Pleuroto- maria,j Euomphalus, Straparollus, Platyceras, Loxonema, Macro- chilina, and Better -option. Besides these Cyclonema, Callonema, and Eunema occur, but die out within the period. Scoliostoma and Philoxene are Devonian only ; Naticopsis and Porcellia appear and range upward. Among Cephalopoda the following continue : Orthoceras, Rf Fig. 61. GROUP OF DEVONIAN FOSSILS. a. Bronteus granulatus. b. Bront. granulatus (pygidium). c. Calceola sandalina. d. Stringocephalus Burtini. e. Pleurotomaria aspera. /. Clymenia striata. Gyrtoceras, Gyroceras, Poterioceras, Gomphoceras, and Trochoceras. The last genus becomes extinct. The Ammonoidea make their first appearance and are represented by Clymenia, Bactrites, and the Goniatitidae, of which the following genera occur : Mimoceras, Anarcestes, Agoniatites, Gephyroceras, Tornoceras, Mceneceras, Branco- ceras, Beloceras, and Prolecanites. Clymenia is found only in the Upper Devonian. Of Pisces, Pteraspis, Cephalaspis, Drepanaspis, Coccosteus, Parexus, and Homosteus occur. DEVONIAN AND OLD RED SANDSTONE SYSTEM 197 Fig. 62. GROUP OF DEVONIAN FOSSILS. a. Cystiphyllum vesiculosum. ft. Heliolites porosus. c. H. porosus (magnified section). d. Rhynchonella cuboides. e. Spirifer Verneuillii. /. Murchisonia angulata. p. Harpes macrocephalus. h, Phacops latifrons. 198 STRATIGKAPHICAL GEOLOGY The following are some of the most characteristic fossils of each division as found in England, France, and Belgium. Fossils of the Lower Devonian Actinozoa. Pleurodictytim problematicum, Zaphrentis oolitica. Echinoderwia. Cyathocrinus megastylus, Ctenocrinus decadactylus. Brachiopoda. Spirifer primsevus, S. auriculatus, S. cultrijugatus, S. paradoxus, S. hystericus, Orthis arcuata, Chonetes dilatata, C. sarcinulata, Rensellseria crassicosta, Leptsena laticosta, Rhynchonella daleidensis. Fig. 63. GROUP OF UPPER DEVONIAN FOSSILS. b. Rhynchonella Partridgise. c. Ptychopteria damnoniensis. e. Cucullsea unilateralis. /. Curtonotus elegans. Lamcllibranchia. Pteria anisota, Pteria spinosa, Pterinsea costata, Pt. (Kochia) capuliformis, Ctenodonta Krachtse. Crustacea. Homalonotus annatus, H. Champernowni, H. gigas, Dalmanites laciniatus. Gastropoda. Bellerophon trilobatus, Platyceras fecundus, Macrochi- lina elongata. Pisces. Drepanaspis gemundensis, Coccosteus angustus, Pter- aspis. Fossils of the Middle Devonian Hydrozoa. Stromatopora concentrica, Actinostroma stellulatum, Ampliipora ramosa. Actinozoa. Pachypora cristata, Phillipsastrsea pengellyi, Meso- phyllura damnoniense, Cyathophyllum caespitosum, C. obtortum, C. hexagonum, Cystiphyllum vesicu- losum, Alveolites suborbicularis, Calceola sandalina, Heliolites porosus. DEVONIAN AND OLD RED SANDSTONE SYSTEM 199 Echinoderma. Sphserocrinus geometricus, Hexacrinus interscapularis, Cupressocrinus crassus. Crustacea, Proetus batillus, Bronteus granulatus, B. flabellifer, Phacops Schlotheimi, Cyphaspis ocellata. Brachiopoda. A try pa reticularis, A. desqiiamata, Stringocephalus Burtini, Pentamerus brevirostris, Meristella plebeia, Spirifer undatus, S. speciosus, Uncites gryphus, Streptorhynchus umbraculum, Cyrtina heteroclita. Lamellibranchia. Megalodon abbreviatus, Conocardium clathratum, Cypricardinia scalaris, Actinopteria placida, Rutotia elliptica. Gastropoda, etc. Loxonema reticulatura, Macrochilina arculatum, Murchisonia turbinata, M. coronata, M. bilineata, Pleurotomaria: delphinuloides, Philoxene Isevis, P. serpens, Porcellia bifida. Cephalopoda. Agoniatites occultus, Anarcestes subnautilinus, Mimo- ceras gracile, Maeneceras decheni, M. moJarium, Trochoceras obliquatum, Orthoceras laterals, Gyro- ceras nodosum. Pisces. Species of Heterosteus, Coccosteus, and Homosteus. Fossils of the Upper Devonian Actinozoa. Acervularia Goldfussi, A. pentagona, Chonophyllum perfoliatum, Phillipsastrsea Hennahi. Echinoderma. Actinocrinus Porteri. Crustacea. Phacops granulatus, P. latifrons (also in Mid. Dev. ), Entomis serrato-striata. Brachiopoda. Spirifera Verneuilli, S. Urei, S. distans, Orthis in- terlineata, Strophalosia productoides, Rhynchonella cuboides, R. laticosta, R. letiensis. Lamellibranchia. Ptychopteria damnoniensis, Cucullsea unilateralis, Ctenodonta ( Palseoneilo) lirata, Myophoria deltoida, Prothyris scalprata, Aviculopecten nexilis, Cardiola retrostriata (Cardium palmatum), Posidonomya venusta. Gastropoda. Loxonema anglicum, Pleurotomaria aspera. Cephalopoda. Gephyroceras intumescens, Tornoceras simplex, Belo- ceras sagittarium, Clymenia undulata, Clymenia striata, Bactrites gracilis. B. BRITISH DEVONIAN ROCKS In Britain the only surface exposures of undoubted marine Devonian rocks occur in Devon, Somerset, and Cornwall; they occupy the northern parts of Devon and Somerset between Morte Bay and the valley of the Tone, as well as the outlying tract of the Quantock Hills, east of that valley. Dipping southward beneath the central Carboniferous area, they reappear in South Devon and North Cornwall (see map, Fig. 64). The strike of these rocks is nearly due east and west, and this appears to be maintained in their subterranean extension through the south of England, for rocks with the same Devonian fossils 200 STEATIGRAPHICAL GEOLOGY have been met with in borings beneath the Cretaceous Beds which underlie London and Essex. They are known to occur in the same position in Belgium, and they emerge again to the surface in the Ardennes, so there is every probability of their being con- tinuous across the intermediate areas. I. Devon and Cornwall The Devonian rocks in England have been so crushed, plicated, and broken by faults that it is no easy matter to make out the original order of succession either in North or South Devon, and it is almost impossible to estimate their original thickness. It will be noticed that the section, Fig. 66, shows a general inclination of the beds towards the north, newer and newer beds coining on in that direction. This is the rule in South Devon and North Cornwall, and yet the prevalent dips actually observed and recorded on the maps of the Geological Survey are toward the south, which would lead one to infer a complete inversion of the beds. These dips, however, are those of the minor plications, for not only is the whole system flexured into a succession of curves, but each flexure is frilled into smaller plications. The following account is compiled from the memoirs of the Geological Survey on the new map of Cornwall and South Devon, and from papers by Mr. W. A. E. Ussher, supplemented by some observations of my own ; for North Devon the publications of Jukes, Etheridge, Hicks, and others have been consulted. The rocks of both the northern and southern areas are divisible into Lower, Middle, and Upper Series, and in order to deal with each of these series separately we must first give a tabular view of the subdivisions which have been made in each area. In con- structing this table a group of rocks occurring in North Devon and known as the Morte slates has been omitted, for reasons which will be mentioned, and because the correlation of the other groups is complete without them. On the map, Fig. 64, these Morte slates are marked as a separate tract of Lower Devonian. Cornwall and South Devon. North Devon. 1 Summit not seen. Upper ^ Styliola and Entomis slates. [ Calcareous slates and limestones. M - -i 1-, /South Devon limestones, with Le \ associated slates and tuffs. {Staddon grits. Meadfoot Beds. Dartmouth slates and Grampound Beds. Pilton Beds. Baggy Beds. Pickwell Down Beds. jllfracombe Beds. Hangman grits. Lynton Beds. Foreland sandstones. (Base not seen.) GEOLOGICAL. MAP OF PARTS OF SOMERSET, DEVON & CORNWALL AFTER , W.A.E.Ussher Upper Culm MiddleCul Lower Culm Upper & Middle Devonian Lower Devonian ^.^.xr -I Sc h i stose Rocks Granite -Fig. 64. Stanftrds Oeoynipf,' Estu&f, ton Jon. 202 STEATIGKAPHICAL GEOLOGY Lower Devonian. The recent work of the Geological Survey in Cornwall has shown that the basal boundary of the Lower Devonian crosses that county along an east and west line from Forth Towan on the west to near Tregony on the east, whence it bends southward to Veryan and Nare Head (see Index map, 4 miles to 1 inch, sheet 25). From this line northward to St. Evan, Bodmin, and Plymouth, the country is occupied by Lower Devonian grits and slates, interrupted only by the intrusive mass of granite which lies to the north of St. Austell and St. Blazey. Along the line of boundary above mentioned Devonian grits and conglomerates rest unconformably on rocks which are either of Ordovician or Cambrian age ; but these grits do not appear to be the oldest Devonian strata. The surveyors think that the Dart- mouth slates are concealed beneath an overlap of the higher beds on to a sloping shore-line, and believe the conglomerates of Grampound, Probus, and Nare Head to be shore deposits belonging to the Meadfoot division of the series. The general structure of North Cornwall is expressed in Fig. 65, which indicates the supposed overlap and the series of monoclinal flexures by which the beds seem to be repeated, but the section must be regarded as merely a diagram of the theoretic structure of the area. The Dartmouth slates form a thick series of purple, red, and green slates with frequent bands of hard and fine quartzitic grit such as are called quartzo-phyllades in France and Belgium. They seem to have a thickness of over 2000 feet, but the base has not been seen because it is not brought up in any of the anticlinal flexures. The general position of these beds, according to the view taken by the Geological Survey, is indicated in Fig. 65. The only definite marine invertebrate yet found in this series is Bellerophon trilobatus, which is stated to be fairly common though badly preserved. Fish remains also occur belonging to the genera Pteraspis, Cephalaspis, and Parexus. The Meadfoot Group is another set of slates and fine grits, but is distinguished from the Dartmouth slates by the colour of the beds, this being always dark grey. In the Looe district the lowest beds are hard grits and quartzites with bands of grey slate, and it is these beds which are believed to pass southward into the coarse grits of Grampound arid Probus. The faults and plications are so numerous that the thickness may be anything from 500 to 1000 feet. The Meadfoot Beds are more fossiliferous than those below, but the specimens are seldom in good preservation. The following are some of those which have been determined : Pleuro- dictyum problematicum, Pachypora ( ? cristata), Spirifer primcevus, Sp. hystericus, Orthotetes hipparionyce, Stropheodonta gigas, Rensellceria DEVONIAN AND OLD RED SANDSTONE SYSTEM 203 f 1 -s s 31 II 1 &s 204 STRATIGEAPHICAL GEOLOGY sp., and Phacops Ferdinandi. The place whence the beds take their name is Meadfoot Beach, east of Torquay, and near that locality they include bands of fossiliferous brown grit (originally calcareous), from which Homalonotus armatus, Leptcena laticosta, Chonetes sarcinulata, Spirifer hystericus, and Sp. paradoxus have been obtained. The Staddon grits, which take their name from Staddon Point near Plymouth, are red and purplish -grey grits with intercalations of red shale and mudstone. Their outcrop is nearly continuous from St. Breock Downs to Plymouth, and thence eastward to the coast south of Brixham ; they are brought in again to the north- ward around Paignton by a local anticline. At Torquay they are exposed on the Warberry and Lincombe Hills which form the centre of a complex and faulted anticline (see Fig. 67), and they have yielded a few fossils, among which are Leptcena laticosta, Bellerophon trilobatus, Streptorhynchus umbraculum, Spirifers, and species of Orthoceras, Cyrtoceras, and Homalonotus. Passing now to North Devon the first point to be noted is the difference in the lithological character of the lowest beds. The Foreland sandstones occupy the position of the Dartmouth slates, and they contain remains of Pteraspis, but they are hard sandstones or grits, of red, brown, and grey colours, mostly of fine grain, but including some coarse and even pebbly beds. They were clearly deposited much nearer to a shore-line which must have lain in the space now occupied by the Bristol Channel. The Lynton Beds consist chiefly of hard slates with some beds of grit and occasional calcareous bands, and are supposed to be from 1200 to 1500 feet thick. Fossils are fairly abundant, but in a bad state of preservation. The following have been recorded : Spirifer primmvus, Sp. hystericus, S. Icevicosta, S. speciosus, Orthis arcuata, Chonetes sordida, and Orthotetes umbraculum. These beds are succeeded by the Hangman grits, which consist of brown, green, yellow, and red sandstones with some coarse red and speckled grits in the upper part. They form the cliffs between Woodabay and Combe Martin Bay, and are probably more than 1500 feet thick. No fossils except a few casts in the highest beds have been found. The Hangman grits correspond in every way with the Staddon grits of South Devon, and the Lynton Beds as clearly represent the Meadfoot Beds. The Morte slates have already been mentioned, and their position is shown on the map (Fig. 64), and in the section, Fig. 68. Jukes was the first to suspect the existence of a great fault along the southern boundary of the Morte slates, 1 but he imagined that these slates were of Carboniferous age, and that DEVONIAN AND OLD RED SANDSTONE SYSTEM 205 5 w"" | I ^ -. g g -S o O tH H< 03 ^ 63 iS cS rfi 1 1 ye fe -.s SBW-^i -fi Smew 206 STRATIGRAPHICAL GEOLOGY the whole of the North Devon rocks were duplicated by the fault, the Pickwell Down Beds being a repetition of the Foreland Beds, and the succeeding beds on each side of the fault being comparable with one another. This view was combated by Mr. Etheridge, 2 who found it easy to show that the fossil evidence made it untenable, for the fossils of the Lynton Beds are unquestionably Lower Devonian, while all the fossils found in beds south of the Pickwell Down Group are with equal certainty of Upper Devonian or Carboniferous age. When Mr. Etheridge wrote, no fossils had been found in the Morte slates, but in 1895 and 1897 Messrs. Hicks and Whidborne 3 found fossils at several places along their outcrop, and the assemblage appears to indicate the existence of a Lower Devonian fauna. In the papers referred to Dr. Hicks stated his belief that not only is there a fault along the southern boundary of the Morte slates, but one along their northern border, that these slates are in fact thrust up between the two lines of fault, and that they are the oldest rocks in North Devon, but the final conclusion is not generally accepted. It is true that the Morte slates differ from any other slates in North Devon, for they are smooth and glossy, while the Lynton slates are mostly rough and gritty ; still it is more probable that they are a facies of the Lynton slates than beds older than the Foreland sandstones. The fossils referred to were obtained from Morte Bay and Mullacott on the coast, and from Treborough in Somerset. Some of the species are described as new (such as Lingula mortensis) ; some resemble Silurian forms, such as Orthis rustica ; and others are Lower Devonian, e.g. Dalmanites laciniatus, Strophomena tceniolata, S. explanata, Chonetes plebeia, and C. sarcinulata. Middle Devonian. In South Devon this consists of a series of grey shaly slates in which masses of limestone are lenticularly developed, and in places both are partially replaced by volcanic materials stratified tuffs and diabase lavas. The chief interest centres in the limestones, and these are best known from their developments at Plymouth, Brixham, Torquay, Totnes, Ipplepen, and Ogwell. The lowest beds, however, are always grey shales, and under the lowest limestone near Torquay these have yielded Calceola sandalina, Spirifer speciosus, Sp. curvatus, Atrypa reticulata, and some other fossils. They evidently correspond with the beds known as Calceola shales in Belgium. The limestones certainly occur on more than one horizon, and Mr. Ussher has classified them as lower and higher limestones. 4 The lower set are generally dark grey or nearly black, are well bedded, and consist mainly of crinoid ossicles, beds containing corals DEVONIAN AND OLD EED SANDSTONE SYSTEM 207 and Stromatoporoids only occurring at intervals and never being more than 3 or 4 feet thick, while the thickness of such bedded limestone may vary from 100 to 300 feet. Such limestones are typically developed round Dartington and Little Hempstead, near Totnes ; they also occur at Ashburton, at Galmpton near Dartmouth, at Torquay, and Babbacombe. Their most characteristic fossils are : Pachypora reticulata, Favosites fibrosus, Cyathophyllum ccespitosum, Oystiphyllum vesiculosum, Mesophyllum helianthoides, and Alveolites vermicularis. The higher limestones are of much lighter colour and often make beautiful marbles, being grey or nearly white, pinkish, or grey mottled with red. They are generally massive and unstratified and are frequently largely composed of Stromatoporoids, some masses of which are from 3 to 4 feet across ; compound corals occur as scattered and generally broken lumps, the commonest being Pachypora cristata and Heliolites porosus. These limestones in the Plymouth and Torquay districts pass up into beds which contain a mixture of Middle and Upper Devonian species. These are also massive but shelly limestones, the chief constituents being crinoids, small Stromatoporoids, broken corals and shells. Of corals the commonest are Pachypora cristata, Alveolites suborbicularis, and Phillipsastrea (Smithia) Pen- gellyi. Of shells there are Stringocephalus Burtini, Rhynchonella cuboides, Pentamerus brevirostris, Spirifer curvatus, Atrypa reticularis, and Cyrtina heteroclita, with, less commonly, the trilobites Bronteus granulatus and Gyphaspis occellata. The fauna of this limestone has been described and figured for the Palseontographical Society by the late G. F. Whidborne. The thickness of massive limestones near Torquay is probably over 200 feet. In North Devon the Middle Devonian consists of a series of grey slates and flagstones with lenticular layers of earthy limestone. The lower ,part is sometimes called the Combe Martin Beds, but they are only the lower part of the Ilfracombe Beds and contain a similar fauna ; the lower beds are rather more sandy and flaggy, while the higher beds are silvery-grey slates. The whole series is intensely plicated, compressed, and cleaved, the observable dips being generally those of isoclinal folds dipping southward so that the actual thickness of the series is nothing like so great as the breadth of basset-surface would seem to indicate, and may not be more than from 1000 to 2000 feet. Like all the other divisions of the Devonian in North Devon and Somerset it occupies a continuous band or tract of country, crossing Exmoor and the Brendon Hills from west to east. It crops out again in the Quantock Hills, and presumably it and the 208 STRATIGBAPHICAL GEOLOGY Lower Devonian had some extension to the northward, but both are absent above the Silurian outcrop in the Mendip Hills, which are only 1 7 miles north-east of the Quantocks. Upper Devonian. In South Devon at certain places, as for instance Plymouth, West Ogwell, and Chudleigh, the massive limestones pass up into red or pink limestone containing Acervularia Goldfussi, A. pentagona, and Ghonophyllum with some- times Rhynchonella cuboides and Spirifer bifidus. In other places a massive grey unfossiliferous limestone is succeeded by red shales with layers of nodular limestone containing the characteristic Goniatites (see p. 199) with Oardiola retrostriata and Phacops cryptophthalmus. These beds are succeeded by red and green shales in which the small Crustacean Entomis serrato-striata is often abundant, with occasionally the small bivalve Posidonomya venusta. West of Dartmoor, in North Cornwall, these higher beds have a much greater development, occupying a broad tract of country extending from between Padstow and Boscastle on the coast round the north side of Bodmin Moor to the borders of Dartmoor between Tavistock and Bickleigh. They consist of grey, green, red, and purple shales in which impressions of the Pteropod Styliola are often abundant as also is the Entomis. Other fossils only occur locally, but quarries near South Petherwin have yielded a large fauna, the following being some of the species : Clymenia Icevigata, C. striata, Orthoceras Phillipsi, Tornoceras lineare, Ctenodonta antiqua, Cardiola retrostriata., Spirifer Verneuilli, Orthis interlineata, Strophalosia subaculeata, and Phacops granulatus. In the cliffs on the west coast it has been possible to make out something like a regular succession in spite of faults and plications. The data on which this is based are given in the Survey memoir on the country round Padstow, but the following table is not a copy of that given in the memoir, as I have omitted the flaggy slates of Camel Quarry, which may be of Middle Devonian age and have separated the lower part of the " Striped Slates." The descending succession then reads thus : 5. Purple, green, and variegated slates. 4. Dark slate and pillow lava. 3. Soft black slates with limestone bands (many fossils). 2. Grey slates and thin limestones and a fossiliferous band yielding many Cephalopoda. 1. Grey shales with scattered phosphate nodules, thin layers of lime- stone and of black grit. The phosphatic nodules in the lowest beds contain Conularia complanata ; the limestone of Dinas Head (in No. 2) contains black cherts with Radiolarian remains ; while the overlying black slates DEVONIAN AND OLD RED SANDSTONE SYSTEM 209 are full of Tentaculites, Styliola, and Buchiola retrostriata, and certain beds have yielded species of Orthoceras, Bactrites, Tornoceras, Mimoceras, Anarcestes, and Agoniatites with remains of Phacops latifrons, P. granulatus, and P. Icevis. The variegated slates contain Entomis serrato-striata and Posidonomya venusta, and they occupy a syncline which forms an oval area around St. Minver and St. Kew. Nothing which exactly corresponds to the Petherwyn fauna has been found on or near the coast, and it is thought that its strati- graphical position is above that of No. 5 ; its absence in the west being due either to an overlap of the basal Carboniferous Beds or to faulting. In North Devon, on the other hand, there appears to be a complete succession, and the difficulty is to determine where the line of division between the two systems should be drawn. The series has been divided into three stages, the lowest being the Pickwell Down Beds, a set of red, purple, brown, and green sandstones with intercalated bands of shale. They are traceable from Pickwell Down near Morte Bay to Wiveliscombe in Somerset, but the only fossils yet found are fish remains and fossil wood. The Baggy Beds consist of green shales and yellowish sandstones and flags, and extend inland from Baggy Point, north of Croyde Bay, by Marwood and Sloly, where are quarries from which many fossils have been obtained, including Ptychopteria damnoniensis, Cucullcea unilateralis, Rhynchonella laticosta, Spirifer Verneuilli, Strophalosia productoides, Chonetes hardrensis, and plant remains (Stigmaria and Knorrid). The Pilton Beds consist of bluish-grey slates with thin bands of limestone and of sandstone. They stretch from Croyde Bay inland by Braunton and Pilton and are highly fossiliferous, con- taining among others Phacops latifrons, Productus prcelongus, Spirifer Verneuilli, S. Urei, Orthis interlineata, Strophalosia productoides, Aviculopecten nexilis, Loxonema anglicum, and other Gastropoda. They appear to pass upward into the Lower Carboniferous shales, and the divisional line has not yet been fixed ; but this point will be discussed on a later page. 2. The South of Ireland The Devonian rocks of Southern Ireland have generally been treated under the head of Old Red Sandstone, but the more complete knowledge which we now possess of their equivalents in Devon, Cornwall, and Brittany makes it probable that they are P 210 STRATIGRAPHICAL GEOLOGY almost entirely of marine origin, and are really comparable with the Devonian Series of these more southern areas. The upward passage from Silurian to Devonian is exhibited in the coast section of the Dingle promontory in Kerry, where the beds with Ludlow fossils are conformably overlain by grey, brown, and purple slates without fossils, followed by alternations of grey grits and slates, the grits at length predominating and forming a thick series of hard, massive green and purple grits. 5 The whole of this Gritstone Series is perhaps 8000 feet thick, and is succeeded by slates of red and purple tints, with beds of conglomerate at Parkmore, which curiously enough enclose pebbles containing Silurian fossils. No other fossils have yet been found, and except for the occurrence of these derived pebbles, the whole of these slates and grits might be taken for a continuation of the Silurian rocks, so similar are they to the underlying fossiliferous series. A similar set of green and purple grits occupies large areas to the south and south-west of Dingle Bay in the Inveragh and Dunkerron promontories, and in the districts of Killarney, Kenmare, and Glengarriff. It has generally been assumed that these " Glengarriff Beds " are the equivalents of the Dingle Beds, but Jukes was doubtful on the point, and Mr. M'Henry is of opinion that the latter are entirely below the Glengarriff Beds. It is possible, however, that the actual base of the latter does occur in the Dingle promontory. The great series known as the Glengarriff Grits consists of several thousand feet of alternating bands of fine-grained grit and slate. The lower part is mainly composed of green grits with subordinate beds of slate ; while in the higher part purplish-red slates predominate over the grits. No minor subdivisions have yet been made, and no fossils have been found in them. The thickness of the whole series has been roughly estimated at 8000 feet, but when allowance is made for folds and faults it may prove to be much less than this figure. 6 The beds have been subjected to such pressures that the shales are cleaved into slates, and the grits rendered nearly as hard as quartzites. They have also been thrown into a succession of regular anticlinal and synclinal folds which range east and west from the west coast of Kerry to the eastern part of Cork. The hard grits of the lower series, though exposed, in anticlines, form long ranges of mountains because of their capacity for resisting subaerial erosion, while deep longitudinal valleys have been excavated through the purple slates and the still softer black slates which overlie them and come into the synclines (see map, Fig. 94, and section, Fig. 72). DEVONIAN AND OLD RED SANDSTONE SYSTEM 211 In Kerry and the west of Cork there seems to be a com- plete upward sequence from the greenish grits through the purple slates to a series of grey and brown grits with intercalations of dark-grey slate, which group is known as the Coomhola Beds. These contain a marine fauna which is similar to that of the Baggy and Marwood Beds of North Devon, and they are succeeded by a set of dark-grey and black slates which resemble the Pilton Beds, but their fossils have not yet been collected with sufficient care to separate the Devonian and Carboniferous portions of the series. Fig. 69 is a copy of the section drawn by Jukes to illustrate the relative position of the beds near Glengarriff Harbour at the head of Bantry Bay. Jukes could find no evidence of unconformity ; but Professor Hull, going over the district in 1878-80, thought he could detect d c c b a Fig. 69. SECTION NEAR GLENGARRIFF (Jukes). Distance about 2 J miles. Feet e. Black slates with calcareous bands ) K f onon d. Black and grey slates / ' ' about 200 o. Coomhola grits and slates 3000 a. Lower } Glen S arriff S rits ' 500 an unconformity between the Coomhola Beds and the Glengarriff Series at Sneem near Kenmare. 7 His section, however, is by no means convincing. In the more recent revision of the country round Cork the Geological Surveyors 8 have failed to find any evidence of a break between the two series, the Upper Glengarriff Beds in that district being succeeded by yellow sandstones and green shales which occupy the place of the Coomhola Beds, and are overlain by shales with Carboniferous fossils. It is true that these yellow sandstones are not the Coomhola Beds and do not contain marine fossils, but only fish and plants of Upper Devonian species. They are in fact the Kiitorcan Beds or Upper Old Red Sandstone of Kilkenny. From the above account the student will understand that there are still several points to be cleared up in the geology of the south-west of Ireland, and in particular the exact relation between the Coomhola and the Kiitorcan Beds remains to be ascertained. If, however, we consider the sequence found in West Cork and Kerry by itself, we may regard it as a succession of presumably 212 STKATIGEAPHICAL GEOLOGY marine Devonian rocks, including representatives of the Middle and Upper Devonian, while the lower series is either absent or still awaits recognition. The correlation of the two areas may possibly be as follows : Ireland. Devon and Cornwall. Lower Carboniferous slates. Lower Pilton Beds. Coomhola Beds. Baggy Beds. Purple slates and grits. Pick well Down Beds. Lower Glengarriff grit, Finally the northern extension of this Devonian Series into Central Ireland must be briefly noticed. The Glengarriff Beds appear at intervals in Limerick, South Tipperary, and Kilkenny, comprising a purple slaty series like the Upper Glengarriff Beds, passing up into a yellow sandstone and shale series like that of Cork. In the Bally houra, Galty, and Knockmealdon Mountains these beds rest unconformably on the Silurian, and hence it is probable that the whole Glengarriff Series is similarly uncon- formable to the older formations, which constitute what may be called the " Caledonian " land. The position of this land is shown by the rapid thinning out of the Glengarriff Series, and the overlap of the uppermost division on to the Silurian in Clare and the north of Tipperary. A similar overlap takes place to the north-east in Carlow, where the Kiltorcan Beds overlap the lower beds on to the granite, and to the eastward in Waterford, where they lie directly on the Ordovician. Of the Kiitorcan Beds it need only be said that they thicken to the north and east, developing a thick mass of conglomerate at their base which has evidently been derived from the erosion of a land lying in those directions. In Waterford and Kilkenny this is overlain by reddish -brown sandstones succeeded by grey sandstones and red shales, above which are yellow sandstones and olive-green shales. These last are fossiliferous, containing the scales of Coccosteus and Glyptolepis, the bivalve Archanodon Ju~kesi y which was supposed to be a freshwater mollusc but has since been found in association with marine Carboniferous Mollusca ; there are also the remains of ferns, Palceopteris hibernica, Sphenopteris HooJceri, Knorria, and other plants. 9 C. DEVONIAN OF THE CONTINENT 1. France Marine sediments of Devonian age are found both in the north and south of France, but are most fully developed in the north- DEVONIAN AND OLD BED SANDSTONE SYSTEM 213 west, Brittany, Normandy, and Anjou, and in the north-east in the region of the Ardennes. It will be convenient to begin with Brittany, which is the nearest to our own Devonian area. The western facies of the system is found in the basin of Finisterre, which extends from the Hade de Brest to Lanfains and Uzel north of Loudeac. For our knowledge of this area we are chiefly indebted to the labours of Professor C. Barrois, who has described the succession as consisting of three natural lithological divisions corresponding with differing bathymetrical conditions of deposition, i.e. shallow, deeper, and deep. 10 The lowest and larger part is arenaceous with a thickness of about 3000 feet ; the middle part consists of gritty slates and mudstones, with lenticular beds of limestone, and is about 1500 feet thick ; the highest part consists entirely of fine shales with thin beds of nodular limestone, and is about 300 feet thick. The last has evidently been formed in much deeper water, for in this small thickness is comprised the whole of the Upper and part of the Middle Devonian. Palaeontologically, however, as many as seven subdivisions can be distinguished in the following descending order : 7. Shales of Rostellec. Black shales with nodular layers, Torno- ceras simplex, Bactrites, Entomis, Posidonomya venusta, and Buchiola retrostriata. 6. Shales of Traouliors. Green and brown, with Rhynchonella cuboides, Rh. pugnus, Pentatnerus globus, and Productus sub- aculeatus. 5. Slates of Porsguen. Greenish brown, with layers of calcareous nodules yielding many fossils of Middle Devonian species. 4. Grey wacke of Fret. Gritty calcareous slates with Phacops Potieri, Spirifer paradoxus, and Sp. aurieulatus. 3. Greywacke of Faou. Grits and slates with Spirifer hystericus and Chonetes sarcinulata, and limestone with Athyris undata. 2. Sandstone of Gahard, white, with Orthis Monnieri and Spirifer Pellicoi about 300 feet thick. 1. Plougastel Beds. Alternating beds of hard dark-green quartz- itic grit, and rough greenish slates. Fossils rare, Bellerophon trilobatus. Orthoceras planiseptatum, Pteria Iwvis, Spirifer octoplicatus. In this succession the Plougastel Beds appear to correspond with our Dartmouth slates and Glengarriff grits ; the Gahard and Faou Beds correspond to our Meadfoot Beds and Staddon grits. The Middle Devonian Series is about 800 feet thick, but is very different from that of South Devon, both it and the Upper Devonian having a greater resemblance to the beds in Cornwall. In the eastern basins of Laval and Angers another Devonian/ facies is found. Combining the work of M. ^Ehlert and the 214 STRATIGRAPHICAL GEOLOGY correlations made by Professor Barrois the succession may be stated as follows : n T y /7. Shales with Entomis and Posidonomya venusta. Jpper ^ Limestone of Fresnaic with Bactrites and Tentaculites. Middle 5. Calcareous shales of Pont Maillet. 14. Gritty slates with Spirifer hystericus and (Dalmanites laciniatus. 3. Limestones with Athyris undata. 2. Sandstone with Orthis Monnieri. 1. Quartzites and slates (Plougastel Beds). Of these groups only 1 to 3 are found in the Laval Basin, and the lowest is much thinner than in the west. In the basin of Angers the Plougastel Beds are absent and the series begins with the equivalent of the Gahard sandstone ; the succeeding beds being all limestones or calcareous shales, and evidently deposited much farther from land than those of Western Brittany. In the basin of Ancenis, still farther south, this change from an arenaceous to a calcareous facies seems still more marked, for the lowest beds there are limestones and shales of Middle Devonian age, but it is rather uncertain whether the absence of Lower Devonian is due to non-deposition or to faulting. There can be no doubt, however, that in passing from west to east we pass away from the land -area which supplied the arenaceous material, and this applies both to England and Brittany. In Normandy only the Lower Devonian comes into the synclines, so that the facies of the Middle and Upper Devonian is unknown. The thickness is not great, for as in the south the Plougastel Beds are absent, and the lowest beds are hard sand- stones with Orthis Monnieri, which represent the Gres de Gahard and rest unconformably on Silurian or Ordovician. Other fossils are rare. The sandstone is succeeded by slates with lenticular beds of limestone which contain Athyris undata, Homalonotus Gervillei, Phacops Potieri, Spirifer Rousseaui, and are clearly the equivalents of the greywacke of Faou and of the Athyris undata limestones of Anjou. The highest beds to be seen are brown mudstones with Pleurodictyum. 2. Spain Devonian strata have been recognised in many parts of Spain, making it probable that nearly the whole of that country was covered by the Devonian Sea. Moreover, although the facies is to a large extent calcareous the fauna is similar to that of Northern France. As an example of the Spanish succession I select that of Asturia, which was described in detail by Professor C. Barrois in 1882 ; 12 this is as follows : DEVONIAN AND OLD RED SANDSTONE SYSTEM 215 f Red and grey marbles with Goniatites. l Unfossiliterous sandstones (500 feet). ( Limestone of Candas with Spirifer Verneuilli. ( Sandstones with Gosseletia devonica (150 feet). Middle, I Limestones of Moniello with Calceola sandalina and Gysti- 1000 feet j phyllum vesiculosum. ^Limestones of Arnao with Spirifer cultrijugatus. /'Limestones of Ferrones with Spirifer paradoxus, Rhynchonella -r I Orbignyana, and many corals. , ,' I Calcareous slates and limestones with Spirifer hystericus, 18^)0? 1 1 ^hynch. pila, and Homalonotus. I Furada grits, red and green grits with green and purple slates I over 600 feet thick. The lowest beds are comparable with the Plougastel grits and slates, and the overlying limestones are calcareous equivalents of the Gahard and Faou sandstones. The limestones of the Middle Devonian correspond with the beds which are known as Eifelien in the Ardennes, and it is curious to find the upper part of the series, which is usually calcareous, here represented by sandstones. In the Upper Devonian again there is a still thicker band of sandstones, proving that the deposition of limestone was temporarily interrupted by the invasion of a sand-bearing current, and that continental land was still not far away. 3. The Ardennes In the Ardennes Devonian rocks occupy the greater part of an area lying partly in France and partly in Belgium, which is over 100 miles in length from west to east, and has a maximum width of about 50 miles (see map, Fig. 70). The area really consists of two basins, one broad and regular (basin of Dinant), the other deep, narrow, and isoclinal (basin of Namur), which are brought into apposition by a reverse fault (see Fig. 71). The throw or displace- ment of this fault amounts to many thousand feet, and the strata of the two basins must originally have been separated by an interval of several miles. That this was so is proved by the fact that while the southern basin, that of Dinant, contains a very thick repre- sentative of the whole Lower Devonian Series (average about 8000 feet), this is entirely absent in the basin of Namur. 13 Lower Devonian. On the south side of the basin of Dinant the Lower Series has a total thickness of about 10,000 feet. At the base, resting unconformably on Cambrian, is a conglomerate overlain by coarse felspathic sandstone or arkose ; this is succeeded by a series of hard grits and slates, generally of green or red tints, but containing very few fossils. A fossiliferous band at Montrepuits 216 STEATIGRAPHICAL GEOLOGY has yielded Homalonotus Rcemeri, Spirifer Mercurii, Pterincea ovalis, and Tentaculites : elsewhere remains of Pteraspis have been found. This group of beds, of which the slates of St. Hubert are the highest member, are known as the Gedinnian. The higher stage of the Lower Devonian is called the Coblentzian ; this has 1500 feet of white sandstone (Gres d'Anor) at its base, followed by the greywacke of Montigny, the black sandstone of Vireux, the red slates of Vireux, and the Burnot conglomerate, the Cambrian. Silurian. Devonian. Carb. Coal- limestone, measures. Fig. 70. MAP OF THE ARDENNES (drawn and lent by Professor J. Gosselet). Scale about 35 miles to an inch. whole being from 5000 to 6000 feet thick. The Anor sandstone has a special fauna, including Spirifer primcevus, Renssellceria crassi- costa, R. strigiceps, Avicula lamellosa, and Kochia capuliformis, and appears to represent a series of beds which are more fully developed near Coblentz, and are known as the Taunus quartzite. The grits and mudstones above this white sandstone are characterised by Spirifer paradoxus, S. primcevus, Rhynch. daleidensis, and Athyris undata, while the higher beds contain Spirifer paradoxus, Chonetes sarcinulata, Sp. hystericus, and Leptcena Murchisoni. The whole series becomes thinner toward the north, and along the northern side of the basin it is represented only by the variegated slates of St. Hubert and the Coblentzian Beds in a DEVONIAN AND OLD RED SANDSTONE SYSTEM 217 rather different facies, the combined thickness being still, however, about 5000 feet. Middle Devonian. This also comprises two divisions, the Eifelian and the Givetian. In the former are now placed the Hierges Beds, flagstones and shales, which are subdivided into a lower zone of Spirifer arduennensis and an upper zone of Sp. cultrijugatus. In the basin of Dinant they are some 500 feet thick, and in the basin of Namur they are the lowest beds (with a basal conglomerate). The next group is that of the Calceola shales; these are calcareous shales including lenticular beds of limestone which sometimes, as at Nismes and Couvin, swell out to a thickness of Coudroz. Fig. 71. SECTION THROUGH THE BASINS OF NAMUR AND DINANT (GOSSClet). //. Coal-measures. C. Dinantian. D. Upper and Middle Devonian. R. Lower Devonian. S, Cambrian and Ordovician. 300 to 500 feet, hence the group varies greatly in thickness, but has probably an average of 1000 feet. Its most characteristic fossils are the little coral Calceola sandalina with Gystiphyllum vesiculosum, Mesophyllum helianthoides, Spirifer speciosus, Meristella plebeia, Phacops Schlotheimi, and Bronteus flabellifer. The Givetian, in its typical locality near Givet, is a mass of dark- grey limestone, over 1200 feet thick, and is specially characterised by certain Brachiopods, viz. Stringocephalus Burtini, Uncites gryphuSj and Spirifer mediotextus, the bivalve Megalodon cucullatus, and the Cephalopods Anarcestes cancellatus, Agoniatites inconstans, and Orthoceras dolatum. The so-called coral reefs of the Givetian are merely thick lenticular masses of unstratified limestone formed by colonies of Stromatoporoids with Favosites, Alveolites, and other corals, such masses being generally surrounded by bedded crinoidal limestones or by calcareous shale and limestone. Upper Devonian. This is also divisible into two stages, the Frasnian and the Famennian. The former consists of shales and M- 2 c-:o6i~. Coccosteus decipiens. d. Pterygotus anglicus. e. Cephalaspis Lyelli. that all three divisions are found together in one area, and it is certain that in most parts of the country only two such series exist, the Upper being unconformable to the Lower. DEVONIAN AND OLD RED SANDSTONE SYSTEM 225 A. FLORA AND FAUNA The fauna of the Old Red Sandstone is not a large one, but in certain localities remains of fish and Eurypterids occur in some abundance ; plants also occur, but the only Mollusc yet found is the Archanodon Jukesi of the Upper Old Red, and this more closely Pig. 75. FOSSILS OF THE UPPER OLD RED SANDSTONE. a. Holyptychius nobilissimus. b. Phaneropleuron Andersoni. c. Archanodon Jukesi. resembles a fresh-water mussel than any marine form. From this limited fauna, and the total absence of any unquestionably marine species, it has been supposed that the Old Red Sandstones were lacustrine deposits, and that the greater part of Northern Europe was at this time a continent, within which lay several large lake- areas comparable to those of North America at the present day. The plants of the Old Red Sandstone are the ancestors of the Q 226 STBATIGRAPHICAL GEOLOGY Carboniferous flora, and are chiefly lycopods and ferns. The former include Lepidodendron, Sagenaria, Knorria, Stigmaria, and Psilophy- ton; ferns are presented by Palceopteris, Neuropteris, and Sphenopteris ; Calamites occur, and coniferous wood is also not uncommon. Fish are the most conspicuous and abundant members of the fauna, and several orders are represented : the Ostracoderms by Cephalaspis, Pteraspis, Scaphaspis, and Holaspis ; the Dipnoids by Coccosteus, Phaneropleuron, Dipterus, and Homosteus ; the Elasmo- branchs by Acanthodes, Cheiracanthus, Diplacanthus ; the Crossoptery- gians by many genera, such as Osteolepis, Glyptolepis, Gyroptychius, Diplopterus, and (in the upper division) Holoptychius. The remains of fish are very abundant at certain horizons in Scotland, but they are not generally distributed through the mass of the formation. The Merostomata are represented by Pterygotus, Eurypterus, Styloneurus, and the only other organic remains found in British strata are the Phyllopod Estheria, two Myriapods, and a Lamelli- branch (Archanodori). In Canada, however, insects of several kinds have been found in rocks of corresponding age and character. The following are some of the more characteristic Old Eed Sandstone fossils : Fossils of the Lower Division Plants. Psilophyton robustum, Parka decipiens. Myriapoda. Archidesmus Macnoli, Kampecaris forfarensis. Merostomata. Pterygotus anglicus, Eurypterus pygmseus, Stylonurus Symondsi. Pisces. Cephalaspis Lyelli, Pteraspis rostrata, Pterichthys cornutus, Pt. Milleri, Mesacanthus Mitchelli, Diplacanthus gracilis, Climatius scutiger, Parexus incurvus. Fossils of the Middle Division Plants. Lepidodendron gaspeanum, Calamites transitionis. Crustacea. Estheria membranacea. Pisces. Coccosteus decipiens, C. minor, Thursius pholidotus, Th. macrolepidotus, Homosteus Milleri, Glyptolepispaucidens, Dipterus macropterus, Osteolepis microlepidotus. Fossils of the Upper Division Plants. Palseopteris hibernica, P. Browni, Cyclopteris Kiltorkense, Stigmaria ficoides, Knorria Bailyana, Sphenopteris flaccida. Mollusca. Archanodon Jukesi. Pisces. Holoptychius nobilissimus, Bothriolepis major, Glypto- pomus minor, Phaneropleuron Andersoni, Asterolepis maxima, and Psammosteus Taylori. DEVONIAN AND OLD RED SANDSTONE SYSTEM 227 B. STRATIGRAPHY OF THE OLD KED SANDSTONE 1. Wales and the Welsh Borders In Monmouth, Hereford, and Brecknock a great thickness of red sandstones and marls intervenes between the Silurian and the Carboniferous, and from these counties extensions of Old Ked Sandstone stretch westward through Carmarthen and South Pembroke, northward into Radnor and Shropshire, and eastward into Worcester and Gloucester. It is evident, therefore, that in all these directions the formation originally extended considerably beyond its present limits. According to recent estimates, its maximum thickness in Brecknock may be 7000 feet, but from this central area it thins in every direction, being only 4200 feet at Abergavenny in Monmouth, and 3500 at Newport ; while north- ward in Shropshire its thickness is believed to be 3700. The area over which the sandstones and marls of this system still extend measures over 100 miles from west to east, and about 80 miles from Cardiff in the south to beyond Bridgenorth in Shropshire. The system is naturally divisible into three parts or series, which were long ago recognised by W. S. Symonds, under the names of (1) Lower or Cornstone Series, (2) Middle or Brownstone Series, (3) Upper Old Red. These divisions have been adopted by the Geological Survey in their recent revision of South Wales, but there remains much doubt whether the Brownstone Series repre- sents any part of the true Middle Old Red. In the eastern parts of the area the whole seems to form a conformable succession, but when followed westward into Carmarthen the Brownstones thin out near Kidwelly, and the conglomerate of the Upper Group overlaps on to the lower series. They recur in Gower to the south, but are only 100 feet thick, and seem to belong rather to the Upper Old Red than to be an independent group. Hence the arrangement of the Geological Survey is as follows : {Yellow and red sandstones, grey grits, and quartz-pebble conglomerates. Brownstones. Red and brown sandstones, with occasional beds of marl and cornstone. Lower or f Green and dull-red sandstones with some beds of marl, corn- CornstoneJ stone ' and con g lomerate - Series I ^ ed marls witn be( ^ s of nodular limestone (cornstone) and V. occasional beds of micaceous sandstone. Cornstone Series. In Brecknock and the east of Carmarthen this red marl and cornstone series has a great thickness, possibly as much as 4500 feet. The greater part of it consists of hard red 228 STRATIGRAPHICAL GEOLOGY sandy marls with layers of hard calcareous nodules which some- times coalesce so as to form beds of compact red or green limestone ; while in other cases the embedding marl seems to have been washed away by contemporaneous erosion, leaving the nodules to form a " cornstone conglomerate." The nodules themselves consist of amorphous carbonate of lime, and nothing but obscure traces of plants have been found in them. They have probably been formed by the agency of lime-secreting algse. The Senni Beds only occur along the northern outcrop, and gradually thin out westward, disappearing a little before the Brownstones. Their maximum thickness is 1200 feet, and they have been included in the lower series because the remains of Pteraspis have been found in them. Both Pteraspis and Gephalaspis have been found in the Eed Marls, and consequently it may be inferred that this Lower Old Eed is the equivalent of the Lower Devonian and that there was some water communication, by river or inlet, between the two areas of deposition. During the resurvey of the Gower district, in the south-west corner of Glamorgan, it was discovered that the underlying Silurian rocks rose to the surface within the Old Red Sandstone area and formed two small inlying tracts ; further that the Cornstone Series, though present, thinned out within the area against the slope of the Silurian rocks. Thus some 300 feet of red marls are visible below Rhossili Down, and are overlain by Brownstones which are succeeded, without any sign of uncon- formity, by 300 feet of quartz-pebble conglomerates (see Fig. 77). On Cefn-y-Bryn, however, the red marls are absent, and Brown- stones (only 100 feet thick) rest on the Silurian and are covered as before by the conglomerates. 16 Still farther west, however, in Pembrokeshire the Red Marl Series regains its normal thickness, being over 3000 feet along its southern outcrop and nearly as much on the northern side. If, therefore, the Gower ridge was continued westward it must have passed some distance to the south of Milford Haven. Upper Old Red Series. The Brownstone Group consists mainly of sandstone, bright red or reddish brown, with subordinate beds of marl and occasional cornstones. In the area round Brecknock, where they form the lower slopes of the Brecon Vans and the Black Mountains (see Fig. 76), it may have a thickness of about 2000 feet, but this decreases in all directions. As already stated, they thin out entirely to the westward and do not occur in Pembrokeshire. No fossils of any kind have been found in them. The highest group of beds is never more than 500 feet thick ; it commences with a conglomerate composed in most places of 3 ' CB- < a; 03 M o B 230 STRATIGRAPHICAL GEOLOGY pebbles of vein-quartz which are so water-worn and rounded that they resemble the pebbles of a shingle-beach. In the northern and eastern part the shingle is succeeded by beds of sandstone, marl, and shale which pass up into the Carboniferous Series, and these beds have in several places yielded remains of fish and plants. The fish belong to the genera Bothriolepis, Holoptychius, and Sauripterus, and shells of Archanodon Jukesi have also been found. In the extreme south and south-west these uppermost beds have a different aspect. Thus in Gower the conglomerate is no less than 300 feet thick and contains, besides the usual quartz pebbles, some of quartzite, together with angular pieces of red jasper, many of which appear to be pieces of silicified rhyolite or other igneous rock. Again in Pembrokeshire, along the northern outcrop and near Tenby, the group consists of quartzitic grits passing up into sandstones, marls, and shales ; and at West Angle Bay, south of Milford Haven, two bands of marine fossiliferous beds occur in this sandstone group. The lower band commences about 125 feet below the line taken for the top of the group, and consists of grey shales and sandstones with thin bands of limestone ; the second is some 50 or 60 feet higher. Both these bands contain marine shells, chiefly Lamellibranchs such as Ptychopteria damnoni- ensis, Cucullcea trapezium, Curtonotus elegans, with Rhynchonella laticosta and other characteristic fossils of the Marwood and Pilton Beds in North Devon. Here, therefore, as pointed out by Dr. Strahan, we have evidence of temporary incursions of the sea into an area which had previously lain within the borders of continental land, and was presumably part of a large lake-basin. 2. Scotland In Scotland there are four principal areas where the Old Red Sandstones are found, and in some of them they attain a much greater thickness than in England. These areas are : (1) parts of Berwick, Roxburgh, and of Northumberland, including the Cheviot Hills ; (2) the area of the central lowlands, underlying the Carboniferous rocks and coming to the surface both on the southern and the northern side of the great basin ; (3) a smaller tract in Argyleshire between the Firth of Lome and Loch Awe ; (4) a large irregular tract in the north-east, bordering the Moray and Cromarty Firths and stretching northward through the east of Sutherland to Caithness and the Orkney Islands. It has been imagined that these four areas were distinct and separate basins of deposition, at any rate during the earlier part of the period ; that they were disconnected from the sea and formed large inland lakes. Sir Archibald Geikie has indeed proposed DEVONIAN AND OLD RED SANDSTONE SYSTEM 231 separate names for these lakes, calling the first Lake Cheviot, the second Lake Caledonia, the third Lake Lome, and the fourth Lake Orcadie. This view, however, is not sufficiently supported by the facts, so far as the three southern areas are concerned.; the Lome and Cheviot areas were probably only extensions or bays connected with one large Caledonian lake basin, which may also have stretched south-westward into the north of Ireland, for patches of Old Eed occur at the very southern end of the Kintyre promontory. The northern area, on the other hand, exhibits a different set of beds which contain a different fish fauna, so that there does seem to have been a Lake Orcadie or at any rate a separate Orcadian basin of deposition. In the central area the Lower Old Eed Series is of great thickness, but no middle division exists, and the Lower Series is overlain unconformably by the Upper Old Red, which passes conformably into the overlying Carboniferous Series. In the Orcadian area there is nothing which can be identified as Lower Old Red, but there is a great thickness of the Middle Series with an abundant fish fauna, and some patches of Upper Old Red resting unconformably upon the beds below. There is a further great difference between the rocks of the two areas. In all the southern districts the sandstones are interstratified with thick sheets of andesitic lava and beds of volcanic tuff and agglomerate, giving proof of volcanic action on a very large scale, and from many volcanic vents, throughout the whole of the Lower Devonian epoch. In the Orcadian area, on the contrary, no such volcanic rocks are found, so that the eruptions do not seem to have continued into the time of the Middle Series. In the following account of the stratigraphic succession of beds these volcanic materials will only be mentioned incidentally, and for more information about them the student is referred to Sir A. Geikie's Ancient Volcanoes of Great Britain (1897). Caledonian Region. The Lower Old Red Sandstone is most completely developed on the northern side of the great central or Lowland trough, and the beds are finely exposed in the cliffs of the Kincardine and Forfar coasts. Until quite recently this series was supposed to include all the red rocks visible in Kincardine, but as stated on p. 163 the lowest set of beds must now be regarded as the highest part of the Silurian System. The base of the Old Red Sandstone is the massive conglomerate which forms Downie Point near Stonehaven, and this is part of a group of coarse sandstones and conglomerates to which Mr. G. Hickling 17 has given the name of Dunottar Conglomerate (see Fig. 78). These beds occupy a long stretch of coast-line which 232 STRATTGRAPHICAL GEOLOGY is here nearly parallel to the strike. Mr. Hickling estimated their thickness to be 5000 feet, and the comparative narrowness of their outcrop is due to their high inclination. The conglomerates of this group consist partly of quartzite pebbles, derived from the quartzites of the Highlands, and partly of angular and subangular fragments of jasper, chert, and " green-rock " derived from the "Margie Series" (see p. 97), showing this series must then have had a considerable northward extension over the Highland region. The Dunottar Group is succeeded by the Sidlaw Group, which includes the contemporaneous lavas of the Sidlaw Hills and the overlying grey sandstones and flagstones ( ? 500 feet), which are followed by beds of coarser grain (the Cairnconnon Group), and perhaps 2000 feet thick. The fourth group is the Red Head Series, which consists of red sandstones with bands of hard red shale, and locally some bluish - grey shales. These beds are overlain by the Auchmithie conglomerate from 500 to 800 feet thick, and mainly composed of large rounded pebbles of quartzite. No fossils have yet been found in any of the beds above enumerated, except a few in the Red Head Series. The succeeding Arbroath sandstones comprise coarse gritty sandstones and some flagstones, both of red and grey colours, with a single bed of marly limestone at Arbroath, probably on the same horizon as a limestone which has been mapped inland to the westward. Fossils have been found at four horizons in these beds, and include Cephalaspis Lyelli, Pterygotus anglicus, Kampecaris forfarensis, and the plant Parka decipiens. The highest beds of the whole series (the Edzell shales) are not exposed on the coast, but only occur in the deeper parts of the synclinal trough of Strathmore (see Fig. 78). The following is a tabular resume of the succession and thicknesses : Feet. Edzell shales . . . . . 1,000 Arbroath sandstones Auchmithie conglomerate . Red Head Series Cairnconnon Group . Carmyllie or Sidlaw Group Dunottar conglomerate 1,200 800 1,500 2,000 1,000 5,000 12,500 [I am indebted to Mr. Hickling for the drawing of the section across Forfar, from which Fig. 78 has been made.] The great series of sandstones, conglomerates, and volcanic rocks above described forms a continuous band across Scotland from the coasts of Kincardine and Forfar to that of Dumbarton on I S 3 r < ^ 8, % ^O N 234 STRATIGRAFHICAL GEOLOGY the Firth of Clyde, a distance of about 120 miles. For the greater part of its extent the breadth of outcrop is about 20 miles, but it narrows in Dumbarton where the thickness is probably less. On the north-west side it is cut off by the great boundary fault ; but in Perthshire and the west of For far this is not a single but a double line of fault, and there are small tracts of the Old Red rocks beyond the second fault-line, resting on the schists and quartzites of the Highland complex (see Fig. 79). These outliers prove that the formation had originally a considerable extension to the northward over the rocks of the Central Highlands. If, indeed, the thickness of the series in Perthshire was as great as it is on the eastern coast, the height of the mountains which then formed the Highland range must have been well over 12,000 feet, and it is clear that the present hills are but the denuded stumps of a Devonian Alpine Range. Further, since the lower slopes of this range seem to have been gradually buried beneath the increasing pile of red sandstones and lava-flows, it is not surprising to find a large tract of volcanic rocks in Lome associated with sediments which prove the whole to be of Lower Old Red age. These rocks form the plateau of Lome, which consists of a succession of terraces and escarpments, rising to 1700 feet above the sea and composed of felsitic and andesitic lavas with occasional layers of agglomerate, conglomerate, and red shale. At the base, and exposed along the coast near Oban and in Kerrera Island, are shales and flagstones with a boulder conglomerate at the base from 20 to 200 feet thick. These beds have yielded remains of Cephalaspis, Pterygotus, Kampecaris, and the plant Psilophyton. Still farther north in Glencoe is another patch of andesites with some shales in which Psilophyton and Pachytheca have been found. On the southern side of the great Lowland syncline the Lower Old Red crops out in a more irregular manner, and is only exposed in disconnected areas in Ayrshire, Lanark, and the Pentland Hills. The beds described on p. 177 as forming the highest member of the Silurian System were formerly classed with the Old Red Sandstone, but the base of the latter is now taken at a band of conglomerate containing greywacke pebbles which is recognisable in all three districts. Near Lesmahagow (Lanark) this con- glomerate rests with apparent conformity on the red Silurians, and is succeeded by chocolate -coloured sandstone containing Cephalaspis Lyelli, but in Ayrshire and in the Pentlands the basal conglomerate is markedly unconformable to this Silurian Series. 18 Fig. 59 is a diagram section showing the general relations of these beds in the Pentland district. DEVONIAN AND OLD EED SANDSTONE SYSTEM 235 Just as the northern belt is cut off by the Highland boundary fault, so the southern tracts are cut off by the boundary fault of the Southern Uplands, which runs from the coast near Ballantrae north-eastwards till it passes beneath the Carboniferous basin of Dalkeith. A small tract of Lower Old Bed near Eyemouth in Berwickshire shows that the Caledonian area of deposition was prolonged southward, and there can be little doubt that the Cheviot area was also a part of it. The Cheviot Hills consist mainly of massive andesitic lava-flows with intercalations of tuff and ashy sandstone, and there are some beds of sandstone and conglomerate at the base, but of no great thickness. Passing now to the Upper Old Ked Sandstone of the Caledonian region we find this everywhere marked off from the Lower Series by a great break and unconformity. It occurs only in isolated tracts, and frequently oversteps the limits of the Lower Series on to the older rocks. The interval of time which elapsed between the two epochs of deposition must have been a long one, for Mr. Hickling finds that in Forfar no less than 8000 feet of material had been removed from the main anticline before the Upper Old Red was deposited ; in other words, the Lower Series was flexured and then denuded to that extent in the interval which is else- where partially filled by the Middle Old Red. The Upper Series occupies several more or less extensive tracts in different parts of the region, the largest being that in the south-east, Roxburgh and Berwick, where it extends from the northern borders of the Cheviots by Jedburgh and Earlston to Lauder and Greenlaw, and then all round the Lammermuir Hills to the coast at Dunbar. A smaller tract is found in the Pentland Hills, and others in Lanark and Ayrshire. On the northern side of the Lowland area two long belts of it lie between the boundary of the Lower Series and the outcrop of the Carboniferous sandstones, the one extending from the Clyde to the Forth near Stirling, the other from Kinross and Loch Leven to Cupar and Leuchars in Fife. There are also some small tracts in Forfar. The following description of the beds composing this Upper Old Red is condensed from the account given by the late J. G. Goodchild, and is applicable to all the more eastern tracts. The series is divisible into a lower and upper stage ; the lower consists of a variable set of conglomerates and sandstones, the coarser beds being at and near the base, and the sandstones becoming finer in grain and more interstratified with marls towards the top ; the sandstones also become more and more largely composed of rounded wind -worn grains. The prevalent colour is bright red; the thickness is variable (from 100 to 500 236 STEATIGRAPHICAL GEOLOGY feet), and fossils are scarce, but scales of Holoptychius occur. The upper or Cornstone Group consists of sandstones and flagstones of pale red or yellow colours with marls and layers of the hard calcareous nodules which are known as cornstones. Fish remains are abundant in places, especially at Dura Den, near Cupar, where most of those mentioned on p. 226 have been obtained. This upper group is just as clearly a lacustrine deposit as the lower part of the series is one of accumulation in dry and wind-swept desert plains. The total thickness in Fifeshire is about 1000 feet. Orcadian Region. The present visible extent of the Old Ked Sandstone has been mentioned on p. 230, but it must have a much greater extension under the bed of the North Sea both in eastern and northern directions, for the general dip of the beds is to the north-east, and their estimated thickness is 16,000 feet. The whole of this thickness is ascribed to the Middle Old Red Series because of the peculiar fish fauna which it contains. It is divisible into four fairly distinct groups : 19 (1) the lowest consists of red conglomerates, sandstones, and mudstones resting on and against the Archaean rocks and largely developed near Berriedale, from which place its outcrop runs N.N.W. to Sandside Bay, west of Thurso ; (2) the Lybster Beds or lower flagstone group dark-blue flags and shales, with occasional beds of sandstone and of black, often bituminous, limestone ; (3) the Thurso Beds or upper flagstones, which are of pale grey or green colours and include but few limestones ; (4) the John o' Groats sandstones of red and yellow tints. In tabular form the succession and thick- nesses are : Feet. John o' Groats sandstones . . . about 2,000 Thurso Flagstone Group . . . . ,, 7,000 Lybster Flagstone Group . . . ,, 5,000 Berriedale Beds ,,2,000 About 16,000 Fish remains are most abundant in the Flagstone Beds, the lower group yielding Thursius macrolepidotus, Dipterus macropterus, and Osteolepis ; the higher beds contain abundant remains of Goc- costeus (G. decipiens and G. minor) with many others (see p. 226). The highest group has yielded plant remains, such as Lepidodendron gaspeanum, Palceopteris Browni, and Calamites transitionis, which occur also in the Devonian of Canada. A fish bed of special importance occurs at Achaiiarras, some 10,000 feet above the base of the formation, while a similar assemblage of species has been found in the Moray area not very far above the base, and if the two beds are really on the same DEVONIAN AND OLD RED SANDSTONE SYSTEM 237 stratigraphical horizon a thickness of about 9000 feet must have thinned out southwards and have been overlapped by the beds containing the Achanarras fauna. Such an overlap had been inferred on other grounds by Sir A. Geikie many years ago, and on this view it is only the highest part of the Middle Old Eed Series which occurs in Nairn and Inverness and in the outliers which are found in Aberdeen. This northern basin includes also some tracts of Upper Old Ked, which rests with a marked unconformity on the Middle Series. From 600 to 800 feet of red and yellow sandstones can be seen. They occupy a tract of some length in Moray and Nairn, and outlying patches occur at Tarbat Ness, round Dorn^ch, at John o' Groats and Dunnet Head, and again in the Island of Hoy. 3. The North of Ireland As already mentioned, several tracts of Lower Old Eed occur in the north of Ireland, and others may be concealed beneath the newer strata. One such tract occupies an area about 30 miles long by 10 wide, between Lough Erne and Pomeroy in Tyrone. " It consists for the most part of dark-red and purple conglomerates, often coarse and massive, and of purple pebbly and fine-grained sandstones, . . . with sandy shales. The pebbles in the con- glomerate, which vary from the smallest size up to blocks over a foot in diameter, consist of purple felstone, grits, schists, and quartzite. Of all these the felstone pebbles are by far in greatest proportion," and have been derived from the breaking up of contemporaneous lava streams, some sheets of which still remain interbedded with the conglomerates. 20 The second area is on the east coast of Antrim between Cushendall and Cushendun. Here again are conglomerates at the base, consisting mainly of quartz pebbles ; above these is a series of red and brown sandstones succeeded by a breccia composed of large blocks of quartz-porphyry ; the whole is estimated to be 5000 feet thick. This tract forms a link between the Tyrone area and the Lower Old Red of the Kintyre promontory. Another tract of similar material has been found in Donegal between Lough Swilly and Mulroy Bay. This consists of a basal conglomerate succeeded by chocolate-coloured sandstones and shales with some pebbly beds, and the materials are all derived from the surrounding quartzites and schists. The thickness seen is about 800 feet. Nothing corresponding to the Upper Series has been found in the north of Ireland, though it undoubtedly occurs in the southern part of the country (see p. 211). 238 STRATIGRAPHICAL GEOLOGY 4. Norway Some small tracts of red sandstone and conglomerate occur on the west coast of Norway between the Sogne and the Ludvig fiords (see map, Fig. 10). They rest unconformably on Silurian and older rocks, but the maximum thickness is not more than 1200 feet. It is very probable that they belong to the western border of the Orcadian basin of deposition, in which case they are likely to be Middle Old Eed Sandstone ; but no fossils have yet been found in them to confirm this assumption. 5. Arctic Regions Sandstones of the Old Ked facies have also been found in several parts of the Arctic regions, notably in Spitzbergen, Bear Island, and Greenland. In Spitzbergen, both Lower and Upper Old Eed Sandstones occur. The lower beds, exposed in Dickson's Bay, consist of red and green sandstones, and have yielded species of Pteraspis, Cephalaspis, and Acanthaspis. The Upper Old Ked has been explored at two localities, the Muner Valley and Grey Hook Point. At the' former the beds consist of red and green sandstones with a band of grey shale enclosing ironstone nodules, in which fish of the genera Psammosteus, Asteroplex, Onychodus, and Holoptychius occur, together with plant remains (Lepidodendron and Bothrodendron ( = Cyclostigma). At the other place (Grey Hook Point) there is a series of black shales and hard grey sandstones, and these have yielded a number of curious Mollusca which appear to be fresh or brackish-water forms ; for no essentially marine genera occur, and they are all bivalves which have been referred by Dr. Kayser to the following genera, Archanodon, Palceomutela, Myalina, Pieria (?), and Nathorstella. 21 The relations of the Upper Series to the Lower have not yet been ascertained, nor the thickness of either, for the whole mass occupies a deep faulted trough which crosses the island from north to south, so that the base is not seen ; but de Geer has estimated the parts exposed to be over 4000 feet thick. Bear Island lies to the south of Spitzbergen, and the greater part of it consists of beds which belong to the Upper Old Eed Series. 22 There is a basal conglomerate resting unconformably on Ordovician limestones and succeeded by a series of sandstones with bands of carbonaceous shale and seams of coal ; these beds have yielded many interesting plants, including species of Archceopteris, Sphenopteris (Sphenopteridium), Cephalopteris, Pseudobornia ursina, DEVONIAN AND OLD RED SANDSTONE SYSTEM 239 Bothrodendron kiltorkense, and four other species with Stigmaria ficoides and others. There are also remains of the fish Holoptychius giganteus and H. monilifer. Again Upper Old Eed has been found on the east coast of Greenland, the beds being red and brown sandstones from which fish of the genera Holoptychius and Asterolepis have been obtained. From these facts it is clear that the Old Ked Sandstone facies has a very large extension in the Arctic regions, and that the continent within which these deposits were formed reached from England to Spitzbergen, and from Greenland at least as far as Sweden, and probably through Northern Eussia where sands and marls with fish remains occur near Archangel, without any marine deposits, but marine Devonian strata do occur in Novaya Zembla. C. GEOGRAPHY AND HISTORY OF THE PERIOD The general geographic conditions under which both the marine Devonian and the Old Red Sandstones were deposited have been fairly well indicated in the preceding pages, and the student will doubtless have realised that the greater part of Northern Europe, together with the whole of the North Sea area and part of the Atlantic Region, formed at this period a large continental tract of land. By some authors the Devonian has been called " the first continental period," regarding it from a European point of view, and it is certainly the earliest period for which we can reconstruct a map of land and sea with anything like an approach to accuracy. I have therefore prepared a map of Northern Europe in Lower Devonian time (Fig. 80) and will briefly recapitulate the data on which the southern boundary or coast-line of the continent has been drawn. Beginning with the western region the great thick- ness of detrital deposits and the prevalence of fine sand in Brittany, Cornwall, and the south-westof Ireland are plain indications that they must all have been on the borders of a large area of land. We find actual portions of this land in Waterford, Wexford, and South Wales, and again in the southern part of Cornwall. Hence we arrive at the conclusion that the area in which the Glengarriff Beds were deposited was a large land-locked bay bounded by continental land on the north, west, and south, and only opening eastward through a channel from 50 to 70 miles in width. From Cornwall the land seems to have projected as a pro- montory into the area now occupied by the English Channel, and its southern shore must have passed outside Brittany and then south-westward toward the north of Spain. In Asturia the lowest ft.. -foC^T A%M x-v DEVONIAN AND OLD RED SANDSTONE SYSTEM 241 600 feet of the Devonian are hard quartzitic grits and slates of the Plougastel and Glengarriff facies, but from their much smaller thickness we may conclude that the Atlantic land did not reach quite so near to Spain as it did to Brittany, the coast-line probably curving away to the west. Returning to England it may be noted that rocks containing Upper Devonian fossils have been found in borings at Turnford, near Cheshunt, and below London, but they do not help much in our present quest. In the Ardennes, however, we find definite evidence of the position of the coast-line in Lower Devonian time, for the entire absence of the Lower Series in the basin of Namur proves it to have been land while that of Dinant was covered by the Lower Devonian Sea. Here it must be pointed out that the supposed existence of land to the south of Dinant and Fepin, with a narrow strait between it and the northern mainland, is founded on a mis- apprehension of the facts. There was land there in Silurian times, and the basal conglomerate and sandstones of the Devonian testify to its destruction, but there is no overlap of successive stages against a slope of older rocks, and there can be little doubt that the whole mass of the Lower Devonian strata in the basin of Dinant passed over the massifs of Kocroi and Charleville ; in other words, that this land of Silurian and Ordovician time sank beneath the waters of the Devonian Sea. The same sea spread eastward over the greater part of Germany, but the great thickness and sandy nature of the Lower Devonian of the Harz Mountains testify to the proximity of land ; and when we reach Poland still clearer indications are found, for there a break and unconformity between Silurian and Devonian presents itself, and we have in the Lowest Devonian deposits an intercalation of marine strata with sandstones of the Old Red facies, as if a large lagoon were being invaded by the sea. The same conditions are present in Eastern Galicia, which must likewise have lain near the margin of the continent, but in the Baltic provinces of Russia we clearly stand on firm Devonian land (see p. 223). The line drawn on Fig. 80 to indicate the eastern border of this continent is, of course, merely an approximation to its true course ; at present we only know that the land did not extend into Central Russia, and that it did not include Novaya Zembla. In Middle Devonian time there seems to have been a general subsidence, enabling the Southern Sea to extend itself over some of the southern and eastern portions of the northern continent. Moreover, it seems that the areas invaded and covered by the Eifelian Sea were of much larger extent in the eastern than in the R 242 STRATIGRAPHICAL GEOLOGY western parts of the continent ; consequently we must suppose that either the subsidence was greater in the east than in the west, or that there were larger areas of low-lying land in the east. The first hypothesis seems the most probable because the lacustrine area of South Wales lay so close to the border of the continent that we can hardly imagine it to have been at a high level above the sea, nor is the intervening isthmus likely to have been very high along its whole length. The Orcadian basin is indicated on the map (Fig. 80), although we have no proof of its existence as a basin of deposition during Lower Devonian time. It may have existed in the form of a wide valley which, by subsequent disturbances and alterations of level in the northern part of the continent, was converted into a large lake-basin. In any case we must suppose that the river which ran out of this lake opened somewhere into the Middle Devonian Sea, for it can only have been by means of the ex-current river that the fish made their way into the lake, ascending it as salmon do rivers at the present day. Whether this river flowed eastward or northward is the question, but unless it traversed the range of mountains which certainly then existed in Scandinavia it must have taken a northerly course into a northern sea. The lapse of time indicated by the slow accumulation of the Middle Devonian shales and limestones must have been very great, and it must not be estimated by the small thickness of such deposits in Brittany and Cornwall, but by their maximum develop- ment, which is probably that to be found in the Ardennes and the Khenish provinces (see p. 215). Again, it is noteworthy that if we had nothing but the succession of marine deposits as a record of the Devonian period, we should unquestionably have supposed that the whole period was one of quiet and uninterrupted subsidence, and that the epoch of the Middle Devonian merged into that of the Upper Devonian without any great physical change. The terrestrial deposits, however, furnish us with evidence of a great change, though it is quite possible that this was accomplished without any great surface or subterranean disturbances. The change probably involved some movement of the earth's crust, but this can hardly have been elevation, for it ushered in the subsequent subsidence of Carboniferous time ; neither was it immediate subsidence, or we should have had marine beds in the place of the Kiltorcan Beds and Upper Old Red of Wales. Rather it seems to have been a change in the climate and in the terrestrial conditions of deposition. It is possible that toward the close of Middle Devonian time DEVONIAN AND OLD RED SANDSTONE SYSTEM 243 dry and desert conditions prevailed over the central parts of the continent, including the British region ; that the lakes shrank to small proportions, if they were not entirely dried up. On this supposition the change which took place may have been a return to more humid conditions, an increase in the rainfall, especially on the mountain ranges, resulting in the birth of torrential streams at certain seasons, which swept out the stones and pebbles that had accumulated in the higher valleys and spread them out on the desert plains, for there is sufficient evidence that the Caledonian area at any rate was such a desert plain. As the annual rainfall increased, lakes were once more formed, and were again peopled with fish, but new kinds of fish had in the meantime been developed in the surrounding sea, so that the fish of the Upper Devonian lakes were a very different assemblage from that which lived in the older lakes. Finally we have evidence of a slow subsidence of the whole region and of the invasion of these lakes by the waters of a Carboniferous Sea. KEFERENCES 1 Jukes, Quart. Journ. Geol. Soc. vol. xxii. p. 320 (1866). 2 Etheridge, Quart. Journ. Geol. Soc. vol. xxiii. p. 568 (1867). 3 Hicks and Whidborne, Quart. Journ. Geol. Soc. vol. lii. p. 254, and liii. p. 438. 4 W. A. E. Ussher, Quart. Journ. Geol. Soc. vol. xlvi. p. 487 (1890). 5 See E. Hull in Quart. Journ. Geol. Soc. vol. xxxv. p. 703 (1879). 6 See Jukes, Quart. Journ. Geol. Soc. vol. xxii. p. 340, and Explanations of sheets of the Geol. Survey Maps of Ireland. 7 See paper by Prof. Hull (above cited). 8 Geology of the Country around Cork, 2nd ed. (1905). 9 See Heer in Quart. Journ. Geol. Soc. vol. xxviii. p. 163 (1872). 10 C. Barrois, Proc. Geol. Assoc. vol. xvi. p. 101 (1899). 11 ^Ehlert in Butt. Soc. Geol., France, for 1889, p. 742; and " Livret- Guide de 1'Excursion en Mayenne," Geol. Congres Internat. , 1900. 12 C. Barrois, "Terr, anciens des Asturies," Mem. Soc. Geol. Nord, 1882. 13 J. Gosselet, Proc. Geol. Assoc. vol. ix. p. 228. 14 See Kayser, Zeitsch. deutsch. geol. Ges., Bd. xxxiii. p. 617 (1881) ; and Koch, Jahrb. k. preuss. geol. Landesanst. xx. p. 237 (1900). 15 F. Seemann, Beitr. Palaont. Oesterr.-Ung. xx. p. 69-114 (1907). 16 " Geology of West Gower," Mem. Geol. Survey (1907). 17 G. Hickling, Geol, Mag. for 1908, p. 396. 18 J. G. Goodchild, Proc. Geol. Assoc. vol. xviii. p. 117. 19 Sum. Prog. Geol. Survey for 1898, p. 87 (with map). 20 J. Nolan in Quart. Journ. Geol. Soc. vol. xxxvi. p. 529 (1880). 21 See Kayser in Stockholm Vet. Ak. Bch. Bd. xxvii. (1901) ; Nathorst in Bull. Soc. Instit. Upsala, vol. x. (1910) ; and A. S. Woodward in Ann. Mag. Nat. Hist. s. 6, vol. viii. (1895). 22 See Heer in Quart. Journ. Geol. Soc. vol. xxviii. p. 163. CHAPTER IX THE CARBONIFEROUS SYSTEM THE materials for a history of this period are more ample than those available for the preceding periods. Carboniferous rocks occupy a larger part of the British Isles than the rocks of any other system, and in the search for coal they have also been more extensively studied and explored, so .that it has become possible to trace the lateral changes which the component members of the system undergo, and to cprrelate the strata of different districts with much greater accuracy. Carboniferous strata have also a considerable extension in France, Belgium, Germany, Silesia, and Russia as well as in Spain. The records of the period are indeed more complete on the continent than they are in the British Isles, for they carry the history of it to a later date in geological time. Continental geologists have found it necessary to divide the whole system into three series, of which only the two older are represented in Britain, so that we have been accustomed to regard the Carboniferous System as consisting of these two, a Lower Carboniferous and an Upper (or Coal-measure) Series. The advance of comparative geology makes this view and our insular nomenclature no longer tenable, and we must either give new names to the British divisions or accept those current on the continent. Dr. Vaughan has proposed the name Avonian for the southern type of the Lower Carboniferous, and the name Bernician has been used for the northern facies. a No good name has yet been proposed for the rest of the British Series. French geologists have adopted the name Dinantian for their Lower Carboniferous Series, but it is less comprehensive than the British Avonian. For the higher stages it will be convenient to adopt their names, and others again are used by Eussian geologists a It was first proposed by S. P. Woodward (1856) for the whole Lower Carboniferous Series, and afterwards by Lebour (1877) for the greater part. 244 THE CARBONIFEROUS SYSTEM 245 as their facies of the system is almost wholly marine. The nomen- clature will therefore be as follows : England. France. Russia. "Westphalian. Stephanian. Uralian. Avonian. Westphalian. Moscovian. Dinantian. Dinantian. In England, France, and Belgium, as well as in Russia, the Avonian consists almost entirely of marine deposits, and in most districts it is mainly composed of limestone. In Germany, on the other hand, it consists of carbonaceous shales with beds of sandstone and has long been known as the Culm. The Westphalian is everywhere (except in Russia) a varied succession of sandstones, shales, clays, and coal-seams, which seem to have been deposited on the borders of a sinking land, in deltas and lagoons, where the water was sometimes saline and sometimes brackish or fresh. The Stephanian is also a Coal-Measure Series of similar character in the west of Europe, but passes eastward into the marine Uralian of Russia. As each of these great series has in some places a thickness of 9000 to 10,000 feet, and as the fauna and flora of each differs con- siderably from that of the others, it will be convenient to describe them separately and in succession. I. THE AVONIAN SERIES A. LIFE OF THE EPOCH Plants are naturally not common fossils in marine deposits, so that the number known from this series is not large. The flora is almost the same as that of the Upper Devonian, and the following are the principal genera : Archceocalamites ( = Bornia), Cheirostrobus, Sphenopteris, (S. ajrfinis and S. elegans), Adiantites, Rhacopteris, Cardiopteris, Heterangium, Lepidodendron, Lepidophloios, and Pitys. The Dinantian fauna is a rich one, both in genera and species. The following is a brief account of it, new generic appearances being marked with an asterisk. Hydrozoa. Ghcetetes. Foraminifera. These become very abundant in the rocks of this period and contribute largely to the formation of some of the limestones, the principal genera being Saccammina, Endothyra, Fusulina, Trochammina, Valvulina, and Lagena. Anthozoa, Of corals the most common genera are Aulo- Fig. 81. GROUP OF CARBONIFEROUS FOSSILS (AVONIAN). a. Syringothyris cuspidata. / Posidonomya Becheri. b. Productus giganteus. g. Modiola Macadami. c. Woodocrinus maerodactylus. h. Phillipsia derbiensis. d. Conocardium aliforme. i. Lithostrotion basaltiforme. e. Glyphioceras sphasricum. in. Aviculopecten sublobatus. n. Euomphalus pentagonalis. THE CARBONIFEROUS SYSTEM 247 phyllum* Lithostrotion* Lonsdaleia* Phillipsastrea, Syringopora, Clisiophyllum* Dibunophyllum,* Cyclophyllum* Cyathaxonia,* Michelinia, Zaphrentis, Amplexus, Caninia, and Cyathophyllum. None of these survive the Carboniferous period in Europe, though a few occur in the Permian of Asia. Echinoderma. Crinoidea are abundant, and their remains often form a large part of the limestones. The chief genera are 3ROUP OF CARBONIFEROUS FOSSILS (AVONIAN). a. Michelinia megastoma. b. Amplexus coralloides. c. Lithostrotion affine. d. Terebratula (Dielasma) hastata. e. Spirifer striatus. /. Productus semireticulatus. Actinocrinus* Platycrinus, Rhodocrinus, Cyathocrinus, Poteriocrinus, Amphoracrinus* and Woodocrinus* None of them, with the possible exception of Cyathocrinus^ survived this period. Blastoidea now seem to take the place of the Cystideans of the older Paleozoic rocks, the commoner genera being Granatocrinus, Cadaster, and Orophocrinus. Of Echinoidea there are Archceocidaris, Palcechinus, and Melonites. Crustacea. The only surviving genera of trilobites are 248 STEATIGRAPHICAL GEOLOGY Phillipsia, Griffiihides, Proetus, Cyphaspis, and Brachymetopus. None are common, all are small, and all died out, in Europe at any rate, before the close of the period. Small Ostracoda are often abundant belonging to the genera Bairdia, Cypridina, Cypridinella, and others. Dithyrocaris and other Phyllocarids also occur. Schizopoda are also represented and are common in some parts of Scotland. Fig. 83. GROUP OF CARBONIFEROUS FOSSILS (AVONIAN). a. Platycrinus Isevis. d. Phillipsia pustulosa. b. Palaechinus gigas. e. Coelonautilus cariniferus. c. Granatocrinus ellipticus. /. Gastrioceras Listeri. g. Orthoceras Gesneri. Brachiopoda. The only common and distinctive Carboni- ferous genus is Seminula, but most of the Devonian genera survive, such as Orbiculoidea, Productus, Spirifer, Spiriferina, Rhynchonella (Pugnax and Hypothyris), Camaroplioria, Dielasma, Athyris, Chonetes, and Orthis. Lamellibranchia. The following are some of the com- moner genera, those with an asterisk making their first appearance : Pterinopecten, Aviculopecten, Crenipecten, Posidoniella,* Parallelodon, THE CARBONIFEROUS SYSTEM 249 Conocardium, Myalina, Schizodus* Leptodomus (Protoschizodus), Edmondia, Cardiomorpha, Pleurophorus, Cypricardiella* and San- yuinolites* Gastropoda. Most of the Devonian genera continue, and the commoner are : Metoptoma, Pleurotomaria, Murchisonia, Bellero- phon, Euomphalus, Straparollus, Naticopsis, Capulus, Loxonema, Fig. 84. GROUP OF CARBONIFEROUS FOSSILS (AVONIAN). a. Orthis resupinata. b. Aviculopecten papyraceus. c. Cardiomorpha oblonga. d. Conocardium hibernieum. e. Euomphalus pentagonalis. /. Bellerophon tangentialis. Macrochilus, and Twbonitella, but very few of these survive the close of the Carboniferous period. Cephalopoda. The following are some of the genera which occur : of Nautiloidea there are Orthoceras, Actinoceras, Cyrtoceras, Poterioceras, Temnocheilus* Ccelonautilus,* Vestinautilus,* Strobo- ceras,* Discites ; of Ammonidea the following genera belonging to the GoniatitidjB, Brancoceras, Pericyclus* Gastrioceras, Glyphio- ceras, Nortiismoceras* Dimorphoceras* Prolecanites* Pronoiites* 250 STEATIGKAPHICAL GEOLOGY Pisces. Remains of fish are very numerous, and a long list of names founded on the teeth and spines of fish might be given, but the following are known from more complete remains, Eurynotus, Rhizodus, Holoptychius, Megalichthys, Coelacanthus, Platy- somus. Many of the Carboniferous fish were Elasmobranchs, and most of the remainder are Teleostomi of the Crossopterygian order. A remarkable fact in the distribution of the marine fauna is the persistence of many species throughout the whole of the Avoniaii Series. Recent studies of the corals and Brachiopoda have, how- ever, shown that a few species are restricted to certain horizons and that others are much more abundant within certain zonal limits than they are above or below, and in this way a zonal classification of the great mass of limestones has been established. The following are some of the most common and characteristic species of the Dinantian Series as a whole : Foraminifera. Saccammina frontiniformis (syn. Carteri), Trocham- mina incerta, Endothyra. Hydrozoa. Chaetetes tumidus. Actinozoa. Amplexus coralloides, Cyathophyllum regium, Lons- daleia floriformis, Clisiophyllum bipartitum, Dibunophyllum turbinatum, Cyathaxonia cornu, Phillipsastrea radiata, Syringopora geiriculata, Lithostrotion affine, L. basaltiforme, L. striatum, Zaphrentis Phillipsi, Caninia cylindrica, Michelinia favosa, and M. megastoma. Bryozoa. Fenestella flabellata, Ceriopora gracilis. Echinoderma. Actinocrinus triacontodactylus, Amphoracrinus an;- phora, Platycrinus laevis, Poteriocrinus crassus, Woodocrinus macrodactylus ; Archaeocidaris Urqi, Pal83chinus sphericus, Pal. gigas, Orbitremites derbiensis. Brachiopoda. Athris (Seminula) ambigua, A. (Seminula) ficoides, A. (Cliothyris) Roissyi, Orthis resupinata, Spirifer striatus, S. clathratus, Syringotliyris cuspidata, Productus seraireticulatus, P. giganteus, Dielasma hastatum, Rhynchonella (Hypothyris) pleurodon, R. (Pugnax) acuminata, Spiriferina octoplicata. Lamellibranchia. Conocardium aliforme, (J. hibernicum, Pterinopecten papyraceus, Pt. sublobatus, Pt. granosus, Edmondia sulcata, Modiola Macadami, Myalina crassa, Cardio- morpha orbicularis, C. oblonga, Posidonomya Becheri. Gastropoda. Euomphalus pentangulatus, Naticopsis plicistria, Pleurotomaria carinata, Murchisonia verneuiliana, Platyceras neritoides, Bellerophon Urei, B. tangenti- alis, Macrochilus ovalis, Capulus vetustus. Cephalopoda. Ccelonautilus cariniferus, Cyrtoceras Gesneri, Ortho- ceras undulatum, Actinoceras giganteum, Poterio- ceras fusiformis, Gly phioceras diadema ( crenistria), G. sphericum, Pronorites cyclolobus. THE CARBONIFEROUS SYSTEM 251 TriloMta. Phillipsia pustulosa, Griffithides globiceps, Brachy- metopus ouralicus. Entomostraca. Leperditia Okeni. Pisces. Eurynotus crenatus, Rhizodus Hibberti, Elonichtlr\ r s Robisoni, Cladodus striatus, Cochliodus contortus, Orodus ramosus, Petalodus acuminatus, Psammodus rugosus. B. BRITISH AVONIAN STRATA The surface areas occupied by the Lower Carboniferous or Avonian Series in the British Isles are shown on most geological maps. They are clearly shown on the cheap (2s.) Index map of the Geological Survey, and most of them are indicated in the maps (Figs. 87 and 94) of this volume. The large extent of the area which they occupy in Ireland is especially noticeable, though the student must remember that in Fig. 94 the superjacent Coal- measures are not shown. In England it is probable that the limestones extend eastward under large parts of the southern and central counties, but they do not appear to exist in the East Anglian counties, where many borings have been made at different times, but have always found rocks of pre-Carboniferous age beneath the cover of Cretaceous strata. With regard to the superposition and relation to the older rocks it may be stated generally that the Carboniferous strata follow those of the Upper Devonian or Upper Old Ked Sandstone in conformable succession wherever the latter are present ; but when these are absent, the Carboniferous rests unconformably upon the older formations. With respect to the subdivisions of the Avonian Series recent studies by Dr. Vaughan, Dr. W. Hind, and Professor Garwood have shown that it is everywhere divisible into two stages, and as French geologists had already made two similar divisions, their names have been adopted, namely Tournaisian for the lower and Vise'an for the upper stage. In England each stage appears to be divisible into three zones, but the delimitation of these zones has not yet been determined in all districts. Where fully developed, the normal succession is as follows : f Dy. Cyathaxonia sub-zone. Visean-! D. Dibunophyllum zone. I S. Seminula zone. f C. Syringothyris zone. Tournaisian^ Z. Zaphrentis zone. [ K. Cleistopora zone. 252 STRATIGRAPHICAL GEOLOGY It should be stated that Dr. Vaughan regards the Cyathaxonia Beds and their equivalents as a sub-zone of the Dibunophyllum zone, but as these equivalents include the Yo red ale Beds of Yorkshire and a great part of the " Pendleside Series " of Dr. Hind, which possesses a peculiar fauna of its own, it seems desirable to give special prominence to them in a general scheme, though they are poorly represented near Bristol. For convenience of description the more or less different facies which the Avonian Series presents in different parts of the British Isles will have to be treated under seven heads, viz. (1) Bristol and South Wales, (2) North Wales and Derbyshire, (3) South Lancashire and Yorkshire, (4) North Lancashire and Westmoreland, (5) Northumberland and Berwick, (6) Scotland, (7) Ireland. 1. Bristol and South Wales The zonal study of the limestone series of the Bristol district by Dr. Vaughan 1 has made it the typical exposure of that series in Britain. The succession is there complete from the base to the top of the Dibunophyllum zone (D 9 ), but no calcareous representative of the Cyathaxonia zone (D 3 and D y ) occurs in the Mendip or Bristol districts, and its equivalent must be sought in the sandstones which form the lower part of the so-called " Millstone grit." The actual thickness of the Limestone Series varies greatly in different parts of the area ; it is thickest in the Mendip Hills, rather less at Clifton, and much thinner to the northward. Litho- logically the series has been divided into four parts, which, with their varying thicknesses, are given below : Mendips. Clifton. Wickwar. Upper limestones and shales .... 1200 1200 830 Middle shales and oolitic limestone . . . 250 300 \ 0n Lower limestones (encrinital) .... 1200 500 J Lower shales 450 400 320 3000 2400 1670 Dolomitisation is prevalent in the middle part of the series, and affects a greater and greater portion as the beds are followed north- ward. The lithology of the strata is, however, only of local importance, while the zonal divisions established by Dr. Vaughan form a standard by means of which the limestones and shales of other areas, whether they are in similar or dissimilar sequence, can be compared with the Avon section. The Cleistopora zone is coextensive with the lower shales, and is divisible into two portions, the lower 100 feet forming the phase THE CARBONIFEROUS SYSTEM 253 of Modiola (cf. lata), and the higher beds containing Productus bassus, and near the top an abundance of Cleistopora geometrica, a small compound coral of Michelinioid aspect. The Zaphrentis zone consists of encrinital limestones charac- terised in the lower part by an abundance of Spirifer clathratiis, and in the upper part by Zaphrentis KonincJci. The Syringothyris zone consists largely of dolomitic and oolitic limestones in which Caninia cylindrica, Michelinia cf. mega- stoma, and Spirifer (Syringothyris) cuspidatus are common and characteristic fossils. Caninia patula marks a lower, and Cyatho- phyllum (sp. <) an upper sub-zone. The Seminula zone consists of limestones with shale-partings in the lower part, and beds of pisolitic limestone in the upper. The index fossil is the Athyrid Seminula ficoides, but Productus corrugato-hemisphericus is common, as also are massive and dendroid Lithostrotions and Oarcinophyllum. The Dibunophyllum zone consists of limestones, shales, and occasional grits. It is characterised by the abundance of highly developed corals : Dibunophyllum (simple in the lower and advanced in the upper sub-zone ; Lithostrotion irregulare, Cyatho- phyllum Murchisoni are common forms ; Lonsdaleia floriformis and Cyathophyllum regium mark the upper sub-zone. At Wick war, near the northern end of the Bristol coalfield, the limestone series is reduced to 1670 feet, and farther north, in the Forest of Dean area, it is less than 600 feet. In the last-mentioned area the highest limestones are of Seminula age, and are succeeded conformably by sands and quartz-pebble conglomerates that have been referred to the " Millstone grit " although they are of con- siderably earlier age than the base of the typical Millstone grit of the Midlands and Yorkshire. In South Wales, at the eastern end of the coalfield, the Avonian Series, with shales at base and summit, is about 1000 feet thick near Cardiff and Newport, but thins northward to less than 500. (The series is, however, incomplete at the top as Millstone grit lies unconformably upon Seminula and Lower Dibunophyllum Beds.) Westward, near Rhymney, and in the Gower Peninsula, the Avonian attains a thickness of 3000 feet. There the black shales and limestones 2 above the Dibunophyllum limestone (see Fig. 77) yield a peculiar fauna, including the Goniatites Glyphioceras bi- lingue, G. spirale, and G. reticulatum. These are characteristic species of the Pendleside Beds of Lancashire and North Stafford, and belong to a zone or sub-zone which lies between the Dibuno- phyllum zone of the Bristol area and the typical Millstone grit of the Dovedale country. 254 STKATIGEAPHICAL GEOLOGY 3 g II & != S 0, THE CARBONIFEROUS SYSTEM 255 In Pembrokeshire all the zones are well developed, and are finely exhibited in the coast sections. At Skrinkle Haven the junction of Old Eed Sandstone and basal Carboniferous is well shown ; the central beds constitute the Tenby anticline, and the " Peiidleside " Beds are found north of that place, where they consist of black carbonaceous shales with large ellipsoidal concre- tions, locally known as bullions. These beds contain Glyphioceras diadema, Pterinopecten papyraceus, and Posidoniella membranacea. 2. North Wales, Sta/ord, and Derby The great difference between the limestone series of this area and that of South Wales and Gloucestershire is that it belongs entirely to the Visean stage, the basal beds being on the horizon of the upper part of the Seminula zone ; so that there is no representative of the Tournaisian stage. The lithological divisions were studied and described by Mr. G. H. Morton and Dr. A. Strahan, while the palseontological zones have recently been worked out by Dr. W. Hind and Mr. J. T. Stobbs. 3 The most complete sequence is found in Flintshire, and this is given in the following table : Lithological Divisions. Zones. Thickness. } Posidonomya Beoheri } 1000 P- Dibun.phjH.un } 450 } Lower Dibunophyll} 1100 About 3000 The absence of Gamma makes it fairly certain that only the top of the Seminula zone comes into the succession. The lower part of the Dibunophyllum zone is marked by the presence of that genus with Cyathophyllum Murchisoni, Productus hemisphericus, and P. cora. The higher part contains all the typical fossils of its Bristol equivalent, and the species of Lithostrotion are common, as are also several species of Spirifer and Schizophora resupinata. The sandy beds at the top of this zone consist of compact fine- grained sandstone, each bed having a medial band of grey or black chert which is full of sponge spicules. These beds and the Aberdo limestone are regarded as a separate sub-zone characterised by the coral Cyathaxonia and Productus longispina. They form a passage into the overlying stage. GEOLOGICAL MAP OP THE NORTHERN PART OP ENGLAND Trias Permian Coal Measures Millstone Grit Low Carboniferous O^d R.M Sandstone Silurian Fig. 87. THE CARBONIFEROUS SYSTEM 257 The true position of the Holywell shales was first indicated by Dr. Hinde, and the group is undoubtedly the equivalent of his Pendleside Series. In Flintshire it consists, in the lower part, of black shales with layers of black limestone and chert, yielding the bivalves Posidonomya and Pterinopecten, while the higher part is composed of soft shales in which Posidoniella Items is common. Southwards all these beds decrease in thickness rather rapidly, and they finally thin out south of Oswestry, only the higher part of the limestone with Prod, giganteus appearing again for a short distance in Shropshire, near Lilleshall. Westward the whole series is faulted down into the Vale of Choyd, and portions of it occur along the coast; thus, in the Orme's Head promontory near Llandudno, the greater part of the limestone series is found with a thickness of 1150 feet, though neither the base nor the summit is seen. In Anglesey tracts occur on the border of the Menai Strait, and a larger one is faulted into the central part of the island ; the lowest limestones here belong to the white division, and in places they include beds of white sandstone and conglomerate from 90 to 100 feet thick. In North Staffordshire and Derbyshire the zonal succession has recently been determined by Mr. Sibly, and found to be as follows : Feet. D 3 4. Shales and thin beds of black limestone (Yoredale Beds) . 400 Do 3. Grey and white limestones containing layers of chert (sub- zone of Lonsdaleia floriformis) ..... 600 (2. Massive white limestones with interbedded lava-flows (toad- stones) = Dibunophyl]um Beds . ..... 600 1. Dark limestones ....... seen for 800 2400 j- ( The actual base is nowhere exposed, and consequently the total thickness of the series is unknown, but it is doubtful whether it exceeds 3000 feet even in the centre of the area. The whole of the series thins out to the south-west before reaching the South Staffordshire coalfield, but continues toward the south-east beneath the small Leicestershire coalfield and wraps round the pre-Cambrian rocks of Charnwood Forest. The limestones which compose the " Mountain limestone " of Derbyshire vary much in lithological character, 4 some consisting mainly of broken fragments of crinoids and shells, others of corals and coral debris, and others being compact and foraminiferal, while in many parts the organisms have been obscured or destroyed by the chemical changes of marmorisation or dolomitisation. The upper surface of D 2 shows signs of contemporaneous S 258 STKATIGRAPHICAL GEOLOGY erosion, and is overlain by a bed of rolled shells and limestone pebbles with occasional quartz pebbles ; it was first found at Castleton, but has since been traced by Dr. W. Hind through the whole of Derbyshire and North Staffordshire. It contains many Brachiopods, trilobites, and teeth of Psammodus, Psephodus, etc., and shows that the current which brought the muddy sediment of the overlying shales was at first strong enough to break up and destroy some of the previously formed limestone, which at that time was, of course, in an unconsolidated condition. The shales above mentioned are termed the Yoredale shales by the Geological Survey and the Pendleside shales by Dr. W. Hind, but as we shall presently see that the two groups are con- temporaneous and not successive, as Dr. Hind supposed, it does Astbury Fault. Limeworks. Quarry. Quarry. Glacial Sands and C/ay Trias""'''/ a bcdef g Fig. 88. SECTION THROUGH CONGLETON EDGE (by W. Gibson and W. Hind). h, i. Millstone grits. b to/. Grit, tuff, limestone, and shale. g. Pendleside shales. o. Main Avonian limestone. not matter which name they bear. The relation of these shales to the limestone below and the Millstone grit above is shown in Fig. 88. 3. South Lancashire and Yorkshire The Avonian Series is brought up again from beneath the Millstone grits in Lancashire and West Yorkshire (see map, Fig. 87) by an anticlinal flexure which runs in a north -easterly direction from near Blackburn by CHtheroe to Skipton, with a more northerly prolongation by Cracoe across Wharfedale by Burnsall. From these places this division of the system spreads northward and occupies a large irregular tract of ground varying from 15 to 30 miles in width through Yorkshire, Durham, and the eastern parts of Westmoreland and Cumberland. The rocks of this large area present two different facies, one in the south-west, which has been called the Lancashire or Clitheroe type, and the other to the north, which is known as the Yoredale type. Moreover, the transition from one to the other is rapid, and coincides closely with the line of the Craven fault, which runs obliquely across the area from near Kirkby Lonsdale by Ingleton, Stainforth, and Linton. South of this line the beds THE CARBONIFEROUS SYSTEM 259 attain a great thickness and their base is nowhere exposed ; north of it the base is exposed in several places, resting on a floor of upturned Ordovician and Silurian rocks, but the series is much thinner (see Fig. 90). For the following comparative view of the succession in the two areas I am indebted to Dr. Vaughan. Clitheroe District. Feet. Ingleton District. Feet. ( Millstone grit Millstone grit M G I (4th grit). (4th grit). Bowland shales . . 50 to 800 Pendle grit . . . to 300 D y . Pendleside limestone . 100 to 400 D 2 . Pendleside shales . . 1200 Yoredale Beds 400 c. o -r* ws-i i 11 T ooArt (} Scar limestones 1 S & D!. Clitheroe knoll-limestones . 2800 | Basement Beds J 800 The Clitheroe and Cracoe limestones have not yet been com- pletely zonalised, but Dr. Vaughan informs me that the main Clitheroe knolls include representatives of D 1 and of S (Seminula zone), and that the beds at the base of the main knolls belong to C (Syringothyris zone), but there is a considerable thickness of limestone below that horizon. Near Ingleborough the lowest beds are not older than the top of the Syringothyris zone, and the total thickness of Visean limestones is only 600 feet. 5 Northward, however, they thicken again con- siderably till near Kirkby Stephen there is 3700 feet of them. Then they again diminish by the rapid thinning out of the lower beds till in the Cross Fell district they are once more reduced to 600 or 700 feet, and much of this is sandstone. Fig. 91 illustrates the position and composition of the Visean Group at Ash Fell and Roman Fell, east of Appleby, and the map (Fig. 89) shows the course of the buried ridge of Ordovician rocks which causes this thinning-out a ridge which forms a natural northern limit to the region we are describing. Returning now to the north-east of Lancashire the higher stage, comprising the Pendleside and Yoredale Beds, remains to be dealt with. The succession of beds forming the Pendleside Group was fully described by Dr. W. Hind, 6 and is given above ; .the black Pendleside shales are characterised by Pusidonomya Becheri and Pterinopecten papyraceus, and the Pendle limestone .by the same with Glyphioceras spirale and G. reticulatum. In the Clitheroe and Pendleside districts the total thickness of this group is at least 2500 feet, but it thins northward, and when the Settle faults are crossed its place is taken by such a different set of beds .that the identity of the two groups has been for a long time in 260 STEATIGEAPHICAL GEOLOGY doubt. This year (1911), however, Mr. Cosmo Johns 7 has published evidence which seems to show that the Pendle limestone is the equivalent of the " main limestone " near the top of the Yoredale Beds. The Yoredale Beds consist of limestones with intercalated bands of shale and sandstone, the limestones being persistent bands which have been traced for long distances, and each band has received a special name. The highest of the Yoredale limestones is known SCOTLAND Stanford's Geay 1 Estab c , Londc Fig. 89. MAP OF THE NORTH-WESTERN COUNTIES TO SHOW THE POSITION OF THE PRE-CARBONIFEROUS RIDGE (Goodchild). as the " main limestone," and the lowest as the " Hardraw lime- stone." Between these at Ingleborough there are three other limestones, and farther north two others come in, so that in Wensleydale there are seven Yoredale limestones. Dr. Vaughan informs me that the fauna of the Yoredale Beds is identical with that of the upper part of the Dibimophyllum zone of Bristol (Dg), and thus, if the beds are the equivalents of the Lower Pendleside Group, the latter must have been formed under different bathymetric conditions. Above the " main limestone " of Wensleydale there are some peculiar beds which, though of no great thickness (160 feet), are THE CARBONIFEROUS SYSTEM 261 262 STKATIGRAPHICAL GEOLOGY remarkable both in their lithological and palseontological characters. They include two beds of limestone, each of which is overlain by a band of chert-beds ; the lower set is known as the " Ked Beds," and the upper as the " Crow limestone and chert." The chert in the Eed Beds is 42 feet thick, and has yielded a remark- able set of fish teeth, an account of which, with a complete enumeration of the beds at Leyburn in Wensleydale, has been published by Mr. J. W. Davis. 4. North Lancashire and Westmoreland The area occupied by Avonian limestones in this district is shown on the map (Fig. 87), and the beds have been recently studied and classified by Professor Garwood, who finds representa- tives of both the Tournaisian and Viseaii stages. 8 The fauna, however, contains some special elements, and the relative abundance of species is different from that prevailing near Bristol, so that he found it well to make different zonal divisions, though these can be roughly correlated with the southern zones. The following table is copied from that given by him : Zones. Subzones. Jl "6 Dibunophyllum Productus corrugato- hemisphericus and Spiriferina, laminosa ! Upper (?) (Productus latissimus Middle- 1 , and { Lonsdaleia floriformis T / Cyathophylliun 81 X ^Murchisoni r y-r f Neniatophyllum minus r Gastropod Beds 1 Lower-! Clisiophyllum keyser- \ lingi at base D 2 a ;| Beds without Lithostrotion. Michelinia megastoma Athyris gldbristria f Upper, Daviesiella carinata \ T f Camarophoria ( Lower l isorhyncha f Upper, ^Seminula grcgaria X Lower, ' Solenopora C 2 j| Ci | g y 1 1 Basement Conglomerate. The lowest beds are shales and conglomerates found in Kavenstone THE CARBONIFEROUS SYSTEM 263 Dale and near Shap in Westmoreland, and it is these which he refers to the top of the Zaphrentis zone. The lowest limestones are highly magnesian, and appear to have been deposited under lagoon-like conditions which encouraged the growth of calcareous algae, especially Solenopora, and these have contributed largely to the formation of these beds. Most of the Visean limestones are rich in Foraminifera, Saccammina being abundant in the D zone. The highest beds are those of Humphrey Head (Morecambe Bay) and of Brampton near Shap. The full thickness is probably over 3000 feet, but lessens northward. 5. Northumberland and Berwick From the buried ridge of older rocks which is indicated in Fig. 89 the thickness of the Avonian Series again increases ; thus the main mass of the Visean or Scar limestone becomes still further split up by the intercalation of beds of shale and sandstone till, near Alston, it has expanded from about 400 to 1600 feet, with twelve distinct bands of limestone. It has, in fact, assumed a facies similar to that of the overlying Yoredale Group. Moreover, the series increases by the incoming of lower and lower beds, some of them representing Tournaisian stages, and at Alston these are 1000 feet thick and about 2000 on the Scottish border. This lower group contains but little limestone, and is largely arenaceous, so that the name of Tuedian was given to it in 1856 by Mr. G. Tate, from its fine development along the valley of the Tweed. Similarly in 1877 Professor Lebour proposed the name Bernician for the compound facies which represents the higher limestones of the Visean and the Yoredalian stage. Probably the most convenient provincial classification of the series is that adopted in recent memoirs of the Geological Survey, but this now requires to be brought into line with the zonal classification of more southern areas, and with the " stage " divisions which have been used in the preceding pages. The zonal comparison has been accomplished in large part by Professor Garwood, 9 and is indicated in the following table, except with regard to the higher beds, which, being the equivalents of the Yoredale Beds, must represent the highest part of the Dibuno- phyllum zone if Mr. C. Johns' views are correct : 264 STRATIGKAPHICAL GEOLOGY Subdivisions. Zones. Stages. Upper Calcareous division (from the Millstone grit ^ to the top of the Oxford limestone). Sand- stones and shales with some coals and one or more beds of marine limestone, 1200-2000 . Lower Calcareous division (from Oxford lime- stone to base of Dun or Redesdale lime- stone). Sandstones and shales with coals and many beds of marine limestone, 800- ! v - , 1000 D 2 r V d. Carbonaceous division (Scremerston Beds). Strata prevalently carbonaceous ; limestones chiefly thin, many containing carbonaceous V matter ; many coal-seams, 800-2500 (c. Upper Tuedian or Fell Sandstone Group : a belt of massive grits with green, grey, and reddish shales ; coals rare and thin, 500-1600 . . SJ I b. Lower Tuedian or Cement-stone Group : sand- g j stones and shales with bands of cement-stone passing into limestones near Rothbury and I Tournais- Bewcastle, 500-1500 C j ian. a. Basement conglomerates (Upper Old Red I Sandstone), 0-500 ? Tuedian Series. a. The basement conglomerates occur locally round the Cheviot Hills, and are by some referred to the Upper Old Red Sandstone, but they pass up into the overlying sandstones, and certainly form a local base of the Carboniferous System. They are of a reddish-brown colour, and contain rolled pebbles of the Cheviot porphyrites. 6. This group consists of yellow and reddish sandstones, grey and purple shales, greenish sandy clays (sometimes full of hard siliceous concretions), and beds of hard cream-coloured earthy limestone or cement-stone. In these beds fossils are rare and badly preserved, but Modiola Macadami occurs. Near the top are limestones in which Professor Garwood has found Syringothyris cuspidata, Athyris glabristria, and Orthotetes crenistria, and these are sufficient to indicate the zonal age. c. The Fell sandstones form a broad belt of high and rugged country in the west of Northumberland, rising into craggy hills such as those of the Chillingham and Simonside Hills, and the Peel and Bewcastle Fells. These sandstones are divided by bands of shale, and a few thin seams of coal occur. The fossils are chiefly plant remains, but Archanodon JuJcesi has been found in the lowest (Harbottle) grit, and Professor Garwood has also discovered a few marine fossils in the lowest beds on the borders of Cumber- land ; these include Camarophoria isorhyncha and Spiriferina laminosa, which in Westmoreland are restricted to the lower part of the Seminula zone. 9 THE CARBONIFEROUS SYSTEM 265 Bernician Series. On the southern border of Northumber- land the Scar limestone of Yorkshire and the Yoredale Beds are still distinguishable as separate stages, but as they are followed northward they lose their distinctive characters and merge into one great series of alternating limestones, shales, and sandstones, with many workable beds of coal. The most remarkable feature of this series is the great develop- ment of mechanical sediments in the centre of the county, some of the lower beds expanding into the "Carbonaceous" Group, and many fresh beds of sandstone and shale with coal-seams coming in above the Main Yoredale limestone to form the upper calcareous division, so that the whole is about 6000 feet thick, but again becomes thinner to the northward. The Carbonaceous division has not yielded many marine fossils, but Professor Garwood has found Spiriferina laminosa with dendroid Lithostrotions, and the group is referable to D r The lower calcareous group contains species of Dibunophyllum with Cyatho- phyllum regium and Lithostrotion junceum, so that its zonal age is evident, and it is noteworthy that several of the limestones are largely composed of the Foraminifer Saccammina Carteri. The Oxford limestone is probably the equivalent of the Hardraw limestone of Durham, and if so the beds above this correspond with the Yoredale Group, but the Pendleside Cephalopoda have not yet been found. On the other hand, the corals Lonsdaleia floriformis and Alveolites septosa, with the Brachiopods Productus giganteus and P. latissimus, first appear in and above the Oxford limestone. 6. Scotland The central lowlands of Scotland display a great thickness of Lower Carboniferous rocks which may be regarded as the equivalents of the Tuedian and Bernician Series of Northumberland, for the Scottish Series is similarly divisible into a lower group, which is mainly arenaceous, and an upper more varied group of shales, sandstones, limestones, and coals. The lower has been called the Calciferous Sandstone Series, and the upper was termed the Carboniferous Limestone Series, but the equivalents of the main mass of the Carboniferous limestone lie entirely below this " lime- stone series" of Scotland. Thus the nomenclature formerly used by the Geological Survey and adopted in most text-books gives rise to misconception and tends to exaggerate the stratigraphical importance of the upper part of the series. For while the upper portion, which has a thickness of only from 1600 to 1800 feet, co Si i 1 >o i sfl O 1 I ll-a is^-c-s 8 S e'S aJ 1 1 THE CAKBONIFEROUS SYSTEM 267 has been divided into three groups, the lower portion, which is nearly 6000 feet thick, was undivided. According to the late J. G. Good child 10 the lower portion is equally capable of division into three groups, which he called (1) the Ballagan Beds, (2) the Gran ton and Hailes sandstones, (3) the Oil-shale Group. He regarded the sandstone group as the equivalents of the Fell sandstones and the Oil-shales as corre- sponding with the Carbonaceous division of the Border counties. The Geological Survey has not entirely accepted this nomenclature, 11 but has included the massive sandstones in the Oil-shale Group because some thin oil-shales have been found in them. In the following table Good child's divisions are adopted as being con- venient and lending themselves to correlation with Northumberland. ThSness Scotch Divisions - Northumberland. 500 /. Upper Limestone Group \ 800 e. Edge Coal Group VYoredale Beds. 500 d. Lower Limestone Group J 2500 c. Oil-shale Group Carbonaceous Group. 1500 &. Granton and Hailes sandstones Fell sandstones. 1300 a. Cement-stone (or Ballagan) Beds Cement-stone Beds. Cement-stone Group. In Edinburgh, Linlithgow, and Fife this consists in the lower part of clays, shales, sandstones, and tuffs with some beds of dull compact limestone (cement-stone) and of siliceous sinter ; this part is about 200 feet thick. Next in the Edinburgh district comes the volcanic episode of Arthur's Seat, a set of basaltic lavas and tuffs with interstratified sandstones and shales, having a total thickness of 750 feet. These are covered by the Abbey Hill shales, which are 300 feet thick and contain marine fossils. The fossils in the lower part of the group are chiefly plant remains and small freshwater Crustacea (Leperditia, Estheria, Bythrocypris, etc.), but at Belhaven, near Dunbar, a flattened form of Modiola Macadami occurs in hundreds. In the Glasgow district this group has been called the Ballagan Beds, and there it consists entirely of sedimentary deposits, sand- stones, shales, and argillaceous limestones, without any con- temporaneous volcanic rocks, though the thickness is still from 1200 to 1500 feet. Granton and Hailes Sandstones. This group consists of two thick sandstones separated by a band of shales. The lower member is the Granton sandstone, probably about 800 feet thick ; the central band consists of the Wardie shales (400 feet), and the upper is the Hailes sandstone, which may be 1000 feet thick. These sandstones form good building stones, but have yielded few 268 STRATIGRAPHICAL GEOLOGY fossils ; but the Wardie shales contain Schizodus pentlandicus, Naiadites obesa, Anthracomya Icevis, and others. These beds gradually thin westward till near Glasgow there is only 100 feet of such sandstone. Oil - shale Group. This group takes its name from the prevalence of black bituminous shales which contain large quantities of petroleum, some bands of them being so rich as to yield from 30 to 40 gallons of oil per ton of shale, besides a considerable quantity of sulphate of ammonia as a by-product in the process of abstraction. Near Edinburgh there is a bed of oil-shale above the Hailes sand- stone which makes a convenient base to the group, and about 900 feet above this is the Burdiehouse limestone, a bed of dark -grey limestone 20 to 30 feet thick and crowded with the cases of the small Ostracod Leperditia OJceni (var. burdigalensis). Another lime- stone band higher up is entirely composed of the shells of Productus semireticulatus and P. longispinus. The group is most fully developed in Fifeshire, where it is over 3000 feet thick, containing many seams of oil-shale and no fewer than eighteen beds of limestone containing marine fossils ; but the mass of the group consists mainly of shales and sandstones which contain the remains of fish, plants, and estuarine molluscs, especially Naiadites obesa. These sediments are interstratified with sheets of volcanic material basalts and basaltic tufts and are pierced by a large number of pipes and bosses of volcanic agglomerate which are the orifices of small volcanic vents. When followed westward by Stirling and Dumbarton to Renfrew and the Isle of Arran, the oil-shales become much thinner and are interstratified with lava-flows, which in some districts seem entirely to take the place of the sedimentary deposits. Lower Limestone Group. Near Edinburgh this group has a thickness of 400 feet to 500 feet, and contains three beds of limestones, but these are seldom more than 12 feet thick, though occasionally swelling out to 30 feet, and in the Bathgate Hills (Linlitligow) to 70 or 80 feet. In Ayrshire the Huiiet, or Main limestone, attains a thickness of 100 feet. Spirifer trigonalis, Productus giganteus, and P. semireticulatus are common fossils, and in some places the limestones are largely crinoidal, i.e. composed of the shelly plates which form the arms and stems of crinoids. These limestones are interbedded with a series of shales, sandstones, and coal-seams, several of the coals being immediately succeeded by a limestone, e.g. the Hurlet coal and limestone, as if there had been times when the shallow lagoons and swamps were suddenly submerged and occupied by marine organisms. THE CARBONIFEROUS SYSTEM 269 Edge Coal Group. This is a shale and sandstone series, and it includes some valuable seams of coal, ironstone, and oil-shale, but no limestones. In Midlothian its thickness is about 800 feet. Some of the beds contain marine shells, others are full of terrestrial plants, showing that the gradual submergence which was in progress was now and then counterbalanced by the deposition of the sedi- ment brought down in such quantity by rivers. Upper Limestone Group. This is about 600 feet in Midlothian ; it contains three bands of limestone and four seams of coal, interstratified with sandstone, shale, and fireclay. The limestones contain the same fossils as those of the lower group, and though seldom more than three or four feet thick have been traced over an area of at least 1000 square miles, thus proving that the physical conditions of the period were very uniform over a large area. In the western part of the area (Glasgow, Eenfrew, and Arran) all these equivalents of the Yoredale Beds are much thinner, the Lower limestones being only 60 feet, the Edge Coal Group about 200, and the Upper limestones about 150 feet. All of them are found in Arran, where the red Corrie limestone is regarded as the equivalent of the Hurlet limestone of the mainland, for many of the beds are stained red in Arran. 7. Ireland In no country can the rocks of the Carboniferous System be better studied than in Ireland. They occupy fully one-half of the superficial area of the country ; the whole of the great central plain, which stretches completely across the island from east to west, consisting of Carboniferous limestones, though its surface is often covered by Pleistocene gravels (see map, Fig. 94). Here and there this limestone plain supports hills of low elevation composed of the Coal-measures, these outliers being the sole remnants of the beds which once spread over the limestone ; their survival being due to the circumstance of their lying in synclinal basins. For the sake of clearness and brevity the chief facts regarding the Avonian Series of Ireland will be given under three heads, (1) Northern counties, (2) The Great Central Plain, (3) Southern counties ; the first and third of these areas each exhibiting a special and peculiar facies of the formation differing greatly from that of the central area. Northern Counties. The Carboniferous rocks of the more northern counties are merely an extension of the Scotch facies, but show a passage into the limestone facies as they are traced into 272 STRATIGKAPHICAL GEOLOGY Leitrim and Sligo. In the counties of Derry and Tyrone a series similar to that of Arran and Ayrshire is found ; the basement beds of red sandstone and conglomerate resting directly either on the Archaean schists or on the Lower Old Ked Sandstone. The Tuedian or Calciferous sandstone is represented by about 2000 feet of strata, about half of which consists of reddish -brown sandstones, succeeded by calciferous sandstones and bands of shale which are well exposed in the valley of the Moyola Kiver near Drapersdown. In these higher beds the commonest fossils are, Leperditia OJceni, Modiola Macadami, and Protoschizodus axiniformis, with Dithyrocaris Colei, Gypris burdigalensis, and remains of fish. Above these beds, at Cullion and Desertmartin, limestones come in which are arenaceous in the lower beds and dolomitic in the higher ; the prevalent fossils are corals (Lithostrotion junceum, L. affinis) and Brachiopods (Productus giganteus and Orthis resupinata). Southwards the lowest beds appear to thin out, while higher beds come in above, so that in Fermanagh, Leitrim, and Sligo there is a more complete succession which, according to Messrs. Hull and Hardman, is as follows : 12 Feet. Black shales with some sandstones (Pendleside Beds) . 600 Grey limestones with nodules of chert .... 500 Carbonaceous shales and sandstones (Calp) . . . 1000 Lower limestones, grey and dark blue .... 600 Yellow sandstones and grey shales .... 200 Brown sandstones and conglomerates .... 500 3400 The basal beds are sometimes called Upper Old Red Sandstone, but they rest unconformably upon the Lower Old Red (see p. 237), and are now generally regarded as Lower Carboniferous because they seem to be the equivalents of similar beds in Derry and of the Lower Tuedian of Berwick. Westward, in Sligo and Mayo, there is a similar succession, but the basal beds are absent and the lower limestones pass laterally into beds of calcareous breccia and conglomerate, indicating a shore-line. The upper limestone rises into hills with bold craggy escarpments, while the softer beds of the Calp have been eroded into broad vales. Great Central Plain. The central part of Ireland, from the shores of Dublin and Meath to Galway Bay and Lough Comb, is a great tract of low undulating land which seldom rises to more than 300 feet above the sea. In consequence of this low elevation, the general horizontality of the limestones below it, and the prevalence of Pleistocene deposits, exposures are few in the central area, and THE CARBONIFEROUS SYSTEM 273 it is only in the outer parts, where the beds rise into hills, that the complete succession can be studied ; the best sections are found on its western, eastern, and southern borders. As the series above described is traced southward through Mayo and Koscommon into Galway it exhibits a change similar to that which takes place in England, namely a thinning of the sandstones and shales and a thickening of the limestones, till the series passes into a great limestone formation nearly 3000 feet thick, with the same black shales above and a still thicker set of basal sandstones. The following is the lithological succession in Galway and Clare according to Mr. Kinahan and the Survey : Upper limestone shales Limestone with layers of chert Upper limestone Lower limestone- Black limestone and chert \ Upper cherty zone / The Fenestella limestone Lower cherty zone Shaly limestone . .Lower limestone shale . Feet. . 80 . 1500 . 400 . 300 . 100 . 400 . 150 About 3000 The zonal distribution of fossils in the limestone of Clare county has recently been ascertained by Mr. J. A. Douglas, 13 and proves to be comparable with that in the Bristol sequence. He tabulates the beds and zones as follows : Survey Divisions. Zones and Sub-zones. Thickness. Black Goniatite shales. Upper Limestone. Bedded crinoidal limestones, both light and dark grey. Chert occurs in layers and nodules, some beds of oolitic lime- stone. ? = Pendleside Beds. D 3 Cyathaxonia. D 2 Lonsdaleia. Dj Dibunophyllum 6. S 2 Productus cora. S x Prod, semireticu- latus. 1900 Lower Unstratified Limestone. Mass- ive grey and mottled limestones, often dolomitic. Bryozoa abundant. C. Syringothyris cuspidata. 450 to 1000 Lower Stratified Limestone. Lime- stone with cherts at top, dark-grey or blue argillaceous limestones. Z 2 Zaphrentis konincki. 650 Lower Limestone shales. Black shales and limestones, green and yellow sandstones. Z 2 Spirifer clathratus. ? Cleistopora zone. Modiola phase. 150 274: STRATIGEAPHICAL GEOLOGY The basal yellow sandstones contain Lamellibranchs only. The succeeding shales have not yielded Cleistopora, but contain Spiri- ferina octoplicata and Productus bassus. The other zones contain assemblages which correspond closely with those of the South Wales and Bristol areas. A noteworthy feature is the great development of the Dibunophyllum and Cyathaxonia limestones, which have been termed the Burren limestone, from the district of that name, where they are magnificently exposed. It is thus described by Jukes 14 : " A range of hills, rather more than 1000 feet in height, sweeps for about 20 miles along the south side of Galway Bay. They are formed entirely of bare rock from the sea-level to the hill-tops, the only soil being found in crevices of the rock, or in patches in the hollows of the valleys. This rock is all limestone, in regular beds, which dip gently to the south, at an angle of 1^ degrees only ; and counting from the lowest bed that rises out on the sea-shore to the uppermost, which caps the summit of the hills three or four miles to the southward, there must be a thickness of at least 1600 to 1700 feet of solid limestone shown here." In Limerick and Tipperary there is a similar succession, and the limestone series attains a thickness of about 3500 feet, its thickest member being the Fenestella limestone (1500 to 1900 feet), while the upper (Burren) limestone has decreased to only 240 feet. The same succession of limestones can be followed eastward through Kilkenny and Carlow, and Fig. 93, across the Castlecomer coalfield, illustrates this part of the area. East of Thomastown the limestones overlap the lower shales so as to rest directly on the granite of Carlow, and a little to the northward, in County Dublin, shales of the calp facies overlap the limestones, thus giving evidence of another shore-line. This is confirmed by the sections on the coast of County Dublin near Lough Shinny and Rush, which have recently been examined by Dr. Matley 15 and Dr. Vaughan. The lowest beds there seen are conglomerates and shales, indicating the close proximity of a coast-line, and these are overlain by limestones belonging to the Dibunophyllum zone. Still higher are limestones with Cyathax- onia, overlain by limestones and black shales with Posidonomya Becheri. Southern Counties. When the Carboniferous rocks are traced to the south-west through Cork and Kerry a still greater and much more sudden change takes place in the constitution of the system. This consists in the introduction of what appears to be a distinct group of shales and slates between the Yellow Sandstone and the Carboniferous limestone, and to which Sir R. Griffith gave the THE CARBONIFEROUS SYSTEM 275 name of Carboniferous slate. This group of beds lias derived special importance from its bearing on the Devonian question, and the whole subject is so associated with the name of Jukes that it would be idle to attempt a better exposition of the interest attach- ing to the Carboniferous slate than is contained in Mr. Jukes's own description quoted below. 16 " If we draw a parallel of latitude through the towns of Ken- mare, Macroom, and Cork, the great development of Carboniferous slate lies wholly south of that line. If we examine the neigh- bourhood of the city of Cork itself, we find the [Upper] Old Red Sandstone with plants in its upper beds, and a very short distance above that we get solid Carboniferous limestone, with some black shales or slates between the two, but not more than 200 or 300 feet in thickness. Passing southwards to the mouth of the harbour of Monkstown or Queenstown, and then by Carrigaline and Cool- more, these intermediate black slates or shales thicken to 2000 or 3000 feet, still having the Old Red [or Coomhola Beds] below and the Carboniferous limestone above ; but going still farther south t>y Ringabella to Kin sale, the dark -grey slates and grey grits thicken rapidly to 5000 or 6000 feet, and are nowhere covered by any part of the Carboniferous limestone, though they show here and there highly calcareous bands." On Whiddy Island at the head of Bantry Bay there are black shales containing Posidonomya Becheri and P. membranaceus, the characteristic species of the shales above the limestone, and it is therefore highly probable that the Carboniferous slate is con- temporaneous with the whole of the Carboniferous limestone, the lower and upper parts of that formation passing laterally into shales, just as its middle part does to the northward. We may therefore regard the Carboniferous slate as representing the whole of this limestone and its underlying shale, i.e. strata amounting to a thickness of 2600 feet in the northern part of Ireland ; such, in fact, was Jukes's opinion. 8. Devon and Cornwall This area has been left till the last, because the equivalents of the Avonian Series occurring therein are in many respects similar to those of the south of Ireland and are very different from the typical Avonian of the Bristol and Mendip area, in spite of the close proximity of the two facies. Devonshire. The delimitation of the Upper Devonian and Lower Carboniferous rocks in Devonshire has not yet been worked out. The junction-beds have not been specially described since 276 STRATIGRAPH1CAL GEOLOGY Mr. Salter compared them with those of Pembrokeshire in 1863, 17 but his opinion was that the Pilton Beds included the equivalent of the lower limestone shale. He says, " Nearer Barnstaple these Pilton Beds begin to trough small patches of a barren softer slate which is only seen well developed south of Pilton and occupying the lower ground east and west of Barnstaple." In these soft slates the prevalent fossils are Phillipsia seminifera, Spirifer bisulcatus, 8. laminosus, S. cuspidatus, Productus Martini, Orthis Michelini, and other exclusively Carboniferous species. Shales with a similar set of fossils occur at Fremington west of Barnstaple, and are succeeded by the Coddon Hill Beds consisting of shales and chert-beds ; the shales yield the Goiiiatites characteristic of the lowest Pendleside Beds, i.e. Prolecanites com- pressus, P. mixolobus, and Nomismoceras spirorbis, and the cherts contain Radiolaria. They are overlain by black shales con- taining Posidonomya Becheri, and black limestones with the same fossil. 18 The combined thickness of all these beds is not very great, apparently not more than 400 to 500 feet, unless a portion of the Pilton Beds is included, and it is impossible to regard them as representing the whole Avonian succession. From the fossils above quoted it looks as if the base and the upper portion of the normal sequence were present, and it is possible that the central beds are cut out by faulting, for the boundary of the Pilton Beds at and east of South Molton is believed to be a fault. In South Devon and Cornwall nothing comparable with the Pilton or with the Fremington Beds has yet been found. The lowest Carboniferous Beds are either unconformable to or faulted against the highest Devonian rocks ; but the cherts and limestones of the Coddon Hill Group are well developed and appear to be rather thicker than in North Devon. West of Dartmoor they are associated with lava- flows, and the general succession according to Mr. Reid is as follows : 19 Feet. Upper lava of Brent Tor, etc. .... about 150 Beds of Radiolarian chert . . . ,, 70 Black shales and lenticular limestones . . ,, 250 Lower lavas .......,, 50 Hard black shale and chert-beds . . . ,, 150 I have elsewhere 20 suggested that the break which here seems to exist between the Devonian and Carboniferous Systems may be a result of the volcanic action indicated by the presence of the lavas and by the intrusive mass of Brent Tor, which is supposed to have been a volcanic vent. The sea-floor may have been slowly raised THE CARBONIFEROUS SYSTEM 277 QQ Q O CO CO CO . CO CO ,3 OQ 03 in CD S O CO CD 1 CO CD 1 03 a _a fl^s i - ^ ^3(3 \ T3 03 CO CO CO rf> CO 03 03 CD 0) CD ; a H d g Pj CO 3 5 CO o 3 o _o 8 CO CO CO CO c6 CD 03 03 CD ^ a j ~ Limestones of Paire and Yvoir with Spirifer \_ Konincki, Caninia patula, and Caninia cylindrica j = Limestones and shales ("Gale- schistes") withS Zaphrentis and Spirifer clathratus ( = S. tornacensis) I = Shales with Spiriferina peracuta and Zaphrentis j Vaughani J Gale-grits and shales with JSumetria, Camarotcechia, \ _ and Ostracods ; the Upper Famennian of Belgian > = authors = Si West of the Meuse in the Dinant area C l and C 2 take on a " knoll " character and are termed Waulsortian ; the rocks are massive mottled limestones with abundant Fenestellids. In these beds characteristic fossils are : Spirifer princeps, S. pinguis, Productus plicatilis, and Amplexus coralloides. East of the Meuse in the Dinant area there is no Waulsortian phase ; the black marble of Dinant ("Marbre noir") belongs, probably, to C 2 . In the Namur basin a large part of the. series (Z to C 2 ) is dolomiitised ("Grande Dolomie "). At Vise only Dibunophyllum Beds occur, and these are of the upper knoll type. This calcareous facies of the Dinantian can be traced eastward by Liege to Limbourg (Aix la Chapelle), but when it reaches the Rhine the limestones are partly replaced by shales, and what we know as the Pendleside fauna extends downward through a greater thickness of beds. The succession near Dusseldorf is : Shales with Posidonomya Becheri . Limestones with a Visean fauna .... Shales with Spirifer clathratus ( = S. tornacensis) . Feet. 1000 300 1500 Brittany and Basse Loire. The Avonian Series of Brittany commences with a volcanic group which is believed to have a thick- ness of more than 1500 feet in the west, consisting of conglomerates and felspathic tuffs with thick masses of andesite ; but these beds only occur on the northern side of the basins of Chateaulin and Laval. The conglomerates rest unconformably on older rocks from Devonian to Cambrian, and as they contain no marine fossils their zonal age is unknown. Above them in the Chateaulin basin is a great thickness of unfossiliferous black shales with layers of felspathic sandstone, 280 STRATIGRAPHICAL GEOLOGY probably 3000 feet thick, and a few fossils (Phillipsia and Productus} nave been found in lenticles of limestone near the top. In the basin of Laval there is a more massive black limestone containing Productus giganteus and other Visean species, overlain by shales, coal-seams, and sandstones with 'plant remains (Bornia transitionis, etc.), and in the centre of the basin is a band of red and green limestone. Only the plant -bearing beds pass into the basin of Ancenis and the Basse Loire district, but these are more than 3000 feet thick, consisting of shales and sandstones with many beds of anthracitic coal. The plants are those mentioned on p. 245. From the above account it will be seen that the Tournaisian does not appear to be represented, except perhaps by conglomerates and lavas. The greater part of the series belongs to the Visean, and the place of the Yoredale Beds is taken by a group of lagunic and estuarine deposits. 2. Germany East of the Rhine in Germany a very different facies prevails, which has long been known as that of the Culm. This consists of a thick series of shales and sandstones ; limestones only occurring as lenticular beds or as layers of calcareous nodules. The series is generally divided into a lower and an upper group, the lower being essentially marine, while the higher seldom contains anything but plant remains. In Westphalia, Hesse, and Nassau the Lower Culm has at the base chert-beds and siliceous limestones about 200 feet thick, containing Prolecanites compressus, Pronorites cyclolobus^ and Orthoceras striolatum ; the overlying beds are dark thin-bedded limestones containing Glyphioceras sphcericum and G. crenistria, with Productus giganteus and Ghonetes papillionaceus. Above this are shales abounding in Posidonomya Becheri, and the total thick- ness is about 1400 feet. The Upper Culm consists of coarser sand- stones, often pebbly, and contains the characteristic plants Knorria imbricata, Bornia transitionis, Lepidodendron veltheimianum. This Culm facies is found also in the Harz Mountains, Tlmringia, the Fichtelgebirge, and over a large area in Saxony. It likewise occupies parts of Silesia and Moravia, where it is of great thickness and has a somewhat different facies. According to Stur, the Silesian succession is as follows : Shales with plant remains. Sandstones and shales with Posidonomya Becheri, Glyphioceras sphoeri- cum, etc. , and other beds with plants. Sandstones, shales, and conglomerates with both marine and terrestrial fossils. THE CARBONIFEROUS SYSTEM 281 Carboniferous deposits of the Culm type are also found in the south-west of Germany, in the Black Forest and in the Vosges district, with which the tracts in Central France were probably continuous. 3. South of France, Pyrenees, and Spain When we pass into Southern France we enter a region where still another facies prevails. As a type of this the succession found in the Montagne Noire, near Corbieres, may be taken. As described by M. Bergeron this is : Grey limestone with Productus giganteus. Sandstones with plants (Lepidodendron veltheimianum). Shales with Spirifer tornacensis and Posidonomya Becheri. Radiolarian chert-beds with phosphatic nodules (Glyphioceras diadema), succeeding Upper Devonian. A similar succession is found at intervals all along the chain of the Pyrenees, with the addition at the base of compact limestones (the " marbres griottes "), variegated with red and brown, which contain Glyphioceras crenistria, Pronorites cyclolobus, and Prolecanites Henslowi. Still farther west in Asturia Professor Barrois has described the following sequence : 22 Feet. Shales and limestones of Lena, with Amplexus coralloides, Lons- daleia floriformis, and Fusulina ...... 240 The Canon limestones with few fossils, Poteriocrinus crassus, Prod, aculeatus, Spirifer striatus ...... 700 ''Marbres griottes" with the Goniatites above mentioned . . 60 It is curious to find limestones with what we regard as Pendleside species of Goniatitidce here conformably succeeding the Upper Devonian ; but there is 110 doubt about the sequence, and we may regard these forms as southern species which did not find conditions suitable to them in the north-west of Europe until a late date in Dinantian time. 4. Russia In this region the Lower Carboniferous Series is a combination of the Culm and the Dinantian limestone facies. The beds lie for the most part horizontally and are exposed in three large areas, that of Donetz in the south, the Moscow basin in the centre, and on the slopes of the Ural Mountains to the east. Ignoring local variations the general sequence in the Ural and Moscow areas is : 282 STKATIGRAPHICAL GEOLOGY 3. Limestones with Prod, giganteus. 2. Sandstones, shales, and coal-seams ; some of the beds yield Prod. giganteus. 1. Sandstones and shales with lenticular limestones containing Prod. mesolobus. These beds are from 5000 to 6000 feet thick and appear to represent both the Tournaisian and Visean stages. In the Donetz area, according to Tschernyschew, nearly the whole series consists of limestones. D. CONDITIONS OF DEPOSITION The geographical conditions which prevailed in the British area during Avonian time are fully discussed in my Building of the British Isles (3rd ed., 1911), but it will be useful here to extend our view to those parts of the European region which have been briefly mentioned in the preceding pages. The complete absence of any trace of Carboniferous strata in Norway, Sweden, Lapland, and Finland seems to indicate the existence of land over a large part of Northern Europe during this period. This inference is confirmed by the character of the deposits found in Northumberland, Scotland, and the north-east of Ireland ; these being mainly detrital deposits carried down by rivers and such as would be accumulated in a large bay or gulf. Hence we conclude that the Scandinavian land extended westward across the North Sea and across the north of Scotland into the North Atlantic region. Again on the eastern side we find a broad band of Carboniferous strata in Russia extending from the Valdai Hills, south-east of St. Petersburg, northwards by the south end of Lake Onega. In this band the lowest beds are terrigenous deposits similar to those of Scotland, and here again, therefore, we are on the borders of continental land. With regard to the southern border of the Scandinavian land there is less certainty, because all evidences of it are buried and concealed beneath the Neozoic deposits of the Germanic region. Probably, however, the coast-line passed across the southern part of the North Sea and across Denmark into the Baltic basin, whence it must have curved north-eastward through the Baltic provinces of Russia. Here it may be pointed out that the absence of Avonian deposits in Poland is no proof that this country was land at this time. The Carboniferous deposits may have been largely removed from that area before Permian time, and seeing the great thickness which they have in Silesia and as far east as Cracow, it would be THE CARBONIFEROUS SYSTEM 283 surprising if they had not extended for some distance to the north of these places. Passing now to the Atlantic region there is reason to think that a large part of this was also occupied by land, and that a part of its coast-line passed through the extreme north-west of Ireland. Again, the'great thickness of shale in the south-west of Ireland and the north-west of France indicates the neighbourhood of land and the debouchures of large rivers which carried much sediment derived from such land. We may therefore picture a continent which almost encircled the British area, lying not only to the north but also to the west and south-west of it; and indeed it seems probable that a promontory of this land actually occupied parts of Cornwall, Brittany, and the intervening channel area during the Tournaisian epoch. Another tract of land which was probably connected with the Atlantic continent seems to have stretched across Central France, for in the basin of the Loire the Dinantian ( = Avonian) deposits are entirely of terrigenous origin, and are probably all of Visean and Namurian age, while to the southward in La Vendee even these are absent, and small tracts of Westphalian Coal-measures rest directly on Archaean rocks. We have seen also that when Dinantian Beds come in again to the southward in Asturia, the Pyrenees, and the Montagne Noire they are of quite a different facies, hence the land above indicated seems to have separated a southern from a northern Carboniferous Sea. From Central France this land may have extended across the valley of the Ehone between Lyons and the Yaucluse, and thence through the Western Alps and Northern Italy, where no traces of Dinantian strata have yet been found. Land also existed at this time over a certain area to the north of Metz and Sarrebruck, for near the latter place Coal-measures rest directly on the Lower Devonian, but this may have been only a large island. The limestone facies of the Avonian Series must have been formed in those parts of the Carboniferous Sea which were not subject to the incursion of currents setting off the surrounding land. It must originally have extended continuously from the west of Ireland, all round an island which existed in Visean time over St. George's Channel, Central Wales, and Central England an island which I have elsewhere called " St. George's Land." Thence these limestones probably extended broadly into the north-east of France and through Belgium into Prussia, nearly as far as the Ehine. There is no valid reason for imagining any special contraction of the sea space on the Franco-Belgian border, as suggested long ago by Professor Gosselet, and still supported by some French geologists. 284 STRATIGEAPHICAL GEOLOGY The trough-like syncline in which the limestones lie is a feature produced by subsequent flexuring ; there is no indication of a shore-line either along the northern or the southern line of outcrop, and we have only to prolong the dip of these limestones to see that they must have extended indefinitely both to the northward and to the southward. Who will deny that there may be another syncline under Holland, and that Carboniferous limestone may occur in it ? From the distribution of the limestones, as above indicated, we may infer that no large rivers entered that portion of the Avonian sea which is now Central Ireland, and also that the island tract of St. George's Land had then a dry climate and was without streams of any size. At the same time there is no ground for supposing that the limestones formed around this island were of the nature of coral-reefs. Crinoids and molluscan shells are the predominant constituents of these limestones as they are of the Devonian limestones (see p. 206), and the corals are distributed through the limestones over wide areas, not concentrated in reef- like masses round the shores of the island. KEFERENCES 1 A. Vaughan, Quart. Journ. Geol. Soc. vol. Ixi. p. 181. 2 "Geology of the S. Wales Coalfield," Mem. Geol. Survey, in ten separate parts. 3 Hind and Stobbs, Geol. Mag., 1906, pp. 385, 445, 496. 4 J. W. Sibly in Quart. Journ. Geol. Soc. vol. Ixiv. p. 34. 6 E. J. Garwood, Geol. Mag. for 1907, p. 70. 6 W. Hind, Quart. Journ. Geol. Soc. vol. Ivii. p. 347. 7 C. A. Johns in The Naturalist for 1911, p. 9. 8 E. J. Garwood, Proc. Geol. Soc., May 18, 1911. 9 E. J. Garwood in "Geology in the Field," Geol. Assoc. (1910), p. 661. 10 Goodchild, Proc. Geol. Soc. vol. xv. p. 117 (1897). 11 "Geology of the Conntry around Edinburgh," Mem. Geol. Surv. 2nd ed., 1910. 12 K. J. Cruise, Mem. Geol. Surv., Ireland, Expl. of Sheets 66 and 67. 13 J. A. Douglas, Quart. Journ. Geol. Soc. vol. Ixv. p. 538 (1909). 14 Jukes, Manual of Geology, 2nd ed. p. 512 (1862). 15 Messrs. Matley and Vaughan, Quart. Journ. Geol. Soc. vol. Ixiv. p. 413. 16 Jukes, op. cit. 2nd ed. p. 510, 3rd ed. p. 588. 17 Salter, Quart. Journ. Geol. Soc. vol. xix. p. 476 (1863). 18 W. Hind, Geol. Mag., 1904, pp. 392, 584. 19 C. Reid in Sum. Prog. Geol. Surv. for 1908, p. 26. 20 Jukes-Browne, Geol. Mag. 1905, p. 353. 21 A. Vaughan, Brit. Assoc. Rep. Sect. G. (1910). 22 Barrois, " Terrains anciens des Asturies," Mem. Soc. Geol. Nord (1882), p. 519. CHAPTER X THE CARBONIFEROUS SYSTEM (continued) II. WESTPHALIAN SERIES A. SUBDIVISIONS AND GENERAL RELATIONS As explained on p. 245 the second of the three great series of strata into which the Carboniferous System is now divided has received the name of WestpJialian, and to it belong all the beds which have been classed as " Upper Carboniferous " in the British Isles. These generally occur in conformable succession to the Avonian Series, but they often extend beyond the limits of the older series, and come to rest on Silurian or other more ancient rocks. Moreover, even when they overlie the Avonian there are sometimes evidences of a slight break between the two series, and there is always a change in the species of plants above this horizon. In the British Isles the "Coal-Measure" Series is generally underlain by a set of massive sandstones which are known collec- tively as the Millstone grit, but this is a very variable group, being sometimes over 3000 feet thick, and sometimes only represented by 300 or 400 feet of sandstone and shale. In the areas where the Coal-measures are most fully developed they reach a thickness of 6000 to 7000 feet, and have been divided into Lower, Middle, and Upper stages. On this basis, therefore, the Westphalian Series is divisible into four stages, and has a maximum thickness of about 10,000 feet. A more scientific division of the series is one based on the different assemblages of plants that are found in this great succession of strata, and by a study of these Dr. Kidston has recognised three different assemblages, 1 though the two higher blend into one another through a set of " Transition Beds." The particular differences in these plant-assemblances will be mentioned 285 286 STRATIGRAPHICAL GEOLOGY on a later page.- The correlation of the two sets of stages is given below. Upper Coal-measures with Fourth flora. Transition Coal-measures Third flora. Middle Coal-measures ,, Second flora. Lower Coal-measures ~\ . , ,, Millstone grits ) First flora B. LIFE OF THE PERIOD 1. The Fossil Flora a. Equisetales. Calamites were gigantic Equiseta or horse- tails, and consisted of a large hollow pith cavity encased iri a thin cylinder of woody tissue. The central part has generally been filled up with mineral matter, forming the casts which are such common Carboniferous fossils. The outer ligneous envelope, being carbonised and friable, is seldom preserved in a perfect state. The name Calamites is at present restricted to these jointed stems, while other names, Asterophyllites and Annularia, are given to different varieties of foliage which belonged to Calamites. The fruit is a scaly spike (Calamostachys and Palceostachya), in which sporangia are arranged like those of Equisetum. Sphenophyllum is the foliage of a plant related to Calamites, but is now known to differ in several respects, and the fruit is a whorled sporophyll resembling that of the recent genus Psilotum. Calamodendra are casts of the central cavity or pith of plants resembling Calamites, but having a thick woody outer coating or bark. Calamodendron thus appears to have been a highly developed Equisetum. 6. Lycopodiales. Lepidodendra were gigantic members of the Lycopodiales, or Club-moss family, having trunks many feet in girth and 40 to 60 feet in height. The trunks branch toward the top into several limbs, which divide again into smaller boughs. The foliage consists of simple linear leaves, which are sometimes found attached to the terminal branches (Fig. 96), and the angular scars on the stem are the marks left by the detachment of these leaves. Ulodendron, Lepidophloios, and Halonia are allied genera. The scaly cylindrical bodies which are known as Lepidostrobi are the cones which contain the sporangia or seed-cases of the Lepidodendra, and were borne at the extremities of some of the branches. Sigillaria. These appear to have been the commonest trees in the swampy portions of the Carboniferous forests, and have evidently supplied the chief mass of the material which forms our Fig. 95. GROUP OF COAL-MEASURE PLANTS. 1. Sphenopteris linearis. 3. Odontopteris Schlotheimi. 2. Sphenopteris affinis. 4. Neuropteris acuminata. 5. Calamites (restored and th of natural size). 288 STRATIGRAPHICAL GEOLOGY coal-seams, while their roots (Stigmarice] ramify in the under-clays beneath. They were tall trees, branching at the top, and rivalling the Lepidodendra in height, with a bare trunk fluted longitudinally, and regularly ornamented by the pits or scars left at the points where the leaf-stalks were attached. These leaves were long and narrow, and similar to those of Lepidodendron. The fruit consists of small sporangia borne at the basis of slightly modified leaves ; Sigillaria was therefore a Lycopod, and allied to Lepidodendron. Fig. 06. GROUP OF COAL-MEASURE PLANTS. a. Calamites cannseformis (root end). c. Sigillaria reniformis. ft. Alethopteris lonchitica. d. Lepidodendron Sternbergi. e. Stigmaria ficoides. Stigmarice. The roots of Sigillaria, Lepidodendron, and Bothro- dendron are so much alike that it is impossible to distinguish them. Consequently they are all known as Stigmaria, long, branch- ing, and tapering, with numerous rootlets spreading out in all directions ; such roots have been found which were upwards of 20 feet in length. Perhaps, like those of the modern mangrove- tree, these roots were partly aerial, rising well above the low-water THE CARBONIFEROUS SYSTEM 289 mark of the swamp, and uniting in a dome-shaped manner to support the trunk. c. Pteridosperms and Primofilices. Fern-like fronds are abundant in the Coal-measures, and the roofs of coal-seams are sometimes covered with beautiful impressions of such fronds. Until recently these were regarded as true ferns, but it has been Fig. 97. LEPIDODENDRON ELEGANS (BROUOH). 1. A terminal branch. 2. Leaf-scars from the stem. found that many of them belong to plants which developed true seeds and not spores like those of existing ferns. They are now regarded as a distinct group of plants, intermediate between ferns and Gymnosperms, and are known as Pteridosperms. True ferns of an ancient type (Primofilices') are also represented among these leaves, for some have been found bearing sporangia, and stems with leaf-scars like those of modern tree-ferns also occur (Psaronius). Fructiferous fronds, however, are rare, and it is U 290 STRATIGRAPHICAL GEOLOGY impossible to distinguish the sterile fronds of these ferns from those of the Pteridosperms ; thus some of the fronds known as Pecopteris and Sphenopteris belong to one group and some to the other, so that these names only stand for certain forms of sterile fronds which are similar in their venation and general characters. In the present state of our knowledge, these names have still to be used generically, and some of the commonest frond-forms thus designated are Alethopteris (Fig. 96, ft), Neuropteris (Fig. 95, 4), Odontopteris (Fig. 95, 3), Pecopteris, Cardiopteris, Linopteris, Mari- opteris, and Sphenopteris (Fig. 95, i and 2). The seeds of the Pteridosperms have long been known under the names of Lagenostoma, Trigonocarpus, Rhabdocarpus, etc. Oordaitales. The stems of plants in some respects resembling the true Coniferse (Pitys and Dadoxylori) have also been met with, especially in the volcanic tuffs of the Scotch Carboniferous Series. The lower slopes of the old volcanoes were doubtless clothed with these trees, their habitat being in such high and dry localities, and not in the swampy ground where the other plants flourished. In the structure of their stems these plants show affinities to the Pteridosperms as well as to the Coniferae. Gordaites was a remarkable plant placed by most botanists in a distinct phylum, but by some with the conifers. The stem grew to a height of 20 or 30 feet, and carried a number of long sword- shaped leaves with parallel venation, something like those of a yucca. The leaves were attached by broad bases and left large leaf-scars on the stem when they fell off. The tree bore bunches or spikes of fruit (Antholites) with seeds somewhat resembling the fruits of a yew, but more heart-shaped, and these are known as Cardiocarpus. Some of the plants above mentioned are more abundant in the lower part of the series and some in the higher parts. A study of their distribution has enabled Dr. Kidston l to recognise four successive assemblages of plants in British Coal-measures, and Zeiller has found the equivalents of all but the uppermost in those of the continent. The principal features of these assemblages are given below. Plants of the Millstone Grit and Lower Coal-Measures Lepidodendra are abundant ; Sigillaria not very common. Galamites are represented by G. Suckovi and G. ramosus. The prevalent fern forms are Alethopteris and Sphenopteris ; common species of the first being A. lonchitica and A. decurrens, of the second 8. obtusiloba. THE CARBONIFEROUS SYSTEM 291 Plants of the Middle Stage In this stage Sigillaria is more abundant than Lepidodendron, S. elongata, S. polyploca, and S. Brardii being common species. Calamites and Sphenophyllum are also very abundant, though no species is restricted to this division. Fern-like fronds abound, Sphenopteris and Neuropteris reaching their maximum development. Of the former S. (Zeillaria) delicatula, S. (Gorynepteres) coralloides are characteristic species. Plants of the Transition Stage These are a mixture of typical middle stage forms, such as Pecopteris Miltoni, Annularia radiata, Sphenophyllum cuneifolium, and Alethopteris lonchitica, with species characteristic of the Upper stage, such as Alethopteris Serli, Neuropteris rarinervis, and Spheno- phyllum emarginatum. Plants of the Upper Stage In these beds Calamites become rare, Lepidodendron is not common, and Sigillaria also less abundant. The most abundant plants are fronds of the Pecopteris form, such as P. arborescens, P. oreopteridea, P. polymorpha, and P. cyathea, but some forms of Neuropteris (N. rarinervis, N. macrophylla, and N. ovata are common. Alethopteris Serli and Cordaites angulosostriatus are also typical Upper Coal-measure species. 2. The Land and Freshwater Fauna a. Mollusca. In the Coal-measure shales, and associated with many of the plants above mentioned, shells somewhat resembling the recent Unio are frequently found. They have been referred chiefly to three genera, Carbonicola ( = Anthracosia), Anthracomya, and Naiadites, and their affinities have recently been studied by Dr. W. Hind, who refers the first two genera to the Unionida and the third to the Mytilidce, but believes that all of them were fresh- water forms. He remarks that their occasional or rare association with marine shells is no proof that they were capable of living in salt water, for their shells may have been carried down by rivers into bays and estuaries. 6. Arthropoda. True Insects, Myriapods and Arachnids, have also been found in the Coal-measures, including crickets, beetles, and a scorpion. Crustacea also inhabited the waters, the small Ostracoda Bairdia, Leperditia, and Beyrichia, and the Phyllopod 292 STRATIGEAPHICAL GEOLOGY Estheria swarming in some localities. Several species of Xipho- sura allied to the modern Limulus (king-crab) have been described by Dr. Woodward, and referred to the genera PrestwicJiia and Belinurus. There were also a few Eurypterids, this being their last appearance in England. c. Fishes. Fish remains, of Crossopterygian and Elasmobranch genera, are everywhere abundant, but it is difficult to separate entirely the freshwater forms from the marine, because some (like the Mollusca) seem to have been equally at home in salt and brackish water ; some of them, like the modern salmon, may have lived in the sea at certain seasons, and in the estuaries and rivers during other parts of the year. The following genera are, however, most frequently associated with the Coal-measure plants : Ccelacanthus, Cheirodus, Ctenacanthus, Ctenoptychius, Gyracanthus, Megalichthys, Platysomus, Pleuracanthus, Rhizodus, Rhizodopsis, and Strepsodus. d. Amphibia. Thirteen genera of Labyrinthodonts are recorded from British Carboniferous rocks, Anthracosaurus, Loxomma, Ophiderpeton, Urocordylus, etc., mostly from the Jarrow coalfield in Leinster, Ireland. Others (Archegosaurus) occur in Prussia and (Baphetes, Dendrerpetori) in Nova Scotia. Footprints have also been observed in the Forest of Dean coalfield and in the limestone series of Dalkeith, Scotland. The larger of these Amphibians are believed to have measured 7 or 8 feet in length, but others were very small ; they were probably fluviatile creatures preying on the fish and Crustacea which swarmed in the swamps and lagoons of the period. 3. The Marine Fauna The Millstone grits are probably marine deposits, for land plants are not common in them, and the intercalated shales some- times contain marine species of Mollusca as well as fish remains. The Lower Coal-measures were swamp deposits formed along the fringe of sinking land areas, parts of which were from time to time invaded by the sea. Hence marine fossils are found in them at certain horizons. The species which have been found in these beds in Britain and also at two horizons in the Middle Coal- measures of Staffordshire are tabulated on p. 293. The first two columns show how many of the species range up from the higher parts of the Avonian Series. C. WESTPHALIAN SERIES IN BRITAIN Just as the Lower Carboniferous Series is in most places brought to the surface along tracts of anticlinal flexure, so the Upper Series THE CARBONIFEROUS SYSTEM 293 i fc a If s PMC5 S*5 !& N 3 d . is Goniatites (Gastrioceras) carbonarius, v. Buck. X x X ,, Listen, Martin X X X (Glyphioceras) bilinguis, Salter . X X X ,, diadema, Beyr. X X ,, reticulatus, Phil. X X x , , spiralis, Phil. . X Ccelonautilus falcatus, Sow. .... X X Pleuronautilus quadratus, Flem. . X X ,, costatus .... X Temnocheilus carbonarius, Foord X ,, concavus, Sow. X Ephippiocerasjclitellarium, Sou'. . Orthoceras Steinhaueri, Phil. X X X X X X X Euomphalus tuberculatus, Flem. . X Pleurotomaria monilifera, Phil. . X X X Macrocheilus michotianus, de Kon. X Bellerophon hiulcus, Sow. .... X X X X ,, Urei, Flem. .... X X X X Nucula sequalis, Sow. ..... X X X x ,, gibbosa, Flem. ..... X X X X X Leda stilla, M 'Coy X X X ,, acuta, Sow. ...... x Otenodonta Isevirostris, Portl. X X X X Myalina Fleming!, M'Coy .... X ,, peralata, de Kon. .... X X X ,, Verneuilli, M'Coy .... X X Schizodus antiquus, Hind .... X x ,, carbonarius, Sow. X axiniformis, Sow. X X X Sedgwickia attenuate, M'Coy X Sanguinolites ovalis, Hind .... X Edmondia accipiens, Sow. .... x , , senilis, Phil. .... X X X Posidoniella laevis, Brown .... X X X ,, minor, Brown .... X X X ,, semisulcata, Hind . X Pseudamusium fibrillosus, Salter . . X X Aviculopecten gentilis, Hind X x Leiopteria longirostris, Hind X X Pterinopecten papyraceus, Goldf. . X X x carbonarius, Hind . X X X Productus semireticulatus, Martin X X X X X ,, scabriculus, Martin X X X x X Chonetes laguessiana, de Kon. X X X X X Spirifer bisulcatus, Sow. .... X X x x ,, Urei, Flem. ..... X X X X Discina nitida, Phil. ..... X X x X x x x x x x ,, squamiformis, Phil. X X X 294 STBATIGRAPHICAL GEOLOGY is generally found in synclinal troughs or basins, the highest portion of the series usually lying in the central parts of such basins. All areas where workable coals exist are called coalfields. Of such coalfields there are no fewer than eighteen in England and Wales, some large and some very small, and if the unpro- ductive Devonshire basin is included there are nineteen. In Scotland there are three coalfields in which Westphalian measures are worked, besides three others in the underlying Avonian Series. In Ireland there are five such basins. In describing the Coal-measures found in these numerous basins it will be convenient to group them into natural areas wherein a similar succession of strata is found, and as their most complete development is exhibited in the basins which lie on each side of the Pennine Anticline (see Fig. 86), these will be placed first. The areas may therefore be taken in the following order : 1. The South Pennine and Midland area. 2. The Newcastle Coalfield. 3. Scotland. 4. Ireland. 5. Bristol, Somerset, and South Wales Coalfields. 6. Devonshire. 1. The South Pennine and Midland Area This area will include the Nottingham and Yorkshire coalfield on the east and those of Lancashire and North Staffordshire on the west, the latter probably being connected under the Cheshire basin with the Flint and Denbigh coalfield (see Fig. 98, and the map, Fig. 87). We shall also include the other Midland coalfields to the southward, those of Shropshire, South Stafford, Leicestershire, and Warwickshire, since they are all portions of a series of deposits which originally formed one continuous sheet. In the basins which flank each side of the Pennine range the complete succession is as follows : Feet. TT mvjr /Red and purple sandstones and marls Upper UM. | W ith thin limestones .... about 700 T -4-- n AT /Grey sandstones and shales, red and 1 UM '\ purple marls ; some thin coal-seams . ,, 1500 TVT-^MI n TV/T fG re y shales and sandstones with many ' U M> \ seams of coal and ironstone . . . , , 2500 T C M /^ re y flagstones and black shales with irUM< t some coal-seams (generally thin) . . 1000 to 2500 n/rn + n +. /'Sandstones and shales, thickening toward Millstone Gritj the north . east / . . 400 to 3500 9500 THE CARBONIFEROUS SYSTEM 295 The Millstone Grit consists of several sets of grits and flag- stones separated by shales of varying thickness. Its thickness increases to the northward, and finds its maximum in East Lancashire and the adjacent part of Yorkshire, where it is said to be over 3500 feet thick. Throughout the northern part of the area the following subdivisions can be traced, though the thickness of each varies greatly : First grit or rough rock. Shales. Second grit. Shales. Third or Chatsworth grit. Shales. Kinderscout grit, or fourth grit. The lowest grit takes its name from the high tableland of Kinderscout in the Peak country, which consists of this rock and rises to a height of 2000 feet above the sea. The third grit forms long conspicuous escarpments on both sides of the central area. The second grit is less persistent than the others, or rather the shales above thin out so that it coalesces with the first grit. The Rough Rock is a very coarse grit, sometimes passing into a con- glomerate of small quartz pebbles. In Lancashire it contains a workable coal-seam called the " Feather-edge coal." The shales below also include thin seams of coal. To the south-west in Staffordshire the middle grits thin out, but the fourth grit appears to be represented on Congleton Edge by a set of grits and shales about 500 feet below the third grit, so that the group here is not less than 1000 feet thick, and has the Pendleside shales below it. Still farther south the grits thin out entirely, and they do not appear in the South Staffordshire or Warwickshire coalfields. In Nottingham also they thin southward and have nearly disappeared at Ashby de la Zouch, in Leicestershire, where only 50 feet of such grit is left at the base of the series. A similar small tongue of grit reaches to the Wellington district of Shropshire and then thins out, Lower Coal-Measures. These are sometimes called the Gannister measures, gannister being the miners' name for a peculiar hard and compact siliceous underclay (or fireclay) which often forms the floors of the coal-seams. Hard grey flagstones are the pre- dominating beds in this division, with subordinate bands of shale. The coals are generally thin, but two or three from 3 to 4 feet thick are often worked. In the vicinity of the coal-seams plant remains are abundant, but the fossils in the other beds are generally marine forms (see p. 293). 296 STBATIGRAPHICAL GEOLOGY I. a I Is PQO In North Staffordshire these mea- sures are 2400 feet thick if the top of them is taken at the marine band above the Mansfield coal, but Dr. Kidston is inclined to place it at the Ash coal, still higher up. They are probably quite as thick in Lanca- shire if the same horizon is taken as the line of division. In Flint they are considered to be about 1000 feet thick and 1800 feet at Wigan, in Lancashire, where they are well exposed in the cuttings of the Wigan and Liverpool railway. Carbonicola acuta, C. robusta, and 0. turgida are common in these measures. On the eastern side of the area there is a similar development of Lower Coal-measures, but the thick- nesses assignable to them depend upon the horizon which is taken to be their upper limit. In Nottingham and Leicester they are said to be about 1000 feet thick, but when the series recurs in Warwickshire they are found to be absent, and must there- fore have thinned out against the ridge of ancient rocks which underlie the Charnwood and Nuneaton dis- trict. Neither are they found in the South Staffordshire coalfield. Middle Coal - measures. This group has a wider extension than those below. It is also a more varied series, consisting of grey shales and grey or white sandstones with many seams of coal and of ironstone (carbonate of iron), this association of coals and ironstones in the same series of beds making it commercially the most important group in the whole series. It is believed to attain its greatest thickness in Lancashire, where if THE CARBONIFEROUS SYSTEM 297 measured from the Worsley 4 -foot coal to the flags below the Arley mine coal, the thickness is about 3500 feet. In York- shire and Nottingham these measures are from 2000 to 3000 feet thick, and are so entirely like those of Lancashire that several of the Yorkshire coal-seams can be identified with their counter- parts on the west ; thus the " Silkstone " of Yorkshire is the " Arley mine " of Lancashire. Layers of nodular carbonate of iron are fre- quent and have been largely worked. Some of the Coal-measure sandstones near Pontefract, Ackworth, and Kotherham in Yorkshire are stained red and were formerly mistaken for beds of Permian age. In Leicestershire these measures are about 1500 feet thick with ten good coal-seams, but in Warwickshire they have shrunk to about 700 feet with only five workable coals. In North Staffordshire they are probably about 1600 feet if the group is restricted within approximately the same horizons, and in South Staffordshire they are about 1000 feet thick at the northern end of the coalfield, lessening to 500 feet at the southern end, but this may not be entirely due to thinning. They contain many good coal-seams, and it is a curious fact that some fourteen seams which are distinct in the north coalesce southward by the thinning out of the intervening measures till they form one bed of coal, from 25 to 30 feet thick, which is then known as the " Thick Coal." In Staffordshire three species of Anthracomya are Common in these beds A. Wardi, A. pulchra, and Adamsi. Transition and Upper Coal-measures. These stages have been most completely studied in North Staffordshire and Flint, where Dr. W. Gibson found reason to include certain red and purple sandstones which had previously been referred to the Permian. For these he proposed the name " Keele Series," but they should be known as the Keele Beds since they are only part of a " series." The following is an abstract of the succession established by Mr. Gibson : 2 Feet. Upper C.M. ( Red and purple sandstones and marls with three thin or -j layers of Spirorbis limestone. Fossils, a few Keele Beds. [ plants and Entomostraca ..... 700 ^Newcastle Group. Grey sandstones and shales with four thin seams of coal and a limestone at the base. Fossils, many ferns and plants (Sigillaria JSrardi). 300 T ... Etruria marls. Red and purple marls and clays with thin bands of green grit and a limestone The CS are lar used ^ mesures, j I Blackband Group. Grey sandstones and marls with blackband ironstones and thin seams of coal. \ Anthracomya PMllipsi and fish teeth . . 300 to 450 Total about 2200 298 STRATIGRAPHICAL GEOLOGY There is a similar set of beds near Manchester, but they have not yet been fully correlated with the above. On the eastern side of the Pennine range these beds are con- cealed beneath the newer rocks, but have recently been proved in a boring at Thurgarton, near Southwell, in Nottinghamshire. The cores indicate a succession like that of North Staffordshire, but with a less thickness. Mr. W. Gibson (op. cit. p. 262) gives it as follows : Feet. Permian (Marl slate). v i -D A f Red sandstone and marl with Neuropteris rarinervis and is \ Pecopteris Miltoni . . . . . . .188 Beds \ Grey sandstones and shales with a seam of coal . 91 Etruria marls : red marls and green grits . . . . . 257 Grey Coal-measures below. N.N.W. Halesfield S.S.E. Hills Lane Fig. 99. SECTION ACROSS THE MADELEY DISTRICT, C'OALBROOKDALE (W. J. Clarke). a. Middle Coal-measures. b. Upper Coal-measures. In Yorkshire near Rotherham there are some red marls and sandstones which are classed as Upper Coal-measures, and probably belong to the Etruria marls, but only 54 feet of them are seen. When these Upper Coal-measures recur in the more southern coalfields of Shropshire, South Stafford, and Warwick, they are found to be more or less unconformable to those below. This is markedly the case in the Coalbrookdale district of Shropshire, where the whole Carboniferous succession is as follows : Feet. Upper f Red and purple sandstones and marls (?Keele Beds) 800 and -! Mottled clays and greenish grits with a limestone Trans. C.M. { containing Spirorbis (Freshwater Beds) . . 440 Middle Coal-measures with a band containing marine fossils . . 600 Lower Coal-measures or Millstone grit, thinning out south . to 100 Avonian limestones and shales (Productus giganteus) . . to 180 Up to 2100 The lower beds, including the Middle Coal-measures, lie in a THE CARBONIFEROUS SYSTEM 299 broad syncline (see Fig. 99), while the Upper Measures are nearly horizontal ; the base of the latter being a mottled clay known as " calaminker " overlain by a bed of gravel or conglomerate called the " Kough Kock." As this series is traced southwards the lower members thin out and disappear, while the Middle Measures over- step the Millstone grit so as to rest on the Old Red Sandstone, and this is their position in the Forest of Wyre west of Kidderminster and around Bewdley. In South Staffordshire there appears to be a similar break between the Middle and Upper Measures, for the diminution in the thickness of the Middle Group toward the south is probably in part owing to an overstep on the part of the Upper Measures. According to Professor Lapworth the succession south of Dudley is as follows : Feet. Upper f Red sandstones and marls (Hunnington) . . 1300 and -| Grey sandstones and marls (Halesowen) . . 800 Trans. C.M. (Red marls and greenish sandstones . . . 800 Middle or productive Coal-measures ...... 500 3400 The same succession is found in Warwickshire. From the preceding account of the strati graphical arrangement of the several members of the Westphalian Series in this region it will be seen that they all become thinner southwards, each stage passing beyond the limits of the one below farther and farther on to the Midland area which was land in Avonian time. Moreover, it is apparent that there was some terrestrial disturbance and uplift in this southern part of the region after the formation of the Middle Coal-measures. It may also be inferred that the main mass of the Westphalian Series above described was deposited in a trough which extended from west to east across what is now the Pennine range, and that this trough was continually deepened by subsidence, while in the area to the south of it subsidence was interrupted by stationary epochs and at one time by an uplift with consequent flexure and erosion. Further, the student should note that the existing coalfields on each side of the Pennine ridge still extend far below the newer beds which cover their outer parts. Thus the Lancashire coalfield probably passes westward below Liverpool Bay, and the Yorkshire coalfield is known to extend as far as Haxey in Lincolnshire, where a boring has been carried to a depth of 3185 feet passing through 1457 feet of Coal-measures. It is believed that the subterranean extension of this coalfield reaches as far eastward as Lincoln and Brigg. 300 STRATIGRAPHICAL GEOLOGY 2. Cumberland and Northumberland Area In the north of England there are two coalfields which, though now far apart, were originally parts of one continuous sheet of Coal-measures which spread over the whole of the Lake District and across the broad Pennine Hills. These are the Newcastle coalfield, occupying parts of Durham and Northumberland, and the Whitehaven coalfield in Cumberland. Remnants of the con- necting strata have been preserved in the small Canonbie coalfield on the border of Dumfries, in a faulted tract south of Haltwhistle in Northumberland, another at Argill near Brough in Westmoreland, and in patches of Millstone Grit on the west side of the Eden Valley. The Newcastle coalfield is the largest of these tracts ; it extends from the northern side of the Tees valley west of Darlington to the mouth of the river Coquet, a distance of about 60 miles. It is limited on the south by an anticlinal flexure which runs from Barnard Castle to the mouth of the Tees, and the Millstone grit runs into the sea near Alnmouth, so that the eastern part of the coalfield passes beneath the floor of the North Sea. Millstone Grit. This division is of small thickness, and it is doubtful what beds should be assigned to it. In Yorkshire the Millstone grits thin rapidly as they are followed northward, and in the North Riding are not more than 600 feet thick. Some of the grits, however, are believed to come in again between Teesdale and Tynedale, and have been mapped as forming broad plateaus on the high fells of West Durham, but in this area they lose their coarse gritty character and become indistinguishable from the sandstones and flagstones of the Gannister measures. Lower Coal-measures. These beds likewise become much thinner in a northerly direction, so that in Northumberland the whole thickness between the Felltop limestone and the Brockwell coal, which is the lowest workable seam, is only about 350 feet. Of this Mr. Lebour assigns 150 feet to the Gannister measures, and in this thickness there are two seams of coal. A few marine fossils such as Aviculopecten papyraceus and stems of crinoids have been found in these beds. The Middle Coal-measures maintain their thickness, which in Durham is estimated at 2000 feet, and they contain some twenty good seams of coal, including the Brockwell seam at the base and the High Main seam, which is the highest known seam in the series. Besides these there are many minor or less valuable seams, bringing the total number to over sixty. Some of the shales con- tain nodules and bands of ironstone which have occasionally been THE CARBONIFEROUS SYSTEM 301 worked, and some of these are so full of Carbonicola shells that they are known as " mussel bands." The sandstones are used for building purposes, and one bed, known locally as the " grindstone post," furnishes the celebrated Newcastle grindstones. No true Upper Coal-measures come into this area, though they may exist under the North Sea. The Cumberland or Whitehaven coalfield borders the western coast from Whitehaven to Maryport and then runs north-eastward in a narrow tract which is faulted down between the Permian Beds and the Carboniferous limestones. The whole area, indeed, is so broken by faults that the thickness of the Coal-measures is not known, but believed to be over 1500 feet. Lithologically the series is divisible into (1) Productive Measures and (2) the Whitehaven Sandstone ; but from an examination of the plant remains Mr. Arber groups the beds in the following manner : Feet. Whitehaven / Upper division = Transition Beds . 418 Sandstone \ Lower division) -- i ^i n A/T f200 "roductive / Upper division/ M . ~\ 450 Measures 1 Lower division = Lower C.M. ? 800 Both divisions of the Whitehaven Sandstone consist of purple sandstones, shales, and conglomerates, which weather to reddish tints, and the upper division includes a bed of Spirorbis limestone. The Canonbie basin though small is interesting because it is deep enough to bring in still higher beds, and because the strata resemble the English Series more than the Scotch Coal-measures. The following abstract is taken from the full account given by Messrs. Peach and Home : Feet. TT /Red and purple sandstones and shales with a Spir-\ i^nn u PP er \ orbis limestone and 2 seams of ironstone . / Middle White sandstones, grey shales and coals . . . 500 Lower Grey sandstones, shales and coals .... 750 AT n ., /"Massive sandstones, some coarse, with shales and) KKn fc \ thin coal-seams / 3400 3. Scotland The Westphalian Series is not so fully developed in Scotland as it is in Central England, but it is curiously different from that of Northumberland, for the Lower Coal-measures are of considerable thickness, while the Middle Division, which is the most productive part of the English Series, has a very different facies in Scotland, its beds being often called the ( ' barren " measures because they contain no coal-seams. There are four separate coalfields occurring 302 STRATIGKAPHICAL GEOLOGY in three basin-shaped tracts, namely that of Midlothian and Fife on the east, of Lanark and Stirling in the centre, and of Ayrshire on the west. Millstone Grit. In the counties of Edinburgh, Fife, Stirling, and Lanark the upper limestones are succeeded by a set of pale red and yellow sandstones in thick beds with occasional layers of shale. This group of sandstones is generally known as the Roslin Sandstone from its fine exposure near that place. Its thickness near Edinburgh and north of the Forth is from 500 to 600 feet, but it thins from the north-east to south-west ; in Lanark it is only from 90 to 100 feet, and farther west, in Ayrshire, it is so thin as not to be definitely separable from the Coal-measures. Eecent examination of the plant remains by Dr. Kidston has led to the recognition of a palseontological break in this set of beds, 3 and the fact is confirmed by Dr. Traquair's study of the fish remains ; so that the lower part of the Eoslin sandstone is now grouped as "Lower Carboniferous" (i.e. Avonian), and the higher part as " Upper Carboniferous " (i.e. Westphalian). In the Edinburgh district the line of division is drawn near a marine band about 230 feet above the highest limestone. The lower beds include several bands containing marine fossils, and Sphenopteris is the only fern-plant ; while in the upper part plant remains are more abundant, species of Alethopteris, Neuropteris, and Mariopteris being the most frequent with several species of Sigillaria which range upward but not downward. Mr. Arber informs me that as a matter of fact no true species is common to the two floras found in the Koslin Sandstone. Lower Coal-measures. These measures attain a consider- able thickness, about 1200 feet in Fife and Midlothian, from 1200 to 1500 in Lanark, and about 1300 in the Clyde Basin. They consist of grey and white sandstones, with shales, coals, fireclays, and iron- stones, and ten of the coal-seams are from 2 to 5 feet in thickness. In Lanark and the Clyde basin the base of the series is taken at the slaty-band ironstone, and there are several other continuous beds of blackband ironstone which are of much commercial importance. These measures contain a flora which is pronounced by Dr. Kidston to be the same as that of the Lower Coal-measures of England. The commonest plants are Neuropteris heterophylla, N. gigantea, and Sphenopteris latifolia. They also include two marine bands, one near the base containing Schizodus deltoideus, Gonularia quadrisulcata, Lingula mytiloides, and Orbiculoidea nitida; and another above the "Ell" coal containing the same Gonularia and Orbiculoidea with Productus scabriculus, Bellerophon Urei, and Aviculopecten papyraceus. THE CARBONIFEROUS SYSTEM 303 Middle Coal-measures. These were formerly referred to the Upper stage because they consist chiefly of red sandstones and shales, and because in some districts they appear to be unconformable to the Lower Measures. From a study of the plant remains, however, Dr. Kidston considers them to belong to the Middle Measures. The group has a maximum thickness of 900 feet in Fife, where it comprises red sandstones, red and purple clays, with some thin seams of coal and ironstone. In Ayrshire the unconformity at the base of these measures is very marked, for they steal across the outcrops of the Lower Coal- measures until they rest on the uppermost limestone of the Lower Carboniferous, the Millstone Grit being absent as above stated. In the Ayr coalfield these beds include a thin band of fine white limestone containing Spirorbis carbonarius. A few Coal-measure plants have also been obtained from them. Around Sanquhar in Nithsdale, Dumfries, there is an outlying tract of Coal-measures which rest directly on the Ordovician rocks south of the boundary fault. It has recently been proved that these include both Lower and Middle Measures, and their occurrence shows that Nithsdale became a strait connecting the Ayrshire and Cumberland basins across a ridge which had previously been an island. 4. Ireland Rocks of Upper Carboniferous age occur in several parts of Ireland, but the principal coalfields are four in number, two in the north and two in the south. Of these the following is a brief account, Southern Counties. One extensive area occupies parts of the counties of Clare, Limerick, Kerry, and Cork, but the coal- seams in this region are few and of variable thickness, so that commercially it is less important than the " Leinster coalfield," which occupies parts of Tipperary, Kilkenny, and Queen's County. The succession in both these areas is similar, but a greater thick- ness of the Middle Coal-measures comes into the western basin. The groups recognised are : Munster. Leinster. 4. Middle Coal-measures with several good seams of coal. Fossils are freshwater Mollusca, Crustacea, and Amphibia ....... 2000 800 3. Lower Coal-measures with thin coals and shale roofs containing marine shells .... 900 600 2. Flagstone group, micaceous flagstones and shales . 500 300 3400 1700 304 STKATIGRAPHICAL GEOLOGY The Flagstone Group is the equivalent of the Millstone Grit of Bristol and South Wales, and the true Lower Coal-measures are the beds which overlie these flagstones, many of the same marine shells occurring in them as in the Gannister Beds of Northern England. 4 The Jarrow collieries in Queen's County are known for the number of Amphibian remains which have been obtained from them. Northern Counties. The most important tract in the north is the Tyrone coalfield, which lies to the north of Dungannon and contains many good seams of bituminous coal. The succession as given by Mr. Hardman 5 is as follows : Feet. Middle Coal-measures. Soft sandstones and shales with coals, fireclays, and ironstones ..... 930 Lower Coal-measures.- Hard sandstones and shales with a few thin coals and ironstones ...... 1000 Millstone Grit. Coarse grits and sandstones . . . 60 to 200 About 2000 Upper Carboniferous rocks occupy a considerable area in the counties of Leitrim and Fermanagh, and their sequence is the same as in Tyrone, but only part of the Lower Coal-measures come in above the Millstone Grit. 5. South Wales and Bristol Area This area exhibits a different development of the Westphalian Series from that found in other parts of the British Isles, and it is not yet certain how far the natural divisions of the series in this southern area can be compared and correlated with those of more northern districts. Moreover, the succession found in the great coalfield of South Wales is not the same as that occurring in Bristol and Somerset, and the differences between the two areas may be tabulated as follows : South Wales. Upper Coal Series ( = Upper C.M.). Pennant Grit Series ( = Transition C.M.). Lower Coal Series ( = Middle C.M.). "Millstone Grit Series " (of doubt- ful age). Bristol and Somerset. M. jUpper C. Trans. C.M. Radstock Beds Farrington Beds Pennant Grits " Vobster Group (Not represented. ) "Millstone Grit" (partly of Avonian age) In the Bristol coalfield the Avonian limestones pass up through shales into the Millstone Grit, the lower part of which probably represents the Yoredale Beds of northern areas (see p. 253). As to the age of the higher part there is no evidence, but there seems to THE CARBONIFEROUS SYSTEM 305 be a break at the top of tlie sandstones, for there is nothing comparable to the Lower or to the Middle Coal-measures of the Midlands, at any rate from a palseontological point of view. In South Wales the so-called Millstone grit may be of any age, for there appears to be a break at the base of it, and the overlying beds which have hitherto been called " Lower Coal-measures " contain the flora of the middle stage, and must be classed as Middle Coal-measures. They everywhere rest on a sandstone which is known as the Farewell Eock, and though this is at present grouped with the Millstone grit it may properly belong to the overlying Coal-measure Group. In the following account the stages found in South Wales will be described first, as the succession is there more complete. Millstone Grits. In South Wales the group is very variable. At the eastern end of the basin north of Cardiff it is about 460 feet thick, consisting of a pebbly grit at the base, of shales and sandstones in the middle, and a massive sandstone (the Farewell Eock) at the top. Westward, however, the group expands to a thickness of over 1500 feet, the greater part consisting of dark- grey shale, but always having the Farewell Eock above, while the basal grits are sometimes present and sometimes thin out, so that in Gower, where it is absent, no line can be drawn between the Bishopston cherts and shales and the overlying barren shales. 6 The' normal succession is well developed in the Ammanford district on the northern outcrop, where the thicknesses are : Feet. Farewell Rock green and white sandstones . . about 200 Dark barren shales with bands of sandstone . . 500 Quartzitic grits with some bands of shale . . ,, 600 In the Bristol and Somerset Coalfield the fine-grained sandstones which have been called Millstone Grit form a complete ring round the Coal-measure basins, but are apparently an upward continuation of the Avonian Series, and the relation of their highest beds to the overlying shales has not yet been ascertained. In most places they are about 1000 feet thick, but in the Mendip district they have thinned to 500 feet. Middle Stage. According to the evidence of the plants this only occurs in South Wales where it varies much in thickness, and like other members of the system expands toward the west and south-west. Thus at the eastern end of the basin it is only from 625 to 850 feet thick, but in the centre it is about 1400 feet, and near the south-west margin upwards of 4000 feet. This expansion appears to be due not only to the increased thickness of sandstones and shales, but also to the incoming of fresh X 306 STRATIGRAPHICAL GEOLOGY I I P , I- t ! m i < *-- o ",' ; * a s v ~^ 1<- Ipl THE CARBONIFEROUS SYSTEM 307 measures, and more coal-seams toward the south-west. Many of the coal-seams are not persistent, and it is not easy to correlate the seams which occur in different parts of the basin. Transition Stage. In South Wales this stage is represented by the Pennant Grit Series, which consists of massive felspathic sandstones or fine-grained grits, with some subordinate bands of shale and seams of coal, and it forms the greater part of the high central moorland plateau which is known as the Mynydd. Near Pontypool on the east it is only about 800 feet thick and the coal- seams are thin, but west of Taff Vale the group rapidly thickens and workable coal-seams set in, so that near Swansea it is over 3000 feet thick, and contains fifteen good seams of coal. In the Bristol coalfield the Transition stage includes two distinct groups, the Vobster Coal-measures below and the Pennant Grit above. The Vobster Beds consist principally of shales with Borne sandstones and many seams of coal, and many of the shale bands contain ironstone nodules, so that these measures are rich in coal and iron. The thickness is about 2000 feet, and in the central part of the area (east of Bristol) there are twenty-six coal-seams, but many of these are thin and disappear both to the north and south. The Bristol measures include several bands which contain marine shells, and from these Mr. H. Bolton has obtained no fewer than forty-three species. The commonest are Nucula cequalis, Posidoniella Icevis, Schizodus antiquus, Lingula mytiloides, Glyphio- ceras carbonarium, and Gastrioceras Listeri. Pterinopecten papyraceus also occurs, and by comparison with the list on p. 293 it will be seen that the shells hardly support the evidence of the plants. Mr. Bolton compares the fauna with that of the Gin Mine band in North Staffordshire, which, according to Dr. Kidston, should be placed in the Lower Coal-measures. The precise age of the Pennant Grits in the Bristol coalfield is not yet known ; they may belong to the Transition or to the Upper Coal-measures. They form a sandstone group from 1000 to 1700 feet thick, the prevalent material being a hard grey felspathic sandstone, which makes a durable building stone, and lies in thick beds, so that it can be cut as a freestone. There are some bands of shale and four workable seams of coal. Upper Coal-measures. These form a variable series of grey shales, light-coloured sandstones, and coals, which appears to be an expanded equivalent of the red Upper Coal-measures of more northern coalfields. In South Wales, near Swansea, the group is very thick, having been estimated at about 3000 feet (Pellengare Series). 308 STRATIGRAPHICAL GEOLOGY All the coal-seams in South Wales undergo a change of character as they are followed from east to west, the seams which are bituminous house-coals in the east becoming hard steam-coals in the central parts of the basin and passing westward into stone-coal or anthracite. It is also found that the coals of the upper measures are more bituminous than those of the lower. In the Bristol coalfield a thickness of about 2000 feet is found, although the summit is nowhere seen ; it is divided into two parts, the lower portion or Farrington Group occurring in both the northern and southern basins, the upper portion or Radstock Group only coming into the southern or Somerset basin. The small coalfield of the Forest of Dean calls for some notice, because previous correlations of the beds with those of the South Wales and Bristol basins have been altered by a recent study of the fossil plants. It was supposed that there was an upper set of Coal-measures underlain by an equivalent of the Pennant grits. It now appears, however, that all the productive Coal-measures belong to the upper stage, and that the supposed Pennant grits are merely a set of massive sandstones in the lower part of this stage, which thus consists of the following members : Feet, Upper shales and coals (10 seams) ..... 1480 Thick sandstones with 3 coal-seams .... 850 Trenchard sandstones and 2 coals . . . J . 77 2407 These beds are directly underlain by the so-called Millstone Grit, and consequently there is no representative of either the Middle or Lower Coal-measures. 6. Devon and Cornwall The " Upper Carboniferous " or Westphalian Series of this area has hitherto generally been called the Culm-measures, but this term has become misleading and should be dropped, for they are simply barren Coal-measures and are not the equivalents of the strata. known as Culm-measures in Central Europe. According to Mr. Ussher 7 the series is divisible into two parts or stages, which he called Middle and Upper Culm-measures, but this is not confirmed by Mr. Arber who regards the whole succession of beds as forming a single lithological division, the so-called " Eggesford Grits " being only a local variation of type which he has recognised in several localities. 8 Mr. Arber describes the series as consisting essentially of fine- grained sandstones or grits and shales in alternating bands. The beds are highly flexured and broken by faults, many of the sharp THE CARBONIFEROUS SYSTEM 309 anticlinal folds seen in the coast-sections being faulted at the crest and sometimes overthrust. The sandstones are hard, of grey, lilac or brownish tints, and usually from 4 to 5 feet thick, but some- times less ; the shales and mudstones are grey or red. These strata include beds of hard impure coal, generally known as culm, and there are also thin unpersistent beds of limestone as well as layers of calcareous nodules. The nodules near Bideford and Instow have yielded marine shells chiefly of Gastrioceras carbonarium, G. Listeri, Dimorphoceras Gilbertsoni, Pterinopecten papyraceus, and Posidoniella Icevis ; and of this assemblage Dr. W. Hind remarks that it resembles that of the Bullion Mine seam in the Lower Coal-measures of Lancashire. In the same district plant remains have been collected by Mr. T. Kogers and Mr. N. Arber, and the latter has shown that the flora is that of the Middle, not of the Lower Coal-measures, so that here, as in the Bristol area, there is a conflict between the evidence of the fauna and the flora. Nothing to indicate the presence of Transition or true Upper Measures has yet been found in Devonshire. In the southern part of the area near Chudleigh and Newton Abbot the base of the Westphalian Series is a conglomerate which lies unconformably on the chert-beds of the Lower Series and in some places appears to overstep them on to the Devonian rocks. It contains pebbles of chert and limestone and a few of granite, which, however, is not like that of Dartmoor. This bed is overlain by a set of alternating grits and shales like those of Bideford, and near Exeter Mr. T. G. Collins has recently obtained fossils which show that they occupy a low position in the series. D. CONTINENTAL EQUIVALENTS 1. Westphalia We take this area first because the succession of beds therein found is regarded on the continent as the typical development of the series. De Lapparent has well remarked the area might be appropriately called the basin of Miinster, since that city overlies the central portion of it. It is only in the southern part of the basin that the Carboniferous rocks come to the surface, the greater part being concealed beneath the Jurassic and Cretaceous strata which form the Westphalian plain. The surface tract extends from near Dusseldorf on the Rhine, by Hagen and Arnsberg to Brilon and Marsburg, a distance of about 90 miles. From this outcrop the beds dip northwards and then extend in a series of undulations beneath the great plain, probably reaching as far as the 310 STRATIGEAPHICAL GEOLOGY Teutoburger Wald before they finally rise again to terminate the broad syncline. The student should refer to Stanford's geological map of Central Europe. At the base of the series is a thick mass of sandstones (Flotzeere Sandstein), some of which are coarse and pebbly. These beds occupy the place of our Millstone grit and are said to be without fossils. The whole series is believed to be over 11,000 feet thick, and has been divided into the following groups : Feet. 6. Group with highly bituminous coals (35 seams) . . 3000 5. ,, less bituminous coals (8 seams) . . 1000 4. ,, good coking coals (19 seams) . . . 1200 3. ,, fairly good coals (7 seams) . . . 400 2. ,, poor coals (21 seams) .... 2500 1. The Flotzeere sandstones 3000 In this succession Cremer has recognised four plant assemblages. Zone d = Q, with Neuropteris rarinervis and N. tenuifolia. Zone c = 5, with Sphenopteris obtusiloba and Lonchopteris. Zone & = 3 and 4, with Sphen. trifoliata and Alethopteris lonchitica. Zone a = 2, with Neuropteris Schlehani. It is specially to be noted that though the succession is com- plete, and every part of it seems to be fully developed, the flora of No. 2 (zone a) is not that of our Lower Coal-measures. It contains, however, several bands in which marine fossils occur, the lowest of these being about 600 feet above the base and yielding Gastrioceras Listeri with other species which in England do not range above the lower division. 2. Belgium and France West of the Ehine, a long but rather narrow tract of Carboni- ferous rocks comes in on the north flank of the high ground formed by the Cambrian and Devonian rocks (see p. 216) and stretches continuously from Aix-la-Chapelle by Liege, Namur, and Mons, and thence still farther under the cover of Neozoic deposits by Valenciennes, Douai, and Bethune to the Boulonnais. Throughout this tract of country the Carboniferous strata are sharply flexured and much broken by faults. Their southern boundary is generally an overthrust fault with a southerly hade and such a great displacement that the Devonian rocks have been carried over the Coal-measures which are in some places actually worked beneath this cover of older rocks. In Belgium the representative of the Millstone grit seems to be very thin, but it forms a constant band of coarse grit from 30 to THE CARBONIFEKOUS SYSTEM 311 40 feet thick, and thus separates the Visean shales (see p. 279) from the overlying Coal-measures. These latter have been divided by Zeiller 9 into the following zones : 4. Zone of the " charbons flenus " (gas-coals) with Pecopteris nervosa, P. dentata, Neuropteris rarinervis, Sphenopteris obtusiloba, and Sigil- laria tesselata. 3. Zone of "charbons gras," with Sphenopteris nummularia, Neurop- teris gigantea, Alethopteris Serli, and Sigillaria polyplcea. 2. Zone of "charbons demigras "Sphenopteris trifoliata, Lonchopteris JSricei, Aleth. Darreuxi, and Sigillaria scutellata. 1. Zone of "charbons maigres "Neuropteris Schlehani, Alethopteris lochitica, Sigillaria elegans, and some Lepidodendra. It will be noticed that the succession of coals and of plant-zones is similar to that of Westphalia. The total thickness of this series near Boussu west of Mons is about 2400 metres (=8000 feet), and near Douai it is estimated to be more than 2500 metres. In the Boulonnais, near Leulingham, Locquinghem, and Ferques, Coal-measures are being worked under overthrust masses of Car- boniferous limestone ; and another trough appears to come in to the north of these overthrusts, part of it lying below Dover where Coal-measures were reached by boring at a depth of 1100 feet from the surface. The same series of troughs and faulted blocks is probably continued westward under the south of England till it emerges again in the coalfields of Bristol arid Somerset. 3. Sarrebruck Area Just as the Belgian coalfields lie in a synclinal trough on the northern flank of the broad tract of Devonian rocks which stretches across the Ardennes and the Khine Valley, so also we find another syncline on its southern flank. The exposed portion of this syncline is known as the Sarrebruck basin, but it is very likely to have a prolongation westward beneath Metz, though probably at too great a depth to be attainable (see Fig. 103). The deposits in this southern basin, however, differ much from those in the northern, for not only is there an entire absence of the Dinantian Series, but the greater part of the Westphalian is also wanting, and the lowest beds in the basin must rest uncon- formably upon the Lower Devonian or still older rocks. The bottom of the basin in its deepest part has not yet been reached, but the lowest beds traversed correspond with the highest part of the Belgian Series (the charbons fle"nus) and are covered by beds which resemble our Upper Coal-measures, while these latter are succeeded by the still higher Stephanian Series. The Westphalian part of the succession is as follows : 312 STEATIGRAPHICAL GEOLOGY Feet. 3. Purple sandstone and shales with thin coals . . about 2000 2. Red sandstones and conglomerate at base . . . ,, 3500 1. Sandstones and shales, both red and grey, with 80 seams of bituminous coal . . . . . . ,, 1500 4. Silesia, Moravia, and Poland No large area of Westphalian strata is found again in Central Europe till we reach Silesia, where there are two basins or troughs, the one known as the Lower Silesian or Waldenburg trough and the other as the Upper Silesian basin. The first runs from the Eulengebirge and Waldenburg westward to Schatzlar in Bohemia, and is partly covered by Triassic and Cretaceous rocks. The second lies farther south, its north-east end resting on Devonian near Zuckmantel in Austrian Silesia, and extending south-west far into Moravia (to near Briinn). The thickness of beds in both these areas is considerable, and those of the Waldenburg trough were divided into two stages by Stur : 2. Schatzlar Beds corresponding with those of Sarrebruck. 1. Waldenburg Beds, with a flora of Culm affinities. He was inclined to class the lower beds with the underlying Culm, but Weiss regarded them as Westphalian, and Tietze has more recently shown that their equivalents in Moravia (the Ostrau Beds) contain a larger number of species belonging to the Schatzlar flora than to that of the Culm. The Ostrau Beds include several marine bands containing Gastrioceras Listeri, Bellerophon Urei, and other species which suggest a correlation with our Lower Coal-measures. Still farther east a similar succession of Westphalian Beds occurs near Cracow in Poland, where it includes a marine band with Spirifer mosquensis, a species which does not exist in Western Europe, and marks a temporary incursion of the eastern or Moscovian Sea. 5. Russia When we reach Central Kussia we find the representatives of the Westphalian Series to be entirely of marine origin, estuarine and semi-terrestrial conditions only prevailing in the southern basin of Donetz. In the Moscow basin the whole series consists of limestones, white or yellow in colour, some beds oolitic and some chalky, and many of them highly fossiliferous. This series is known as the Moscovian, and is about 1500 feet thick; it is specially char- acterised by Spirifer mosquensis and Streptorhynchus (Meekella} eximius, with the large Foraminifer called Fusulina cylindrica, which is sometimes so abundant as to form a Fusulina limestone. THE CARBONIFEROUS SYSTEM 313 These limestones are covered by other beds which also probably belong to the Westphalian Series, for they contain some species common in our Avonian Series, such as Spirifer striatus, Productus com, and P. semireticulatus, together with species of Syringopora, Lonsdaleia, and other corals. The Moscow basin, though it occupies an area of 13,000 square miles, is probably only a part of a great tract of Carboniferous rocks which underlies the broad Russian plain, for the beds pass in every direction underneath newer strata of Permian, Jurassic, and Cre- taceous ages. Moreover they crop out again along the western flanks of the Ural Mountains in an almost continuous strip, and this (partly covered by Permian) is believed to extend northward to the shore of the Arctic Ocean. In the Ural Mountains the Moscovian Series consists of lime- stones with bands of calcareous shale. Spirifer mosquensis and Productus cora are still characteristic fossils, but Cephalopoda are more frequent and include Glyphioceras striolatum, Pronorites cyclo- lobus, and Gastrioceras marianum. 6. Mediterranean Region A few tracts of Westphalian or Moscovian strata occur in Southern Europe, and they all resemble the Russian type more than that of the north-west. Thus in Austria sandstones and shales, with bands of Fusulina limestone, are infolded with Trias on the western borders of Styria ; and again in Carinthia the Gailthal slates (Dinantian) are overlain by sandstones and shales succeeded by Fusulina limestones. A similar series is found to the south in Carniola, Croatia, Bosnia, and Herzegovina. Westphalian Beds are said to occur in Corsica, and in Spain (Andalusia) there are Coal-measures with Dictyopteris Bronyniarti and interbedded limestones with Spirifer lineatus and Sp. mosquensis. Again in Asturia the Sama slates with Dictyopteris and intercala- tions of marine beds with Schizodus and Posidonomya probably belong to the same series. III. THE STEPHANIAN SERIES As stated at the beginning of last chapter (p. 244), no representa- tives of the highest portion of the Carboniferous System occur in Britain. This gap in the geological record appears to be due to a general uplift of the whole of North- Western Europe, whereby the low-lying areas in which the older (Westphalian) Coal-measures had been accumulated were raised into dry land. This upheaval, moreover, was accompanied by volcanic disturbances and by crustal 314 STRATIGRAPHICAL GEOLOGY compression, which resulted in the production of several systems of flexures, but especially of a great series of folds having a general direction from west to east. This series of flexures is known as the Armorican or Hercynian System, and it extends from the south-west corner of Ireland through South Wales, Devon, and Cornwall. Thence it passes under the whole of Southern England and through the north-east of France and Belgium. Another part of the same series passes through Brittany and Normandy and under the Paris basin till it emerges in the Ardennes. The combined series then extends through the Ehine Province of Germany, through Westphalia and the rest of Germany to Poland and the northern border of Bohemia. These Armorican flexures are narrow and the synclines are deep ; the folds are sometimes overthrust and often broken by faults. North of this series of flexures in England we find another set, or rather two sets which cross one another nearly at right angles, and it is to the intersection of these two sets of folds that the pro- duction of Coal-measure basins is due. The one set runs approxi- mately north and south, and is exemplified in the ridge of the Malvern and Abberley Hills, where Upper Coal-measures are included in one of the flexures, and yet the folding is shown to be pre-Permian by the position of Permian strata on a plane of erosion cut across these flexures. Farther north is the parallel anticline of the South Pennine area, which separates the flanking synclines occupied by the Coal-measures of Nottingham and Yorkshire on the one side and of North Stafford and Lancashire on the other side. Here again the pre-Permian age of the main flexures is indicated by the position of the Permian Beds overlying them in Notting- hamshire, though there may have been further uplifts in Triassic time, and still later in Tertiary times. The North Pennine faults in West Yorkshire, Westmoreland, and Cumberland were also in part pre-Permian. The other set of flexures runs approximately from W.S.W. to E.N.E., and may be termed the Lancastrian Series. They differ markedly from the nearly parallel Armorican folds in being very broad and comparatively shallow undulations. Such is the broad anticline of the Eibble Valley and its continuation across Yorkshire, which passes eastward under the Permian of that county. For a fuller account of these flexures the reader is referred to the new edition of my Building of the British Isles (1911), but which of the two series is the older has not yet been determined. While uplift, disturbance, and erosion were going on in the north of Europe, deposition was still in progress farther south, and more especially over the eastern part of Europe. In the western THE CARBONIFEROUS SYSTEM 315 region deposition seems to have been restricted to certain shallow- water areas of a lacustrine character, and only portions of these remain in small isolated basins. It was only in the east over Kussia and the Balkan provinces that open sea prevailed. Conse- quently there are two very different facies of this series, a western Coal-measure Group (with plants and freshwater fossils) to which the name Stephanian has been given from St. Etienne in France, and an eastern marine facies known as the Uralian from its development in the Ural Mountains. We can only give a brief account of these two facies. A. THE STEPHANIAN FACIES 1. The Flora. This is characterised by the abundance of certain species of Pecopteris (P. arborescens, P. cyathea, etc.), of Callipteridium, Odontopteris, of Cordaites and Calamodendron with the foliage known as Annularia, the non-ribbed Sigillarias, and finally by the appearance of Tceniopteris, of the Conifer Walchia and the Cycads Pterophyllum and Plagiozamites. Three phases or stages have been recognised by M. Grandeury as existing where the Stephanian Series is complete ; these are : 3. Phase of Calamodendrons. 2. Phase of Annularia stellata and Pecopteris cyathea. 1. Phase of Ann. sphenophylloides and P. neuropteroides. The Fauna is not a large one, consisting of some small fresh- water Crustacea (Estheria, Leaia\ a few Molluscs (Anthracosia), a number of insects, especially beetles (Eoblattina), and Phasmidae (Protophasma, Dictyoneura) ; also some Fish and Amphibians. 2. Stratigraphy. In France the Stephanian deposits occupy a number of small isolated basins or troughs, most of which are situate on the borders of the "Massif Central," though some occur in the middle of it (see map, Fig. 102). Many of these patches are doubtless portions of much larger areas or basins of deposition and were originally continuous, but no attempt has yet been made to ascertain which may have been connected with one another, and indeed most French geologists seem to assume that they were all disconnected. The typical district of this region is that of St. Etienne in the Department of the Loire (see Fig. 104), where the succession is stated to be as follows : Feet. r i (5. Bois d'Aveize Group 600 measures! 4 ' Berard Grou P 110 es (3. St. Chamond Group 1800 2. Sterile Beds, conglomerates and sandstones .... 1500 1. Breccia resting on the Archaean rocks 600 316 STKATIGRAPHICAL GEOLOGY To the north, in the Morvan district, the Dinantian is overlain Fig. 102. M4.P OP CENTRAL PLATEAU OF FRANCE. Reproduced by permission from de Lapparent's Traite de Geologie. 5. Jurassic rocks. 4. Triassic rocks. 3. Carboniferous. 2. Permian. 1. Archaean and Igneous rocks. by flows of quartz-porphyry which may be of Westphalian age ; and above these are beds with a Stephanian flora, the lowest being THE CARBONIFEROUS SYSTEM 317 sandstones and shales with some beds of coal (from 160 to 300 feet thick) overlain by a mass of sandstones and conglomerates more than 2000 feet thick. Another small basin in the north, that of Commentry (Allier), is interesting in three respects. One is that it contains a very thick bed of coal (40 to 60 feet thick) which in certain directions splits up into 5 or 6 beds, separated by shales and sandstones, just as in the case of the "thick coal" of Staffordshire. Another point is that a shale 15 feet above this coal has yielded a large number of insect remains. Thirdly, one of the beds of sandstone passes at one place into a mass of breccia formed of angular blocks of gneiss and granulite which seem to represent a contemporaneous landslip from neighbouring high ground. A few small tracts of Stephanian measures occur in Brittany and Normandy, and are mainly of importance because they occur within the region of Armorican flexures and show that deposition had not completely ceased in that region. The most northern of these is at Littry, south-west of Bayeux, the Coal-measures here resting directly on Brioverian slates and being conformably over- lain by Permian Beds, and they belong, therefore, to the latest phase of the Stephanian. In Germany the most western tract of Stephanian Coal-measures in the Germanic region is that of the Palatinate near Sarrebruck (see Fig. 103), where they are known as the Ottweiler Beds. Here they succeed the Westphalian Beds without any break, though they overlap the latter on to the Devonian ; and they are also conformably overlain by Permian Beds. The greater part of the series consists of felspathic sandstone and conglomerate, testifying to the upheaval and rapid detrition of the country to the north of the basin ; but both at the base and again at the top are a few hundred feet of finer sediments, chiefly grey shales with seams of coal. The total thickness is from 2000 to 3000 feet. The small coalfields of Pilsen and Kradno in Bohemia are specially interesting because they exhibit a still more complete transition from Carboniferous to Permian and include a bed from which remains of many insects, fish, and Amphibia have been obtained by Dr. Fritsch. The plants in and above this bed are Carboniferous, while the vertebrate animals are closely related to forms which have been found in beds of Permian age : fish of the genera Amblypterus, Acanthodes, Palceoniscus, and Xena- canthus ; Amphibians of the genera Dendrerpedon, Sphenosaurns, Ghelidosaurus, Limnerpedon, Keraderpedon, Urocordylus, and Hyalonomus, some of which are figured in the Geological Magazine for 1885. THE CARBONIFEROUS SYSTEM 319 The following is a tabular view of the strata in the Pilsen basin, modified from that given by Feistmantel : 10 p , f Red felspathic sandstones with Araucarites. \Shales with ironstone nodules (Permian plants). f Shales with Carboniferous plants, a coal-seam and the Stephanian-J Nurschau "gas-coal" (oil-shale). [Lower shales with a coal-seam. The plants in the Stephanian shales include Calamites (three species), Sphenopteris Schlotheimi and eight other species, Odontopteris obtusiloba, Dictyopteris Brongniarti, Sigillaria distans, and fragments of Walchia. Both Feistmantel and Homer correlate these beds with those of the Ottweiler Group. Further particulars will be found in the paper referred to. B. URALIAN FACIES 1. The Fauna. This is entirely marine in the more typical areas. Foraminifera play an important part in the formation of Uralian limestones, the principal genera being Fusulina, Swagerina, and Doliolina. A spiral Bryozoan, Archimedes, is also a common and characteristic fossil. Brachiopoda are abundant, especially the Terebratuloid Dielasma, the Rhynchonellid Camarophoria, and the Orthotetid Meekella. Spirifers are still common, and the persistent species Productus cora is abundant ; Chonetes is represented by G. uralicus, and shells of the genus Spiriferina make their appear- ance, together with some which are closely allied to Waldheimia. Of Mollusca the Lamellibranchs Conocardium, Aviculopecten, Schizodus, and other Carboniferous genera occur. The Gastropod Omphalotrochus is common, but Cephalopoda are not numerous ; Gastrioceras marianum occurs throughout, and in the highest beds Agathiceras uralicum appears. A few trilobites also survive. 2. Stratigraphy. The typical development of this series is found on the south-west side of the Urals in the basin of the river Ai (Province of Ufa), where the whole consists of limestone, and three stages or zones are recognised by Tschernyschew, viz. : 3. Zone of Swagerina princeps and Spirifer sarana. 2. Zone of Productus cora and Meekella eximia. 1. Zone of Omphalotrochus Whitneyi. The same succession is found in the Moscow basin and in the Tinian district in the north of Russia. The limestones of the middle zone are generally dolomitic. On the eastern slopes of the Ural Mountains there is a much greater variety of sediments indicating the near proximity of land, 320 STRATIGRAPHICAL GEOLOGY limestones being interbedded with marls and sandstones, clays and layers of gypsum ; and the highest beds contain a Stephanian flora (Pecopteris arborescens, Annularia longifolia). The Uralian Sea must have occupied the greater part of Russia, passing southward over the Crimea and south-eastward through the Balkans, Bosnia, Herzegovina, Carniola, and the Julian Alps. Farther west, in the Carnic Alps between Italy and the Tyrol, there is an interesting intercalation of the Uralian and Stephanian facies, i.e. of freshwater and marine beds. The following series has been described by Schellwein at Krone : n Shales with Spirophyton. Dark-grey limestones with Fusulina and Swagerina. Sandstone with Pecopteris and Annularia stellata. Bed containing marine shells. Shales with Annularia stellata and other plants. Shales and calcareous sandstones with Camarophoria, Productus, and Phillipsia. Shales with Pecopteris oreopteroides. Conglomerates at base (about 200 feet thick). The total thickness of these beds is about 900 feet, and they are overlain by another Fiisulina limestone which is of a light-grey colour and contains a Lower Permian fauna. How much farther west the Uralian Sea extended is not yet known, for no marine strata of this age have been found in Italy or Spain, though deposits with a Stephanian flora occur in both countries. C. THE FORMATION OF COAL-SEAMS The most important point connected with the deposition and distribution of the Coal-measures is the formation of coals ; what they are made of, and how they were accumulated. The reader will find this subject well treated in a little book, On the Natural History of Coal (Camb. Univ. Press), by Mr. N. Arber, but a brief summary of the current theories and modern views respecting the origin of coal may here be given. A complete explanation of coal-seams must account for (1) the source of the carbonaceous material ; (2) the different existing varieties of coal ; (3) the frequent alternations of coals, clays, shales, and sandstones ; (4) the great extent of some single beds of coal. The reader is probably aware that there are many kinds of coal which go by different names according to their chemical com- position and physical peculiarities. There is so-called " Brown- coal," which is little more than lignite or fossil wood ; there is glossy Cannel coal, which does not soil the fingers and burns with a clear, continuous flame ; there are the Humic or so-called THE CARBONIFEROUS SYSTEM 321 Bituminous coals, which vary in their gas-producing qualities ; and lastly, there are the hard coals Steam-coal and Anthracite the last differing in being smokeless when burnt. It is now generally agreed that the original substance of all kinds of coal was vegetable matter of some kind, and that many kinds owe their peculiarities to differences in the nature of the vegetable matter of which they were originally composed. Thus the Toula coal of Russia consists entirely of sheets of the bark of the tree called Bothrodendron, and this is so little altered that, when exposed to the weather, it splits up into thin layers. Other coals appear to have been composed of a mixture of woody tissue with leaves and fronds of fern-like plants. Some beds consist of the spores or sporangia of Carboniferous plants ; thus Cannel coal is mainly composed of Lycopod-spores, and some Boghead coals of closely packed cellular bodies which are either spores or gelatinous algse arising from their decay. Some other varieties of coal, however, are due to the amount of chemical change which the substance has undergone since its original deposition and burial in the earth's crust, as well as to the varying degrees of pressure to which the strata have been subjected at different places. These influences appear to have led to a con- centration of carbon in the older coals, or rather to an increase in the proportion of carbon to hydrogen. Let us next consider the conditions under which the coal-seams are likely to have been accumulated, and it is evident that on this question we can learn much from a study of the associated sedi- ments, that is to say from the general characters and contents of the whole series of measures of which the coal-seams form a part. These beds include sandstones which are generally of fine grain, and often show current bedding; shales which are consolidated silts, i.e. sandy or micaceous muds deposited in successive layers or laminae ; and clays which are more homogeneous muds. All these beds usually contain only remains of plants, and sometimes of insects, with bivalve shells such as Anthracosia and Carbonicola, which were of fresh or brackish -water habitat. Occasionally, how- ever, there are bands which contain marine shells, and from these we may safely infer that the measures in which such beds occur were formed in swamps or marshes liable to incursions of the sea. There is, in fact, no escape from the conclusion that the Lower and Middle Coal-measures of the British region, and probably most of those in the Continent, were formed in extensive swamps on the borders of sinking land, where the area of alluvial levels, swamps, and brackish lagoons was being continually enlarged at the expense of the sea by the process of silting up and by the outgrowth of Y 322 STRATIGRAPHICAL GEOLOGY plants which, like modern mangroves, did not mind having their roots covered by salt water. So far as to general conditions, but we have still to deal with the formation of* the individual coal-seams, and to consider the way in which the coal-material may have been accumulated so as to form a seam or bed, for these are sometimes of great thickness, as much as 20 or 30 feet. There are a certain number of facts which seem to indicate that the vegetation from the decay of which a bed of coal was formed grew on the spot, while other facts have suggested the theory of drift-origin, i.e. that the vegetable materials were drifted by a current into their present situation. Some writers have maintained the one, and some the other theory ; but as Mr. Arber remarks, it is very likely that both theories are equally true, some coals having been formed in the one way and some in the other. He also points out that the significance of certain facts, often appealed to as evidence for one or the other theory, has been greatly exaggerated and misunderstood. Thus the presence of an underclay containing many Stigmarian rootlets is no proof that the coal above has been formed in situ from plants which grew in that clay ; neither does the fact that the roof -bed of a coal contains drifted plant remains prove the coal to have been formed by an accumulation of such drift. The formation of each bed of coal must be individually considered, and its own structure must especially be ascertained and noted ; coals which consist entirely of one kind of material, such as the Cannels, were doubtless formed in situ from local growths, while some of those which consist of many sorts of plants may have been formed of drifted materials. Some seams may have been semi-terrestrial growths like modern peats, others were formed on the floors of large swamp lakes or in brackish-water lagoons, and in some cases these water spaces were clearly near to the estuary of a large river. Finally, a few remarks may be made on the succession of plant- assemblages in the Coal-measures. That there is such a succession in the northern part of England admits of no doubt, and this same succession prevails over a large area, but whether it prevailed all over Europe, wherever the sequence was complete, has not yet been proved. Thus we have seen that what is known at present as the flora of the Lower Coal-measures cannot be recognised in South Wales, nor in the Bristol and Somerset coalfields, and further that it has not been found on the Continent except in a narrow zone of so-called Millstone grit. From the absence of this particular assemblage of plants, or rather from the absence of certain species, it has been inferred that the Lower Coal-measures are not represented in the more Southern areas. THE CARBONIFEROUS SYSTEM 323 It is doubtless true that from a botanical point of view the Millstone grit of the Pennine area is only the basal portion of the Lower Coal-measures, and that in other areas, such as Bristol and South Wales, the latter might be represented only by sandstones. Further, it is almost certain that the Millstone grit episode marks a general uplift of the whole British region, and a shallowing of all the areas in which deposition was taking place. This uplift may have been greater in some parts of the region than in others, and deposition may possibly have been stopped for a time in Southern England ; but when we pass over into Belgium, and are asked to believe that divisions which have a thickness of 3000 to 4000 feet in England are entirely absent in that country, the case is different. There is no stratigraphical evidence of such a break, and if the absence of beds was due to uplift into dry land, signs of such a break should be clear and conspicuous. Another and very different explanation presents itself which should be carefully considered by the botanists ; this is that the absence of the species in question may be due to geographical limitation. It may be that the particular assemblage found in the lower stage of the northern region did not extend into the southern region, and that the southern assemblage did not extend into the northern region until the epoch of our Middle Coal-measures. In other words, the deposition of Coal-measures may have been con- temporaneous in both regions throughout Westphalian times, but the succession of plant-assemblages was different in the two regions, and dependent on slight climatal differences north and south of a certain line of latitude. The truth of this theory must be tested by more careful zonal study both of the plants and of the shells. REFERENCES 1 Kidston, Proc. Roy. Phys. Soc., Edin., vol. xii. p. 183. 2 Gibson, Quart. Journ. Geol. Soc. vol. Ivii. p. 251. 3 Peach and Home, Trans. Edin. Roy. Soc. vol. xl. p. 835 (1903). 4 " Geol. of Edinburgh," Mem. Geol. Survey, 2nd ed. (1910). 5 Hardman, Mem. Geol. Surv. Ireland, Expl. of Sheet 35. j 6 "Geology of S. Wales Coalfield," Parts vii. and viii., Mem. Geol. Survey. 7 Ussher, in Proc. Som. Arch, and N. H. Soc. vol. xlvi. p. 1. 8 Arber, Quart. Journ. Geol. Soc. vol. Ixiii. p. 6 (1907). 9 Zeiller, Bull. Soc. Geol. Fr. (3) xxii. p. 483 (1894). 10 Feistmantel, Geol. Mag. for 1877, p. 105. 11 Schellwein, Palceontographica, xxxix. CHAPTER XI THE PERMIAN SYSTEM A. NOMENCLATURE AND SUBDIVISIONS IN the early days of geological research all the strata which inter- vened between the Carboniferous and the Lias were classed together as the New Red Sandstone, but in 1841 and subsequent years Sir R. Murchison saw good reason to divide them into two great series and to believe that the lower series, on account of the Palaeozoic aspect of its fauna, formed the summit of the Palaeozoic succession, while the higher strata might be regarded as marking the beginning of Neozoic time. Further, he found that this lower set of post- Carboniferous rocks was more fully developed in Russia than in any other part of Europe, and was especially well exposed in the province or ancient kingdom of Perm, whence he proposed to call them Permian. Although the name Permian has been adopted in England, France, and Russia, it made its way very slowly in Germany, mainly because Murchison made mistakes in his grouping of the German rocks. In that country the strata which are really of Permian age fall into two strongly contrasted divisions, and just as the overlying Trias was named from its consisting of three such groups, so the term Dyas was proposed by Marcou in 1859 from the duality of the underlying system. This name was adopted by Geinitz, and was for a time much used on the continent, but Professor Kayser and others have more recently preferred the older name as on the whole a better one. We have seen that where the Carboniferous System is complete and the Stephanian Series is fully developed there is a passage from it into the Permian. Further, where the Permian System is fully represented it consists of three stages or series, viz. (1) a series of beds like the underlying Coal-measures and containing a flora which does not differ much from the Stephanian ; this has been 32-1 THE PERMIAN SYSTEM 325 termed Autunian by the French geologists ; (2) a mass of red sand- stones well developed in Germany, where they are known as Rothlie- gende, but for which the name Saxonian has been proposed ; (3) a group of limestones and marls, known in Germany as the Zechstein, but for which Renevier has proposed the name Thuringian. In the British Isles there are several different facies of Permian, and there is little doubt that these include representatives of the Saxonian and Thuringian, but it is uncertain whether we have any equivalent of the Autunian. In Russia and Southern Asia the whole system consists of marine deposits with a much more varied fauna, and Russian geologists have proposed special names for parts of their succession. The correlation of English, German, and Russian divisions is shown below : English. Germany. Russia. Magnesian limestone. Thuringian. Thuringian. Penrith sandstone ? Saxonian. Kostromian. Shropshire Permian ? Autunian. Artinskian. B. LIFE OF THE PERIOD 1. The Flora The Permian flora is on the whole very like that of the Carboniferous, fern-like Pteridophytes (Pecopteris, Odontopteris\ Calami tes, and Tree Ferns (Psaronius) being still the prevalent forms, and many of the same genera being present. At the same time it differs both in the absence of certain genera, in the greater relative development of others, especially Callipteris, and in the presence of numerous Mesozoic types, especially Coniferae, Cycads, and Ginkgos. Thus Sigillaria and Lepidodendron have almost disappeared, while Walchia is abundant together with the new Coniferous genera Ullmannia. Among (?) Pteridosperms Callipteris, Schizopteris, and (in India and the southern hemisphere) Glossopteris are new and characteristic forms. Cordaites and Cordioxylon ( = Araucarites) are not uncommon. Cycads are represented by Pterophyllum and Plagiozamites. Leaves of the curious tree called Ginkgo, still existing in China, are also found, as well as those of Baiera, a long extinct genus of similar affinities. The following are some of the most characteristic species of plants : Pteridospermce. Callipteris conferta, Neuropteris auriculata, Schizopteris Gumbeli, Tseniopteris multinervis. Equisetales. Calamites gigas, Anmilaria stellata. Conifer ales. Walchia piniformis, W. hypnoides, W. filiciformis, Ullmannia Bronni, U. Geinitzi. 326 STRATIGRAPHICAL GEOLOGY 2. The Fauna The Permian fauna is interesting because, as in that of the Carboniferous, it may be said that we possess remains not only of the marine fauna of the period, but also of the inhabitants of its great salt lakes as well as of the fresh waters and of the land. The typical marine fauna is only found in the east of Europe, b Fig. 105. GROUP OF PERMIAN FOSSILS. a. Walchia hypnoides. b. Schizodus Schlotheimi. c. Productus horridus. d. Spirifer alatus. e. Monotis speluncaria. /. Strophalosia lamellosa. g. Camarophoria Schlotheimi. Ji. Fenestella retiformis. extending thence into Asia and America, but in later Permian time arms of this sea extended into Central Europe and into the British Isles, forming landlocked inland seas and large salt lakes resembling the Black Sea and the Caspian of the present day. ,In these areas we find a limited and dwarfed assemblage of marine species. This marine fauna is largely composed of species which belong to Carboniferous genera, but with these are associated some new forms. The following is a list of the prevalent genera, those to which an asterisk is appended being new introductions : THE PERMIAN SYSTEM 327 Bryozoa. Fenestella, Synocladia, * Acanthocladia. Brachiopoda. Dielasma, Camarophoria, Spirifer, Productus, Stropha- losia, Aulosteges.* Lamellibranchia. Allorisma, Bakevellia,* Gervillia,* (Hoernesia), Schizodus, Pseudomonotis, Pleurophorus, Gastropoda. Bellerophon, Loxonema, Naticopsis, Pleurotomaria. Cephalopoda. Medlicottia,* Popanoceras,* Thallassoceras,* Cyclolo- bus, Temnocheilus. Pisces. Acrolepis,* Acanthodes, Cselacanthus, Amblypterus, Palseoniscus, Xenacanthus. * Fig. 106. SOME PERMIAN FISH. a. Platysomus striatus. &. Ccelacanthus gvanulosus. c. Palseoniscus comptus. The freshwater and terrestrial fauna is of special importance because it includes the earliest representatives of the Reptilia. The most conspicuous members of this fauna are : Crustacea. Uronectes ( = Gampsonyx). Amphibia. Archegosaurus, Actinodon, Branchiosaurus, Dasyceps, Lepidotosaurus. Eeptilia. Two orders of Reptilia are represented, viz. : (1) Rhyn- chocephala, by Proterosaurus and Palseohatteria ; (2) Theromorpha, by Pareiasaurus, Elginia, Gordonia, Geikia. 328 STRATIGRAPHICAL GEOLOGY The following are some of the most characteristic species of the English and German Permian : Bryozoa. Fenestella retiformis, Synocladia virgulacea. Brachiopoda. Camarophoria Schlotheimi, Strophalosia lamellosa, St. Goldfussi, Productus horridus, Dielasma elongata, Spirifer alatus, Lingula Credneri. Lamellibranchia. Pseudomonotis speluncaria, Pleurophorus costatus, Gervilia (Bakewellia) antiqua, B. ceratophaga, Schizodus obscurus, S. Schlotheimi. Gastropoda. Loxonema fasciatum, Natica minima, Pleurotomaria antrina, Turbo mancimiensis. Cephalopoda. Temnocheilus Frieslebeni. Pisces. Platysomus striatus, P. gibbosus, Palaeoniscus comptus, P. macropomus, Coelacanthus granulosus, Acrolepis Sedgwickii, Rhabdolepis macropterus. C. BRITISH PERMIAN EOCKS In spite of the fact that the British Permian succession is nowhere complete, and that it is uncertain whether the oldest part of the system is represented, Permian rocks cover considerable areas in England, and it will be convenient to describe them before dealing with those of the continent. The oldest Permian Beds must be those which are most nearly conformable to the Upper Carboniferous, and there can be no doubt that those of the Mid- land counties occupy this position. Those found in Devonshire resemble the Rothliegende of Germany, and may be of the same age ; while there can be no doubt of the Thuringian. age of the Magnesian limestones of the north-east of England. We shall therefore take these several areas in the above order. 1. Midland Area Beds which appear to be of Permian age, though no fossils have been found in them, occur in Worcestershire, Staffordshire, and Shropshire, extending as far north as Shifnal, and as far west as Shrewsbury and Alberbury, but not reaching into Denbigh nor into North Staffordshire. The Keele Group of the Upper Coal- measures was formerly regarded as part of them, and so closely are they connected with it, and so little evidence is there of any unconformity, that Mr. Cantrill regards them as an upward continuation of the Coal-measures. 1 He also considers them to be passage beds into the Permian and Trias, but they cannot bridge the whole interval between British Coal-measures and Trias. If they are Carboniferous they might be of Stephanian age, but until this is proved it is better to regard them as Permian and as a s flH 'r4 * arls oid b rates 330 STRATIGRAPHICAL GEOLOGY possible equivalent of the French Autunian. They can hardly be newer than that. Where fully developed this Midland Permian consists of three members or divisions as below. Feet. 3. Upper (or Enville) marls . . . 100 to 150 2. The Trappoid Breccia . . . 50 to 450 1. The Conglomerate Group . . .200 to 330 500 to 800 The thicknesses given are those found in the typical districts of Shropshire and South Stafford. When followed southward the conglomerate group thins out and the marls are overstepped by the Trias, only small patches of the breccia being seen at intervals below the latter. This Permian Series follows the eastern border of the Coal- measures through Shropshire and Worcester, passing southward on to the Silurian rocks of the Abberley and Malvern Hills. To the eastward it borders the southern half of the South Stafford- shire coalfield, but does not extend to its northern extremity. The Midland Permian, as above restricted, attains its fullest development in Shropshire, and is well exposed in the neighbour- hood of Enville on the Staffordshire border (see Fig. 107). Another typical district is that of the Clent Hills in Staffordshire, where the trappoid breccia attains its greatest thickness, but the upper marls are not seen, being presumably concealed by the overstep of the Bunter pebble beds (Trias). The following account refers more particularly to the Enville district, and has been compiled from a paper by Mr. W. W. King, 2 who has made a special study of that area. The Conglomerate Group. This consists of soft red sand- stones and red marls with three interbedded bands of calcareous sandstone and conglomerate, the latter consisting of various pebbles embedded in a sandy matrix, which is cemented by carbonate of lime. The second of these bands is the thickest, being sometimes a massive conglomerate 75 feet thick, and sometimes including lenticular beds of red sandstone with a total thickness of 150 feet. Both conglomerates and sandstones thicken to the north-west and become thinner to the south-west. In the Clent Hills this group is represented by red marls with three bands of sandstone and compact limestone or cornstone, without any conglomerate, but containing small fragments of the older rocks. The nature of the pebbles varies considerably in different parts of the district. Thus in Shropshire about half the pebbles consist THE PERMIAN SYSTEM 331 of dolomitic Silurian limestone like that which occurs in the Abberley Hills, the other half consisting largely of Carboniferous limestone and sandstone. In Staffordshire, especially near Baggeridge on the west, and near Barr on the east side of the coalfield, the majority of the pebbles (60 per cent) are of Carboni- ferous limestone, most of the remainder consisting of Wenlock limestone. It is thus a remarkable fact that these conglomerates are essentially limestone conglomerates, but contain the debris of various rocks down to and including the Woolhope limestone. The Trappoid Breccia was so called because it is largely made up of angular blocks of the compact felspathic lavas and tuffs which were formerly known as " traps." They are really rhyolites, hornstones, felspathic tuffs, grits, and agglomerates, and have evidently been derived from pre-Cambrian rocks like those which are known to occur in the Lickey Hills, at Barnt Green near Birmingham, and at Nuneaton. With these are associated many fragments of Llandovery sandstone and a few of Woolhope limestone. Mr. King remarks that the difference between the assemblage of rocks found in the conglomerates and that in the breccia may be explained on the supposition that during the accumulation of the former valleys in the adjacent country were being cut down through the Carboniferous and Silurian to the Woolhope limestone, while in the later epoch erosion was continued through the Llandovery Beds, and deep into the underlying Archaean rocks. It is also possible that the materials of the two deposits were brought from different directions. Another remarkable fact connected with this breccia is that many of the blocks are very large, and that some of them -are scratched or striated, the striations resembling those of glaciated stones, so that Sir A. Eamsay was led to infer the existence of glaciers and glacial moraines to account for them. This view, however, has not been sustained, and the striations are now attri- buted to the grinding of rock-fragments against one another during the many subsequent earth-movements which they have experienced. This breccia can be traced southward through Worcestershire, where it is often called the Haffield breccia, from its occurrence on Haffield Hill near Great Malvern. This is its most southerly outlier, but others occur on the Abberley Hills, resting on Silurian and Old Eed Sandstone. The Enville marls are only found in Shropshire, and are not seen farther south than Enville, though they doubtless extend much farther southward beneath the Trias. They include a band of Breccia which near Enville is 50 feet thick, but thins out to 332 STRATIGRAPHICAL GEOLOGY the north-west, while the thickness of marl increases in that direc- tion. It is clear, therefore, that the materials of this breccia, as of the main mass, came from the south and south-east. In Leicestershire there are a few small but interesting deposits of Permian age, which have been carefully studied and described by Dr. H. T. Brown 3 and Professor Bonney. 4 They consist of grey breccias, grey sandstones, and red marls, not more than 60 feet thick in Leicester, but as much as 200 at Polesworth in North Warwick. The breccias thicken to the south and the marls to the north, and their thickness may originally have been greater, as they are overlain unconformably by the Trias. The pebbles in the breccias consist of the following kinds of rocks : (1) The larger number (about 60 per cent) are felspathic grits or quartzites, derived from the Cambrian quartzites, but differing slightly from those exposed at Hartshill. (2) Gritty slates from the same series, averaging 17 per cent. (3) Flinty slates and argillites, probably Charnwood rocks. (4) Volcanic rocks, some being felsites and andesites from the pre-Cambriaii (Caldecote) Series, some of a rock which may have come from Charnwood, and diorites like those in the Cambrian near Nuneaton. (5) Fragments of Carboniferous grits, ironstones, and hematite ; but pebbles of Carboniferous limestone are rare, except at Poles- worth. In most of the exposed breccias over 80 per cent of the fragments come from the Cambrian quartzite series, and Dr. Brown shows that they have probably been derived from a buried ridge of these rocks which underlies Market Bosworth and runs parallel to the well-known Hartshill ridge. This buried ridge appears to be a faulted anticline, and so exact is the parallelism between these ridges and the principal faults of the Leicestershire coalfield that they may safely be attributed to the same period of earth-movement. These faults are known to be post-Carboniferous, and the undis- turbed way in which the breccias lie, bridging over the faults, makes it very probable that they are of Permian age. The view that these are terrestrial deposits of early Permian age will account for their proximity to the different set of deposits near Nottingham which belong to a late epoch in Permian time. 2. Devon and Somerset Area This area is taken next because its Permian rocks have a strong resemblance to the Saxonian fades, and especially to the Eothliegende of the Hunsruck district between Treves and Bingen, as pointed THE PERMIAN SYSTEM 333 out by Herr von Reinach. 5 The Devonshire Series is much thicker than that of the Midlands, but thins northward and appears to have been formed in an isolated lake basin which persisted throughout both Permian and Triassic times till it was merged into a larger inland sea (see Fig. 64, p. 201). The succession, where most fully developed in. South Devon near Teignmouth, Dawlish, and Exeter, is shown in Fig. 108, and h*as been described by Mr. Ussher 6 as follows : Feet. Red marls with occasional sandstones (? Permian) about 500 Red sandstones 250 to 400 Red conglomerate and breccia .... 900 to 1000 Dark-red Watcombe clays (local) . . . . 100 to 150 Maximum about 2000 The Watcombe clays occupy some little space north of Torquay by Watcombe, Barton, and Daccombe, and they reappear between Teignmouth and Bishopsteignton. They consist of fine dark-red clay which has long been used for the manufacture of terra-cotta ware, and also for the making of tiles and hard bricks. The deposit seems to fill a broad hollow in the Palaeozoic rocks, for it is overlapped by the breccias both to the south and north as well as westward. The overlapping breccias consist in the lower part of con- glomerate and breccia, in which fragments of Devonian limestone are abundant. These beds are well exposed in the fine cliffs between Babbacombe and the estuary of the Teign, and their thick- ness is estimated at from 400 to 500 feet, but the frequent faults make the estimate uncertain. The higher breccias differ in con- taining a larger proportion of derived volcanic rocks, chiefly of a peculiar red quartz-porphyry, and some of the boulders are of large size. This portion has a thickness of from 500 to 600 feet, so that the total thickness of breccia is about 1000. These breccias stretch northward beneath the Haldon Hills to Dunchideock and Ide. South-west of Exeter there are sheets of trachytic and andesitic lava at their base, indicating the contemporaneous existence of volcanoes, but the material of these lavas is quite different from that of the blocks in the breccia, which must have belonged to another set of volcanic eruptions, probably of rather earlier date. A similar breccia is found in the Crediton valley forming a long tongue which runs westward across the valley of the Tawe, and has an outlier at Hatherleigh (see Fig. 64, p. 201). North of Silverton the breccias pass into loose gravels which contain a different assem- blage of stones, evidently derived from rocks to the northward, and the whole series becomes thinner as it is traced into Somerset. 334 STEATIGKAPHICAL GEOLOGY Near Dawlish and Exeter the breccias graduate upward into sandstones, and these are succeeded by red marls with lenticular beds of sandstone, which seem to be an upward continuation of the series. These marls may be of Thuringian age, and consequently homotaxial equivalents of the Magnesian Limestone Series described below, though never actually connected with that series. How far these Permian rocks extend eastward beneath the Trias and newer strata we have no means of knowing, but they certainly had a considerable extension southwards, for patches of them occur between Bolt Head and Plymouth, and fragments of red breccia have been dredged from many par.ts of the sea-floor south- east of Cornwall, in such fresh condition as to indicate the existence of submarine reefs of this rock. 3. North-Eastern Area In this area we have undoubted representatives of the Thuringian stage in the form of fossiliferous magnesian limestones with intercalated red marls. The outcrop of this series is shown on the map (Fig. 87), from which it will be seen that although the strike of the Permian Beds is at first nearly in accordance with that of the underlying Coal-measures, yet they do not participate in the anticlinal flexures which give an east and west strike to these measures in South Durham. Near the valley of the Tees the Permian oversteps the Coal-measures, and rests first on the Mill- stone grit, and then on Carboniferous limestones. Farther south it again crosses the Millstone grit onto the Coal-measures of York- shire and Nottingham, but it is everywhere unconformable to the latter. There is a marked lithological change in this series as it is followed from north to south. In Northumberland and Durham it is composed almost entirely of magnesian limestones (i.e. dolomites), and the complete succession of beds, according to Dr. Woolacott's 7 recent examination of them, is as follows : Feet. Red marls with thin limestones and beds of salt .... 500 Upper limestones, bedded for the upper 100 feet, con- cretionary for some 200 feet, with a bed of flexible lime- stone at the base (10 feet) 300 Middle limestones bedded yellow limestones on the coast replaced westward by compact fossiliferous limestones, both often much altered and brecciated . . . 150 to 300 Lower bedded brown limestones . . . . . 40 to 200 The Marl Slate (a brown calcareous shale) ..... 3 The Yellow sands (soft and without fossils) . . . to 150 Total about 1200 THE PEKMIAN SYSTEM 335 The fauna is restricted (about 140 species) and peculiarly distributed ; fossils are most abundant in the Middle limestones, which seem to have formed a shell bank, and only a few Lamelli- branchs and Gastropods survive into the Upper Beds. Fish remains occur at two horizons the Marl Slate and the flexible limestones, and in the beds just above these deposits. Productus horridus is characteristic of the Lower and Middle limestones. For other common fossils see Fig. 105. In Yorkshire, near Pontefract and elsewhere, the limestones are thinner, but the marl slates are thicker than in Durham, and sometimes rest on white and yellow sands, which in turn rest on an eroded surface of red marls and sandstones belonging to the Upper Coal-measures (see p. 298). In South Yorkshire the succession may be summarised as follows : Feet. Upper red marls ....... 50 Upper (Brotherton) limestone . . . . 50 to 120 Middle red marls 30 to 50 Middle limestone, thick-bedded . . . . 150 to 200 Lower limestone about 120 Blue limestones and shales . . . . 5 to 15 Quicksands and breccia . . . . . 10 to 20 From 430 to 570 Traced southward into Nottinghamshire, the Upper limestone thins southward and is at the same time overlapped by the Trias, so that the thickness of the Permian at its outcrop is very much diminished, being as follows : Feet. Middle marls and sandstones . . . . 20 to 30 Magnesian limestone (lower) . . . . . 60 to 70 Marl slates . 30 to 50 Southward near Kimberley the limestone passes into a yellow calcareous sandstone about 30 feet thick, resting on 15 to 20 feet of shaly marl, with a few feet of coarse breccia at the base. This change and diminution in thickness is evidently an indication that we are in this direction approaching the southern shore of the Permian Sea. The breccia contains angular fragments of sandstone and shale derived from the Coal-measures on which it rests, together with pebbles of slate, quartz, and quartzite, and it is not unlikely that these were derived from a northerly prolongation or the Charnwood rocks. From borings made in search of coal near Haxey in Lincoln- shire, and near Newark on the eastern border of Notts, it is known that the Permian limestones and marls exist in full force below these places. At Haxey the Permian was entered at 1183 from 336 STRATIGRAPHICAL GEOLOGY the surface, and its base is supposed to be at 1728, but may be 68 feet lower. The beds traversed were : Feet. Upper marls with gypsum ... 58 Upper limestones Middle marls with gypsum Lower limestones Doubtful beds (sandstones) 84 134 270 At Newark it consisted mainly of magnesian limestones, with 118 feet of red marl above them, the total being 519 feet, and the base is there clearly marked by a bed of grit and breccia. 4. North-JVestern Area and Ireland On the western side of the Pennine ranges only a few isolated tracts of Permian strata are found, but as they include fossiliferous magnesian limestones, there can be little doubt that they were originally connected with the eastern area and are remnants of an extensive series of deposits which covered a large part of Northern England and spread across the space now occupied by the Irish Sea into the north of Ireland, where patches still exist as far west as Tyrone. Beginning with the southern part of this area a strip of Permian is found near Stockport and Manchester, consisting of : Feet. 2. Marls and limestones with marine fossils . . 230 1. Soft red and variegated sandstones . from 300 to 1000 Fossils found in the upper beds at Fallowfield include Pleuro- phorus costatus, Schizodus Schlotheimi, Bakevellia antiqua, Aucella Hausmanni, and Turbo helicinus. Still farther north several small outlying patches of Permian have been found, one east of Preston, another near Clitheroe, where it lies on Carboniferous limestone, and a third rather larger area in the centre of the Burton coalfield ; these occurrences show the beds to be completely unconformable to the Carboniferous rocks, just as on the eastern side of the Pennine anticline. The largest Permian tract is that on the western side of the Vale of Eden, extending from Kirkby Stephen in Westmoreland by Appleby and Penrith to within about 3 miles of Carlisle, where it is faulted down below the Triassic sandstones. Here the lowest member of the series is a bright-red sandstone, known as the Penrith sandstone and having a maximum thickness of 1500 feet. In the southern part of the Vale this red sandstone includes THE PERMIAN SYSTEM 337 beds of breccia and conglomerate (locally called brockram) con- sisting of fragments of limestone embedded in a sandy matrix. These breccias are thickest in the south-east, being sometimes over 100 feet thick, and they thin out northwards till at and north of Penrith they have quite disappeared, and only red Penrith sand- stone remains. The surfaces of some of the sandstone beds bear numerous foot- prints, some of which seem to be tracks of Labyrinthodonts or other amphibians, and some of reptiles (the lacertoid prints). Above these sandstones are shales and magnesian limestones ; at Hilton Beck, near Appleby, the former are about 40 feet thick and contain remains of the following plants : Ullmannia Bronni, U. selaginoides, Alethopteris Goepperti, Sphenopteris Naumanni, N. dichotoma, Cardiocarpum triangulare, and Odontopteris sp. These Penrith Beacon. Melmerby. Fig. 109. SECTION THROUGH THE PERMIAN BEDS NEAR PENRITH (HarknCSS). A. Red sandstone. B. Red shale. C. Magnesiau limestone. .D. Penrith sandstone. E. Carboniferous. F. Ordovician. shales thin out northwards, but the magnesian limestone extends rather farther ; where thickest there is only 25 feet of it, but probably it is only the base of a much thicker mass since removed or else concealed by the unconformity of the Trias. Other small patches of these Permian deposits occur in the west of Cumberland, as on the coast at St. Bees Head, and inland at Egremont and Arlecdon. It should also be mentioned that the Permian Series above described is everywhere overlain by a set of red shales and marls with beds of gypsum, and by some these are classed as Permian, but Mr. Goodchild has pointed out that they rest sometimes on the limestone, sometimes on the Hilton shales, and sometimes on Penrith sandstones, so that there is a decided unconformity, and the marls should be regarded as the base of the Trias. Permian Beds are found also in the Isle of Man and again at three places in Ireland, one at Cultra on Belfast Lough, another z 338 STRATIGRAPHICAL GEOLOGY near Ardtrea in Tyrone, at both of which they include fossiliferous magnesian limestones ; at the third place (Armagh) there are only boulder-beds and breccia resting on Carboniferous limestone. 5. Scotland In Dumfries and Ayrshire there are tracts of bright-red sand- stone which have much resemblance to the Penrith sandstone, but till recently no evidence for their identity was forthcoming, and consequently they were regarded by some as belonging to the Trias, and were so classed in the last edition of this book. Mr. G. Hickling, 8 however, has compared the Reptilian footprints of the Dumfries sandstones with those of the Penrith Beds, and finds that though not identical they are similar, and differ from those on Triassic sandstones, and further that two of the types seem to be identical with prints in the Permian sandstone of Mansfield (Nottingham). Dumfries. In this district the three largest tracts are (1) one extending from the Solway Firth by the mouth of the Nith to and beyond the town of Dumfries, (2) round Lochmaben in Annandale, and (3) round Thornhill in Nithsdale. Professor Harkness was of opinion that the general succession of the beds near Dumfries and Lochmaben is as follows : Feet. 4. Thin-bedded sandstones with beds of clay . . . 300 3. Hard massive breccias and conglomerate . . . 300 2. Soft red current-bedded sandstones\ -QQ 1. Coarse sandstones and breccias J ' The breccias consist of angular blocks and fragments of various Palaeozoic rocks, with some of granite and other igneous rocks, embedded in red sand. The lower sandstones are brick-red in colour, frequently flaggy and divided by thin seams of dark-red clay, and their surfaces often bear the footprints of Labyrinthodont and other reptiles. Such footprints were first described by Professor Buckland from specimens found in the quarries on Corncockle Moor near Lochmaben, but they are equally common near Dumfries ; they also occur in the upper sandstones, the beds of which are separated by rather thicker layers of clay. In the Thornhill basin farther north there are no breccias, and the greater part of the tract consists of brick-red sandstone, but in the northern part lava-flows and volcanic detritus are interstratified with the lowermost sandstones. The higher sandstones contain beds of shale and clay, and seem comparable to the higher beds near Dumfries. THE PERMIAN SYSTEM 339 Ayrshire. In the centre of the Ayrshire coalfield near Mauchline lies a pear-shaped area, occupied by rocks similar to those of Thornhill (see Fig. 110), and associated with the volcanic breccias of this district are some beds of a purple sandy limestone, but no fossils have been found in them. The igneous zone consists of a number of lava-flows which mark successive eruptions, and have probably proceeded from more than one vent, for though they are all of a basic character they vary much in lithological composition. Some of them are basalts, others are magnetite-felspar rocks, while the rock of Mauchline Hill is a picrite composed chiefly of olivine and augite. Magnetite is prevalent in all of them, and its oxidation has given them a red colour, on which account they have been described as " porphyrites." Besides these interbedded volcanic rocks there is other evidence of the number and activity of the volcanoes. Outside the existing tracts of Trias the Coal-measures are pierced by numerous large e Fig. 110. SECTION THROUGH THE PERMIAN BASIN OF AYRSHIRE (Geikie). f. Basaltic dyke. d. Red sandstones. 6. Lava-flows. e. Volcanic necks. c. Volcanic tufls. a. Carboniferous rocks. pipes filled with a coarse unstratified agglomerate, and usually appearing at the surface as small rounded hills or hillocks. Some- times the agglomerate includes a core or column of lava, and the coal-seams in their neighbourhood are often charred or altered into columnar coke. There can be little doubt that these columns or necks are some of the vents through which the volcanic materials were ejected. The necks are especially numerous in the Dalmel- lingtoii coalfield, but they occur also in Eenfrewshire and Lanark- shire, piercing the highest Coal-measures in those counties. Moray. The only other area where Permian Beds have been recognised is in Moray, one small tract lying to the east of the town of Elgin and another near Cummingstone. The rock is a coarse red false-bedded sandstone, consisting of rounded quartz-grains, and containing sand-worn pebbles ; the total thickness of the beds, according to Mr. D. M. S. Watson, 9 being from 400 to 500 feet. Apparently it was formed in a depression or lake in a wind-swept desert region, for near Elgin it has yielded bones of the character- istic Dicynodont reptiles Gordonia, Elginia, and GeiJcia. The similar sandstone near Cummingstone has not yielded bones, but shows footprints which closely resemble some at Mansfeld in Notts, 340 STRATIGRAPHICAL GEOLOGY and in the Permian of Thuringia. Mr. Watson suggests that they may have been made by Gordonia. D. CONTINENTAL PERMIAN STRATA 1. France Small tracts of Permian rocks occur at many places round the great Central Plateau of France (see map, Fig. 102), most of them abutting against the ancient rocks of this plateau, and passing out- ward beneath the newer deposits which surround it. They appear to belong to two separate areas of deposition, the one group on the north extending from a point south of Berry by Buxiere and Moulins, and reappearing on the Morvan near Autun where there are two basins or troughs. The other area lay to the south and south-west of the plateau, patches of its deposits occurring near Brive, near Rodez in Aveyron, near Carmaux in the Tarn, and again from Broquies in a long strip to Grassessies on the Montagne Noire. The beds of the Autun basin form the type of the Autunian stage. They overlie the Stephanian measures, and the lowest set of beds which are classed as Autunian are really passage-beds, con- taining a flora like that of the beds below, except that it includes Walchia piniformis and Callipteris conferta. These beds are about 1300 feet thick and are entirely of freshwater origin, consisting of sandstones and bituminous shales, with beds of magnesian lime- stone full of Cyprids ; and they form the zone of Igornay. They are succeeded by another group of similar rocks (zone of Muse) with two valuable beds of oil-shale, and a more typical Permian flora. Lastly there is an upper stage (zone of Millery) about 1500 feet thick, consisting of shales and thin coal-seams, one of the latter being the famous Boghead coal. The whole series is about 4000 feet. This Autunian Series has yielded a rich harvest of vertebrate remains ; the Amphibia have been described by Gaudry, and include Protriton petrolei, Pleuroneura Pellati, Actinodon Froissardi, and Stereorachis dominans. Fish are numerous, and silicified stems of tree-ferns (Psaronius) are common in the upper beds. At Bourbon and Buxiere the series is not so thick, and consists more largely of felspathic sandstones derived from the erosion of the neighbouring granitic and gneissic rocks. The lowest beds, however, include bands of oil -shale and black limestone, and doubtless belong to the Autunian lake-basin. Wherever the summit of the Autunian is exposed, it is over- THE PERMIAN SYSTEM 341 lain by sandstones of Saxonian age, grey in the lower part, red and mottled in the upper part, with a combined thickness of about 1600 feet. Nothing comparable with the Zechstein or Magnesian Limestone Series has been found in any part of France. Of the more southern Permian facies that of Aveyron and Lodeve may be taken as typical, the succession in that district being : Saxonian, /Fine sandstones and black shales passing down into red 1500 feet \ sandstones and marls with Walchia and Callipteris. A ( Lodeve slates with plant remains. -OoT^ I Black shales with Pal <* oniscus ' [ Sandstones, shales, and a basal conglomerate. The small basin of Littry in Calvados deserves mention because of its proximity to England, and because the deposits which there overlie the Stephanian are referred by French geologists to the Autunian. They consist of red sandstone, succeeded by a bed of magnesian limestone and shales from which remains of Palceoniscus and Amblypterus have been obtained. 2. Germany Permian rocks occupy much larger areas in Germany than they do in France or Britain, and two of these areas may be specially noticed, i.e. that of the Sarre district and that of Saxony and Thuringia. Sarre Valley. The most extensive area of the Lower Kothlie- gende or Autunian is that of the Sarre and Nahe district, where they occupy a wide space between the Devonian of the Hunsruck on the north and the outcrop of the Trias on the south. Here, as shown in Fig. 103, the Ottweiler Beds are overlain conformably by the Ousel Beds, consisting of red and grey sandstones and shales with thin layers of coal, and as at Autun they have a Carboniferous flora to which a Permian aspect is imparted by the presence of Walchia piniformis, Callipteris conferta, and Galamites gigas. The succeeding Lebach Group comprises sandstones at the base, overlain by shales containing ironstone nodules in which are fish remains, scales of Archegosaurus and the Crustacean Gampsonyx. Interbedded with these sediments are sheets of porphyritic andesite and melaphyr. The Saxonian is represented by a series of red conglomerates, sand- stones, and shales, the materials of which have been mainly derived from the detrition of the neighbouring Devonian and Carboniferous rocks. These " Upper Rothliegendes " have a much wider extension northward and eastward than the Autunian, overlapping the latter on to the older rocks. 342 STRATIGRAPHICAL GEOLOGY Central Germany. In the central part of Germany (Thuringia and Saxony) the passage-beds are absent, and the Permian is more or less unconformable to the Carboniferous ; but the Lebach Group is well represented, and has yielded many amphibian remains, e.g. of Branchiosaurus, Archegosaurus, and Pelosaurus. Its chief develop- ment is in the Erzgebirge basin near Chemnitz and Zwickau, where it consists of coal-bearing beds, with the usual plants in the lower part, together with beds of volcanic tuff and sheets of trachytic lava ; and these beds are overlain by a set of red breccias, conglomerates, and sandstones, largely composed of the debris of contemporaneous volcanic rocks. The Saxonian of this region is a thick mass of conglomerates, sandstones, and marls, the predominant colour of which is red or reddish brown. These beds have a thickness of about 1600 feet in Saxony and nearly 2000 in Thuringia. No fossils occur in them except plant remains and the footprints of reptiles and amphibians, Fig. 111. SECTION THROUGH THE ERZGEBIRGE BASIN (by Siegert). v. Sandstones and tuffs. r. Lebach Beds. s. Silurian. P, t. Lavas and tuffs. c. Coal-measures. p. Archaean. so that the whole formation up to the top of the Saxonian appears to be of lacustrine and terrestrial origin. The succeeding division is largely marine, and includes the Zechstein limestone, but as that is really only a small portion of it the name Thuringian is a much better one. This group of beds occupies a considerable area in Central Germany, spreading round the mountain ranges of the Thuringerwald, the Frankenwald, the Eeisengebirge, and the Harz, and passing everywhere below the overlapping Trias and Jurassic strata. In its typical development it is a series of limestones and shales of no great thickness, from 400 to 500 feet, but in certain localities it is expanded by the intercalation of enormous masses of gypsum and rock-salt. In the Mansfeld district it has the following components : Upper /Red clay and gypsum. 200 feet\ Clays with lenticular beds of limestone. Middle /Dark bituminous limestone and shale. 150 feet\ Gypsum and dolomitic breccia. I" Magnesian limestone (the Zechstein). I Bituminous shale (Kupferschiefer). * ^Calcareous sandstone and conglomerate. THE PERMIAN SYSTEM 343 The similarity between this series and the " Magnesian lime- stone " of England will at once be apparent. The layer of copper- bearing shale which overlies the basement beds is only about 2 feet thick, and is the analogue, if not the actual continuation, of our " Marl slate " ; it yields similar fish remains (Platysomus gibbosus and Palceoniscus Frieslebeni being common species), with scales of Proterosaurus, and occasionally star -fish and Brachiopods, which clearly indicate the invasion of the area by an outer sea. The fossils of the Zechstein are similar to those of the English limestones, many of the species being the same, and the only Cephalopoda being a few species of Orthoceras and Nautilus. Some of the beds are almost entirely composed of Bryozoa, species of Fenestella, Acanthocladia, and Phyllopora, and these beds are called " Bryozoon-reefs " by German geologists, a misleading term suggest- ing reef- formation which does not exist, for they are merely lenticular beds of limestone resulting from local Bryozoon growths, like the crinoidal and coralliferous limestones of Devonian and Silurian age. In Hesse the composition of the Thuringian differs little from the above, but north of the Harz district thick beds of gypsum and rock-salt come in, which are worked at Stassfurt, Egeln, Vienenburg, and other places. The rock-salt is sometimes over 1000 feet thick, and at Sperenberg (south of Berlin) it is over 3000. The typical Zechstein facies of the Thuringian extends also eastward into Silesia (north of Gatz). 3. Russia In Kussia, as already stated, the Permian is largely marine, but the limestones with marine fossils are often interbedded with sand- stones, shales, and marls of estuarine or lacustrine origin. According to Nikitin, in the typical district of the ancient kingdom of Permia now the provinces of Perm, Samara, and Ufa the series is divisible as follows : 10 ( Red marly sandstone. Thuringian -| Red marly limestone. ( Grey limestones and marly sandstones. v , . f Red sandy marls (marine fossils). Kostromian ( Marly limestone without fossils. relays with gypsum and beds of limestone. Artinskian 4 Dolomitic limestones with Permian fossils. [Dolomitic limestones (Perino-Carboniferous). The stages were not established when Nikitin wrote, but I have bracketed his subdivisions to show how they probably correspond with the grouping made in the Ural region. On the western slopes of the Urals the Fusulina limestone of the Uralian is succeeded by a series of sandstones, limestones, and 344 STEATIGKAPHICAL GEOLOGY marls, partly lacustrine and partly marine, the fauna including some survivals from the Carboniferous, such as Productus cora, P. punctatus, and a species of Phillipsia, but associated with a new and peculiar set of Ammonoids, i.e. Medlicottia, Popanoceras, Agathiceras, and Thalassiceras. The plant remains again are chiefly Permian forms, including Galamites gigas and Callipteris conferta, with several species of Walchia and Ullmannia. This set of beds was called the Artinskian stage by Karpinsky in 1874, and it is clearly a marine representative of the western Autunian. In the same area, overlying the Artinskian, is Sibertzew's n Kosstroma stage, which is evidently the equivalent of the Saxonian. It presents two facies, the one of marls and sandstones (some grey and some red), containing plant remains with Carbonicola and bones of Palceoniscus and Archegosaurus ; the other consisting of various kinds of limestone (oolitic, compact, marly, etc.), with a marine fauna, including Productus Cancrini, Athyris Roissyi, Dielasma elongatum, and Bakevellia parva. Still higher are limestones with many Zechstein fossils, but it is noteworthy that Productus Cancrini takes the place of P. horridus, and Strophalosia horrescens that of 8. Goldfussi. These are overlain by red sandy marls, containing freshwater Unionidse (Naiadites and Palceomutela), which Nikitin proposed to call the " Tartarian stage." If adopted as an equivalent for the whole of the Kussian Thuringian the name might be useful. In Vologda (Northern Eussia), near the sources of the Little Dwina and Suchona, certain marls and sandstones belonging to the upper stage have acquired importance from Amalitsky's discovery in them of the remains of many reptiles Pareiasaurus, Dicynodon, Rhopalodon, Elginia, Gordonia, and others, associated with many Unionidae (Palceanodon, Palceomutela, Oligodon, Carboni- cola, etc.), and a flora of mixed Asiatic and European types, i.e. species of Glossopteris and Gangamopteris, with Callipteris, Tceniopteris, Schizoneura, etc. 12 E. HISTORY OF THE PERIOD From the stratigraphical facts, which have been summarised in the preceding pages, we may gather that the Permian period was a natural continuation of the Carboniferous, and that the earliest Permian sediments were accumulated in the same areas and under almost the same physical conditions as those of the preceding Stephanian epoch. The western half of the European area was clearly part of a large continent which extended westward far into the Atlantic THE PERMIAN SYSTEM 345 region, and eastward to the borders of Russia. The northern part of this continent seems to have been a hilly and mountainous region, wherein little deposition of any kind took place until quite the later part of the period. Across the centre of the continent, and traversing its whole breadth from west to east, ran the great Armorican and Hercynian ranges, probably consisting of many mountain ranges with intervening valley-like depressions. To the south of these ranges the country sank to lower levels, and here there were large lakes and swampy depressions in which the Autunian deposits were formed. In these tracts and along the valleys of the rivers which carried the rainfall from the hills to the lower lands grew a rank flora of Carboniferous aspect ; and here dwelt many kinds of Amphibia, both small and large, and some primitive Reptilia, the larger of which doubtless fed on the fish which inhabited the lakes and rivers. Such is the picture that we may form of Europe in early Permian time. The nature of the Saxonian deposits, and the great contrast which they present to the underlying Autunian, indicates that a great change took place ; and we may infer that this was caused by a general uplift of the whole continental region, whereby the lakes and marshes were drained, and its lower levels were converted into broad plains separated by ranges of hills and mountains. This change produced a corresponding alteration in the climate of the country, so that its more southern parts passed from a condition of great humidity with mild winters to one of comparative dryness and very cold winters. Great terrestrial erosion and detrition was another result of the upheaval, leading to the production of the thick deposits of sand- stone and conglomerate which form the chief Saxonian accumula- tions, and to the formation of breccias round the mountain ranges. These breccias are in themselves strong evidence of peculiar climatic conditions. They have been compared, both by Mr. R. D. Oldham and by Professor T. G. Bonney, with the fringes of angular gravel and breccia which flank the hill ranges of Persia and other parts of Central Asia at the present day. Referring to the Permian breccias Professor Bonney 13 remarks that the resemblance above mentioned suggests that the climate, both of Germany and Britain, was also similar to that of modern Persia, hot in summer but cold in winter. He also quotes Professor Garwood as pointing out that the distance to which some of the large blocks have been transported can be explained on the supposi- tion of snow-slopes, without bringing in the agency of glaciers as some have thought necessary. Professor Garwood had observed in Spitzbergen that the talus-slopes on the hillsides were deeply 34:6 STRATIGRAPHICAL GEOLOGY covered by snow in the winter, and that fragments falling on this snow-slope slide down to its outer edge. This process is repeated year by year, so that the talus fringe is continually being extended, and blocks of large size are conveyed to a much greater distance from their sources than would be possible without the snow-slope. Another accompaniment of this Permian upheaval was the outbreak of volcanic action and the establishment of numerous volcanoes, from which frequent eruptions of lavas and ashes took place. Nor are there wanting evidences of earthquakes, especially in Devonshire, where round Brixham and Torbay the Devonian limestones are riven by fissures from a few inches to 3 feet in width, and these are filled with red sandstone similar to the Permian sandstones, of which patches still remain on the surface of the limestone plateau. This epoch of high elevation may have lasted for a long time, but at length another change took place ; volcanic activity ceased, and as so often happens in such cases a general sinking in and subsidence ensued. Some of the great plains on which the red sandstones and breccias had been accumulated sank below sea-level, and were invaded by the Eastern Sea, becoming a large gulf or land- locked " Mediterranean " -sea in which deposits of dolomitic lime- stone, shale, and marl were formed. Toward the close of the period, however, this inland sea became isolated, probably by local warping and other irregular crust-move- ments, so that it was cut off from the outer sea and converted into a great salt lake like the Caspian Sea of the present time. From time to time during dry seasons portions of this sea were dried up, and the salts in solution were precipitated to form the thick masses of gypsum and rock-salt which are such conspicuous members of the highest Permian deposits. REFERENCES 1 Cantrill, Quart. Journ. Geol. Soc. vol. li. p. '528. 2 King, Quart. Journ. Geol. Soc. vol. Iv. p. 97. 3 H. T. Brown, Quart. Journ. Geol. Soc. vol. xlv. p. 22. 4 Bonney, Midland Nat. vol. xv. p. 25 (1892) ; and Quart. Journ. Geol. Soc. vol. Iviii. p. 188. 5 Reinach in Ussher, Geol. Mag. vol. xviii. 6 Ussher, Quart. Journ. Geol. Soc. vol. xxxiv. p. 459. 7 Woolacott, Quart. Journ. Geol. Soc. vol. Ixvii. p. 312. 8 Hickling, Mem. Manch. Lit. and Phil. Soc., 1909. 9 Watson, Geol. Mag., 1909, p. 102. 10 Nikitin, Gongris geol. internal. (1897), Guides Nos. 1 and 2. 11 Sibertzew, Mem. Com. Geol. Russe, xv. (1899). 12 Araalitsky in Geol. Mag., 1901, p. 231. 13 Bonney, Quart. Journ. Geol. Soc. vol. Iviii. p. 185. CHAPTEK XII THE TRIASSIC SYSTEM WITH the close of the Permian period the last remnants of Palaeozpic life became extinct in the British area, and when after a long continental period the western and northern parts of Europe were again submerged beneath the sea, the waters were tenanted by a very different set of generic forms. The Palaeozoic Groups of Cystid and Blastoid Echinoderms, Eugose Corals, Eurypterids, Trilobites, Goniatites, and most of the Palaeozoic genera of Nauti- loidea had died out, their place being taken by the Euechinoids, Aporose Corals, Siphonate Gastropods, and Cephalopods of the Ammonite and Belemnite type ; many orders of Eeptiles appear for the first time, and the earliest Mammal yet known occurs in the upper beds of the Trias. The fish of the Triassic and Jurassic periods also exhibit remarkable changes and developments. The Rhizodont Crossop- terygian fishes of the Devonian and Carboniferous, and the hetero- cercal Palaeoniscids of Carboniferous and Permian time now become scarce ; while a new type of fish (the Protospondyli} with a nearly homocercal tail and a more bony endoskeleton having a tendency to form vertebrae, make their appearance and lead up to the com- pletely vertebrate Teleosteans. Even where a continuous series of marine deposits bridges over the gap in the history of marine life which exists in Northern Europe we find that the above remarks are still practically true, the only exception being the existence of a few Gastropod genera and a few species of Orthoceras in the Trias of the Alps. For reasons which cannot yet be fully explained the Triassic period seems to have been one of rapid organic change and development, and in no class is this more beautifully illustrated than in that of the Cephalopoda, the Alpine Trias presenting us with a remarkable assemblage of genera which practically connect the Palaeozoic Gonia- 347 348 STRATIGRAPHICAL GEOLOGY tites with the Ammonites which are so characteristic of the Mesozoic periods. A similar dying out of old forms and a rapid spread of newer types took place in the vegetable world. The Palaeozoic Lyco- podiales Lepidodendron, Bothrodendron, and Stigmaria have now all disappeared, while the Cycads which appeared in Permian time now become much more numerous, and are associated with the coniferous Voltzia. Among ferns, though some of them have been referred to Pecopteris and Neuropteris, it is very doubtful whether any belong to these Palaeozoic genera ; most of them are rather referable to the more recently established genera Chladophlebis and Todites. Many other new genera make their appearance. Here, therefore, at the base of the Trias is drawn the line between the Palaeozoic and Neozoic eras, and here begins that division of Neozoic time which is often called the Mesozoic. It must not be supposed, however, that there is any great break between the two series of formations, for though there is a partial break in Britain, there is complete conformity between the Permian and the Trias, with a passage from one to the other in the Alpine region, and in most parts of Germany. THE TRIASSIC SYSTEM A. NOMENCLATURE AND CLASSIFICATION The Triassic System resembles the Devonian System in that it presents us with two very different facies, (1) a terrestrial and lacustrine facies formed within the limits of a large continent, and (2) a marine facies of the usual kind. The first forms the upper part of what used to be known as the New Red Sandstone (see p. 6), but was called the Trias by von Alberti in 1834 on account of its obvious tripartite arrangement in Germany. In England, however, it has only two parts, and these are both of terrestrial or lacustrine origin ; in Germany these persist, but the middle beds are lime- stones containing marine fossils, and consequently there are three series, which have received the following names : 3. The Keuper (red marls and sandstones). 2. The Muschelkalk (marine limestones). 1. The Bunter (variegated marls and sandstones). For the normal marine facies of the Trias we must go to the Alps, and the student cannot understand the true relations of the Triassic System to those which precede and follow it until he has learnt something about the important marine facies of the forma- THE TRIASSIC SYSTEM 349 tion which exists in the Tyrol and extends throughout Southern Europe. Moreover, this marine facies has a wide extension over other parts of the world ; it spreads through Southern Europe from Spain to Hungary and the Balkans ; it is found in Turkestan, Central Asia, and the Himalayas, and a similar marine Triassic System occurs in Siberia, Japan, North America, and again south- ward in Columbia, Peru, New Zealand, and Australia. In South Africa, in South India, and in Argentina red sandstones with reptilian remains are found, and these regions seem to have formed part of other Triassic continents. In this volume, however, it is only possible to give a brief account of the European Trias. The name Trias, as will be seen, is not a good one, being really applicable to the German facies only, which is a limited and excep- tional one. A nomenclature capable of wider application will doubtless be eventually proposed, and must, of course, be based on the pelagic development of the system. From the recent work of Mojsissovics, Waagen, and Diener on the distribution of the fossils in the pelagic facies, it appears that the system falls more naturally into two great divisions, a lower and an upper, each of which is again divisible into two series. B. LIFE OF THE PERIOD 1. The Marine Fauna The fauna of the Alpine and Tyrolian Trias, though consisting mainly of Neozoic forms of life, yet includes some survivors of Palaeozoic types ; the proportion of species belonging to genera which are essentially Paleozoic being stated as only one-tenth of the total number of species composing the fauna. The Cephalopoda are a remarkable assemblage. A single species of Orthoceras with some of Temnocheilus and Pleuronautilus connect it with Palaeozoic faunas, while the genus Nautilus (as now restricted) makes its first appearance. The Arnmonoidea, however, are the dominant forms, and include many genera with nearly simple lobes and saddles, such as Ceratites, Lobites, Tirolites, Trachyceras, Tropites, as well as genera with more complex lobes, such as Arcestes, Cladiscites, Ptychites, Pinacoceras, Phylloceras, Rhacophyllites, and Megaphyllites. Dibranchiate Cephalopoda make their first appear- ance and are represented by precursors of the Belemnitidae in Aulacoceras and Atractites. Gastropoda are abundant, and include representatives of the Palaeozoic genera Murchisonia, Loxonema, Naticopsis, and Euom- 350 STRATIGRAPHICAL GEOLOGY phalus, with the Neozoic types of Trochus, Turritella, Scalaria, Ccelostylina, Cerithium, and Pseudomelania, as well as peculiar forms which have been named Naticella, Chilocyclus, and Ptychostoma. Of Lamellibranchs Posidonomya, Megalodon, and Myophoria (which is probably inseparable from Schizodus) connect the fauna with Palaeozoic life, while certain genera' are especially Triassic, such as Halobia ( = Daonella\ Monotis, Cassianella, Trigonodus, and Anoplophora. There are many representatives of Pteria, Gervillia (Hcernesia), Pecten, and Lima ; and the following genera make their first appearance, Velopecten, Plicatula, Mytilus, Lucina, Sphceriola, Cardita, Palceocardita, Unicardium, Protocardium, and Homomya. The Brachiopoda include the last survivors of the ancient genus Spiriferina, with some specially Triassic forms, such as Koninckina, Thecospira, and the Athyroids which Bittner has separated under the names of Tetractinella and Pentactinella. Besides these, species of Terebratula and Rhynchonella are abundant. Among Echinoderms Cidaris, Hemicidaris, Mesodiadema, and Tiarechinus make their appearance, and the last is restricted to the Trias. Of Crinoids Dadocrinus, Isocrinus, and Encrinus are found. Corals (Anthozoa) also occur, especially in the limestones of the Upper Trias, and belong chiefly to Cladophyllia, Calamophyllia, Thecosmilia, Isastrea, and Thamnastrea. Calcareous algae, however, are the chief limestone builders, and have been described under the names of DiplopoYa and Gyroporella. 2. Lacustrine and Terrestrial Fauna This is found in the Bunter and Keuper Series of England, Germany, and France, the animals being those which lived in or near the large inland salt lakes and lagoons, the plants those which grew on their borders and by the rivers which ran into them. Plantss. The ferns are represented by the following genera : Anomopteris, Chelipteris, Chladophlebis, Lacopteris, Coniopteris, Neuropteridium, and Tceniopteris. The Cycads Zamites, Otozamites, and Pterophyllum are common plants. Conifers are represented by Voltzia, Albertia, Baiera, and Schizolepis. The horse-tail Equisetites also makes its appearance, as well as a new genus Schizoneura. Mollusca. Shells of this class are very rare, but a few resem- bling Unio and Anodon have been found, and others in the English Keuper which resemble Thracia, Goniomya, and Pholadomya. Crustacea. This class is represented only by the Phyllopod Estheria, which is abundant in the Keuper. Pisces. Remains of many kinds of fish occur, especially in the Keuper Beds, the more important genera being Acrodus, THE TRIASSIC SYSTEM 351 Hybodus, Ceratodus, Dipteronotus, Gyrolepis, Semionotus, and 6 ^w . 7 8 I 10 Fig. 112. GROUP OF FOSSILS FROM THE ALPINE TRIAS. 1. Pseudomonotis Clarai. 2. Naticella costata. 3. Tetractinella trigonella. 5. Ptychites Studeri. t>. Ceratites cassianus. 7. Trachyceras aon. 8. Halobia rugosa. 23. Dadocrinus gracilis. 10. Myophoria Kefersteini. 11. Cidaris dorsata (spine). 12. Chemnitzia gradata. 14. Arcestes tornatus. 15. Pinacoceras Metternichi. 16. Tropites subbullatus. 22. Monotis salinaria. Saurichthys. The ridged teeth of Oeratodus are abundant in some 352 STRATIGRAPHICAL GEOLOGY parts of the German Trias, and closely resemble those of the living species of that genus. Amphibia are still represented by Labyrinthodonts, the chief genera being Labyrinthodon, Mastodonsaurus, and Trematosaurus, but this order does not survive the Triassic period. Their tracks are five-toed, like the rough impress of a man's hand. Reptilia. Remains of true reptiles also occur, and include no Fig. 113. GROUP OF GERMAN MUSCHELKALK FOSSILS. a. Encrinus liliiformis. ft. Terebratula (Coenothyris) vulgaris.' c. Gervillia (Hoernesia) socialis. d. Myophoria A'ulgaris. e. Temnoeheilus bidorsatus. /. Ceratites nodosus. fewer than seven orders, several of which attained their maximum development in the succeeding Jurassic period. The Sauropterygia are represented by Nothosaurus, the Ichthyosauria by Mixosaurus, and the crocodiles by Stagonolepis, Erpetosuchus, Ornithosuchus, and Belodon. Dinosaurs occur for the first time in the forms known as Palceosaurus, Zanclodon ( = Teratosaurus\ Thecodontosaurus, and Cladyodon. Most of these creatures appear to have walked on their hind feet, after the fashion of a kangaroo ; and as some of them THE TRIASSIC SYSTEM 353 had only three toes, their footprints greatly resemble those of gigantic three-toed birds, and were supposed for a long time to have been made by such birds ; other genera had four or five toes. Still more characteristic of the Trias are reptiles belonging to the divisions known as Placodontia, Rhynchocephalia, Dicynodontia, and Theriodontia. The genus Placodus is found in Germany. Fig. 114. TRIASSIC VERTEBRATA. a, b, c. Footprints, skull, and tooth of Mastodonsaurus. d. Dipteronotus cyphus. The Ehynchocephalia are represented by Hyperodapedon (which is allied to the modern Sphenodon of New Zealand), Teler- peton, Leptopleuron, and Rhynchosaurus. The Dicynodonts made their appearance in Permian time (see pp. 327 and 348), but the typical genus Dicynodon was first found in the Lower Trias of South Africa ; it has not yet been found in the Trias of Europe. The following are some of the more characteristic fossils of the German and English Trias : 2 A 354 STRATIGRAPHICAL GEOLOGY Fossils of the Bunter Plants. Schizoneura paradoxa, Voltzia heterophylla, Albertia elliptica, JEthophyllum speeiosum. Mollusca. (In Germany only) Myophoria costata, Gervillia Murchi- soni, Pleuromya mactroides, Beneckeia tenuis. Amphibia. Bones of Trematosaurus and Labyrinthodont footprints. Fossils of the MuschelkalJc Echinoderma. Encrinus liliiformis. Brachiopoda. Terebratula (Coenothyris) vulgaris, Plicigera trigonella, Rhynchonella decurtata. Lamellibranchia. Myophoria vulgaris, Gervillia (Hcernesia) socialis, Pecten discites, Lima striata, Trigonodus Sandbergeri. Gastropoda. Coelostylina gregaria, Undularia scalata, Dentalium torquatium. Cephalopoda. Temnocheilus bidorsatum, Ceratites nodosus, Beneckeia Buchi. Pisces. Ceratodus Kaupi, Hybodus major, Acrodus Gaillardoti. Reptilia. Placod.us gigas, Nothosaurus mirabilis. Fossils of the Keuper Plants. Equisetites arenaceus, Pterophyllum Jsegeri, Voltzia heterophylla, Schizoneura paradoxa, Danseopsis mar- antacea, Yuccites vogesiacus. Crustacea. Estheria minuta. Mollusca. Myophoria Goldfussi, Anoplophora donacina (German). Pisces. Acrodus minimus, Dipteronotus cyphus, Hybodus Keuperi, Palseoniscus superstes, Semionotus Bergeri. Amphibia. Mastodonsaurus giganteus. Reptilia. Hyperodapedon Gordoni, Telerpeton elginense. Mammalia. Tritylodon Fraasi (German). C. THE TRIAS IN BRITAIN The general extent of the Triassic deposits in England can be seen from any good geological map, and I assume that the student possesses the smaller Index Map of the Geological Survey. With, respect to the relations between the Trias and the Permian, recent observations have tended to show that the break between them is not of such magnitude as was formerly supposed. The outcrop of the Bunter follows the strike of the underlying Permian more or less conformably, especially in Devonshire, and, again, in the north-east of England ; but there is unquestionably a break between them, accompanied by overstep which is uncon- formity. Some physical changes took place at the close of the Permian period, but the actual lapse of time between the cessation of Permian conditions and the formation of the Bunter Basement Beds may not have been great. In order to avoid the repetition of detail, and to impart a THE TRIASSIC SYSTEM 355 clearer idea of the component strata, it will be best to deal separately with each of the two great divisions of the Trias, for only two exist in Britain, the Bunter and the Keuper. THE BUNTER SERIES This set of beds has a more restricted range than the Keuper Series, a fact which is not always realised because it is concealed on a geological map by the overlap of the Keuper and the continuous outcrop of the latter all across England from south-west to north- east. The Bunter is not similarly continuous, but was deposited in two separate basins, a northern and a southern. The northern was much the larger of the two, extending from Worcestershire through the Midland counties, and including a large part of Northern England with probably a branch or gulf which reached into the north-east of Ireland and the south-west of Scotland. The southern basin was on the site of the pre-existent Permian lake, covering parts of Devon, Somerset, and the area of the English Channel. The typical district of the British Trias is that of the Western Midlands, an area embracing Cheshire, South Lancashire, Shrop- shire, Staffordshire, and Worcestershire. Here the Bunter Series is divisible into three stages as below : The Upper Mottled Sandstone, from 200 to 600 feet. The Pebble Beds, ,, 300 ,, 1000 feet. The Lower Mottled Sandstone, ,, 200 600 feet. In describing these Bunter deposits it will be convenient to begin with this typical district, next to trace their extension to the north and north-east, and, finally, to notice their equivalents in the south of England. v 1. Central Area The Lower Mottled Sandstone. In Cheshire and Shrop- shire this is usually variegated in bright red and yellow colours. It generally has a basal breccia about 2 feet thick lying uncon- formably on the Permian or the Coal-measures, but is otherwise devoid of pebbles. It always exhibits much oblique bedding, which has the aspect of current bedding, but is not a proof of deposition in water, for the sands may have been aeolian, and the bedding be due to the action of wind, for in some of the beds the component quartz grains have the worn face and elongate oval shape which is characteristic of modern desert sands, the grains resembling millet-seed. to, 356 STRATIGRAPHICAL GEOLOGY This Lower sandstone attains its greatest thickness (650 feet) near Bridgenorth, but is still about 400 feet near Chester. It ranges northward to the Mersey and then east- ward along the southern border of the Lanca- shire coalfield, but thins in that direction, and does not occur at Manchester or Stockport. More- over, it appears to thin out in every other direc- tion, for, though it is found in places on the west | side of the South Staffordshire coalfield, it does not appear on the east side. Southward, again, JS S it extends to Stourport, and is probably con- 2 g tinned underground as far as Worcester, but is .!? overlapped in every direction by the succeeding 1 1 Pebble Beds. Clearly, therefore, this sandstone was accumu- lated within a restricted area, which was prob- ably a desert plain, bounded on the west, south- east, and east by higher ground, but possibly reaching for some distance to the north-west n- below the Irish Sea. The Pebble Beds. These beds are most (3 fully developed in East Shropshire and South I'l.'l o 2 g Staffordshire. They pass eastward beyond the M-'l " outcrop of the Lower sandstones, resting uncon- formably sometimes on the Permian, sometimes 011 the Coal-measures, as in Cannock Chase (see Fig. 116). In this central area they form a mass O f conglomerate which is from 300 to 400 feet thick, and consists of well-rounded pebbles, the majority of these being brown and chocolate- coloured quartzites, with which, however, are pebbles of Carboniferous limestone and sandstone, of purple grit, and of some volcanic rocks. At the surface the sandy matrix is generally soft, but when traversed in deep borings it is very hard and highly calcareous. Near Bridgenorth it has a breccia of angular fragments at its base about 60 feet thick, but this does not seem to extend far. It recurs, however, at Holt, on the river Dee, where it is succeeded by soft red-streaked sandstone over- lain by pebbly sandstone, the whole series resembling that at Bridgenorth. Near Liver- pool and St. Helens they attain a great thickness, THE TEIASSIC SYSTEM 357 consisting of two parts, (1) hard, red -brown, pebbly sandstones, about 600 feet thick, and (2) hard, reddish sandstones without pebbles, about 400 feet. Eastward, by Wigan, Manchester, and Stockport, they overlap .the Lower Mottled sandstones, so as to rest on the Dyassic and Carboniferous rocks. On the east side of the South Staffordshire coalfield the Pebble Beds occupy a broad strip of country from Birmingham to the river Tame north-west of Tamworth, and they recur in a narrow strip along the western side of the Ashby coalfield in Leicester- shire. In the same way from the borders of North Staffordshire coalfield, they pass eastward to Ashbourn and Derby, lying with complete unconformity on the edges of the Lower Carboniferous rocks. The Upper Mottled Sandstone. This resembles the Lower sandstone, but is generally of a bright red, variegated with yellow and white. From Shropshire, north of Shrewsbury, its outcrop runs northward by Ellesmere along the flanks of the Peckforton Hills, which rise to heights of 800 and 900 feet, and thence through Delamere Forest to Euncorn and "Warrington on the Mersey. In the Peckforton Hills the outcrop is repeated by several strike faults which have led to the formation of as many nearly parallel escarpments, the western slopes of which are occupied by this soft, reddish sandstone, while their summit ridges consist of the hard basement conglomerate of the Keuper. In the southern part of the Midland area all the Bunter stages thin out toward the south-east, one overlapping the other, a fact which may be regarded as proof that the pre-Triassic floor sloped from the south-east to the north-west. The following table indicates this attenuation and overlap : l Shropshire. West Side of South Stafford- shire Coalfield. East Side of South Stafford- shire Coalfield. North Warwick. Upper sandstone . Pebble Beds . Lower sandstone . 500 400 650 300 300 to 100 100 200 Absent Absent Absent Absent 1550 700 300 Nil 2. North-Eastern and Northern Areas Having described the typical development of the Bunter Series, it will suffice to indicate briefly their occurrence beyond the limits of this Midland area. THE TRIASSIC SYSTEM 359 Though overlapped by the Keuper marls in the valley of the Trent east and south-east of Derby, the Bunter is probably con- tinuous as a narrow strip beneath them, since it recurs near Nottingham and extends northward for a considerable distance. The Upper Mottled sandstone, however, is absent throughout, this eastern area. Near Nottingham the Lower sandstone has the usual bright-red and yellow tints, but is there only 25 to 30 feet thick ; northward, it thickens to about 100 feet at Mansfield. As in the western area, it seems to occupy a tract of limited width, and to be overlapped eastward by the Pebble Beds, which are the lowest recognisable Triassic deposits in deep borings at South S carle near Newark, and at Haxey in North Lines. The Pebble Beds form the rock on which Nottingham Castle stands, where they are about 200 feet thick, and consist of a coarse yellowish sandstone full of quartzite pebbles. They range north- ward through Sherwood Forest by Worksop, Tickhill, and Doncaster. Along this tract the thickness of the Bunter is about 450 feet, and in a boring near Selby it is 620 feet ; but beyond this place the pebbles disappear, and the whole Bunter division is supposed to thin out near Knaresborough, though it may extend eastward underground beneath South Yorkshire. There is a similar disappearance of the pebbles on the western side of the Pennine Anticline in North Lancashire, but the combined Bunter sandstones persist and are about 1000 feet thick near Preston and Garstang. They recur in the promontory of Furness on the western side of Morecambe Bay, and extend along the western side of Cumberland to St. Bees (see map, Fig. 39, p. 130), where they are known as the St. Bees' sandstone. This St. Bees' sandstone is the sole representative of the Bunter in the Carlisle basin, and along the valley of the Eden, unless the underlying gypsiferous marls are included in the Trias (see p. 337). These latter vary in thickness from 100 to 300 feet, and pass up into the overlying sandstone by intercalations of sandstones and marls. The St. Bees' sandstone is sometimes variegated in the lower part, but most of it is of a dull, uniform, brick- red colour. It contains some bands of pebbly sandstone, and its thickness is estimated at over 1000 feet. 2 A similar sandstone has been found by borings in the northern part of the Isle of Man 3 ; and again in Ireland, near Belfast, where the Keuper is underlain by about 800 feet of soft red and yellow sandstones with some beds of grey and buff shale. Rocks of Triassic age occur again in the Isle of Arran, occupying much of the central and southern part of that island. The lower portion of them is coloured as " Lower Trias " on the recent map UJ I II 3 1 a !<*> I 11 " ll I o si ^. 60S II H c w HII 1 III C /T, F^ JT THE TRIASSIC SYSTEM 361 of the Geological Survey, though it was described as Permian in the "Summaries of Progress" for 1896 and 1898. These lower beds are about 1000 feet thick, and consist of bright-red sandstones with conspicuous current bedding, thus resembling the Penrith Beds rather than the St. Bees' sandstone. They are overlain and overlapped by conglomerates, which are probably of Triassic age, though they seem to follow in conformable succession. 3. South-Western Area In this district there is nothing comparable with the Lower Mottled sandstone of the Midlands, and in the typical section of the south coast of Devon, near Budleigh Salterton, the Bunter is less than 500 feet thick. It is composed of the following beds : 2. Coarse red sandstones . . . from 350 to 400 feet. 1. The Pebble Bed or conglomerate . about 80 feet. The Pebble Bed rests directly on the red Permian marls mentioned on p. 333, but without any sign of erosion, for the lowest pebbles are embedded in a sandy clay which is continuous with the marl below. The mass of the Pebble Bed consists of well-rounded oval pebbles of quartzite and hard grit embedded in a matrix of red sand ; it exhibits current bedding and includes lenticular layers of sand. Some of the pebbles contain Devonian fossils, and others are Silurian and Ordovician, resembling the Gres armoricain and the Gres du May of Brittany and the Ordovician rocks of Cornwall. Dark, compact, tourmaline rocks also occur, like some which occur in the metamorphosed zones round the granite masses of Devon and Cornwall. According to Professor Bonney, however, the majority of the pebbles do not resemble Devon and Cornwall rocks, while there are some of a hard red quartz-felspar grit which resembles the Torridon sandstone of Scotland, but is equally like some parts of the Gres felspathique at the base of the Ordovician in Normandy. It is, in fact, now generally conceded that this assemblage of pebbles was brought by a river or rivers from the south, i.e. from Normandy and from land which then occupied the southern part of the English Channel. Oval quartzite and quartz-grit pebbles have, in fact, been dredged from the floor of the Channel at many points from 15 to 50 miles south of the Eddystone. This conglomerate can be traced inland through Devonshire into Somerset, but the pebbles become smaller in that direction, and the quartzites becomes less and less abundant, till at Burlescombe the bed passes into a sandy gravel composed chiefly of small quartz and grit pebbles. 4 362 STRATIGRAPHICAL GEOLOGY The overlying sandstones are coarse, current-bedded, red sand- stones, and in the lower 100 feet they contain a few pebbles of quartz and quartzite, with occasionally subangular fragments of granite, felspathic grit, and lumps of hard red marl. The higher beds are less coarse and more massive, and from these bones of Hyperodapedon have been obtained. These beds occupy the coast from the mouth of the Otter to Sidmouth (see Fig. 117). THE KEUPER SERIES This series has a much wider extension than the Bunter, and its higher beds were formed in a great salt lake which covered both the Bunter areas of deposition as well as all the intermediate parts of Central England. Borings at several places in the Midland and Eastern counties have proved that the Keuper sandstones and marls extend eastward beneath the Lias and newer rocks till they thin out against the western slope of the great mass of Palaeozoic rocks which underlies the south-east of England. Thus, a boring at Burford in Oxfordshire passed through 428 feet of Keuper Beds, and entered Coal-measures at 1184 feet from the surface. Borings near Northampton found only 60 to 70 feet of Trias above the older rocks, and one at Culford near Bury St. Edmunds proved its absence. It is also absent below Kichmond and London, but appears to pass under the Wealden area as beds underlying Lias, and believed to be of Keuper age, were traversed by a boring at Brabourne in Kent. As in dealing with the Bunter, we shall first describe the complete and typical development of the Keuper Series in the Midland area, and then indicate such stratigraphical variations as occur when it is traced eastward, northward, and southward. 1. Typical Midland Area In this area the Keuper Series is divisible into three very unequal parts : Feet. Red marls with rock-salt . . . . . . from 1000 to 3000 " Waterstones," flaggy sandstones and shales . . about 200 Basement beds, sandstone, and breccia . . . from 100 to 250 Lower Sandstone or Basement Beds. The actual base of the series is generally a bed of calcareous breccia, but sometimes a hard pebbly quartz grit ; this rests on an eroded surface of the Bunter sandstone (see p. 357), and marks a sudden change in the physical conditions, though the break does not amount to an THE TRIASSIC SYSTEM 363 unconformity, for there is no proof that this basement bed rests on different parts of the Bunter Series in different places. The overlying sandstones are rather coarse grained and irregularly bedded, generally of bright colours, red or yellow, but sometimes pink or white. The current bedding is remarkable, and the lammoe often dip at high angles and change within short distances. 5 The thickness is variable, but often about 200 feet, and the group is persistent throughdut Cheshire, Shropshire, Staffordshire, and Worcestershire. To the south-east, however, it thins out, and is absent in Warwick. Fossils have recently been found in the upper part of these beds at Bromsgrove in Worcester 6 ; they include both plants and animals. The plants have been identified by Mr. Arber as Equisetites arenaceuSj Yuccites vosgesiana, and Voltzia heterophylla (twigs and cones). The animals are Hyperodapedon Gordoni, a Labyrinthodont (? Mastodonsaurus) ; the fish, Dipteronotus and Acrodus, Estheria minuta, and fragments of Arthropods which are probably scorpions. The Waterstones are brownish, even-bedded micaceous sand- stones and flagstones, with layers of red or green sandy shale ; they are therefore very different from the underlying beds upon which they rest with a sharp line of demarcation. 7 Their surfaces exhibit ripple -marks, sun-cracks, and they contain casts of salt- crystals, all clear indications that they were deposited in shallow water. Their average thickness is about 200 feet, but there is no definite summit to the group as it passes up by intercalations of shale and marl into the overlying Keuper marls. When traced south-east into Warwickshire, they overlap the Lower sandstones and rest directly on the Carboniferous rocks. In some places the waterstones have yielded bones and teeth of Labyrinthodon, Cladyodon, and Hyperodapedon, and also teeth of the fish Dipteronotus cyphus, but these remains are rare. Reptilian and Labyrinthodont footprints, however, are not uncommon. The Keuper Marls, though the thickest and most important members of the series, are in some respects the least interesting, because they consist of a monotonous succession of red clays (or " marls ") varied only by nodular layers and beds of gypsum and occasionally beds of rock-salt, which are sometimes from 70 to 100 feet thick. The thickness of these saliferous marls in Cheshire is from 2000 to 3000 feet ; in Shropshire it is from 1500 to 1800, thinning eastward to 800 and 700 in Staffordshire, and about 600 in Warwickshire and Leicestershire. In the last county they overlap the sandstones and are banked against the Archaean rocks of Charnwood Forest, partially filling the valleys between the hills. 364 STRATIGEAPHICAL GEOLOGY In the Worcester, Warwick, and Leicester areas a bed of grey sandstone, about 20 to 30 feet thick, occurs near the top of the series and is known as the Shrewley sandstone, from Shrewley between Birmingham and Warwick, where it is well exposed. It has yielded remains of fish (Palceoniscus superstes), with Estheria minuta and some casts of bivalve Mollusca, which, according to Mr. E. F. Newton, resemble Goniomya, Pholadomya, and Thracia. 2. North-Eastern Area Basement Beds. Near Nottingham these are repre- sented by an irregular deposit of white sand, which rests on an eroded surface of the Bunter Pebble Beds and varies from a few inches to many feet in thickness. Mr. J. Shipman has described certain pebbly sandstones more than 100 feet thick, which he regards as representing this deposit. 8 The Waterstones which succeed generally have a pebbly or conglomeratic bed at the base, and this rests sometimes on the white sand deposit, and sometimes directly on the Bunter, but always on an eroded surface, so that it seems to be unconformable to the beds below. The pebbles are chiefly quartz and quartzite, but Mr. Shipman has detected fragments of trap, slate, chert, and magnesian limestone. The waterstones themselves consist of soft brown sandstones interstratified with red marls, which cause plentiful springs by throwing out the water collected in the sand- stones. Eipple-marks and sun-cracks are frequent in these beds. The Keuper Marls. These are considered to begin when the marl bands become thicker and the interbedded sandstones become harder and generally white or bluish-grey. Of such beds there is about 150 feet. Gypsum occurs in the marls, and the higher beds are chiefly marl with sheets and nodules of gypsum. The whole is about 900 feet thick (see Fig. 118). In Yorkshire the Keuper appears to become thinner and not to exceed 700 feet in thickness ; it is still capable of division into an upper red marl group, about 400 feet thick, and a lower series of red and white sandstones from 200 to 300 feet thick, but its lower limit is somewhat uncertain in the northern part of the county. 3. Cumberland and Irish Sea Area Beds of Keuper age overlie the St. Bees' sandstone in the Carlisle basin, and so far resemble the Midland type that they consist of a lower sandstone division and an upper gypsiferous marl group. The former is known as the Kirklinton sandstone THE TRIASSIC SYSTEM 365 and is about 500 feet thick ; the latter (Red Marls) are estimated at 950 feet. The Kirklinton sandstone is only seen in the eastern part of the basin, its outcrop elsewhere being concealed by the Drift deposits. It consists of soft sandstones, which are in some parts red, in others white, or sometimes mottled red and white, and they are probably the equivalents of the Cheshire " waterstones." The Keuper marls are also for the most part buried beneath Drift, but near Carlisle the sandstones are covered by red and grey marls, which are exposed in several places on the banks of the Eden. West of Carlisle there is believed to be a fault of consider- able magnitude, for borings have shown that a great thickness of red shaly marls with gypsum and rock-salt is brought into the western and deeper part of the basin. Keuper marls also occur beneath the northern part of the Isle of Man, borings having proved a thickness of 557 feet with 21 beds of rock-salt, one of these being 16 feet thick. Below are 800 feet of sandstone, part of which is probably also of Keuper age. In Ireland the largest area of Trias occurs in the valley of the Lagan, and extends from Lurgan to Belfast, and thence along the northern shore of Belfast Lough as far as Black Head. The beds are also seen at intervals along the coast to the northward as far as Red Bay near Cushendall. They are thickest in the southern part of the area near Belfast. Rock-salt is found at Carrickfergus, where the general succession is as follows : Feet. j' Red marls with gypsum ..... 700 Keuper- Rock-salt and blue clays . . . . .150 ( Brownish sandstones . . . perhaps 20 p f Soft red and yellow sandstones with grey er \ and buff shales 800 Writing of the Trias in the Isle of Man, Professor Boyd Dawkins remarks that " the unexpected presence of the Saliferous Marls is a link uniting the Triassic strata of the north of the island with those of Barrow on the one hand, and Carrickfergus on the other. It is very probable that these three salt-fields belong to one great saliferous basin underlying the Irish Sea, extending south- ward and eastward to join that of Fleetwood [in Lancashire], and to the south in the direction of the great salt-field of Cheshire." 10 4. Mendip and Bristol Channel Area It is convenient to deal separately with this area because its connections with the areas to the north and south are narrow, and because it exhibits a somewhat peculiar local facies of the Keuper. 366 STRATIGRAPHICAL GEOLOGY As stated on p. 359, this area was land during the formation of the Bunter Beds, and it was not covered by the waters of the Keuper lake until a late epoch in the period ; indeed, the Mendip Hills were never completely covered by these waters, and the flanks of these hills display an interesting arrangement of littoral deposits. These deposits consist of breccias and conglomerates with a calcareous matrix, and are known as the "Dolornitic Conglomerate." They are mainly composed of pebbles of Carboniferous limestone, TOO 00 OQ^ o O o*> o o oo Fig. 120. GEOLOGICAL MAP OF BACKWELL HILL BETWEEN BRISTOL AND WRINGTON (Surveyed and drawn by Professor S. H. Reynolds.) Scale 1 inch to a mile. d. Lias, c. Keuper. ft. Dolomitic conglomerate, a. Carboniferous limestone. most of these being about the size of a hen's egg, but ranging up to boulders of 2 or 3 feet in diameter, the whole being compacted by a calcareous cement which is generally dolomitic. These beds dovetail into calcareous sandstones and red marls (see Fig. 121), the latter of which fill the wider spaces between the islands of older rocks. Remains of Palceosaurus and Thecodontosaurus have been found in the conglomerate deposits. The manner in which the conglomerates run up into the lime- stone hills along the lines of the old valleys is illustrated in Fig. 120, and, from the contours indicated on this map, it will be seen that these old inlets have no relation to the modern valleys. 11 THE TRIASSIC SYSTEM 367 A continuation of these beds is found in South Wales, where they occur in patches from Cardiff westward as far as Forth Cawl and Kenfig, with a small outlier in Gower which indicates a still further extension to the west. The beds surround islands of Old Red Sandstone and Carboniferous limestone, and their thickness, of course, varies, but there is generally a border or base of dolomitic breccia which passes up into red marls with layers of red and white limestone ; these are succeeded by red clavs, and, finally, by green clays or marls. 12 On the Somerset side of the Channel a similar set of red breccias, sandstones, and marls is found all round the Quantock Hills, and along the coast from Watchet to Porlock, where they occupy an ancient strait between the Brendon Hills and that called North Fig. 121. SECTION TO SHOW THE RELATIONS OF THE TRIAS TO THE CARBONIFEROUS ROCKS NEAR SHEPTON MALLET (Somerset). 1. Lower Lias. r. Rhaetic Beds, fc. Keuper marls. c. Conglomerate. 4. Millstone grit. 3. Carb. limestone. 2. Carb. shale. 1. Old Red Sandstone. Hill on the coast line. The Quantock Hills were an island in the great lake, and the Trias is bedded into a similar old valley or strait on the west of them. 13 5. Devonshire South of Taunton the thickness of the Keuper Series becomes greater, though the actual outcrop is narrowed by the overstep of the Cretaceous strata which form the Blackdown Hills. The whole series, however, is well exposed on the cliffs of the south coast, from Sidmouth to Seaton and Axmouth. Here the following succession of beds is found : Feet. I' Red and green marls . . about 300 Keuper-| Red marls with gypsum . . ,, 1000 (Red sandstones . . . . ,, 75 Keuper. According to Professor Hull and the Rev. A. Irving, 14 the base of the Keuper division should be taken at a bed of hard breccia about two feet thick which can be seen below the footbridge 36'8 STRATIGRAPHICAL GEOLOGY over the Sid at Sidmouth. Above this are pale-red and grey sandstones with seams of marl, succeeded by alternating bands of sandstone and marl, the beds in which sandstones predominate being about 75 feet thick. The succeeding beds are red marls with thin sandstones in the lower 150 feet, but passing up into massive marls with a con- choidal fracture containing strings of gypsum and pseudomorphs of rock-salt. These form the lower part of the cliffs as far as Branscombe mouth, but near Beer Head a syncline carries them below the beach. Higher beds occur in the Axmouth and Bindon cliffs, where they finally pass up into the Rhsetic Beds. 6. Scotland Rocks of Triassic age occur in several parts of Scotland, but, though they resemble the English Trias in consisting largely of red sandstones and marls, their stratigraphical arrangement is different, and this area has therefore been left till the last. The possibility that the Bunter is represented in Arran has been mentioned on p. 361, but in the more northern districts it is probable that all the beds are of Keuper age. Arran. The beds described on p. 361 are succeeded by red sandstones, and these are overlain by red shales and marls. More- over, recent exploration of the agglomerates which fill up the vent of a huge Tertiary volcano in the central part of the island has led to the discovery of large masses of rock which show a sequence from red marl through Rhcetic shales into Lower Lias, thus proving the red marls to be of Keuper age, and to have been succeeded by Rhcetic and Liassic Beds, though all other traces of these beds have been destroyed by erosive agencies during Tertiary times. 15 Western Coast. Small patches of Trias occur at the base of the Neozoic rocks in Morvern and Ardnamurchan (Argyleshire), in the Isles of Skye and Raasay, and on the coast of Gruinard Bay in Ross-shire. They have been described by Professor Judd, who remarks that the materials forming the breccias, conglomerates, and coarser sandstones have evidently been derived from the older rocks on which these strata repose, usually either the Torridon sandstone or some member of the Highland gneissic series. A special feature in these Triassic rocks is the abundance of calcareous matter which enters into their composition ; frag- ments of compact limestone (? Durness limestone) abound in the coarser beds ; the sandstones often pass into calcareous grits, and bands of cornstone are not unfrequent, while all the beds are frequently traversed by veins of calcspar. As a rule, breccias and THE TRIASSIC SYSTEM 369 conglomerates prevail in the lower part of the series, sandstones, marls, and limestones in the upper part, but the succession of beds is very irregular and inconstant. At Gruinard Bay the total thick- ness exceeds 1000 feet ; at Raasayit is probably less than 500 feet, though the general succession is the same ; at Lussay, in Skye, it is reduced to a few feet, and southward, in parts of Mull and Morvern, it varies from 50 to 500 or 600 feet. The following is the succession at Gruinard Bay : Feet. Soft reddish argillaceous and sandy beds with bands of hard white and greenish sandstone 200 Red and variegated marls with thin beds of sandstone and concre- tionary limestone ......... 200 Red argillaceous sandstones and conglomerates, alternating in the lower part with masses of coarse breccia-conglomerate . . 500 Eastern Coast. Small tracts of Trias occur in Elgin and on the coast of Sutherland. The tract in Elgin extends from Spynie, north of Elgin, to the coast at Lossiemouth and Stotfield Head. 16 It rests on the O.R.S., but its base is not exposed. The rock seen in quarries is a white sandstone, massive, fine-grained, and fairly soft, with a thickness of about 200 feet, apparently dipping north, and overlain at Stotfield by a hard, whitish, cherty limestone, which may be 30 feet thick. This sandstone has yielded remains of eight different reptiles : three Rhynchocephalians, Hyperodapedon, Teler- peton, and Stenometopon ; two Crocodiles, Stagnolepis and Erpeto- suchus ; and two Thecodonts, Ornithosuchus and Scleromochlus. The same two kinds of rock occur below Dunrobin Castle on the Sutherland coast, i.e. white sandstone dipping north-east, and passing below a cherty rock like that of Stotfield, above which are beds of Liassic age. D. THE TRIAS IN EUROPE 1. Germany Triassic recks occupy large areas in Central and Southern Germany, in Thuringia, Hesse, North Bavaria, Wurtemberg, and Lorraine ; they are also found in the Eifel district, in Hanover, Brunswick, and Magdeburg, and finally in the island of Heligoland. From the Rhine district they extend westward to Luxemburg, and through the Vosges Mountains into France. They are everywhere divisible into the three series of Bunter, Muschelkalk, and Keuper. The German Trias was formed under the same physical and geographical conditions as that of Britain, with which it is prob- 2B 370 STRATIGRAPHICAL GEOLOGY ably continuous under the bed of the North Sea ; in other words, the British area was part of a much larger Germanic region. The German Bunter and Keuper may both be regarded as salt-lake or lagoon deposits, and though the fossils of the Muschelkalk are marine, yet its fauna is not a large one, and is evidently that of an inland sea which was only temporarily connected with the open sea to the south. The area over which these limestones occur may be regarded as including what was the deepest part of the great depression occupied by this inland sea. The subdivisions of the German Trias and their average thicknesses are given in the following table : Feet. f "Steinmergel," marls without gypsum . 300 to 500 Keuper-j Gypsiferous red and green marls . . 500 to 700 [Kohlenkeuper 200 Muschelkalk in three stages 600 to 1000 ( Upper Bunter marls . .... 300 Bunter ! Middle Bunter sandstones . . . 700 to 1000 [Lower shaly sandstones .... 300 to 500 3000 to 4000 The Bunter. This is essentially a sandstone series. The lower stage in Northern and Central Germany consists of dark- red flaggy sandstones, which succeed with apparent conformity the clays of the Upper Zechstein, but in Western Germany they overstep the whole of the Permian, and rest on the crystalline schists of the Odenwald and the Black Forest. The rest of the Lower Bunter consists of fine-grained micaceous sandstones of various colours, often speckled, and often including angular fragments of red clay ; there are also some layers of dolomitic sandstone (Rogensteiti). (See Fig. 122.) Fossils are rare. The main mass of the Bunter sandstone (or Middle Bunter) consists of coarse quartzose sandstones, which generally show current bedding and have a maximum thickness of 1000 feet. No fossils except tracks of Labyrinthodonts have been found. The Upper Bunter or Roth consists of variegated red and green marls with beds of gypsum and rock-salt. In Thuringia these beds contain layers of dolomitic limestone with marine fossils (Myophoria costata, M. vulgaris, and other species which also occur in the Muschelkalk). Westward in the Eifel, the Vosges, and Lorraine the Roth passes into fine argillaceous sandstones containing plant remains, Voltzia^ Schizoneura, and Anomopteris. Muschelkalk. This is a limestone series, and takes its name from the abundance of bivalve shells (" muscheln ") in many of its beds. It is divisible into three stages. THE TRIASSIC SYSTEM 371 The lower stage is about 300 feet thick, and consists of thin- bedded marly limestones (Wellenkalk) with yellowish porous lime- stones (Schaumkalk). The chief fossils in the lowest beds are Myophoria orbicularis, Natica gregaria, and Dentalium torquatum, but the higher part of the Wellenkalk is one of the richest fossili- ferous zones in the whole series, its fossils including Myophoria vulgaris, Gervillia costata, Pecten discites, Terebratula (Ccenothyris) vulgaris, T, (Aulacothyrifi) angusta, and Beneckeia Buchi. The middle stage consists chiefly of dolomitic limestone and marl with beds of gypsum and rock-salt. Fossils are rare. The upper beds are hard limestones, consisting mainly of Encrinite stems, overlain by thin -bedded limestone characterised by Ceratites nodosus. These upper beds are rich in fossils, some of the commonest being Encrinus liliiformis, Lima striata, Gervillia (Hcernesia) socialis, Terebratula (Cosnothyris) vulgaris, Myophoria vulgaris, and Temnocheilus bidorsatus. In the Vosges country some of the Muschelkalk limestones are replaced by sandstones and marls, and the whole division gradually dwindles till it finally thins out and disappears. The KLeuper. The main area of the German Keuper lies in Franconia and Swabia (Wurtemberg), but it also occupies a con- siderable space in Luxemburg and in Alsace-Lorraine. It occurs also in the middle of the Thuringian basin and to the north-east of the Teutoburger Wald (in Hanover and Brunswick). The series is everywhere divisible into three groups, which have a combined thickness of from 900 to 1200 feet (see Fig. 122). The Kohlenkeuper consists of dark-grey shales and sandstones with thin beds of shaly coal, but the fossils are marine and brackish water, some of the beds being crowded with Estheria minuta and Lingula tenuissima, with the teeth of fishes (Ceratodus, Acrodus, and Hybodus). The highest beds of the group are yellow dolomitic limestones some compact, some oolitic and these contain Myo- phoria Goldfussi, M. transversa, and Anoplophora donacina. The central beds, forming the main mass of the Keuper, are red and green marls with beds of gypsum and some of rock-salt. Fossils are rare in these marls, but in South Germany (Stuttgart) they include plant-bearing sandstones with Equisetites arenaceum and Pterophyllum Jwgeri. The Steinmergel seems to correspond with the highest red and green marls of the English Series. They are free from gypsum and rock-salt, but contain thin layers of sandstone and, in Lorraine, beds of dolomite and layers of dolomitic concretions. From these beds many Reptilian remains have been obtained, e.g. Aetosaurus and Belodon, and the Dinosaurs Dimodosaurus and Zanclodon. THE TRIASSIC SYSTEM 373 2. The Alpine Fades The Triassic System has no great thickness in the Western or Central Alps, but attains a great development in the well-known Dolomite Mountains of the South Tyrol and Carnic Alps, and also in the Vorarlberg, in North Tyrol, Salzburg, and Bavaria, on the northern side of the main crystalline axis of the Alpine chain. The succession of beds forming the lower and middle parts of the system in the northern districts differs considerably from that in the southern, but the higher part from the Raibl Beds upwards is similar in both regions. From this and from differences in the Cephalopodan faunas of the two areas it was inferred that there were at first two marine provinces which were more or less separated until the later part of the Triassic period. The northern province has been called the Juvavian, and the southern the Mediterranean province. The former includes Bavaria, the Salzburg district, and Austria proper, extending eastward to the Carpathians ; the latter comprises the Lombardy Alps, the Southern Tyrol, Carinthia, with the whole of Italy and the whole Dalmatian and Bosnian region. The progress of research, however, has shown that the differences are less than were supposed, and that, so far as the fossils are concerned, it is rather a matter of relative abundance of species than of difference in fauna. The thicknesses of the several divisions vary much in different districts, but the total thickness in the North Tyrol is supposed to be about 6000 feet, while in the South Tyrol it is probably about 9000 feet. The following table shows the general succession in the two regions : Southern Region. Northern Region. 8. Lower Dachstein Dolomite . . Main Dolomite \Keunpr 7. The Raibl Beds .... Raibl or Cardita Beds / . P 6. The St. Cassian Beds . . . Partnach Beds 5. The Wengen Beds 1 -o -m- r 4. The Buch e enstein Beds) ' ' Keifflmg limestone 3, 2. The Virgloria limestone . . Guttensteiu limestone 1. The Werfen Beds . . '. Werfen Beds . . Bunter. The whole system has also been divided into zones and into stages or series of more equal paleeontological value, but much difference of opinion still exists as to the number of zones and as to their grouping, and many names have been proposed for the different groups or stages. The following table gives the principal zones, which correspond to the local subdivisions as numbered above, and the names which seem most applicable to the 374 STRATIGRAPHICAL GEOLOGY stages, the latter being almost the same as those proposed by Mojsissovics : 17 Stages. Zones. E. Juvavian 8. Avicula exilis. Upper . Carinthian (J; Norian {; $% a^helau, TJ TV f 3. Ceratites trinodosus. Lower f B ' Dmarian | 2 . Ceratites binodosus. | A. Werfenian 1. Tirolites cassianus. More recently he has proposed to combine the Norian and Carinthian into one stage or series, for which he adopts the name Tirolian, but the advantage of this is not very apparent. The "Werfenian, or zone of Myophoria costata, is fairly constant in its lithological and palaeontological characters through- out the Alpine and Austrian region. Its prevalent facies is one of red sandy micaceous shales, but in the Carnian Alps there are intercalated beds of marly and dolomitic limestones, and in the Tyrol it also consists of shales and limestones which have been divided into two groups, the Seiss Beds and the Campil Beds. The Ammonoid Tirolites cassianus is found in both provinces, associated with Myophoria costata, Naticella costata (Fig. 112, No. 2), and Pseudomonotis Clarce (Fig. 112, No. 1). In the South Tyrol the thickness of this stage is only 300 to 500 feet. Dinarian, or Alpine Muschelkalk. It is at this stage that the differences between the southern and northern provinces become pronounced. In the Southern Alps it is represented by the Virgloria limestone and its equivalent, the Mendola Dolomite, and in the South Tyrol there is a lower set of bedded limestones over- lain by Mendola Dolomite. In the more northern districts there are two groups or zones, the Guttenstein limestone or zone of Ceratites binodosus and the Reiffling limestone or zone of C. trinodosus. The Dinarian contains many of the same species as the German Wellenkalk, notably Caznothyris vulgaris, Terebratula (Aulacothyris) angusta, Tetractinella trigonella (Fig. 112, No. 3), Gervillia (Harnesia) socialis, Myophoria vulgaris, and Encrinus liliiformis. The thickness varies from 400 to 1200 feet. The Norian. In the southern province this comprises the Buchenstein shales, with calcareous nodules containing Trachyceras Curionij and the Wen gen Beds, which include black shales and calcareous grits containing Trachyceras archelaus and Halobia Lommeli, succeeded by thin-bedded limestones and shales with Posidonomya wengensis. In the Juvavian province these are replaced by the higher part of the Reiffling limestone and by the THE TRIASSIC SYSTEM 375 Wetterstein limestones, so that there the stage is wholly calcareous. A large number of peculiar Ammonoids come in with this series, such as Pinacoceras Metternichi, P. parma, Cladiscites tornatus, Arcestes ruber, and species of Diphyllites and Megaphyllites. The thickness varies from 2000 to 3000 feet. The Carinthian. The separation of the two provinces appears to have continued through this group up to the phase of the Raibl Beds, when similar conditions began to prevail over the whole region. In the Mediterranean province the Wengen Beds are overlain by the St. Cassiaii Beds, a series of fossiliferous marls, shales, and limestones with occasional beds of volcanic tuff. Among the large number of fossils which these beds afford the following may be mentioned : Trachyceras aon (Fig. 1 1 2), Choristo- ceras eryx, Orthoceras elegans, Naticopsis neritacea, Halobia cassianus, Anoploplwra Munsteri, Cidaris dorsata, Encrinus cassianus, Isocrinus propinquus, Thamnastrea Zitteli, Koninchina Leonhardi. The St. Cassian Beds are succeeded by the Kaibl Beds, dark-grey marls with lenticular beds of limestone and some of gypsum, the chief fossil being Myophoria Kefersteini (Fig. 112, No. 10). In the St. Cassian area of the Tyrol a mass of dolomitic limestone, the Schlern Dolomite, from 1000 to 1400 feet thick, comes in between the St. Cassian and the Raibl Beds, 18 and this thickens southward, replacing more and more of the beds below and above till in the Schlern and Rosengarten districts it rests directly on the Mendola Dolomite of the Norian stage, and is then about 3000 feet thick with a small equivalent of the Raibl strata at the top. Consequently it then represents both the Norian and Carinthian (see Fig. 123). In the North Tyrol and Bavaria the St. Cassian Beds are represented by the Partnach Beds with the overlying Raibl Beds, but in the Austrian Alps to the eastward there is a different succession, consisting of (1) the Raugraben shales, (2) the Linz sandstone with plant remains, (3) the Oppoiiitz limestone. Juvavian. Finally, over the whole region we have the massive dolomitic limestones known in Austria as the Hallstadt and Dachstein limestones and in Lombardy as the Great or Main Dolomite. These masses of limestone are several thousand feet thick, but fossils are rare in them, except casts of the large bivalve Megalodon ; Gervillia exilis and Worthenia solitaria, though less common, are characteristic species. These limestones pass up into the Kossen or Rhsetic limestones and are by some authors regarded as Lower Rhaetic, but it is only the Upper Rhsetic or zone of Avicula costata which overlies the Keuper of Germany, France, and England. Hence it is almost certain that the zone 376 STRATIGKAPHICAL GEOLOGY of A. exilis was contemporaneous with the highest part of the Keuper. The probable correlation of the several series of beds in England, Germany, and the Tyrol is shown in the following table : BRITAIN GERMANY THE TYROL KEUPER JUVAVIAN ~^~^^ ~~ ^ ^ BUNTER \ \ \ \ CARINTHIAN \ \ PERMIAN % \ \ NORIAN \ \ \ \ \ \ DINARIAN h. \ WERFEMAN PERMIAN Vertical Scale 6000 feet to an inch. 3. France The surface areas occupied by Trias in France are small, though the Keuper probably has a considerable subterranean extension both in the northern and southern parts of the country. From the Yosges district and the Sarrebruck area in Alsace a broad tract of Trias reaches westward into France, passing under the Jurassic strata which occupy the great basin between the plateau of the Ardennes and that of the Morvan. All the members of the German Trias are here present, but both the Bunter and the Muschelkalk thin rapidly to the westward ; their attenuation can be followed along the northern outcrop in Luxemburg, and both are very thin before they are overlapped by the Lias north-west of Arlon. There can be no doubt that a similar thinning out takes place to the south-west of the Vosges, for when the beds emerge on the borders of the Morvan only Keuper remains, but this is exposed at intervals round the Morvan and along the north side of the Central Plateau from the valley of the Allier to a point about 15 miles west of La Chatre. Though the beds are of no great thick- ness, they seem to represent the Lower or Kohlenkeuper as well as the higher marls. They can be traced southward along the eastern border of the Plateau to Macon, Villefranche, and Lyons, the beds consisting of coarse quartzose and felspathic sandstones from 50 to 100 feet thick, surmounted by a pink dolomitic lime- stone with Myophoria Goldfussi, which is succeeded by Keuper marls. At Lyons the sandstone rests directly on the ridge of Archaean THE TRIASSIC SYSTEM 377 rocks which here crosses the valley of the Rhone between that place and Valence, and it is probable that this was part of the barrier which separated the northern and southern basins of deposi- tions. South of Valence similar deposits set in again and form a long strip by Privas and Largentiere, and again in the Lodeve district ; they also occur in the Pyrenees and in Eastern Spain, but to find a complete Triassic Series one must go to the Maritime Alps, where lower beds come in and the series expands eastward into the Apennine facies which is similar to that of the Eastern Alps. The only other Triassic tract in France which need be noticed is that in Normandy, occupying a bay -like depression in the Archiean rocks from Valogne and St. Waast on the north to St. Lo and Littry on the south, but thinning out to the west of Caen. The beds consist of gravel and conglomerate at the base, passing up into a dolornitic conglomerate which is succeeded by red sandstones and marls. The maximum thickness is not more than 200 feet, and the whole is probably of Keuper age, the beds being shore deposits, like those round the Mendip and Quantock Hills in England. E. THE HISTORY OF THE EUROPEAN TRIAS From the preceding descriptions of the Triassic strata which are found in different parts of Europe the student will have gathered that considerable geographic changes took place during the course of the period ; though these seem to have been accom- plished slowly and without the forcible uplift of any mountain hain like that of the Armorican ranges in Carboniferous time. The change from Permian to Triassic conditions seems to have been effected by a slow upheaval of the whole of Northern Europe, so that all those parts of the Permian Sea which lay within that region were shallowed and some portions of them were converted into sandy plains. This seems especially to have been the case in the British area where the north-west gulf of the Permian Sea became a sandy desert, swept by winds which probably blew chiefly from the north, like those which blow over Persia and Beluchistan at the present day at certain times of the year. Through this plain may have meandered a river fed by the streams which flowed off the western and northern mountains, so that it was a strong and rapid river even though it may have received no affluent after leaving the gap between Ireland and Scotland. Such is the theory by which the accumulation of the " Pebble Beds " in the English Midlands has been explained, for the majority of these 'pebbles are quartzites which are unlike any 378 STRATIGRAPHICAL GEOLOGY Welsh quartzites, but are identical with those found in. the Old Red conglomerates of Scotland, the stones themselves having been originally derived from the Torridon sandstone. 19 For a discussion of this matter the reader is referred to the author's Building of the British Isles (1911). Here it is only necessary to remark that these pebble beds do not signify more than a brief episode in the Bunter epoch ; a few abnormally wet seasons would suffice to pile up all the pebbles which are now spread over Shropshire, Staffordshire, and Worcestershire. They had doubtless been accumulated in the higher tributary valleys of the river-system, and when heavy rains turned the streams of these valleys into rushing torrents, the pebbles were swept down into the main river, which, swollen by the floods, was strong enough to carry them for several hundred miles before they finally came to rest in a bay-like curve of the great plain, for the river must have flowed out westward into and across what is now the basin of the North Sea. There it was doubtless joined by other rivers, all combining to form a still larger one .which curved southward and ran into the German lake or lagoon. That the episode of the English Pebble Beds was not one of much physical importance is proved by the fact that it was followed by a reversion to the preceding conditions, for the Upper Bunter sandstone is so similar in character to the Lower sandstone that outside the area of the Pebble Beds it is difficult to distinguish them, and the sandy facies of the Pebble Beds merges outward into the Upper sandstone. In Germany the characters of the Lower Bunter sandstone, it& ripple marks, sun-cracks, and rain-prints, show that it was deposited in a large lake, probably of fresh water, and the Middle Bunter must be either lacustrine or terrestrial ; but the case is different with the Upper Bunter, for its red gypsiferous marls are the deposits of saline water, and in Thuringia the inclusion of a marine limestone marks an incursion of the Southern Sea. The epoch of the Bunter was brought to an end by a general subsidence of the whole region, the result of which was to bring the sea over the whole of the low-lying Germanic plain. This is the epoch of the Muschelkalk, and the higher level of the British area at this time is indicated by the fact that it was not reached by the Muschelkalk Sea. That sea passed over Hanover and the Island of Heligoland, and probably reached some distance farther west, terminating somewhere below what is now the bed of the North Sea. Thus for a time the plains and lakes of the Bunter were con- verted into arms of the Southern Sea, and communication with THE TRIASSIC SYSTEM 379 that sea was established across Bavaria, and perhaps also through Silesia to the east of the Bohemian Highlands. With regard to Russia little evidence is available, because any Triassic rocks which may exist in the western and central parts of that country are concealed beneath a thick cover of Jurassic and Cretaceous strata. In the Astrakan. province, however, the Bogdo Mountains are capped by outliers of a limestone which is believed to be of Mus- chelkalk age, and beds belonging to the Upper Trias occur in the Crimea, so that it is very likely that the Southern Sea spread over a large part of Southern Russia during Muschelkalk time. We have seen that a great thickness of rock-material was accumulated in this Southern Sea, and that as time went on the deposits formed therein became, at any rate in some parts, more and more calcareous. At the same time it does not seem to have been a deep sea, for apart from Molluscan shells (chiefly Lamelli- branchs and Gastropods), the principal components of the lime- stones are calcareous algee (Diplopora and Gyroporella), which at the present day do not flourish much below 150 fathoms. Hence, to account for the much greater mass of deposit in Austro- Alpine region, about 5000 feet as compared with 1000 in Germany, we must suppose that subsidence was more continuous in the southern than in the northern region. The truth of this inference is confirmed by a consideration of the area where the German and Austrian facies of the Trias approach most nearly to one another. Between the Schwartzwald and Bohemia the German Keuper dips southward under the Jurassic strata of the great Danubian plain ; while south of this plain, and cropping out from below the same Jurassic and Cretaceous rocks, are the highest members of the Austrian Trias, extending from the Vorarlberg to the Salzburg district. The breadth of this interven- ing tract, which conceals the passage of one facies into the other, between Tubingen and the Vorarlberg is only 100 miles. There is no good reason for supposing that this tract conceals a ridge of Palaeozoic rock, which could have formed a land-barrier between the two regions ; it is much more probable that the one facies passes into the other with a comparatively rapid change, and the facts can be explained on the supposition that the crust- movements were of a counterpoising nature, the southern region continuing to subside while the northern at the beginning of Keuper time underwent a slight uplift with a north-west tilt, so that while the uplift was sufficient to cut off the northern area from the Southern Sea the inland waters were able to extend themselves farther west than those of the Muschelkalk had done. In this manner was produced the great salt lake or inland THE TRIASSIC SYSTEM 381 sea in which the red marls and salt deposits of the Keuper were formed ; and judging from the relative thickness of the Keuper Series at different places the deepest parts of this inland sea were near Berlin in Germany (Sperenberg boring) and below Cheshire in England. In Fig. 124 I have attempted to restore the geography of Europe in the Keuper epoch, and though I have left an opening between Switzerland and Bohemia, this can only have been open during the time of the Kohlenkeuper, and must afterwards have been closed by local upheaval or in some other manner so as to shut out the waters of the Southern Sea from the northern basin. The physical aspect of the country surrounding the inland sea must have combined the features of the Syrian desert around the shores of the Dead Sea with those of the country near the Caspian ; that is to say, the Anglo-Germanic Sea was bordered in some places with rough and rocky hills, in others by sandy plains, and again in some places by reedy marshes and salt lagoons. We must not picture the whole region as completely barren and arid, for we know it was inhabited by various kinds of reptiles and Labyrintho- donts ; moreover the continued existence of the lake would depend on the water-supply carried into it by rivers, though it is clear that this supply was often less than the amount which evaporated from its surface. REFERENCES 1 Lapworth, "Geology of the Birmingham District," Proc. Geol. Assoc. vol. xv. p. 384 (1898). 2 Holmes, "Geology of Cumberland," Proc. Geol. Assoc. vol. xi. p. 231. 3 Boyd Dawkins, Quart. Journ. Geol. Soc. vol. Iviii. p. 647 (1902). 4 Martin, Geol. Mag., 1909, p. 160. 5 Strahan, Geol. Mag., 1881, p. 396. 8 Wills, Geol. Mag., 1907, p. 28. 7 Strahan, op. cit. 8 Shipraan, " Geology of Nottingham," Proc. Nott. Nat. Hist. Soc., 1889 ; see also the Geological Survey Memoir on Sheet 126 (1908). 9 Holmes, op. cit. 10 Boyd Dawkins, op. cit. p. 655. 11 Morgan and Keynolds, Bristol Nat. Hist. Soc. for 1908, p. 5. 12 Strahan, "Geology of South Wales," Geology in the Field, p. 842. 13 See W. A. E. Ussher, Quart. Journ. Geol. Soc. vol. xxxii. p. 367, and vol. xxxiv. p. 459. 14 Hull and Irving, Quart. Journ. Geol. Soc. vol. xlviii. pp. 60 and 68. 15 Judd, Quart. Journ. Geol. Soc. vol. xxxiv. p. 686 (1878). 16 Watson, Geol. Mag. for 1909, p. 102. 17 Mojsissovics, Sitz. Akad. Wien, 1892, p. 769, and 1895, p. 1279. 18 Ogilvie, Quart. Journ. Geol. Soc. vol. xlix. p. 1 (1893). 19 Bonney, Quart. Journ. Geol. Soc. vol. Ivi. p. 287 (1900). CHAPTEE XIII THE JURASSIC SYSTEM A. NOMENCLATURE AND DIVISION IN the early days of Geology part of this system was called the " Oolitic Series" (by Conybeare and Phillips in 1822) from the prevalence of oolitic limestones in the central and upper parts of the system in England, and the Lias was described as a separate series ; but on the Continent the equivalents of the former were called the "Jura limestones" by Brongniart and Boue in 1829 from the fact that beds of this age form the entire range of the Jura Mountains between France and Switzerland. In Germany the name " Jura " was adopted by von Buch for the whole system (1839), who divided it into Lower, Middle, and Upper Jura, and this nomenclature is still used in that country. It appears to have been d'Orbigny (1840-46) who first applied the term "Terrain Jurassique " to the whole succession, including the Lias ; and at the same time he divided the system into a number of stages, but did not group these into series as von Buch had done. The arrangement now generally adopted is a division into three series, which are practically those of von Buch, and a further sub- division into stages for which we in England have used local English names, while French geologists employ the latinised names proposed by d'Orbigny with some additions to their number. As a matter of fact it would be more convenient, and more in accordance with the nomenclature of older systems, to employ latinised names for the three great series themselves. a The English and French subdivisions are shown in the following table : For the Lower Series the term Lias or Liasian is already in general use ; for the Middle Series I long ago proposed l the name Glevanian from the Eoman name of Gloucester, and that of Clavinian for the Upper Series from the Roman name of Wey mouth. 382 THE JURASSIC SYSTEM 383 fPtirbeck Beds ) T, ,, ,. I TI .LI j 13 j >1 ortlandian. Portland Beds J Upper Jurassic A Kimeridge clay Kimeridgiari. Corallian Beds Sequanian. I Oxford clay Oxfordian. . f Great Oolite Bathonian. Middle Jurassic ( Inferior ^^ E^o G i^. ? Upper Lias Toarcian. Middle Lias Charmouthian. It will be noticed that I follow French geologists in regarding the Rhaetic or zone of Pteria contorta as the basal part of the Liassic Series, while the Germans have generally included it in the Trias. These beds were briefly referred to on p. 375, and reason was there given for restricting the name to this zone, which succeeds that of Gervillia exilis in the "Eastern Alps but spreads far and wide beyond the latter over large parts of Europe. There is no doubt that it marks an epoch when the sea once more occupied the Triassic lowlands of Northern and Western Europe, and that over these lowlands it is a mere zone, from 30 to 40 feet thick, which is more intimately connected with the overlying Lias than with the underlying Keuper. In dealing with the stratigraphy of this system it will be convenient to take each of the three great divisions or series separately, and to give a condensed account of the component strata of each series, both in Britain and on the Continent, before proceeding to the next one. We can afterwards review the general history of the whole period, and indicate the geographical changes which took place at its close. B. LIFE OF THE PERIOD The reader may be reminded that throughout Triassic times our islands formed part of a continent which occupied the northern part of Europe, and some part at least of the North Atlantic region, and that it was not until the Rhaetic epoch that the inland lakes of this continent were finally submerged beneath the waters of an open sea. Further, the fauna of the British Rhaetic beds or zone of Pteria [ = Avicula] contorta in Northern Europe contains but a small number of species, and it was not till the time of the Lias that conditions were favourable for the establishment of a new and abundant marine fauna over the whole region. We are thus suddenly presented with an assemblage of fossils w 7 hich is very different from those of the Carboniferous and Permian periods, and 384 STRATIGRAPHICAL GEOLOGY for the ancestors of these we must search the marine Trias of Southern and Central Europe. The rate of change in the organic world and the differentiation of species appears to have become much more rapid in Jurassic times than it had been in Palaeozoic times, so that any given group of species has a much less vertical range, and consequently a much smaller set of strata requires separate description. It will, therefore, be convenient to describe the prevalent genera and the characteristic species of each series separately, but in the first place the distinguish- ing features of the Jurassic fauna as a whole may be indicated. In the first place it was an " age of Reptiles" and it is specially characterised by the remarkable development of the orders Dino- sauria, Pterosauria, Ichthyopterygia, and Sauropterygia. These orders survived into the Cretaceous period and then became extinct. The Linosauria were land animals, and many were of gigantic size (from 30 to 70 feet long) ; in many cases their hind limbs were much longer and stronger than the fore legs, so that they could walk in an upright position like a kangaroo. Some were carnivorous (as Megalosaurus) and some herbivorous (as Cetiosaurus). The Ptero- sauria are reptiles adapted for the purpose of flight in the air, and thus possess many bird-like characters, such as pneumatic bones, keeled sternum, and more or less bird-like skull ; their membranous wings were attached to the sides of the body and to the greatly elongated fifth digit. The Ichthyopterygia and Sauropterygia were specially adapted for life in the sea, both pairs of limbs being modified into the form of paddles or flippers. Representatives of the orders Orocodilia and Chelonia also occur. Birds also make their appearance, but are known by two specimens only, found in the lithographic limestone at Solenhofen in Bavaria. They are referred to the genus Archceopteryx, which exhibits strong reptilian affinities, possessing teeth and a long narrow tail of twenty separate vertebrae, each of which apparently carried a pair of feathers ; the wings are bird-like, and the creature was about the size of a common pigeon. It is in Jurassic strata also that the earliest mammals have been found, and these all belong either to the sub-class Prototheria (which includes the living duck-billed mole, Ornithorhynchus, and the spiny ant-eater, Echidna} or to the sub-class Metatheria (which includes the opossums and kangaroos). No placental mammals have been found in the Jurassic System. Mammalian remains have been found at three horizons ; the oldest (Microlestes) in the Rhaetic Beds at the very base of the system, others (e.g. Phascolotherium, Amphitherium), in the Great Oolite, and others (10 genera, e.g. Triconodori) in the very highest group (Purbeck Beds). THE JURASSIC SYSTEM 385 Another special feature of the fauna is the great abundance and variety of the Ammonoid Cephalopoda. A few genera such as Phylloceras appeared in Triassic times, but a large number occur in Jurassic rocks, and many are confined to the period. The genus NautiluSj which also dates from the Triassic period, is the only representative of the Nautiloidea. Belemnites, a genus of Dibranchiate Cephalopods, appears in the Lias and reaches its maximum development in Jurassic times. Gastropods are numerous, especially Nerincea, Pleurotomaria, Cirrus, Pseudomelania, Natica, and Amberlya. Among Lamellibranchia the Ostreidse, Pectinidee, Limidse, and Plioladoinyidre become abundant, and the genus Gryphcea is very characteristic of the period ; Trigonia also is abundant. Among Echinoderms the Euechinoid Group makes its appearance in England, and is very abundant both in species and individuals. The first irregular Echinoid is found in the Upper Lias. Some representatives of the Asteroidea, Ophiuroidea, and Crinoidea also occur. Corals also abound in most of the limestones. Plants occur at several horizons, and Cycad-like fronds, stems, and "flowers" are so abundant that botanists have called the Jurassic period the "age of Cycads." The chief genera are Zamites, Otozamites, Nilssonia,, Pterophyllum, and Williamsonia. Ferns continue to be common, and conifers were represented by genera allied to the modern Araucaria, Cupressus (cypress), and Taxus (yew). The following are some of the genera which are not known in Britain or France before Lower Jurassic time, though some occur in the Alpine Trias. Those preceded by an asterisk did not survive the Jurassic period. Anthozoa. Astroccenia, *Heterastrea, Isastrea, Montlivaltia, *Stylastrea, Thamnastrea, Thecosmilia, Trochocyathus. Echinoderma. *Pentacrinus, Isocrinus, Ophioderma, Ophio- lepis, Acrosalenia, Cidaris, Khabdocidaris, Hemipedina, Pseudo- diadema. Crustacea. *^Eger, *Eryma, Eryon, Glypheea, Penseus, *Scapheus. Brachiopoda. Waldheimia, Thecidium, Terebratella, Cado- mella. Lamellibranchia. Astarte, Cardinia, Cardita, Cardium, Proto- cardia, *Ceromya, Grammatodon, Cucullaea, Trapezium, *Gresslya, Gryphaea, Velopecten, *Hippopodium, Homomya, Inoceramus, Myoconcha, Opis, Ostrea, Plica tula, Perna, Pinna, Pholadomya, Goniomya, Pleuromya, *Tancredia, Thracia, Trigonia, Unicardium. G-astropoda. Actseonina, Amberleya, Cerithium, *Ceritella, 2C 386 STBATIGRAPHICAL GEOLOGY *Cryptaenia, Cylindrites, Dicroloma ( = Alaria), *Exelissa, Littorina, Moiiodonta, Nerita, Neritopsis, Patella, Pseudomelania, *Purpurina, Solarium, Scalaria, *Trochotoma, Turritella. Cephalopoda. *^Egoceras, *Amaltheus, *Arietites, *Ccelo- ceras, *Dactylioceras, *Deroceras, Harpoceras, *Haugia, *Hildo- ceras, *Grammoceras, *Liparoceras, Lytoceras, Holcostephanus, *0phioceras. *0xynoticeras, *Paltopleuroceras, *Phylloceras, *Psiloceras, *Schlotheimia, Belemnites. Pisces. Acrodus, *Dapedius, *Eugnathus, Hybodus, Lepi- dotus, *Leptolepis, *Pholidophortis. Reptilia. *Dimorphodon (a Pterodactyle), *Scelidosaurus (a Dinosaur), *Steneosaurus and *Pelagosaurus (Crocodiles), *Plio- saurus, Ichthyosaurus, Plesiosaurus, and Eretmosaurus (the last three, however, are represented in the Rhaetic bone beds). I. THE LIAS OR LOWER JURASSIC SERIES A. THE BRITISH LIAS Subdivisions. This series has on the whole a very uniform lithological aspect throughout England. It consists almost every- where of four successive sets of strata, and might with good reason be divided into four stages instead of the three which are usually adopted in England, for such a subdivision is not inconsistent with the range of the fossils, and is actually in use on the Continent (see table below). Thus (1) its lowest beds are more or less calcareous, and consist generally of thin-bedded limestones and shaly clays ; (2) above these comes a thick mass of dark-grey clays ; (3) these are succeeded by sandy clays and sandy limestone (marlstone) ; (4) while at the top are thin limestones and marly clays. A further subdivision of the Lias into zones has resulted from a careful study of the distribution of its fossils. These zones are each characterised by one or more species of Ammonites, and though some of them are more or less local, and cannot be traced all across England, they can be grouped into nine principal zones which are so traceable. The following table shows the divisions in general use, the principal zones, and the continental divisions : Lithology. Zones. Cont. Equiv. C Grammoceras aalense "\ Dumortieria radians Upper ( Clays and I Lillia Lilli \ Toarcian Lias \thin limestones | Grammoceras striatulum I Dactylioceras commune V Harpoceras serpentinum J THE JURASSIC SYSTEM 387 Lithology. Zones. Cont. Equiv. Middle f Maiistones and f Paltopleuroceras spiuatum \p, ,, . Lias \ sandy clays \ Amaltheus margaritatus j u ( ^Egoceras capricornus ~\ I Deroceras armatum Lower! Cla y S 1 Uptonia Jamesoni fSmemurian. Lias 1 I Oxynoticeras oxynotum } Limestones f Coroniceras Bucklandi \., and clays \ Psiloceras planorbis J M With respect to the Upper Lias there has been much discussion and difference of opinion as to the horizon at which the line should be drawn between the Lower and Middle Jurassic Series. Where the two series are fully developed, as in the south-west of England, there is often a considerable thickness of sand between the Lias clays and the limestones of the Inferior Oolite, and as the sands were clearly passage-beds, some have included them in the Upper Lias, and others have grouped them with the Inferior Oolite. The recent zonal work of Mr. S. S. Buckman has made it clear that the sands are not always on the same zonal horizon, but may be of any age from the Lilli to the aalense zone. It is also clear that they pass laterally into Upper Lias clays, and that they are as a whole more closely connected with the Lias than with the over- lying Inferior Oolite. Moreover, in Yorkshire there is a distinct stratigraphical break between the two series, and it occurs at the base of the "Dogger" sandstone which overlies the zone of Grammoceras aalense. Hence it is convenient to place the divisional line at the summit of the zone of Gram, aalense^ and to consider that of Lioceras opalinum as forming the base of the Middle Series. By adopting this line we are in agreement with the Committee who drew up the scheme of the International Geological map of Europe, and in England it includes all the sands in the Lias except about 5 feet at the top of them near Bridport. In some localities several zones are condensed into a limestone bed of a few feet or even inches in thickness, and this happens sometimes with the lowest zones and sometimes with the highest. Fossils of the Lower Lias Actinozoa. Montlivaltia rugosa, Heterastrea Tomesi. Echinoderma. Pentacrinus briareus, Isocrinus basaltiformis, Cidaris Edwardsi. Brachiopoda. Spiriferina Walcotti, Khynchonella variabilis, R. rimosa, Waldheimia (Cincta) numismalis. Lamcllibranchia. Pteria (Oxytoma) cygnipes, Cardinia Listeri, Gryphtea arcuata ( incurva), Hippopodium ponderosum, Lima (Plagiostoma) gigantea, Pseudomonotis decussata, Ostrea liassica. 388 STRATIGBAPHICAL GEOLOGY Gastropoda. Pleurotomaria anglica, Cryptsenia expansa, Amberleya elegans. Cephalopoda. The following Ammonites characterise zones or sub-zones in ascending order : Psiloceras planorbe, Schlotheimia angulata, Coroniceras Bucklandi, Arnioceras semi- costatum, Arietites Turneri, Asteroceras obtusum, Oxynoticeras oxynotum, Echioceras raricostatum, Fig. 125. GROUP OF RH^TIC FOSSILS. a. Protocardium rhseticum. tl Myophoria postera. ft. Pecten valoniensis. e. Orbiculoidea Townsliendi.. c. Pteria contorta. / Ceratcdus parvus. g. Neiuacanthus monilifer. Deroceras armatum, Uptonia Jamesoni, Rhacoceras ibex, ^Egoceras capricornus, Nautilus striatus, Belemnites acutus, B. elongatus, B. clavatus. Pisces. Dapedius punctatus, Hybodus reticulatus, Acrodus Anningse. Reptilia. Ichthyosaurus communis, Plesiosaurus dolichodeirus. Fossils of the Middle Lias Echinoderma. Ophioderma Egertoni, 0. Milleri. Brachiopoda. Rhynchonella tetrahedra, R. acuta, Terebratula punctata, Waldheimia (Aulacothyris) resupinata, ^^ r aldheimia. (Zeilleria) cornuta. THE JURASSIC SYSTEM 389 Fig. 126. LIASSIC AMMONITES. . Dactyhoceras commune (). d. Amaltheus margaritatus (A). 1). Harpoceras serpentinum (). , e. Coroniceras Bucklandi (i). c. JSgoceraa capricornus G). /. Psiloceras planorbe (i). 390 STEATIGRAPHICAL GEOLOGY Lamellibranchia. Pteria (Oxytoma) insequivalvis, Protocardia truncatum, Gresslya intermedia, Lima punctata, Modiola scalprum. Gastropoda. Trochus lineatus. e.'/ Fig. 127. GROUP OF LIASSIC TOSSILS. a. Pseudomonotis decussata, 6. Gryphsea arcuata. c. Hippopodium ponderosum. d. Pleurotpmaria anglica. e. Belemnites elongatus. /. Dactylioceras communis. g. Nautilus striatus. h. Lima gigantea. m. Leda ovum. n. Rhynchonella tetrahedra. Cephalopoda. Amaltheus margaritatus, Paltopleuroceras spinatum, Lytoceras fimbriatum, Belemnites paxillosus. leptilia. Pliosaurus propinquus. THE JURASSIC SYSTEM 391 Fossils of the Upper Lias Echinoderma. Pentacrinus jurensis. Brachiopoda. Rhynchonella jurensis, Waldheimia (Cincta) Lycetti, Discini reflexa, Cadomella Moorei. Lamellibranchia. Inoceramus dubius, Gresslya donaciformis, Leda ovum, Pecten pumilus, Posidonomya Bronni. Gastropoda. Trochus lineatus. Fig. 128. GROUP OF LIASSIC FOSSILS. a. Otopteris obtusa. ft. Pentacrinus briareus. c. Ophioderma Egertoni. d. Spiriferina Walcptti. e. Waldheimia nuraismalis. /. Rhynchonella rimosa. Cephalopoda. Harpoceras serpentinum, Harp, aalense, Harp, striatu- lum, Dactylioceras commune, Hildoceras bifrons, Dumortieria radians, Belemnites abbreviatus, B. Voltzii. Pisces. Lepidotus elvensis ( gigas), Leptolepis concentricus. Eeptilia. Ichthyosaurus acutirostris, Steneosaurus Chapmanni. Stratigraphy. The Lias occupies the coast of Dorset from Lyme Regis to Bridport harbour, it stretches inland to Ilminster, 392 STRATIGRAPHICAL GEOLOGY Petherton, and Yeovil, and thence passes northward to the Mendip Hills, where all the divisions become very thin. North of Bristol it thickens again and forms a broad tract running through the counties of Gloucester, Warwick, Northampton, Leicester, Lincoln, THE JURASSIC SYSTEM 393 and York. Another fine coast section is exposed in the cliffs of the Yorkshire coast between Kedcar and Whitby. Outlying tracts of Lias occur in Staffordshire, at Needwood Forest and north of Abbots Bromley ; in Shropshire, between Wem and Audlem, and in Cumberland, near Carlisle, proving that the formation originally extended far to the north-west of its main line of outcrop. There is little doubt, indeed, that the Lias was once coextensive with the Trias, but there is no indication that it ever extended very far beyond the limits of the Keuper marls. In Ireland Liassic rocks crop out in certain places from beneath the Cretaceous rocks of Antrim, and in Scotland remnants of Lias occur above the Trias on the west coast and on the east coast of Sutherland. With regard to the eastward subterranean extension of the Lias in England, it is known that its upper beds, as they pass beneath the higher members of the Jurassic System, thin very rapidly. Thus in a boring at Burford, in Oxfordshire, the Middle and Upper Lias are together less than 30 feet thick ; the Lower Lias is there about 620 feet thick, and the Rhsetic is 10 or 12 feet ; but between that place and Richmond the Lias has wholly disappeared, allowing rocks of the Middle Jurassic Series to rest directly on rocks which are either of Triassic or of Old Red Sandstone age. It would appear, therefore, that between Richmond and the Malvern Hills the width of ground now occupied by the Lias is less than 100 miles. No trace of Lias has been found in any of the deep borings that have been made in the eastern counties. Northward it may underlie part of Cambridgeshire, but it thins out beneath Suffolk, a boring at Culford, near Bury St. Edmunds, having proved Cretaceous Beds to rest directly on the Palaaozoic rocks. The chief zonal and provincial features of the British Lias may be indicated under the head of four provincial areas : (1) the south-west province, including Dorset, Somerset, Gloucester, and Glamorgan ; (2) The Midland Counties from Oxfordshire to the Huniber ; ,(3) Yorkshire ; (4) Cumberland, Ireland, and Scotland. 1. The South- West Province An excellent section of the whole Liassic Series is exhibited in the cliffs of the south coast of Dorset, from Lyme Regis on the west to Bridport on the east (see Fig. 130), and this may be regarded as the typical section of the English Lias, though some of the zones in the Lower Lias are not so fully developed as they are in the Midlands. THE JURASSIC SYSTEM 395 Lower Lias. The succession of beds which constitute the Lower Lias at Lyme Regis is as follows in descending order : 2 Feet. C Bluish - grey marly clays with Rhacoceras\ Green Ammonite) Loscombei \ ^Q- Beds J Bluish-grey marly clays with Platypleuroceras j V. latcccosta J D i <- -R A J Pale-grey marls with some marly limestones ; Ls \ few fossils 80 r Shaly maiis with Eckioceras raricostatum. . 15 pi i oh i ' Shaly marls with Oxynoticeras oxynotum . . 15 1 Dark-grey shales and limestones with Asteroceras {, obtusum and Asteroceras Brookei . . . 170 /'Shales and marls with Arnioceras semicostatum 18 rn T | Limestones and shaly clays with Coroniceras 1 Bucklandi and Schlotheimia angulata . . 67 V. Limestones and shales with Psiloceras planorbe 20 I " White Lias," limestones, and shales . . 25 Rhretic Beds -j Black shales with Pteria contorta ... 32 ( Grey marls and limestones (passage-beds) . 30 577 The Rhaetic Beds are exposed to the west of Lyme Regis at Charton Bay and Culverhole (see Fig. 131). Throughout Dorset and Somerset there are a set of passage-beds between the Keuper marls and the Rhsetic Beds, and these have recently been called the Sully Beds by Mr. L. Richardson. 3 The base of the Pteria contorta zone is marked by a layer of calcareous grit full of the bones and coprolites of fish and small reptiles ; this is also of general occurrence, and is known as " the bone bed." In the White Lias or Upper Rhsetic Ostrea liassica and Modiola minima are the commonest fossils. The succession of zones and sub-zones is normal up to that of Ech. raricostatus, but above that the typical Ammonites are rare. The Belemnite Beds are supposed to represent the armatum, Jamesoni, and ibex zones, but the two latter species of Ammonites do not occur. The succeeding beds may be regarded as the equivalent of the zone of JEgoceras capricornus, and its sub-zone of J&g. Henleyi. In Somerset the beds forming the Lower Lias are similar to those of the Dorset coast, but the higher zones are seldom exposed. Near Watchet on the Bristol Channel there is a good section of Rhadtic Beds (Black Shales and White Lias, 46 feet) and of Lower Lias for about 150 feet, including the planorbe, angulatum, Buck- landi, and semicostatum zones. The limestones of these lower zones have been quarried at many places from Ilminster eastward, and are upwards of 200 feet thick, eg / T s| S ll THE JURASSIC SYSTEM 397 about half of this belonging to the planorbe zone and the rest to those of Schlotheimia angulata and Coroniceras Bucklandi. The quarries at Queen Camel and Street have yielded many fine specimens of Ichthyosaurus, Plesiosaurus, fish, and other fossils, and at the base are shaly limestones which contain remains of insects and Decapod Crustacea (Eryori). Evidence that the higher zones persist is found in the occurrence of the Ammonites semicostatus, obtusus, and oxynotus, but it is doubtful if the Lower Lias in Somerset is more than 400 feet thick. Northwards, however, there is a rapid thinning, so that at Shepton Mallet the same four zones have a thickness of only 33 feet, overlain by 36 feet of clays and limestones referable to the obtusum and oxynotum zones. Bound the Mendip Hills there are only irregular patches of limestone and clay to represent the whole of the Lower Lias. To the north of the Mendips, however, it soon regains its normal aspect and thickness. Westward, in South Wales, there are fine exposures of the lower beds in the cliffs between Penarth and Dunraven Castle. At Penarth there is a normal sequence of Ehaetic Beds with a thickness of about 40 feet, and above them are the usual limestones and shales for 135 feet, but when followed westward they exhibit a remarkable lithological change. The Rhsetic Beds first pass into a massive oolite, and finally, near Bridgend and Pyle, into pale-green and yellow sandstones with a medial band of green marls, shales, and limestones, the whole being from 40 to 50 feet thick. The shales and limestones of the overlying Lias pass westward into massive pale-coloured marly limestones which overlap the Rheetic sandstones so as to rest, with a conglomerate base, on the Carboniferous limestone at Sutton and Brocastle near Bridgend. In these beds are corals of the genera Astroccenia, Montlivaltia, and Thecosmilia, together with Pseudomonotis fallax and Ostrea liassica. Ammonites are scarce, but the higher beds contain Schlotheimia angula,ta> Arietites BucJclandij and Arnioceras semicostatum. Middle Lias. On the Dorset coast this division is exposed in the cliffs between Charmouth and Bridport, and presents the following succession : Feet. Zone of Palt. / Ferruginous limestone (marlstone) ... 1 spinatum (Yellow sands and sandy clays .... 77 /"Laminated micaceous sands and clays . . . 72 I. | F Zone of AmaL | Blue micaceous clays with limestone nodules . 160 maryaritatus 1 " Three tiers" of calcareous flagstone with 1. sandy clays between ..... 35 345 398 STRATIGRAPHICAL GEOLOGY The blue clays contain. Amaltheus margaritatus, Belemnites elongatus, Plicatula spinosa, etc., and at their summit is a bed of calcareous sandstone known as the Starfish Bed, from its having yielded two species of Ophioderma. The Marlstone is here very thin. When followed northward into Somerset the whole group becomes thinner, its thickness near Glastonbury being little over 200 feet, but still having a bed of Marlstone rock at the summit, from 1 to 2 feet thick. Eound the Mendip Hills the Middle Lias becomes very attenuated, and is represented sometimes by a few feet of marly shale, and sometimes by conglomerate resting directly on Carboniferous limestone ; in other places it is over- lapped by the Inferior Oolite. Towards Bath the normal succession is again found, but with a thickness of only 50 or 60 feet. Northwards through Gloucester- shire the beds thicken considerably, till near Stinchcombe the Marlstone is 15 feet, and the underlying sands and clays are about 150. The Marlstone here forms a conspicuous rock -bed which is often quarried. It contains Paltopl. spinatum, Belemnites paxil- losus, Rhynch. tetrahedra, Pecten cequivalvis, and other fossils. Upper Lias. In Somerset and Dorset this stage may be described as consisting of three unequal portions (1) the Junction Bed, (2) the Upper Lias clays, (3) the Upper Lias sands. The first is a bed of pale-grey argillaceous limestone (1^ to 2^ feet thick), which is welded to the top of the Marlstone and contains in successive layers the characteristic species of three zones, viz. Harpoceras falcife?'um, Hildoceras bifrons, and Harpoceras striatulum. The clays or shales are about 70 feet thick and contain but few fossils, but species of Dumortieria have been found in them. They pass up into fine yellow sands containing large burrstones or lumps of calcareous sandstone. Of such sand there is nearly 200 feet, but most of it yields no fossils ; between 40 and 50 feet from the top a band contains Dumortieria radians and I). Moorei, and then come 30 feet with species of the Harpoceras aalensis type, and these are succeeded by about 5 feet of sand with opalinoid forms, so that the passage from one stage to the other here takes place very near the top of the sands. 4 In Somerset, near Yeovil, there is a similar succession, but the base is formed by a band of shaly clay containing many small Brachiopoda such as Leptcena Moorei y L. Bouchardi, and Zellania liassica. This is overlain by a bed of limestone with Harp. falctferum, and remains of fish, Crustacea, and insects. Above are blue clays succeeded by sands with Ammonites of the radians type (Dumortieria), and in the higher part of these sands a THE JURASSIC SYSTEM 399 lenticular mass of oolitic freestone is developed with a local thick- ness of 80 feet and containing Dum. Moorei ; this is known as the Ham Hill Stone. Northward the sands thin out entirely, so that near Bath the radians zone is represented entirely by clay, but is underlain by sands which Mr. Buckman assigns to a sub-zone intermediate between that of Lilli and that of radians, and at its base there is a condensed limestone representing lower zones. In Gloucestershire there is another change and a different succession : (1) clays representing the falciferum and bifrons zones ; (2) the Cotteswold sands, which include the zones of Harp. Lilli and H. variabilis ; and finally a Cephalopoda Bed representing all the higher zones in successive layers of limestone. The clays vary from 50 to 130 feet, the sands are on the average 150 feet, and the limestone varies from 2 to 8 feet. 2. The Midland Counties Bhaetic Beds. These form a narrow but continuous band between the Keuper marls and the dark shaly clays of the Lias, but they are rarely exposed in Worcester. In Warwickshire there are exposures of the White Lias at Knowle (near Solihull) and at Harbury, Rugby, Southam, and Eatington, and a good section has been described at Wigston, near Leicester, where the following succession was seen : Feet. 2. Blue shales with two thin beds of limestone and sandstone . 22 1. Black shales containing Schizodus cloacinus, Protocardium rhwti- cum, Pteria contorta, and fish teeth ...... 18 A similar succession is found in Rutland and Lincolnshire. The Lower Lias is seldom well exposed in Oxfordshire or Warwickshire, but a boring has proved its thickness to be 450 feet, and fairly good sections are found near Harbury and Fenny Compton, where all the zones are seen except those of oxynotum and armatum. In the counties of Leicester, Rutland, and Lincoln the Lower Lias is both more accessible and more fully developed, and the following succession of beds has been described : (Blue clays with septaria (^Egoceras capricornus}. Blue clays with thin limestones and septaria (Rhacoceras ibex and Uptonia Jamesoni). B "I Sandy clays and sands, occasionally indurated into a stone (Deroceras artrwtum). .Blue clays with small septaria (Oxynoticeras oxynotum}. 400 STKATIGRAPHICAL GEOLOGY ( Ferruginous limestones with Arnioceras semicostatum. Lower Lias, J Blue clays and grey limestones (Ooroniceras Bucklandi*) . A | Dark- blue pyritous clays (Schlotheimia anyulata). ^Limestone and shales (Psiloceras planorbe). The zones of planorbe, angulata, and Bucklandi present the same features as elsewhere, and have a combined thickness of about 200 feet. They are succeeded by several beds of hard ferruginous limestone with interbedded shales (the semicostatum zone), which thicken northward till they are 27 feet thick at Scunthorpe and Frodingham, where they are worked for ironstone. The upper part (B) of the Lower Lias is about 470 feet thick in the Vale of Belvoir, and a boring at Grantham showed the Lower Lias as a whole to be about 700 feet thick. The oUusum zone is thin or absent, but clays with oxynotum and raricostatum occur, and are about 90 feet thick. Northwards, however, the thickness diminishes. The lower zones (A) maintain their thickness, but the zones of oxynotum, armatum, and Jamesoni are represented by clays from 100 to 140 feet thick, surmounted by a bed of ironstone about 4 feet thick, which is known as the " Pecten Bed," from the abundance of several species of Pecten. This bed contains a mixture of Lower and Middle Lias species, and is by some regarded as the base of the Middle Lias, but its Ammonites are armatum, Henley i, and striatum, while the clays above it contain ^Egoceras capricornus throughout and not Amaltheus margaritatiis. These clays are about 70 feet thick, and consequently the total thickness of Lower Lias in this district is from 400 to 440 feet. Middle Lias. In the counties of Oxford and Warwick the Middle Lias is about 150 feet thick, and Mr. H. B. Woodward 5 remarks that " the Marlstone Rock-bed is perhaps nowhere better developed than in the country around Banbury. It covers an extensive area to the north-west, forming a plateau that rises gradually from about 500 feet at Banbury to the escarpment of Edge Hill, 710 feet high, and this area is intersected by several deep valleys." The Marlstone consists of more or less ferruginous and sandy limestone, which varies from 12 to 25 feet in thickness. The outer parts are always brown from the oxidation of the iron, but the centres of large blocks are generally grey. The commonest fossils are Terebratula punctata and Rhynchonella tetraliedra, which often occur in clusters. Upper Lias. This division is well developed in North- amptonshire, where it has been studied and described by Mr. B. Thompson. 6 He divided it into 8 zones, but the highest of these forms a passage into the overlying series, and would perhaps be better included in the latter ; his 7th zone is a nodule bed at the THE JURASSIC SYSTEM 401 base of this, so that the restricted Upper Lias would consist of the following : Feet. Leda Ovum Beds (zone of Am. Lilli) .... 36 Cerithium Beds (Cerithium armatum) .... 36 Zone of Dadylioceras fibulatum and Hildoceras bifrons . 76 ,, Dadylioceras subcarinatum (Commune Beds) . 5 ,, Harpoceras falciferum (Serpentinum Beds) . . 4 ,, Am. latescens (Fish Beds) 2 159 A similar set of beds is found in Rutland, but in Lincolnshire the highest beds seem to have suffered erosion before the deposition of the overlying sandstone, and the highest remaining part of the Lias belongs to the bifrons zone. On the other hand the serpentinum Beds thicken northward to about 40 feet, and the total thickness of the stage at Lincoln is about 100 feet. 3. Yorkshire In the south of Yorkshire, from Brough to Kirby Underdale and Howsham, the Lias is similar in all respects to that of North Lincoln, but as soon as the valley of the Derwent is crossed a change begins to be manifest, both the Middle and Upper Lias swelling out till they again attain a development equal to that of South Lincolnshire. From Howsham the outcrop curves round by Thirsk and Northallerton to the coast between Redcar and Whitby, which exhibits an excellent section of the whole formation except the zone of Psiloceras planorbe. The following is a resume of the coast section : 7 _ U PP er ' Blea Wyke sands with Harpoceras aalensis, 50 feet. Blue shales, with Harp, striatulum and H. variabilis, 70 feet. "Alum shales" with Dactylioceras commune, Hildoceras 330 "feet < bifrons, and Leda ovum, 90 feet. ' Blue shales with calcareous doggers and layers of jet in the lower part. Harpoceras serpentinum, Belemnites tubu- laris, Inoceramus dubius, 90 feet. | Grey shales with Dactylioceras annulatum and Belemnites v cylindricus, 30 feet. {Ironstone (Cleveland main seam) and shales below, zone of Amaltheus spinatus, 30-50 feet. Ironstone (lower seam) or Pecten Bed, with beds of marly sandstone below ; contains Amaltheus margaritatus and Pecten Icevis, about IdO feet. 2D Lower, 714 feet 402 STRATIGRAPHICAL GEOLOGY Clays with ^Egoceras capricornus, 130 feet. Shaly micaceous clays with Uptonia Jamesoni, 225 feet. Clays with Asteroceras obtusum, Oxynoticeras oxynotum, and JSchioceras raricostatum, 100 feet. Shales and limestones with Arietites Turneri and Arnioceras semicostatum, 66 feet. Shales and thin limestones, zone of Arietites BucJclandi, 113 feet. Shales and thin limestones, Schlotheimia angulata, 30 feet. Zone of Psiloceras planorbe below, but not accessible ( ? 50 feet). Mr. E. H. Kastall has shown 8 that it is only at Blea Wyke on the coast that a complete passage from Lias to Inferior Oolite is seen. At other localities the ferruginous sands which form the base of the latter (see p. 420) rest upon an eroded surface of the Lias and usually upon some part of the commune zone. In many places currents have cut channels in the Lias, and these are often filled with sand at the base of which is a layer of pebbles and phosphatic nodules. 4. Cumberland, Ireland, and Scotland These areas may well be considered together, for there can be no doubt that the Liassic Series was originally continuous across the seas which now divide them. Lower Lias. It is interesting to find an outlier of Lias in the Carlisle basin, because it confirms the view that the northern Pennine Hills were not then in existence, but were part of the Triassic plain which became first the floor of the Keuper Lake and subsequently that of the Liassic Sea. It is believed, however, that the Rhsetic Beds are absent, and that the zone of PsiL planorbe rests unconformably on the Keuper marls. The whole country is covered with glacial drift, so that there are few exposures, but a boring near Great Orton passed through 210 feet of dark shales with bands of limestone in the lower part. In Ireland the Lower Lias is found at intervals along the coast of Antrim between the Trias and the Cretaceous rocks, but is frequently cut out by the overstep of the latter. The zone of Pteria contorta is well developed, the lower part consisting of grey shales and sandstones with fish remains, the middle of black shales with the usual fossils, and the upper of grey shales with layers of oolitic limestone, and the whole is about 80 feet thick. These beds are succeeded by Lower Lias with the same lithological facies as in England, but the series is cut off in or above the zone of Echioceras semicostatum. The mapping of the Isle of Arran by the Geological Survey THE JURASSIC SYSTEM 403 revealed the fact that a similar succession of beds existed there in Tertiary times, for in the conglomerate which fills a large volcanic orifice or neck (see p. 368) fragments and tabular masses of Rhaetic and Liassic limestones were discovered, the fossils found in them leaving no doubt about the age of the rocks. Farther north on the coast of Argyleshire and in the islands of Mull, Skye, and Raasay, a nearly complete succession of Liassic deposits is found, which differ in several respects from those of England. They were fully described by Professor Judd in 1878, 9 from whose account the following is mainly taken. The Rhaetic zone is thin and seems to be represented by about 8 feet of calcareous sandstones, limestones, and shales. The over- lying beds are very hard bluish-grey limestones, alternating with calcareous grits and conglomeratic sandstones. The commonest fossil is Ostrea irregularis (a variety of 0. liassica), but corals Thecos- milia Martini and Isastrea Murchisoni also occur, together with Gardinia concinna and spines of Acrosalenia. Ammonites are not found, but the beds may be regarded as equivalent to the planorbe and angulatum zones ; their thickness is about 200 feet, and they are well seen at Applecross and on the opposite coast of Raasay (see Fig. 132). The succeeding zone of Arietites Bucklandi is well developed, and consists of shelly limestones and shales crowded with fossils of the same species which are common in England ; its thickness is 250 feet. Above are dark micaceous shales containing Arn. semicostatum, Asteroceras BrooJci, and Belemnites acutus, 150 feet. The separate existence of the oxynotum zone is not determined, the fossils of this zone only occurring at Tobermory in Mull, where they are mingled with others that belong to higher horizons. The higher part of the Lower Lias consists of sandy and micaceous shales called the Pabba shales by Professor Judd. At the base of these there are some beds in which Deroceras armatum occurs, but the greater portion of the group represents the zones Jamesoni and capricornus ; these shales contain Uptonia Jamesoni, Platypleuroceras brevispina, Deroceras Davcei, Modiola scalprum, and Gryphcea cymbium, and are about 250 feet thick. In the north-east of Scotland beds of Liassic age occur in the reefs below Dunrobin Castle on the Sutherland coast. Here the basement beds consist of coarse sandstones and conglomerates, con- taining fragments of the cherty rock mentioned on p. 369. These beds are succeeded by a remarkable set of estuarine deposits, consisting of sandstones and shales with thin layers of clay and coal, the whole attaining a thickness of between 400 and 500 feet ; no fossils have been found except in the uppermost beds, 404 STRATIGEAPHICAL GEOLOGY where a few dwarfed Gryphcea and Cardinia prove the occasional presence of salt water. As these beds are overlain by clays with Oxynoticeras oxynotum, they must be the estuarine equivalents of the planorbe and BuMandi zones. The overlying marine beds exhibit the following succession : 10 Feet. 3. Dark-blue micaceous clays with septaria, Belemnites acutus, Platypleuroceras brevispina, Uptonia Jamesoni, Hippopodium ponderosum, and Gryphcea cymbium 80 2. Micaceous sandstones and clays with Gryphcea obliquata, Pecten cequivalvis, and Belemnites acutus ..... 34 1. Blue clays and impure limestones with Oxynoticeras oxynotum, Arnioceras caprotinus, Bel. acutus, Cardinia hybrida, Lima pectinoides, and Gryphcea obliqua ...... 60 Middle Lias. There can be little doubt that deposits of this age succeeded the Lower Lias over the whole region of the Irish Sea and the Hebridean Sea, but the only remnants of them which have escaped destruction are in some of the Western Isles. Here they are shallow- water deposits, consisting chiefly of sandstones. They are well exposed in Scalpa and Kaasay, where grey calcareous sandstones, with some beds of calcareous shale, have a thickness of 200 feet. They contain such characteristic fossils as Amaltheus margaritatus, Paltopleuroceras spinatum, Belemnites acutus, and Rhynch. tetrahedra. In Mull they are represented by soft greenish sandstones with few fossils. In Sutherland no such beds are exposed in situ, but their former existence is proved by the occurrence of boulders of calcareous sandstone, like those of the western coast, in the boulder-clays of Moray and Elgin. These boulders enclose such fossils as Belemnites paxillosus, Modiola scalprum, Pteria incequivalvis, and Panopcea elongata, and others which occur in the Scalpa sandstones of the west coast. It is quite possible indeed that such rocks (of Marlstone age) may still exist in situ beneath the vast masses of glacial clay which mantle the surface of Elginshire. Upper Lias. This also is only found in the Western Isles, where it presents the usual shaly facies with much pyrites and jet ; the lower portion yields Harpoceras serpentinum and Dumor- tieria radians, and the upper beds Dactylioceras commune, with Belemnites Voltzi and Posidonomya Bronni. Sections are generally obscured by slips and grass, but the thickness averages from 75 to 80 feet. B. CONTINENTAL LIASSIC STRATA Jurassic rocks form a nearly complete ring round the great Cretaceous basin of the north of France, the continuity of their THE JURASSIC SYSTEM 405 outcrop being interrupted only in the north-east where they are concealed beneath the overstepping Cretaceous deposits between the Ardennes and the Bouloniiais. It is unnecessary to burden the student with details of the local variations exhibited by the Lias in this and other parts of France, and it will suffice to give a brief account of its development on the eastern borders of the country in Alsace, Lorraine, and the Jura district, while the very different facies found in Provence will be referred to under the head of the Alpine region. French geologists divide the Liassic Series into five stages, making the Rhaetic one of them, but as it has really only the value of a zone it is better to include it in the overlying stage (Hettangian). If this is done we have the following stages in descending order : 4. Toarcian. Zones ofjurense, commune, and falciferum. 3. Charmouthian. Zones of spinatum, margaritatus, capricornus, and armatum. 2. Sinemurian. Zones of raricostatum, obtusum, Turneri, and Buck- landi. 1. Hettangian. Zones of angulata, planorbe, and Pteria contorta. The French also include the zone of Lioceras opalinum in their Toarcian, but the Germans do not, and in England its equi- valent is usually regarded as the base of the Inferior Oolite. How far the Hettangian zones extend under the Paris basin is unknown, but probably all the higher stages are continuous beneath that basin and the English Channel, passing into their equivalents in the south of England. The tipper Lias has recently been reached by a deep boring at Rouen. Hettangian. This takes its name from Hettange in the south of Luxemburg, in which district this part of the series has an unusual thickness because it consists largely of estuarine beds, as below : Feet. Li s /Sandstone with marine shells and plant remains . 200 ' \Blackclaysandblacklimestones ... .30 t>i ( ,. /Red clays with one or more bone beds . .18 /lc \Yellow micaceous sandstones . . . . 25 to 30 The black beds contain few fossils, but must represent the zone of Psiloceras planorbe, and the Hettange sandstone that of Schlo- theimia angulata. In Alsace and Lorraine the Rhsetic presents a similar facies and maintains a thickness of from 30 to 40 feet, but both the zones of Psiloceras planorle and Schlotheimia angulata are so thin as to be scarcely recognisable. Sinemurian. In the north-east of France this stage consists principally of grey limestones in regular beds full of Gryphcea arcuata, 406 STRATIGKAPHICAL GEOLOGY so that they are known as the " calcaire a Gryphees" but they also contain Goroniceras Bucklandi, Arnioceras bisulcatum, and other fossils. These beds are from 100 to 130 feet thick, and are over- lain by some 80 feet of uiifossiliferous shales which occupy the place of the higher zones. Charmouthian. In Alsace and Lorraine this stage consists mainly of calcareous marls with ferruginous nodules, surmounted by hard limestones with Paltopleuroceras spinatum, the whole being about 200 feet thick. To the north-west in Luxemburg and the Ardennes the lower beds become sandy, and near Mezieres the succession, according to Professor Gosselet, is as follows : Feet. Ferruginous limestone with Paltopleuroceras spinatum . 130 Marls with Bel. clavatus and ^Jgoceras capricornum . .160 Sandy limestone with Microceras planicostum ... 60 350 Toarcian. In Alsace and Lorraine this group consists in the lower part of bituminous shales full of Posidonomya Bronnij but also yielding Hildoceras bifrons and Harpoceras falciferum; the higher beds are marls with Lytoceras jurense from 10 to 12 feet thick, while the shales below vary from 30 to 160 feet. The details of the Toarcian have been more fully worked out in Normandy by Messrs. Brasil and Buckman, 11 who have recognised all the Dorset sub-zones, though the whole stage is only 32 feet thick and consists entirely of shales and clays. It is directly overlain by limestone with Lioceras opalinum. In the district of the Jura Mountains the Liassic Series is not so fully developed as the higher parts of the Jurassic System, but the whole of it is exposed in different places and has the following composition : Feet. Toarcian. Marls and Posidonia shales 150 Charmouthian. Marls with Belemnites ..... 100 Sinemurian. Blue clays and Gryphcea limestones ... 60 Hettangian. Limestones ........ 6 Rhcetic. Sandstones, marls, and limestones with a bone bed at the) base / 366 1. Germany The Lias is chiefly exposed in the south of Germany, its outcrop extending in a long strip from the foot of the Black Forest on the west through Swabia and then curving northwards by Nuremburg to Bamburg, Coburg, and Baireuth. In this area it has been THE JURASSIC SYSTEM 407 studied by Oppel, Quenstedt (1850-88), Struckmann (1881), Waagen (1864), and S tuber (1893). Its thickness is here compara- tively small, but most of the zones established in England and France have been recognised, and the general succession of beds is as follows : Upper, /Grey marls with Lytoceras jurense. 30 feet ^Bituminous shales with Posidonomya. f Limestone with Pallopleuroceras spinatum. Middle, f Clays with Amaltheus margaritatus. 100 feetl Hard marls with Deroceras Davcei. Delays with Rhacoceras ibex and Uptonia Jamesoni. f Clays with Asteroceras obtusum and Oxynoticeras oxynotum. Bituminous shales and Pentacrinus Beds. Limestones with Gryphcea and Coroniceras Bucklandi. 14oV"t"l Sandstones with Sdilotheimia angulata. I Limestones with Psiloceras planorbe. VRhaetic sandstones and shales (20 to 40 feet). In Northern Germany there are irregular outcrops of Lias along a tract extending from the Teutoburger Wald eastward through the south to Hanover and Brunswick ; the component beds being very similar to those of Swabia and Franconia. Moreover, it appears to extend northward and north-eastward below the great plains of Hanover and Prussia, for its existence has been proved by a boring at Hermsdorf near Berlin and by another at Kammin in Pomerania. 2. Sweden The Lias recurs on the north side of the Baltic in Scania, where the Eheetic position has an estuarine facies, consisting of a series of sandstones, shales, and clays with beds of lignitic coal and of clay ironstone ; in fact a " carboniferous " facies, with a thickness of about 300 feet. Plant remains are very abundant, including species of Camptopteris, Lepidopteris, Thaumatopleris, Nilssonia, and Dictyophyllum with many Cycads and some coniferous trees. Over- lying these deposits are shales and sandstones with marine fossils, as well as plant remains, and these appear to represent the zones of Psiloceras planorbe and Coroniceras Bucklandi. In Central Scania to the north-east the Rheetic Beds have disappeared, and the marine shales are represented by massive sandstones which lie directly on the Archaean rocks and are made up of detritus derived from the latter. 3. The Alpine Region In this region and in the south of Europe generally the Lias assumes a completely different facies from that which prevails in Germany and the north of France, for it passes into a series of 408 STRATIGRAPHICAL GEOLOGY massive limestones of which the thickness often amounts to several thousand feet. They indicate, in fact, a continuance of the pelagic conditions which prevailed in the same region during Triassic times. In the Western Alps Haug has described the Lias as forming three distinct bands round the core of the Cottian Alps ; the innermost consisting of crystalline limestones, often brecciated and containing large blocks of the older rocks, but frequently fossili- ferous and characterised by a fauna of Corals, Lamellibranchs, and Gastropoda; the second formed of compact limestones (never crystalline), characterised by the abundance of Ammonites ; and thirdly, bedded limestones with Crinoids and many Brachiopoda. He regards these three facies as replacive, one outside the other, probably formed in zones of increasing depth of water. The outer belt is well developed round Digne and in the Basses Alpes, where all the French divisions can be recognised, and the several thicknesses are as below given : Feet. Toarcian (bifrons and radians zones) .... 660 Charmouthian (limestones and marls) .... 860 Sinemurian limestones . 200 Hettangian (180) and Rhsetian (120) . . . .300 2020 In the Tyrol and Eastern Alps the Lias is represented by red and variegated limestones and spotted marls (Fleckenmergel'), all of which abound in Cephalopoda. The Bhsetic zone of Pteria contorta here forms a link between the Trias and the Lias, being in some localities a part of the great Dachstein limestone, but the typical Kossen Beds are marls containing P. contorta and black limestones with many Brachiopoda. In the Adneth and Salzburg district the highest member of the Bhaetic is a limestone full of the coral Calamophyllia ( = Lithodendrori), and this is overlain by variegated limestones containing many Ammonites of the genera Arietites, Phylloceras, and Lytoceras which must represent the Hettangian and Sinemurian stages. In the Bavarian Alps the Middle Lias is represented by varie- gated marls with Amaltheus margaritatus, but elsewhere it does not seem to be specially distinguishable in the mass of limestones and marls. The Upper Lias (or Toarcian) is more clearly definable, its equivalent in the Bavarian Alps being red and grey limestones with Hildoceras bifrons, and in the Tatra district near Vienna by fleckenmergel with the same fossil. Farther east the whole formation passes into a littoral facies of THE JURASSIC SYSTEM 409 sandstones, marls, and shales, with occasional coal-seams, and this extends through Hungary to the Carpathian Mountains. It is evident, therefore, that the sea shallowed rapidly to the east as well as to the north. Beyond this in Russia no representative of the Lias is known, the lowest beds 'of the Jurassic System in Central Russia belonging to the Cornbrash or the Callovian. Liassic deposits, however, are found again to the south-east in Servia, Croatia, and Dalmatia, and again in the Caucasus, while littoral beds with plant remains (possibly of Liassic age) have been found at Isjoum on the Donetz in South Russia. II. MIDDLE JURASSIC SERIES A. STAGES AND ZONES This series has always bgen divided into two stages ; a lower, known as the Inferior Oolite Group in England and the Bajocian in France, and an upper, best called the Bath Oolite Group and in France the Bathonian. The precise limits of these stages have, however, been matters of dispute, and are not yet settled. As stated on p. 387, the most convenient horizon at which to draw the line between the Lower and the Middle Series seems to be at the base of the opalinum zone. This places the base of the Inferior Oolite at an horizon which can easily be followed all across England, and it is at this horizon also that the base of the series is taken on the International Geological map of Europe. With respect to the upper limit of the Bajocian or Inferior Oolite, there is a growing prevalence of opinion that it must be placed at a lower horizon than formerly, and that the zone of Parkinsonia ParJcinsoni should be included in the Bathonian ; for the fauna of this zone is more closely related to that of the beds above it than to that below, and in England there is a marked stratigraphical break at its base. Again, there can be no doubt that whatever line is taken as the plane of separation between the two stages, this line should be drawn at the same horizon both in England and France, so that the stages should be completely correlative. During the last ten years much detailed zonal work has been done both in England and France, and a large number of zones and sub-zones (or hemerae) have been proposed. For those who make a special study of this series such work is of much value, but for the general student it will suffice to indicate the broader zones which can be traced over wide areas in England, France, and Germany. These are given in the following table : 410 STRATIGRAPHICAL GEOLOGY Bath Oolites or Bathonian Inferior Oolite or Bajocian Zone. Clydoniceras discus. Waldheimia digona. Oppeliafusca and Macrocephalites subcontracts. ParJcinsonia ParMnsoni. Cceloceras Blagdeni. Witchellia Ammonites. Lioceras concavum. Ludwigia Murchisonce. Lioceras opalinum. B. LIFE OF THE PERIOD Marine Fauna. The fauna of the Middle Jurassic rocks is closely related generically to that of the Lias, but differs princi- pally in the relative abundance of Corals, Echinoderms, and Gastro- pods. This, however, is sufficiently explained by the fact that these limestones were evidently formed in clear water, and some- times in the neighbourhood of coral reefs. Bracniopoda, too, are individually very abundant, though the number of species is not greater than in the Lias, and the genera are the same, except that Cadomella and Spiriferina are absent. The following are the more important genera which now appear, and those to which an asterisk is prefixed did not survive the Jurassic period : Porifera. Corynella, Craticularia, Holcospongia, Lymnorella, Peronidella, Verrucocaelia. Phyllocsenia (^Adelastrea), *Anabacia, Cryptocrenia, Cyathophora, Stylina. Apiocrinus, *Clypeus, Collyrites, Nucleolites, Hemici- daris, Holectypus, *Hyboclypeus, Pedina, Pygaster, Pygurus, Stomechimis. Terebratulina. Lamellibranchia. Corbis, Corbicella, Cyprina, Isocardia, *Pachyrisma, Pteroperna, Sowerbya, Spondylus, *Trichites. Gastropoda. Brachytrema, Bulla, *Cirrus, Dephinula, Fissurella, Nerinsea, Pileolus, *Purpuroidea, Rissoina. The Ammonite genera Cosmoceras, *Haploceras, Lioceras, Ludwigia, Macrocephalites, Oppelia, *Parkinsonia, Perisphinctes, Sonninia, and Sphsero- ceras. Aspidorhynchus, Microdon, Strophodus, Ganodus, Ischyodus. Eeptilia. Streptospondylus, Teleosaurus. The following are lists of the fossils which may be regarded as specially characteristic of the groups indicated : Fossils of the Bajocian (Inferior Oolite) Actinozoa. Anabacia hemispherica, Thecosmilia gregaria. Echinoderma. Stom echinus germinans, Pygaster semisulcatus, Hybo- clypeus agariciformis. Anthozoa. Echinoderma. Brachiopoda. Cephalopoda. Pisces. THE JURASSIC SYSTEM 411 Brachiopoda. Rhynchonella (Acanthothyris) spinosa, Terebratula fimbria, T. maxillata, T. perovalis, T. Phillipsii, Waldheimia (Aulacothyris) carinata. Lamellibranchia. Astarte elegans, A. excavata, Ceromya bajociana, Gresslya abducta, Gryphsea sublobata, Pholadomya fidicula, Trigonia costata, T. denticulata, T. hemi- spherica. Gastropoda. Cirrus nodosus, Nerinsea cingenda, Pleurotomaria ornata, Pseudomelania procera, Bourguetia striata. ?Fig. 133. GROUP OF INFERIOR OOLITE FOSSILS. o. Parkinsonia Parkinsoni. d. Trigonia costata. b. Stepheoceras Humphriesianum. e. Pholadomya fidicula. c. Nautilus sinuatus. /. Rhynchonella spinosa. g. Terebratula perovalis. Cephalopoda. (For Ammonites see ante, p. 410) also Stepheoceras Humphriesianum, Sonninia Sowerbyi, Nautilus truncatus, N. sinuatus. Fossils of the Great Oolite and Forest Marble Actinozoa. Anabacia complanata, Isastrea explanatuia, Apio- crinus Parkinsoni. Echinoderma. Acrosalenia hemicidaroides (ranges to Cornbrash), Hemicidaris minor, Clypeus Ploti, Collyrites ringens. 412 STRATIGRAPHICAL GEOLOGY Brachiopoda. Khynchonella concinna, R. obsoleta, Terebratula. maxillata, T. (Dictyothyris) coarctata, Waldheimia digona. Lamellibranchia. Pteria (Oxytoma) costata, Lucina bellona, Lima cardii- formis, Grammatodon hirsonensis, Ostrea acuminata, 0. Sowerbyi, Pachyrisma grande, Trigonia Goldfussi, T. Moretoni. Gastropoda. Dicroloma (Alaria) armatum, Amberleya nodosa, Cylindrites acutus, Purpuroidea Morrisi, Nerinsea Voltzi, Patella rugosa, Nerita costulata. Fig. 134. GROUP OF BATHONIAN FOSSILS. a. Pterophyllum comptum. &. Hemicidaris minor. c. Rhynchonella concinna. d. Waldheimia digona. e. Lima cardiiformis. /. Trigonia Goldfussi. g. Purpuroidea Morrisi. h. Nerinsea Voltzi. Cephalopoda. Parkinsonia Parkinson!, Zigzagiceras arbustigerum, Macrocephalites subcontractum, Oppelia fusca, Nautilus Baberi. Pisces. Strophodus magnus, Pholidophorus minor. Reptilia. Steneosaurus brevidens, Cimoliosaurus erraticus. Fossils of the Cornbrash Actinozoa. Anabacia complanata. Echinoderma. Holectypus depressus, Nucleolites clunicularis. THE JURASSIC SYSTEM 413 Fig. 135. GROUP OF CORNBRASH FOSSILS (WITH APIOCRINUS FROM THE BRADFORD CLAY). a. Acrosalenia hemicidaroides. b. Apiocrinus Parkinsoni. c. Terebratula intermedia. d. Gresslya peregrina. e. Pleuromya decurtata. /. Pholadomya lyrata. g. Clydoniceras discus. h. Macrocephalites macrocephalus. i. Nucleolites clunicularis. 7c. Ostrea flabelloides. 414 STKATIGKAPHICAL GEOLOGY Brachiopoda. Terebratala intermedia, T. Bentleyi, "Waldheimia obovata. Lamellibranchia. Pseudomonotis echiiiata, Pecten vagans, Lima dupli- cata, Pleuromya decurtata, Gresslya peregrina, Pholadomya lyrata, Ostrea flabelloides. Gastropoda. Pseudomelania vittata. Cephalopoda. Clydoniceras discus, Macrocephalites macrocephalus. Fig. 136. RESTORED SKELETON OF MEGALOSAURUS BUCKLANDI (^). Actual length supposed to be about 30 feet. Terrestrial and Freshwater Fauna. The estuarine beds which occur occasionally in these rocks furnish us with some interesting remains of the plants and land animals. The borders Fig. 137. JAWS OF (1) PHASCOLOTHERIUM BUCKLANDI AND (2) AMPHITHERIUM ERODE RIPI, FROM STONESFIELD. Natural size, with enlarged views of teeth above. of the rivers and estuaries must have been clothed with a dense vegetation, ferns of the genera Matonidium, Laccopteris, Coniopteris, Cladophlebis, Todites, Sphenopteris, Twniopteris ; cycads of many genera, Williamsonia, Ctenis, Nilssonia, Anomozamites, Otozamites, etc. ; coniferous trees, such as Araucarites, Brachyphyllum, Baiera, THE JURASSIC SYSTEM 415 and Ginkgo. Among this vegetation insects abounded, such as bettles, flies, and dragon-flies. In the water lived molluscs of the genera Cyrena, Anodonta, crocodiles of the Gavial type, and turtles (Protochelys). On land there were small marsupials, Amphitherium, Amphi- tylus, Amphilestes, and Phascolotherium (Fig. 137), and the Protothere Stereognathus. The huge Dinosauria, Megalosaurus and Ceteosaurus, were also terrestrial creatures (see Fig. 136). C. THE MIDDLE JURASSIC OF ENGLAND Geographical Range. Like the Lias, the Middle Jurassic Series ranges completely across England from Dorset to Yorkshire. It occupies the Dorset coast between Bridport and Weymouth, and runs inland by Yeovil, Sherborne, and Bruton to the Mendips. In Gloucestershire it gives rise to the well-known scenery of the Cotteswolds, and spreads over a considerable width of country. Thence it passes through the counties of Oxford, Northampton, Rutland, and Lincoln, forming the long escarpment known as the "cliff" in Lincolnshire, which runs due north from Grantham by Ancaster, Navenby, Lincoln, and Kirton to the Humber. In South Yorkshire it makes but little show, the beds being thin, but north-west of Malton they thicken, and, sweeping round to the north of Helmsley, form the high ground of the Yorkshire Wolds, which range from west to east and terminate in the cliffs between Whitby and Scarborough. The component members of this series in England are a variable series of oolites, shelly limestones, ironstones, clays, and sands ; none of the strata continuing long of the same thickness, and many of them thinning out altogether within comparatively short distances. The most persistent stratum is that which forms the summit of the series and is known as the Cornbrash. In the south of England, where the most complete succession of purely marine deposits is found, the series has been divided into the following groups and zones : Zones. /Cornbrash Clydoniceras discus. Bath Oolites I -^ ores * Marble Waldheimia digona. (Bathonian) | fulkr? Earth } Cceloceras subcontractum. IClypeus and Trigonia grits Parkinsonia Parkinsoni. (Terebratula Beds Witchellia. Inferior OoliteJ Gryphite and Trigonia grits Lioceras concavum. (Bajocian) j Freestones and Pea-grit Ludwigia Murchisonce. lOpalinum band Lioceras opalinum. 416 STRATIGRAPHICAL GEOLOGY The stratigraphical variations of this series will be best explained by taking first the Bajocian and following it across England from south to north, and then doing the same for the Bathonian. A tabular correlation will be found on p. 427. Inferior Oolite or Bajocian Southern Counties. In Dorset the limestones of this stage are thin, and the lowest zone, that of Lioceras opalinum, is repre- sented partly by sand and sandstone, the beds being about 1 1 feet thick at Chideock, near Bridport, and 5 feet at Burton Bradstock (see Fig. 130). The overlying zones are brown, ironshot limestones, which at Chideock are about 8 feet thick, but at Burton they are barely 3 feet. 12 Farther north, however, near Beaminster, the opalinum zone is about 16 feet and the limestones above about 5 feet, making a total of 2 1 feet ; but the uppermost zonal horizon is not everywhere the same, and the surface of the uppermost bed, whatever it may be, is always eroded, either planed off and coated with oysters or worn into irregular hollows. Between Yeovil and Sherborne the limestones properly belonging to the Inferior Oolite are again very thin, though well exposed in numerous quarries, and the succession given by Mr. Buckman is : Feet. Limestone with Stepheoceras Humphriesianum . . . 1 to 5 ,, Lioceras concavum . . . . . 3 to 4 ,, Ludwigia Murchisonce . . . . 1 to 5 Sandy stone with Lioceras opalinum 1 to 2 About 10 In the Mendip district by Bruton, Doulting, and Frome the Inferior Oolite is entirely absent, having apparently been raised into a broad, low anticline, the upper part of which was eroded and destroyed in the interval between the Bajocian and Bathonian epochs, so that the beds of the latter rest directly and unconform- ably on those of the Upper Lias or Toarcian. North of the Mendip ridge the lower beds of the Inferior Oolite come in again with a pisolitic limestone at the base, succeeded by oolitic freestones and ragstones of much greater thickness than in Dorset or Somerset. The following is a generalised table of the succession found in the Bath and Cotteswold areas : 13 THE JURASSIC SYSTEM 417 Names of Beds. Zones. Thickness. 36t . 20 to 35 Cap of Cephalopoda Bed Lioceras opalinum . 1 to 2 From 100 to 250 The Cephalopoda Bed was described on p. 399 as a condensed representative of several zones and forming a "junction bed " between the Toarcian and Bajocian of the Cotteswold district. Above it there are generally some beds of brown ferruginous limestone, and then conies the " Pea-grit," a coarse pisolitic lime- stone, 4 to 20 feet thick, which extends from Dursley to Nottingham Hill near Winchcombe, a distance of about 30 miles. The freestones are soft pale-coloured oolites in thick beds, and are locally separable into lower and upper freestones by a band of yellowish oolitic marl full of Terebratula fimbria. The beds which form the zone of Lioceras concavum are locally called "grits," but this is a misnomer as they are really rough shelly limestones. The Trigonia Grit is characterised by Trig. costata, T. striata, and T. sculpta, and the Gryphite Grit by Gryphcea sublobata. Other common fossils in these beds are Astarte elegans, A. excavata, Ceromya bajociana, and Pholadomya fidicula. The Notgrove freestone occurs only in the northern part of the Cotteswold area, thickening northward from 4 to about 30 feet. It and the overlying beds are grouped together by Mr. Buckman, who regards them as occupying a lower horizon than the zone of Stepheoceras Humphriesianum, which appears to be absent in the Cotteswolds. Above the beds containing Terebratula Phillipsi there is a marked plane of erosion, the succeeding " Upper Trigonia Grit," or Parkinsoni zone, cutting across all the beds in several directions, but especially eastwards, till it lies directly on the Lias. In the northern part of the Cotteswold area, between the Vale of Winch- combe and Stow-iii-the-Wold, the Inferior Oolite has been flexed into a syncline, while the Vale of Moreton to the east coincides with an anticline, the axes of these flexures running approximately north and south. These flexures were produced and a plane of erosion cut across them before the deposition of the Bath Oolites, as shown in Fig. 141 which is modified from one drawn by Mr. Buckman. Midland Counties. In passing from Gloucestershire into 2 E CD/ I 111 g THE JURASSIC SYSTEM 419 Oxfordshire the Inferior Oolite undergoes a great change. South of Oxford, as above stated, there is no Inferior Oolite, its beds having been destroyed ; and near Fawler the base of the Bathonian con- tains small blocks of the older limestones covered with Plicatulse and pierced by lithodomous molluscs. These conditions continue to and beyond Charlbury, but north-west of that place near Chipping Norton the Bathonian rests on a representative of the opalinum zone, consisting of a sandy limestone about 6 feet thick in which Lioceras opalinum, Rhynch. cynocephala, and other fossils occur ; moreover this has a pebbly basement bed resting on the Upper Lias clay. This sandy limestone may be regarded as the beginning of the Northampton sandstone, which is found again to the eastward at Steeple Ashton and Worton near Deddington, where it consists of calcareous flagstones and sands with ironstone nodules, the beds being 15 or 16 feet thick. East of the Cherwell valley we enter Northampton, where the whole Inferior Oolite passes into an arenaceous facies with very little limestone of any kind, though with a valuable oolitic iron- stone at the base. The lower beds are marine deposits and appear to represent the opalinum and Murchisonce zones, but the higher beds are of estuarine origin. Near Northampton the succession is : Feet. ( White sands, including a bed with plant remains . 12 to 30 Lower | Yellow sandstone and brown sandy limestone or"j Estuarinel ragstone, a variable set of beds, sometimes \- 10 to 30 V. replaced by white sand as at Spratton ) T ( Brown ferruginous sandstone and ironstone, with ironstone J ^ oceras opalinum, Astarte elegans, Rhynch. [ cynocephala, etc 30 to 40 Average 60 feet These beds continue to be the sole representatives of the Inferior Oolite till we reach the valley of the Welland, where the thin end of a lenticular mass of limestone appears above the Lower Estuarine Beds and rapidly thickens northward, becoming the important formation known as the Lincolnshire limestone, which in South Lines is not less than 150 feet thick. At its base in South Lincolnshire are some beds of fissile sandy limestone, which form a passage from the Lower Estuarine Beds ; these contain ferns (Laccopteris polypodioides), with Lucina Wrightii, Gervillia acuta, Trigonia compta, Tr. impressa, Astarte elegans, and Malaptera Bentleyi, but no Ammonites. At Colly- weston (near Stamford) these beds are 12 feet thick, but they thin northward and are hardly distinguishable at Ancaster. 420 STRATIGRAPHICAL GEOLOGY The Lincolnshire limestone frequently contains beds in which corals are very abundant, and which have evidently been accu- mulated in the neighbourhood of coral reefs. The upper layers, too, invariably exhibit much current bedding, indicating shallow water conditions. Ammonites are rare, but Echinoderms are not uncommon. Gastropods (Natica, Nerinwa, Patella, Pleuro- tomaria, Trochus, Trochotoma) are common. Rhynchonellce and Lamellibranchs of many species are abundant. The fauna has eminently a shallow-water and coral-reef facies. This limestone appears to represent the higher part of the Inferior Oolite of the Cheltenham district from the Lower Trigonia Grit upward, for the Ammonites quoted from it include Cceloceras Blagdeni, C. sub- radiatum, Stepheoceras cf. Humphriesianum, and Oppelia Truelli, all characteristic of the higher zones. Mr. Hudleston also considered its Gastropod fauna to indicate the same part of the stage. In North Lincolnshire the limestone diminishes to about 60 feet, and in South Yorkshire the lower part is represented by shaly limestones and the upper part by oolitic limestones, each set being about 30 feet thick. Yorkshire. When they again emerge in North Yorkshire the series has undergone much transformation, and the mass of the Inferior Oolite consists of estuarine sandstones and shales. 14 Marine beds, however, occur at the top and bottom, and also as a thin zone in the midst of the Estuarine Series, the succession being as follows, and the thicknesses those in the cliff sections : Feet. 5. Scarborough limestone 20 to 90 4. Middle Estuarine Beds . . . . . 50 to 100 3. Millepore Beds (marine) 20 to 40 2. Lower Estuarine Beds 200 to 280 1. The " Dogger " sands and sandstone . 30 to 36 Average about 400 1. Above the Liassic shales are yellow and grey micaceous sands containing Grammoceras aalensis, Lingula Beani, Rhynchonella cynocephala. They are surmounted by a sandy oolitic ironstone, which contains many fossils Ludwigia Murchisonce, Terebratula trilineata, and Nerincea cingenda ; these sands and sandstone may be regarded as the equivalent of the Northampton sands. 2. The Lower Estuarine Series is exposed in the cliffs from Eobin Hood's Bay to Huntcliff (see Fig. 140). The shales and oolitic ironstones contain plant remains in some abundance. In the midst of this series is a band of flaggy sandstone and ironstone with marine fossils known as the Eller Beck Bed. It yields Gervillia acuta, Astarte minima, Pholadomya Heraulti, and other fossils. M 3 5 s- ~>> if C H XO BC CO O o3 00 1 ' CH p^ OQ S ra O ^ .-- '- * ^ 3 <- 1 i -* , 6 , < - ^ > | 1 (D o3 ^c4 c g S 5 pq 03 Pi g e8 p3 S s ,2 s ^ CO O) be i 1 w S r O 1 ^ g S 66 8 9 ft ^& Q O p *J efl | __ - r ( 0) ;- - 1 H |s I o v^ v ' O3 T3 H d =3 Rutland and Lincolnshire. Cor nb rash O o> 1 | 6 Great Oolite Limestone B ^ w '"d IP 4 1 Lincolnshire Limestone Lower Estuari Beds Northampton S. o 0) c6 g ^ la -S PQ f3 O o 1 B ,5 M S 3 2 'S 1 .3> 03 C -^ l.^S 1 CO ,5 "S CD 02 a P x o O O 1 2 o ^ 1 QQ -J , SS. ^ 5 I o 2" H 1 o O Coriibrash Forest Marble o 1 1 f Stonesfield - a l i Clypeus and Upper Trigonia Grits r Limestones of the Witchellia and concavum zones Freestones and Pea-grit S3 a d 1 O J oq CO 2 -*^ Q} co 1 -*- 3 a S 'w >_i o3 g 1 1 J w i 1 ^H OQ g J <> If Is a d ID O 3 _3 H^ O cS i ' V. 428 STRATIGRAPHICAL GEOLOGY On the Yorkshire coast the whole group is represented by shales and sandstones of estuarine origin, capped by a thin band of Cornbrash. In Gristhorpe Bay this Upper Estuarine Series is 120 feet thick, consisting of marine sandstones in the lower part and shales with thin sandstones above. Farther north they are over 200 feet thick. A few ferns are the only fossils. The Cornbrash is very thin, only 5 or 6 feet, but has the usual characters, being a grey rubbly ironshot limestone full of the ordinary fossils, among which are Macrocephalites Herveyi, Pseudo- monotis echinata, Lima rigidula, and Waldheimia lagenalis. D. THE MIDDLE JURASSIC IN SCOTLAND 1. West Coast " Eocks of this age," says Professor Judd, 16 " must have origin- ally had a very wide distribution, for traces of them are found at various points from the Shiant Isles in the north to Ardnamurchaii in the south. The best exposures of their strata, however, are those which occur in the islands of Skye and Raasay" (see Fig. 132). Here the Inferior Oolite consists of the following beds : Feet. Beds of white sandstone with thin shaly hands containing remains of ferns and cycads i ' J . . . . . . . 60 Beds of sandstone and shale, the former passing into shelly lime- stones, Steph. Humphriesianum, Belemnites giganteus . . 160 Sandy micaceous shales alternating with calciferous sandstones, Ludwigia Murchisonce, Harpoceras corrugatits, Belemnites giganteus, Lucina Wrightii, etc. . . . . . .120 340 These beds are succeeded by a great thickness of strata belonging to the Bathonian Group. At the base are marine limestones almost wholly made up of comminuted shells, but containing perfect Ostrea Sowerbyi, Waldheimia lagenalis, etc. They probably represent the zone of Parkinsonia Parkinsoni and are 45 feet thick. Above them are estuarine sandstones and shales fully 500 feet thick but not well exposed. Farther south, however, in the islands of Eigg and Muck this estuarine group is still thicker and is accessible, the following succession being seen : A. The lowest beds are fine conglomerates and shelly limestones full of freshwater shells, but containing also remains of Plesiosaurus, turtles, and other reptiles. Above these are black shales and limestones abounding in Cyprids, fish remains, and shells of Cyrena, THE JURASSIC SYSTEM 429 Cyclas, and Viviparus. The thickness of the whole group is 200 feet B. Grey and white sandstones, sometimes passing into calcareous grits, with large concretionary siliceous masses ; current-bedding and ripple - marked surfaces are frequent, but excepting plant remains fossils are rare. Thickness probably more than 500 feet. C. The highest beds are black shales with thin bands of argillaceous limestone and bands of fibrous carbonate of lime ; they are sometimes crowded with Cyrena and Cyclas, and in other places with Viviparus, Melania, and other Gastropoda. Beds entirely composed of oysters (0. hebridica] also occur, resembling those in the Parbeck Series (see postea). Thickness 150 feet. 2. East Coast Beds of Middle Jurassic age occupy a small space on the coast of Sutherland south of Brora, and form two small patches at Burgh Head and Stotfield, on the coast of Elgin. They consist entirely of estuarine strata, which are lithologically comparable to those of the Great Oolite Group on the western coast. The descending succession near Brora is as follows : Feet. " Roof bed," calcareous sandstone (Kellaway) . . . \ o i Coal composed of crushed Equisetites . . . . . J Black shales with thin coal-seams in the upper part, plants, and crushed shells of Unio, Ostrea, Cyrena, etc., abound ; also teeth and scales of fish, Lepidotus, Pholidophorus, and Hybodus 26 Black, brown, and grey clays with layers of argillaceous lime- stone, Cyrena and Ostrea ....... 96 "White sandstones with some beds of clay ; in this a few obscure marine shells occur . . . . . .110 The base is not seen, and whether any beds comparable to the Inferior Oolite of the west coast exist below is, therefore, unknown. At Stotfield there are soft greenish -white sandstones containing Ostrea Sowerbyi, Tancredia axiniformis, and Astarte hombordalisi, and are probably near the base of the Bathonian Series. E. CONTINENTAL EQUIVALENTS 1. France The Middle Jurassic strata of the Parisian basin are a continua- tion of the British Series, and their outcrop is almost continuous round the basin, though concealed here and there beneath the transgressive Cretaceous deposits. The districts where the succes- sion is most clearly exposed are that of Bayeux and Caen in e3 (M PQ 430 STRATIGRAPHICAL GEOLOGY Normandy, the southern border of the Ardennes, and the broad outcrop on the borders of France and Germany, extending south- ward to the Jura Mountains. In Normandy the succession is as follows : Marly limestone and blue clay, 12 feet ( = Cornbrash). Bryozoan limestone with Waldheimia (Zeilleria) digona, Terebra- tula flabellum, and Eudesia cardium, 30 feet ( F. Marble). Compact oolitic limestones with Lucina bellona, 30 to 60 feet ( = Great Oolite). Caen limestone with Rhynchonella (Acanthothyris} spinosa, pass- ing northward into Bessin limestone with Oppelia fusca, 100 feet ( = Fuller's Earth). White Oolite with Parkinsonia Parkinsoni, 60 feet. Ferruginous Oolites with Cosmoceras subfurcatum and Strenoceras garantianum, 2 feet. Brown Oolite with Cceloceras Blagdeni, 1 foot. Glauconitic limestone with Sphceroceras, 1 foot. White limestone with flints, Ludwigia Murchisonce, and Lima heteromorpha, 9 feet. Limestone with Lioceras opalinum and a layer of phosphatised Ammonites derived from the Toarcian at the base, 2 feet. On the other side of France, in the French Ardennes and thence eastward into Lorraine, the series presents a different facies with a much thicker development of limestones. The succession there is : Tlaty limestone with large Oysters (Ostrea flabelloides}, 15 feet. Marly limestone with Waldheimia digona, Eudesia cardium, and Nucleolites ( Echinobrissus} clunicularis, etc., 100 feet. White chalky limestones and oolites, 200 feet. Oolite with Clypeus Ploti, 30 feet. Yellow sandy limestones with Parkinsonia Parkinsoni and Ostrea acuminata, 60 feet. - . fYellow limestone with Pteria ornata. jji "S Oolitic limestones and marls with Cceloceras Blagdeni. ' ^ -[ Marly limestones with corals, Belemnites giganteus and Sonninia 'o?o Sowerbyi. pq co ^Thin bed with Lioceras opalinum. In the French Jura the succession is somewhat similar, but differs somewhat in zonal detail, being as follows : Tlaty limestones with large oysters, Waldheimia (Zeilleria} digona and Nucleolites clunicularis, 26 to 120 feet. Compact white limestones with Rhynchonella decorata, 200 feet ( = Great Oolite). Marly limestone with Pholadomyas \ Fuller's Earth Marls of Vesoul with Ostrea acuminata / (130 feet). ^ White Oolite with Cosmoceras subfurcatum, 20 feet. _~ . ( Marls with Belemnites giganteus. 03 "S Crinoidal limestones with Emeleia polyschides, Witchellia, and ' & -j Lioceras concavum. 'o? I Ferruginous limestone with Ludwigia Murchisonce and Lioceras ^ ^ \ opalinum. THE JURASSIC SYSTEM 431 2. Germany In the south of Germany the outcrop of this series follows that of the Lias from the Black Forest through Swabia and Franconia. In the south-west of this tract, near Tuttlingen and Nusplingen, the facies is calcareous and similar to that of the Northern Jura, but eastward the limestones are largely replaced by clays, and the Bathonian portion becomes much thinner as if approaching a shore- line. In East Swabia the succession is as follows : Clays and marls with JRhynchonella varians and Parkinsonia fj "S ferruginea, 16 feet. o > 4 Limestone with Oppeliafusca passing into blue clay, with Parkin- +jj o I sonia Parkinsoni, 10 to 100 feet. pa Idays and limestones with Cosmoceras subfurcatum, 10 feet. 'Zone of Witchellia Eomani, limestones with Stepheoceras subco- ^ t ronatum and clays with Belemnites giganteus. -g Zone of Sphceroceras Sauzei, blue limestones. 2 - Zone of Lioceras concavum, clays and marls. 'S 5 .^ Zone of Ludwigia Murchisonce, sands, clays and ironstones, 80 feet. Zone of Lioceras opalinum, dark clays, 200 feet. Apart from the lithological composition of the series the most noteworthy fact is the great expansion of the opalinum and Murchi- sonce zones, as compared with the lesser thickness of the overlying limestones. In Northern Germany there is a similar succession of shales with very little limestone, and the whole series is not more than 100 feet thick. It probably underlies the greater part of the Hanoverian and Prussian plains, and is known to occur in Pomerania ; eastward in Silesia the lowest Jurassic Beds belong to the Parkinsoni zone ; while in Poland near Cracow there are sands and clays of Bajocian age overlain by 100 feet of dark-grey clays with Parkinsonia wurtemburgica (Bathonian) and succeeded by a ferru- ginous oolite with Rhynchonella varians and Belemnites bessinus ; but farther east these beds thin out, leaving the Callovian to form the base of the Jurassic System. 3. Alpine and Mediterranean Facies In the western sub- Alpine region, Savoy and Lower Provence, the Middle Jurassic has a thick and special development, and is over 1000 feet thick. The Bajocian consists of regular alternating beds of marly limestone and shaly marl, characterised by the abundance of Ammonites of the genera Lytoceras and Phylloceras. The succession of zones is similar to that of the Jura and Swabia, and the same species can be taken as indices, but each zone is also characterised by 1 one or more species of Phylloceras, and this is a 432 STRATIGKAPHICAL GEOLOGY special feature of the Mediterranean and Alpine facies. Near Digne the Bajocian is 660 feet thick. In the same region the Bathonian consists mainly of black shales which are divisible into two zones, viz. : 2. Black marly shales with Posidonia alpina, 300 to 400 feet. 1. Black shales and marly limestone with Oppelia fusca, 170 feet. Eastward, however, both stages pass into the condition of marly limestones, and in a more or less metamorphosed form they show themselves at intervals through the Swiss and Italian Alps, and recur in the Apennines. In the Eastern Alps (Klaus Alp) Bathonian shales with Posidonia alpina occur, and again near Vienna there is Bajocian with Phylloceras and Bathonian with Posidonia. These beds extend to the Carpathians, and occur also in Servia, where (near Milanovitz) the Bajocian is absent, and beds with Oppelia fusca and Zigzagiceras arbustigerum rest directly on crystalline schists. III. UPPER JURASSIC SERIES A. SUBDIVISION AND NOMENCLATURE In England, where fully developed, this series is divisible into five groups or stages two great argillaceous formations (the Oxford and Kimeridge clays), each overlain by a group of sand- stones and limestones (the Corallian and the Portland Beds) ; while at the summit are the Purbeck Beds, a set of estuarine and lacustrine strata which may be regarded as formed during a continuance of Portlandian conditions in the purely marine areas. On the continent the thick clays are in many places repre- sented by limestones and marls, so that the British names are not literally applicable ; but in. Germany and Northern France the general succession is similar, and many of the English names have been employed in a latinised form. In Southern France and the south of Europe generally there is a very different facies, wholly calcareous and marine with a different assemblage of fossils. The following table shows the nomenclature used in Northern Europe, but the French stages do not exactly correspond witli those of England and Germany, as will be shown in the sequel. England. North France. North Germany. 5. Purbeck Beds ) r> n j- Munder Mergel 4. Portland Beds/ and Plattenkalk. 3. Kimeridge clay Kimeridgian Kinieridgian. 2. Corallian Beds Corallian Corallian. 1. Oxford clay and /Oxfordian\ Oxfordian Kellaways Beds \Callovian / THE JURASSIC SYSTEM 433 B. LIFE OF THE PERIOD The Upper Jurassic fauna is merely a continuation of that which preceded it, the genera being the same, though the species are for the most part different, only about sixty species passing from the Cornbrash into Oxford Clay. The most impressive feature of the later fauna is the abundance of reptilian remains. The Jurassic period has been termed the Fig. 142. SKELETON OF SCAPHOGNATHUS CRASSIBOSTRIS (J) FROM SOLENHOFEN, GERMANY. " age of reptiles," and these creatures seemed to have reached the climax of their dominion towards the close of the period. Their numbers and the variety of their modifications are quite extra- ordinary. Not only did they swarm in the seas and rivers, but they peopled the land with gigantic herbivores and filled the air with winged forms. They thus adapted themselves for all kinds of food and all conditions of life with an elasticity of organisation which is truly remarkable, and to which no parallel can be found in the subsequent history of the world, unless it is among the modern Mammalia, though no member of that class possesses powers of flight comparable to those of the Pterodactyles. 2F 434 STRATIGRAPHICAL GEOLOGY Of Upper Jurassic reptiles, Cimoliosaurus, Peloneustes, Pliosaurus, Ichthyosaurus, with the crocodiles Teleosaurus, Steneosaurus, Dako- saurus, and Goniopholis were the chief aquatic genera. Ceteosaurus, Gigantosaurus, Iguanodon, Omosaurus, and Cryptosaurus were ter- restrial Dinosaurs, and in America other remarkable genera, Bronto- saurus and Atlantosaurus, both of enormous size, have been found. Fig. 143. GROUP OF OXFORDIAN FOSSILS. a. Gryphaea dilatata. d. Cosmoceras Jason. 5. Anatina undulata. e. Cardioceras excavatura. c. Dicroloma compositum. /. Belernnites hastatus. g. Leptolepis macrophthalmus. Of Pterosauria species of Pterodactylus, Scaphognathus (Fig. 142), and Rhamphorhynchus were prevalent. That small Mammalia existed throughout the Jurassic period on terrestrial surfaces there can be little doubt, but their remains are not found in purely marine deposits. The Purbeck Beds, however, have yielded no fewer than twelve genera and twenty species of marsupials, the most important genera being Plagiaulax, Belodon, Spalacotherium, Triconodon, Stylodon, and Amblotherium. Birds also of a primitive kind had now come into existence, THE JURASSIC SYSTEM 435 but the only known genus is Archwopteryx, whose remains occur in limestones of Portland age at Solenhofen in Bavaria. This bird had strong reptilian affinities in the structure of its head and tail, . 144. GROUP OF CORALLIAN FOSSILS. c. Thecosmilia annularis. b. Acrosalenia decora ta. c. Gonioniya litterata. d. Trigonia clavellata.'. e. Cerithium muricatum. /. Aspidoceras perarmatum. (/. Belemnites abbreviatus. h. Cidaris florigemma. i. Pholadomya aequalis. and in possessing teeth, but it had feathered wings, a feathered tail, and feet adapted for grasping boughs of trees. Fish were evidently abundant in the seas, their teeth and scales being often found, and sometimes the entire skeleton ; the chief genera are Leptolepis (Fig. 143), Aspidorhynchus (Fig. 147), Pholi- dophorus, Gyrodus, Lepidotus, and Hybodus. 436 STRATIGRAPHICAL GEOLOGY Of Invertebrates very few new genera make their appearance except among the Echinodermata and the Ammonoidea. The Echinoderm genera Peltastes, Acrocidaris, Glypticus, Magnosia, and Cyphosoma make their appearance. Among the Ammonoids many of Middle Jurassic forms die out, while the following appear : AspidoceraSj Cardioceras, Craspedites, Holcostephanus, Peltoceras, Fig. 145. GROUP OF KIMEBIDG1AN FOSSILS. a. Rhynchonella inconstans. bluish sandy clays . n ^Nothe Grits, calcareous grits, and marls . 30 to 40 20 to 35 The Nothe Grits and Clays contain a mixture of Oxfordian and Corallian species Gryphcea dilatata, Cardioceras cordatum, Ostrea gregaria, and Trigonia perlata. The Osmington and Trigonia Beds represent the Coral Kag, but corals are rare ; the rags contain Trigonia clavellata, Gervillia aviculoides, Nerincea Goodhalli, Am. plicatilis, and many other fossils. The Sandsfoot Clays and Grits yield Astarte supracorallina, with Thracia depressa, Ostrea deltoidea, Bhynchonella inconstans, and Belemnites nitidus, which range into the Kimeridge Clay. Near Abbotsbury (see Fig. 148) the Sandsfoot Grits pass into ferruginous sandstone and oolitic ironstone, the latter quarried as ore. Similar ironstone occurs at Westbury in Wiltshire, in which county the general succession is : Feet. Upper I San* and clays } U PP er Calcal>eo s Grit - - to 20 Corallian 1 Rubblv oolite and coral beds "I n , T> I /~x i "* r ^orill Xttl^ ^(J tO OU l^Calne freestone \ Lower Corallian, sands with calcareous burrstones . . 50 The freestone thins out rapidly north of Calne, and is not present at Faringdon, where the Upper Corallian is not more THE JURASSIC SYSTEM 443 than 20 feet thick. In Oxfordshire this stage consists of Lower Calcareous Grit (50 to 60 feet) and Coral Bag (30 to 40 feet), the former consisting of sands with calcareous burrstones, the latter of shelly and coralliferous limestone. But near Stainton St. John these beds change rapidly and appear to be replaced by clays. In Buckingham, Bedford, and Huntingdon the whole of the Corallian stage is represented by the Ampthill Clay, named from the small town of Ampthill near Bedford. This clay has not yet been separated on the maps of the Geological Survey from the clays above and below, but is shown on a small map illustrating Mr. Rastall's account of Cambridge and Bedford (Geology in the Field, vol. i. p. 141), reproduced in Building of the British Isles (1911, p. 299). Its fauna includes a mixture of species be- longing to the Oxford and Kimeridge Clays with a few that are more especially Corallian. The chief Ammonites are achilles, cordatum, plicatilis, and vertebralis ; with these are found Alaria bispinosa, Exogyra nana, Ostrea deltoidea, 0. discoidea (allied to deltoidea but more circular), Gryphwa dilatata, Cidaris florigemma, and G. Smithi. As seen near Ampthill this division consists of grey marly clay with a bed of nodular limestone 4j feet thick at the base, and a band of septaria at the top. Near Cambridge and St. Ives the Lower Calcareous Grit is represented by a bed of dark-grey ferruginous oolitic limestone, known as the Elsworth Rock, from 6 to 8 feet thick, and sometimes in two beds with clay between. This is usually overlain by clays of the Ampthill type, 19 but at Up ware, between Cambridge and Ely, there is an isolated mass of undoubted Coral Rag and Coralline oolite, which seems to be part of a true coral reef. One pit exposes coral limestone, with Thamnastrea arachnoides and Isastrea explanata, Cidaris florigemma, Opis,Plicatula,Lithodomus,a,nd other inhabitants of the reef ; while another pit is opened in soft yellowish oolite, without corals, but containing Nucleolites scutatus and Holectypus di'pressus. This coral reef is not of large extent, for no such rock was found in a boring 2 7 y miles north-west of Upware. The Ampthill Clay was recognised by Mr. Roberts in Lincoln- shire. It forms a narrow band between the Oxford and Kimeridge Clays from Bardney on the Witham to Briggs and Wrawby, and it yields the same fossils as in Bedford and Cambridge. Its thickness is not less than 20 feet, and may be more. In Yorkshire the Corallian attains an unusual thickness and occupies a large area of ground round the Vale of Pickering. It has been specially studied by Messrs. Blake and Hudleston and by Mr. Fox-Strangways, from whose memoir the following tabular view of the succession near Pickering has been taken : 444 STRATIGRAPHICAL GEOLOGY Feet. Upper Calcareous Grit, shales, and grit. Cardioceras alternans, Belemnites nitidus, and Goniomya literata 30 to 40 Upper limestones, coral rags, and oolites with Peri- sphinctes plicatilis, BourgiLetia striata, Trigonia per- lata, Cidaris Jlorigemma, Thamnastrca concinna, and other corals . . . . . . . . 40 to 50 Middle Calcareous Grit with Perisphinctes plicatilis . 40 to 80 Lower limestones with Cardioceras cordatum and Nucleo- lites scutatus 30 to 60 Graystones, coarse, gritty, and cherty limestones with Aspidoceras yoliathus, Gervillia aviculoides, and Rhynchonella Thurmanni . . . . . . 25 to 40 Lower Calcareous Grit, gritty limestones, and soft cal- careous sandstone with Aspidoceras pcrctrmatum, Rhynclionella Thurmanni, and Collyrites bicordatv.s . 50 to 130 It ^ ^ From 215 to 400 3. The Kimeridge Clay. This formation takes its name from Kimeridge Bay near St. Albaii's Head in Dorset, where it has a thickness of no less than 1000 feet and is divisible into Feet The Upper Kimeridge or zone of Perisphinctes biplex . . . 600 The Lower Kimeridge or zone of Cardioceras alternans . . . 400 In Dorset and Wilts the lower part consists of dark clays and shales with layers of septaria and cement-stones ; these beds yield Cardioceras alternans, Perisphinctes mutabilis, Ostrea deltoidea, and Rhynchonella inconstans. The upper beds are black bituminous shales and grey papery shales, often full of broken and compressed shells ; they yield Perisphinctes biplex, Protocardia striatula, Thracia depressa, Exogyra virgula, and Lucina minuscula. Reptilian bones occur throughout the Kimeridge Clay, and among them are remains of two species of Pterodactyles. There are few good inland sections of the Kimeridge Clay, and its thickness is greatly diminished in passing through Dorset and Wiltshire, for at Swindon it is estimated to be only about 300 feet ; and in Berkshire and Oxfordshire it is not much more than 100 feet, but is still divisible into the two zones. Near Aylesbury the Kimeridge Clay is not more than 100 feet thick, and merges upward into a sandy clay containing Portlandiaii fossils (the Hartwell Clay). Near Leighton Buzzard, and thence through Bedfordshire, most of the Kimeridge Clay is concealed by the Cretaceous sands, and does not emerge again till we reach the neighbourhood of Papworth in Cambridgeshire, where it appears above the Arnpthill Clay, and runs in a narrow belt by Knapwell, Boxworth, and Cottenham to the Fens. Near Ely this clay has been largely dug, and both zones (upper THE JURASSIC SYSTEM 445 and lower) have been recognised. The lower (from 80 to 90 feet) contains Cardioceras alternans, Astarte supracorallina, Ostrea deltoidea, etc. ; the upper, of which only 16 feet is seen, contains Exogyra virgula and Orbiculoidea latissima, but still higher beds may be concealed beneath the Cretaceous rocks. In Lincolnshire both lower and upper divisions are well de- veloped, and the whole stage is probably about 600 feet thick. The lower beds are exposed near Horncastle, Wragby, Market Rasen, and Wrawby near Caistor. At Rasen they yield many finely preserved fossils which include Cardioceras alternans, Perisphinctes mutabilis, Triyonia Juddiana, Inoceramus rasenensis, Thracia depressa, and Nucula Menkei. The upper beds are mainly shales, seen near Spilsby, Fulletby, and West Ashby, and yielding Per. biplex, Protocardia striatula, Lucina minuscula, Lingula ovalis, and Orbiculoidea latissima. In Yorkshire the Kimeridge Clay is not well exposed, for though it doubtless underlies a large part of the Vale of Pickering, it is concealed by Glacial drifts, and only the highest beds are exposed on the shore at Speeton Gap. These consist of brown and black shales containing Perisphinctes biplex (?), Ostrea gibbosa, Discina latissima, and Lingula ovalis. 4. The Portland Beds or Portlandian. Like the Kimeridge Clay, these beds are thickest in Dorset and thin north- wards. They are well exposed between Durlstone Head and St. Alban's Head (see Fig. 149), where they are divisible as follows : Feet. Upper Portlandian f Freestone Beds . . . 40 to 50 zone of Hole, giganteus \ Cherty Beds . . . . 60 to 75 Lower Portlandian f Sandy marls, sands, and calcareous zone of Perisph. gigas \ sandstone 130 to 160 230 to 285 The hard sandy marls of the lower group contain Perisphinctes biplex, Exogyra bruntutana, Trigonia incurva, Tr. Pellati, and Rhyn- chonella portlandica. The "cherty beds " are brown earthy limestone with irregular layers and nodules of flinty chert ; Pecten lamellosus and Protocardia dissimilis occur in these beds. The freestones are oolitic limestones, which are largely quarried for building stone ; they contain Trigonia gibbosa, Perna mytiloides, Oerithium port- landicum, and Holcostephanus giganteus. At Portland the succession is similar and the total thickness about 220 feet, but at Upwey, north of Weymouth, a different facies presents itself, the whole Upper Portlandian having passed into a white chalky limestone with black and grey flints. This 446 STRATIGRAPHICAL GEOLOGY ^ .2 . \ * S s= & ; 0.2 S U OQ Sis? ?5 r So / | H { H \ t s x 1 I S . ^p . Is 3 *f*, rf'odl- .5P rt ^3 ^3^ I'M', '( i THE JUEASSIC SYSTEM 447 is not more than 30 feet thick, while the sandy beds are about 60 feet, so that the whole is less than 100 feet. In the Vale of Wardour both chalky and oolitic limestones occur, and each division of the stage is from 50 to 60 feet thick. The next good exposure of the Portland Beds is at Swindon, where the limestones are partly replaced by sands. The beds are exposed in railway cuttings and quarries, and the combined succession is given by Mr. H. B. Woodward as follows : Feet. I Marly and oolitic limestones . . . . 6 to 12 Whitish sands with lenticular layers of cal- careous sandstone ..... 20 to 25 Marly and oolitic limestones with Trigonia gibbosa and Cerithium portlandicum . . 3 to 7^ ( Blue clay weathering brown . . . . 14 to 20 p H \ Marly sandstone with Exogyra bruntutana . 6 to 8 1 [ Sands with doggers of calcareous sandstone . 30 to 40 About 100 In the Midland counties the only good exposures of the Portlandian are in Buckinghamshire near Thame, Brill, and Aylesbury, where they exhibit a special facies different from that of Dorset or Wiltshire. The succession is as follows : Feet. Grey marls and thin limestones, Holcostephanus giganteus, Ostrea expansa, and Trigonia gibbosa . . . . . . . 8 to 12 Upper Portland, 26 feet Marly and slielly limestones, Trigonia gibbosa . 5 to Greenish-yellow sand . . . . . 5 to 6 Marly and rubbly limestone with fossils . 10 to 15 Yellowish-green sand with a bed of small pebbles of quartz and lydianite at the base . . . 8 to 10 (Fine sandy clay with Astarte hartwellensis, Car- \ dium morinicum, Perna Bouchardi, Trigonia { Pellati 20 to 30 Near Thame there are yellow and green sands below the pebble- bed, but these thin out or pass into clay northward ; the Hartwell Clay being apparently the equivalent of the Portland Sands. The most northerly outlier of these beds is at Stewkley Warren, 4 miles west of Leighton Buzzard, where about 15 feet of them, capped by a few feet of Purbeck Beds, was at one time exposed. Beyond this they are not seen either in Bedford, Cambridge, Norfolk, or Lincoln, but derived and phosphatised Portlandian fossils are of frequent occurrence in the Lower Cretaceous sands, and consequently it is most probable that they originally extended through these counties, but were destroyed at the beginning of the Cretaceous period. In Yorkshire, at Speeton, the Kimeridge Clay is surmounted 448 STRATIGEAPHICAL GEOLOGY by a bed of black phosphatic nodules, tlie "coprolite bed," and the clays above this contain Ammonites which resemble Portlandian species, so that the beds have been classed by some as Portlandian. As, however, the Ammonites are not identical with Portlandian forms, and as the Neocomian Belemnite (B. lateralis) occurs throughout, with Hoplites regalis in the higher part, the beds are probably of Lower Cretaceous age (see p. 484). 5. The Purbeck Beds. The typical Purbeck Beds are only found where the Portland Beds are completely developed. They are a variable set of beds exhibiting alternations of terrestrial, freshwater, brackish -water, and estuarine conditions, the changes from brackish to freshwater deposits being generally gradual, while the reverse changes are abrupt, indicating sudden inroads of the sea. In Durlston Bay, near Swanage, these beds are 400 feet thick, and have been divided into Lower, Middle, and Upper Groups, characterised respectively by Cypridea purbeckensis, C. granulosa, and G. punctata (see Fig. 149). The following is a summary of the cliff section : Upper, 80 feet Middle, 155 feet Lower, 170 feet Grey and purple marls with Viviparus cariniferus Shales with beds of Paludina limestone (Purbeck marble), Cypridea abundant . . . . Unio Beds and shelly limestones with Viviparus . Alum-shales and thin limestones with layers of gyp- sum, Corbula, Cyrena, Cyclas, etc. .... Hard shelly limestones with partings of shale, fossils of marine and estuarine species .... Cinder bed, a mass of Ostrea distorta . . Thin limestones, marls, and shales with remains of insects, fish, and freshwater shells ; at the base is a black shale with mammalian remains Marly freshwater beds, Physa and Planorbis Marls and marly limestones with Cardium, Corlula, Cyrena, and Cypridce ...... Marly limestones and shales, with Cypridce . Beds of broken slaty limestone ..... Brown bituminous limestones with a layer of dark earth (dirt bed) resting on Portland stone Feet. 14 45 15 64 50 43 7 92 36 15 19 409 The same series can be seen in Worbarrow Bay, Mupe Bay, and Lulworth Cove, but it becomes thinner to the westward, 250 feet at Mupe Bay, only 176 feet at Lulworth, and at Eidgeway Hill 190 feet. 20 The lowermost beds are well exposed in the quarries at Portland, where the most remarkable stratum is the upper "dirt bed," which is a dark-brown loamy earth an actual soil or terres- trial surface, in which cycads and coniferous trees are still rooted. THE JURASSIC SYSTEM 449 The Purbeck Beds are cut off and faulted against the Chalk and Greensand by the great Ridgeway fault, and they are not seen again till they crop out in the Vale of Wardour (Wilts). The succession in this area, as described by Mr. W. E. Andrews and myself in 1894, 21 with such subsequent corrections by Mr. H. B. Woodward 22 as we accept, is as follows : . Feet. [Clays, marls, and shelly limestones with Cypridea, Upper -| punctata, Cyrena media, and Unio . . . .10 ( Yellow sand and grey clay ...... 12 ( Sandy and marly limestones, shelly marls, and cinder Middle 4 bed with Ostrea distorta and Trigonia gibbosa . . 12 (Limestones and shelly marls, Cypridea granulosa . . 11 ? Marly limestones (locally known as Lias) with layers -r J of shaly marl, Cypris purbeckensis on Pale yellow oolitic limestones and marls .Grev laminated marls and limestones Total about 110 i I i T Fig. 150. DIAGRAMMATIC VIEW OF THE swiNDON QUARRIES (after Professor J. F. Blake). g, h, i. Purbeck Beds. c, d, c, f. Portland limestones. b. Sand with calcareous doggers a. Portland limestone. The Lower Purbeck Beds are seen again at Swindon, where they rest partly on the Upper Portland and partly on the Portland sands (see Fig. 150), and their basement bed contains rolled lumps of stone derived from the Portland Beds. In the Midlands the Purbeck Beds only emerge from beneath the Cretaceous strata in Oxford and Bucks. Around Thanie, Aylesbury, Brill, and Whitchurch the Portland Beds are succeeded by a variable group of beds with freshwater fossils, which are generally considered to be of Purbeck age. They consist of a variable series of thin-bedded limestones, marls, shales, clays, and calcareous sands, but do not exceed 30 feet in thickness, and probably represent the Lower Purbeck only, the higher beds having been removed by Cretaceous erosion. They have yielded remains of Plants, Insects, and Cyprides, with Viviparus, Cyrena, Modiola, and the fish Lepidotus minor, Pleuropholis serrata, Aspidorhynchus, and Mesodon. 2G 450 STRATIGRAPHICAL GEOLOGY The only other part of England where Purbeck Beds have been found is in Sussex north-west of Battle, where narrow strips of them are brought up by anticlinal flexures from below the Wealdeii Beds. They consist of shales with beds of limestone which were formerly quarried, and the succession proved by shafts and given by Mr. Topley 23 is : Feet. Shales with beds of limestone and some of calcareous sandstones, Cyp. punctata and C. granulosa . . . . . .80 Black, grey, and greenish shales with ironstone nodules . . 130 Bluish-grey limestones and shales, with Ostrea, Cardium, Cyrena, and other fossils ......... 75 Shales with many beds of gypsum 130 415 Purbeck Beds have also been traversed by many of the borings which have recently been made in Kent, and the records show that the group thins both to the north and east. Thus at Pluckley, about 16 miles north-east of the outcrops in Sussex, they are only 100 feet thick, and at Brabourne, about 10 miles farther east, they have diminished to about 60 feet. They were not found in any of the borings to the north or east of that place, though this may be due to the planing off of beds below the Vectian sands. D. THE UPPER JURASSIC SERIES IN SCOTLAND "West Coast. The only member of the Upper Jurassic series which is found on the west coast of Scotland is the Oxford Clay. This overlies the great estuarine series in the islands of Skye and Eigg, and consists of blue clays containing Cardioceras cordatum, Card, excavatum, Quenstedticeras Lamberti, and others. Above it are Cretaceous rocks. East Coast. In Sutherland a much more complete series is found ; the beds are seen at intervals along the coast for a distance of 16 miles, and form a narrow strip of low ground which is bounded inland by a powerful fault, and has an extreme width near Brora of about 2 miles. In this small space is found a succession of beds representing the whole of the Oxford Clay, Coral Rag, and part of the Kimeridge Clay, as below : Feet. Light-coloured sandstone ; no fossils .... 100 Shelly limestones, black shales, and grits with Belem- nites abbreviatus, Cardioceras alternans, Perisphinctes biplex, and many others ; ferns, cycads, and conifers 500 Kimeridge Clay White sandstones and carbonaceous shales ; estuarine beds with few fossils 200 Grits and sandstones ; Perisphinctes biplex, Per. muta- bilis, ReinecTcia eudoxus, and Belemnites obeliscus . ? 200 Corallian Oxford THE JURASSIC SYSTEM 451 Feet. Grey limestones and sandy clays in two bands, separated by 60 feet of white sandstone ; Cardioceras cordatum, C. excavatum, Trigonia corallina . . . . ? 150 White sandstones with bands of lignite and occasional layers containing casts of marine shells ; estuarine beds 400 White cherty sandstones with many fossils, Aspidoceras ^ perarmatum, Perisphindes achilles, etc. ... 25 Sandy clays and black shales with Belemnites Oweni, Bel. hastatus, Cardioceras ornaturn, Cosmoceras Duncani . 300 Sandy shales, with Nucula nuda, Kepplerites gower- Clay ianus, and Kepplerites calloviensis Calcareous sandstone with many fossils, fanna of Kel- laways Rock ........ E. CONTINENTAL EQUIVALENTS 1. Northern and Eastern France In the great Parisian basin the general facies of the Upper Jurassic Series is similar to that of England, except that it is more calcareous. It can be divided into a similar series of zones and sub-stages, but unfortunately the original English divisions were not founded on zones but 011 local lithological differences, and though the English names have been widely adopted on the Continent, they have become attached to groups of strata which do not exactly correspond with the English divisions. Some adjustments of nomenclature will be required on both sides of the Channel in order to bring the names of the several stages into accord. Thus French geologists have created two stages, a Callovian and an Oxfordian, but the latter is the zone of Aspidoceras perarmatum and equivalent to the Lower Corallian of England ; while the Callovian includes the ornatum clays, though these have since been separated as a sub-stage under the name of Divesian. Thus their Callovian is practically our Oxfordian. Again there has been much confusion about the correlation of the equivalents of the Corallian and Kimeridge Clay in France, for the beds which represent our Upper Calcareous Grit are known as the Astartian or Sequanian, and have been classed in some places with the Corallian and in others with the Kimeridgian limestones. De Lapparent, following Tombeck, dropped the name Corallian altogether, and advocated an expanded Sequanian as its substitute ; but Messrs. J. F. Blake and T. Roberts, who both carefully studied the Upper Jurassic Series of Eastern France and the Jura, agree in thinking that the Astartian is more closely allied to the Kimeridgian than to the Corallian. 24 On the other hand, they agree that the English Kimeridge 452 STRATIGRAPHICAL GEOLOGY Clay is an exceptional argillaceous facies, and that it is represented in France and in Germany by thick masses of limestone and marl which belong to several sub-stages or zones. Even in Normandy and the Boulonnais the highest part of our Kimeridge Clay is represented by calcareous sandstones which are faunistically linked to the Portlandian stage. Lastly, French geologists regard the Purbeck Beds as a subordinate portion or sub-stage of an expanded Portlandian stage, and they have good reason for this because the marine equivalents of the Purbeck Group contain a fauna which is closely related to that of the Portlandian. The student will now be able to understand the following tabular correlation of the succession found in France and in the Jura Mountains with the English subdivisions, these and the French divisions having the value of sub-stages. England. Paris Basin. Jura. Purbeck Beds Purbeckian Marls and limestones. Bononian Massive limestones. [Clays with Ex. virgula Virgulian Compact limestones. 3.-J Clays with A. altervans Pterocerian Pale limestones. (Sandsfoot Beds, etc. Astartian Grey limestones and marls. 2 /Corallian limestone Coralliau Corallian limestones. 'V Lower Calcareous Grit Oxfordian Oxfordian limestones. , /Oxford Clay Divesian Ironstone and clay '\Kellaways Beds Callovian(restr.) Rough limestone. It will be seen that a good correlative grouping of these sub-stages into four stages would be made in accordance with the brackets, and that these stages could be called respectively (1) Oxfordian, (2) Corallian, (3) Kimeridgian, (4) Portlandian. The total thickness in the Jura is not more than 1200 feet. The most important and interesting groups are the Corallian, the Pterocerian, and the Purbeckian. The Corallian swells out in Lorraine, Burgundy, and the Jura, and includes thick masses of coralliferous limestone, often over 300 feet thick, which the French regard as true coral-reefs; they contain Diceras arietinum, D. eximium, with Comoseris mceandroides, and other corals. The Pterocerian in the Ardennes, Meuse, and Aube consists of clays, marls, and limestones with Perisphinctes decipiens and Exogyra virgula, not very different from the corresponding part of our Kimeridge Clay ; but southward they are replaced by massive limestones and marls containing a different set of fossils, especially Harpagodes Oceani, Natica gigas, Pholadomya Protei, and Hemi- cidaris Thurmanni. The Purbeckian is divisible into two parts, the lower, about 150 THE JUKASSIC SYSTEM 453 feet thick, consisting of dolomitic limestones containing Cyrena rugosa, Oorbula inflexa, and other marine species, the upper being a few feet of marl with brackish and freshwater fossils ; and these beds pass up into the Valenginian, which is classed as Lower Cretaceous, but may represent the highest part of our Purbeckian as well as some of the Wealden. 2. Germany The Upper Jurassic Series is well developed in Hanover, and extends westward into Westphalia beneath the Cretaceous deposits. The succession established by the labours of Seebach and Credner has been compared with the English Series by Dr. C. Struckmann, and the following table is compiled from his memoir,' a translation of which will be found in the Geol. Mag. for 1881. p. 546. [Serpulite limestone, 10 to 60 feet ) T , . r> /. Port- ) Mundermergel, gypsiferous marls/ Lacustnne Beds ' landian 1 Plattenkalk of Eimbeckhauser (platy limestones). I Limestones and marls with Perisphinctes gigas. Kime- i ^ r "' tl ^ a Beds, compact and oolitic limestones. ridffian I Pteroceras Beds, limestones with Harpagodes Oceani. \ Nerinoea Beds ( = Astartian), limestones. Corallian / Upper Cor. limestones (zone of Pecten (Lower Cor. limestones (zone of Ostrea rastellaris). Oxf rd' n / Hersumer Beds. Sandy beds with Aspidoceras perarma- \ turn, Oardioceras cordatum, and Gryphcea dilalata. p 11 f Clays with ornatus Ammonites. \Macrocephalus Beds, clays and ironstones. It will be seen that the greater part of the series consists of limestones with some clays at the bottom and a remarkable development of marls near the top ; these are of red and green colours and much resemble the marls of the Keuper ; they are from 400 to 500 feet thick in Hanover, but thicken westward to over 1500 feet in Westphalia, where they include limestones full of Corbula, Vivipara, and Cyclas, as well as beds of gypsum and rock-salt. The Serpulite limestone is full of small Serpula tubes, but also contains Oorbula. In the south of Germany (Swabia and Franconia) the Upper Jurassic Series is again found, but has a different palseontological facies. Its thickness is only from 800 to 1000 feet, and the succession of zones is as below : Feet. p , /Zone of Oppelia steraspis and Perisphinctes gigas, '"' \ (Plattenkalk and Solenhofen limestones) ... 80 {Zone of Exogyra virgula, including the coralliferous limestone of Nattheim about 150 Zone of Reineckia eudoxus (limestones) .... 100 Zone of Oppelia tenuilobata (marly limestones) . .130 454 STRATIGRAPHICAL GEOLOGY Feet. Cor. Zone of Peltoceras bimammatum .... about 80 f. ,/. /Zone of Peltoceras trasversarium and Waldheimia \ impressa (marls, clays, and limestones) . . 140 to 350 p, a l fZone of Ornatus Ammonites (clays), with Mac.ro- \ cephalus limestone at base. The most interesting member of this series is the lithographic limestone of Solenhofen, which must have been deposited in very quiet and clear water and not far from land ; for though it is of marine formation and contains several species of Ammonites and Belemnites with Crabs, Starfish, and other sea-creatures, it also encloses entire skeletons of Pterodactyles and the remains of the curious bird Archceopteryx (see p. 384), as well as the remains of dragonflies, cockroaches, wood-wasps, and ordinary flies (Musca and Empidia) ; all in the most beautiful state of preservation owing to the fine grain and slow deposition of the limestone. 3. The Alpine Fades In the Alpine region the lower part of the " Upper Jura " has but a small development ; the Callovian, Oxfordian, and Coralliaii being only represented by bands of limestone which do not always contain the characteristic species of Ammonites. The higher portion of the series, however, swells out into a massive series of limestones from 1000 to 2000 feet thick, which form a special facies of the Kimeridgian and Portlandian stages. At the base of these beds are limestones which frequently contain Oppelia tenuilobata, and consequently represent the Lower Kimeridgina (or Astartian). The middle part of the series consists of reddish limestones so full of Terebratula cliphya (now called Pygope jani- tor] that they are known as the Diphya limestone. The upper beds are massive pale-coloured limestones, often called the Stramberg limestone from Stramberg in Moravia, and they also contain Pygope janitor as well as Ammonites, Diceras, Nerinaeas, and Corals. Fig. 151.-TEREBRAT0LA DIPHYA. These Diphya limestones are sometimes called the Tithonian or the Tithonic facies of the Portlandian, for they contain the Portlandian Ainmonoids Oppelia steraspis and Oppelia lithographica THE JURASSIC SYSTEM 455 as well as other peculiar species, such, as Phylloceras ptychoicum and Waagenia hybonota. In the Stramberg limestone Phylloceras silesiacum and Perisphinctes transitionis also occur, and this limestone may be the equivalent of our Purbeckian. This facies extends from the Central Alps and the north of Italy through the Tyrol and Austria to the Carpathians and is found also in the Balkan States. 4. The Russian Facies As already stated the oldest Jurassic deposits in Russia are of Oxfordian age, and the fauna of these and the succeeding Corallian and Kimeridgian Beds is similar to that of their equivalents in Western and Central Europe ; the highest part of the series forms a special Russian or Boreal facies, differing in so many respects from the normal Portlandian that Nikitin proposed the name Volcjian for this set of beds. Although the thickness of the group is very small only a few feet of sand and shale with layers of phosphatic nodules several Ammonite zones have been detected in it, and it passes up into beds of Lower Cretaceous age. The beds were evidently formed in shallow water and under the influence of strong currents. The prevalent genera of Ammonites are Perisphinctes, Virgatites, Holcostephanus, and Craspedites, with Belemnites of the absolutus type and shells of the bivalve Aucella. The best sections are near Moscow and in the province of Simbirsk, especially between Simbirsk and Syrzan on the Volga, where, however, the beds are very thin, the succession according to Professor Pavlow being as follows in descending order : 4. Calcareous sands with Craspedites subditus . . . . ? 6 feet 3. Glauconitic sand with Holcostephanus Bladei, Hole, triplicatus, and Perisphinctes cf. gigas . . . . . . . 1 ,, 2. Sand with phosphate nodules and shales with Virgatites virgatus, Belemnites absolutus, etc. . . . . . 3 , , 1. Shales with Perisphinctes Bleicheri and Aucella Pallasi . 2 The subditus zone is correlated by Pavlow with the Purbeck Beds, but this is uncertain (see p. 491). It will be noticed that the underlying Portlandian is only 6 feet thick, but near Moscow the corresponding beds have a thickness of 26 feet. HISTORY or THE JURASSIC PERIOD Liassic Time. We have seen that it is convenient to date the beginning of this period from the subsidence which permitted the Southern Sea to make its way into the great Inland Sea which 456 STRATIGRAPHICAL GEOLOGY occupied so large a part of Northern Europe in later Triassic time. The level of the water in the Inland Sea seems to have been some- what below that of the outer ocean, because the Rhsetic Beds here and there pass beyond the limits of the Keuper, but the difference of level cannot have been great, and the invasion of the area by the outer sea seems to have been accomplished quietly and with little disturbance of previously formed deposits. The sea-waters simply occupied the basins and embayments of the great lake, in which shales and limestones began to be deposited. It is clear, however, that the mingling of sea and salt -lake waters resulted at first in the death of many of the creatures which inhabited both. The fish and marine reptiles which came in from the sea could not exist in the highly saline mixture, and their remains are found in the bone-bed which so often occurs at or near the base of the Rhsetic Beds. Neither could all the Mollusca which came in with them flourish in the land-locked waters, and those that managed to exist are small in size, as if dwarfed by unfavour- able conditions. In some localities there are several bone -beds at successive horizons, suggesting that there were several invasions of the sea over the barrier which separated the inner from the outer waters. Such irruptions may have taken place at conjunctions of high tides and southerly winds. Moreover, in Alsace and Lorraine, and again in Glamorgan, the Rha^tic sandstones and bone-beds are overlain by a band of red marls exactly like those of the Keuper, from which we may infer that communication with the outer sea was not yet permanent, and that there was a time when the area of the inland sea was again diminished by evaporation so that portions of it returned to the condition of separate salt lakes. At length a further subsidence brought in such a body of sea- water that permanent marine conditions were established, and many forms of life were able to inhabit the northern water-spaces. It was under these conditions that the shales and limestones of the Lower Lias began to be deposited. The material of the shales was probably obtained from the erosion of Palaeozoic shales, and more especially from those of the Coal-measures, large tracts of which must then have existed as part of the continental land to the west and north of the Liassic Sea. During the Permian and Triassic periods the climatic conditions had been like those of Central Asia at the present day, with a small rainfall, but in Liassic time it is evident that the climate was greatly changed, and that much more rain fell over the whole region, so that active erosion went on, and the rivers carried much mud into the Liassic Sea. THE JURASSIC SYSTEM 457 With respect to the limestones, it is noticeable that they are thin-bedded and compact and mostly of an earthy or argillaceous nature, so that they must have been originally more or less calcareous muds. Hence it is probable that the calcareous matter had a chemical or sedimentary rather than an organic origin. Mr. H. B. Woodward thinks they may have been derived mechanically from the waste of Palaeozoic limestones, 25 but it is more likely that the conditions were favourable to the chemical precipitation of carbonate of lime from solution. The geographic arrangement of land and sea in Liassic time was at first only a modification of that which prevailed in Keuper time (see Fig. 124) ; but the sea gradually extended itself over a larger area. There is no evidence that the Lower Lias occurs under the northern part of the Paris basin, and there is no trace of it in the Boulonnais. In Normandy it only occupied what may be called the bay of Carentan, and does not occur near Caen. The Middle Lias, however, extends much farther south, patches of it being found at intervals in the Departments of Calvados, Orne, Sarthe, and Maine, while Upper Lias has been reached in a deep boring at Kouen. Thus it appears that by the time of the Middle Lias a channel of communication had been established between the British Sea and that which spread over the greater part of France. There were, however, two large islands in this western part of the Liassic Sea. One of these was the central plateau of France ; proof of this is found in the fact that the Charmouthian deposits, in the form of sands and conglomerates, overlap the Sinemurian at many places round its borders, and rest directly on the Archaean rocks. Similarly in Poitou, near Thouars, the Toarcian in its turn passes beyond the Charmouthian as a consequence of continued subsidence. The other island lay farther north ; it comprised a part of Eastern England and the southern part of the North Sea, with the north-east of France and the whole of Belgium. The Liassic Sea seems to have occupied the whole of Germany and Central Europe, as well as Italy and the greater part of Spain, and it extended eastward through Hungary to the Carpathian Mountains, and through Transylvania to Kronstadt, but it does not seem to have reached so far east in Northern Europe, for no Liassic deposits have yet been found in Poland or Western Russia. Glevanian Time. In the west of Europe and in those places where the succession of marine deposits is most complete, as in the south-west of England, the north of France, and the Jura Mountains, there is a more or less rapid change from Liassic Clays through sands to limestones. There is also a considerable change in the fauna, 160 to 100 Ashdown Sand 500 to 200 1940 to 800 The Ashdown Sand consists of soft buff or white sand and sandstone : lit is 400 feet thick in Ashdown Forest, and forms the high ground of Crowborough Beacon (800 feet high). Near Hastings and Fairlight the lower part of this sand is replaced by clays with interbedded sandstones. Plant -remains are the only fossils, and it is the Fairlight Clay which has yielded the species mentioned on p. 471; the stems of Tempskya, formerly called Endogenites erosa, being common. The Wadhurst Clay, though never more than 160 feet thick, is palaeontologically important, as it is rich in fossils, and contains a bone-bed near Battle, from which many reptilian bones and teeth have been obtained. It consists of clays and shales, with bands of irregular layers of hard calcareous sandstone (Tilgate stone). The fossils are Viviparus fluviorum, Cyrena media, teeth of Lepidotus Mantelli, Hybodus basanus, Goniopholis crassidens, with bones of Iguanodon and Megalosaurus. The Tunbridge Sands are similar to the Ashdown Sands. Near Hastings their thickness is only 150 feet, but it increases westward, a set of mottled clays and shales appearing in the middle of the sands, and becoming of sufficient importance to deserve a separate name, the Grinstead Clay. In the upper sands of the Cuckfield district, which is often called the Tilgate Forest, there are large concretionary masses of calcareous grit or sandstone, such as were termed Tilgate stone by Dr. Mantell. It was in these masses 476 STRATIGEAPHICAL GEOLOGY that the remains of Iguanodon and Hylceosaurus were first dis- covered by Dr. Mantell, and the specimens which he figured and described are now in the South Kensington Museum. The uppermost member of the Wealden Group consists mainly of brown and blue clays, with layers of shelly limestone and occa- sional beds of sand and calcareous sandstone. The limestones are known as Sussex marbles, and are composed almost entirely of Viviparus shells, those in the lower part of the clay consisting chiefly of V, sussexensis, while the most constant band (the Pet- worth and Bithersden marbles) consists of V. fluviorum. The minute freshwater Crustaceans (Cypridea) are also very abundant in some places, the surfaces of the more shaly beds being crowded with them. Unio and Cyrena also occur. At Haslemere the highest beds, just below the Atherfield Clay, contain Corbula and Mytilus, as well as Cyrena and Melanopsis, indicating the change from freshwater to marine conditions. ' In the Isle of Wight Wealden Beds are exposed both on the eastern and western sides (see map, Fig. 172). The equivalent of the Hastings sands is not visible ; the lowest beds seen are those at Brook Point in Coinpton Bay, and consist of a pale sand- stone overlain by red and green marls in which the broken trunks of pine trees are so numerous that it would seem as if a raft of waterlogged trees had here sunk to the bottom, a frequent occur- rence in the deltas of large rivers like the Mississippi ; the spot is known to local geologists as the "pine -raft." The higher beds, consisting of variegated marls or clays with thin beds of sand- stone, can be seen by walking along the coast to the southward ; they have yielded bones and footprints of Iguanodon, Hypsilophodon, and other reptiles, with Unio valdensis, but other fossils are rare in this lower division, which has a thickness of about 700 feet. The higher beds are well exposed between Cowleaze Chine and Atherfield Point ; they consist almost entirely of grey shales with some beds of sandstone, ironstone, and limestone. These beds are much more fossiliferous, containing species of Meta, Cypris, Cypridea, and Candona, with Cyrena, Viviparus, and Vicarya (a sub-genus of Potamides). Small oysters also occur near the top. This series of beds is about 200 feet thick, and is directly overlain by the " Perna Bed " mentioned below. The whole of the Wealden comes to the surface near Swanage in Dorset, and extends thence along a broad valley to Worbarrow Bay on the other side of the Isle of Purbeck. At Swanage the estimated thickness of the formation is about 2300 feet, 3 but the junction with the Purbeck is not exposed. The lower beds consist of variously coloured sands and clays red, yellow, grey, and THE CRETACEOUS SYSTEM 477 white. The succeeding beds are red, purple, and white clays with beds of sand, and they contain layers of lignite and, in one place, the broken trunks of trees. At the top are shales (Punfield Beds) like those of the Isle of Wight, but only 34 feet thick. The greater part of the Wealden is exposed in the cliffs of Worbarrow Bay, but here its total thickness is reduced to about 1200 feet, and it consists entirely of alternating sands and clays with much lignite. The formation continues to diminish west- ward, and where last seen at Ridgway the thickness exposed is only 350 feet, but this may not be the full amount. Wealden Clays are found over a small space in the Vale of Wardour, but are only 60 or 70 feet thick. Probably they do not extend much farther northward, but their boundary line is concealed beneath the Upper Cretaceous rocks. Vectian. In the Wealden district the lithological characters of the beds composing this stage are very variable, but four sub- divisions are generally recognised (see p. 475 and Figs. 160, 161). Along the north side of the Weald the thickness of the Vectian varies from about 500 feet in Surrey to 240 near Sandgate, and about 130 in the boring at Dover. The Aiher field Beds are exposed on the Kentish coast between Sandgate and Hythe, and are traceable thence all round the area occupied by the Weald Clay, except between Lewes and Eastbourne. In Surrey they are traversed by the railway cuttings at Sevenoaks and Red Hill, Panopcea, plicata, Exogyra sinuata, Perna Mulleti, and Trigonia dcedalea being common fossils ; along this line their thick- ness varies from 30 to 50 feet. At Haslemere it is 60 feet, and their junction with the Wealden was observed by Mr. Salter, who says " the change from dark greyish-blue clay (Wealden) to the purely marine deposit of brown clay was marked no less by the fossils than by the change in the colour of the bed." Within a few inches of the Wealden Clay he found Pleuromya plicata abundantly embedded in the vertical position which such molluscs occupy when alive. This shows how quiet and gentle was the change from estuarine to marine conditions of deposit. In the Isle of Wight the Aiherfield Clay has at its base two beds which are together known as " Perna Bed," because Perna Mulleti is common in them and does not occur above. The lower bed consists of sandy clay with a thin basal seam of coarse grit containing rolled fragments of fossils and many broken bones and teeth of fish ; this clay is 2 J feet thick and is succeeded by a bed of brown calcareous sandstone also 2j feet. Both layers contain many fossils. The rest of the subdivision consists of pale blue clay with many flat calcareous concretions, and the upper part is 478 STRATIGRAPHICAL GEOLOGY known as the "lobster bed," from the occurrence of Meyeria magna. Other fossils found in these beds are Hoplites furcatus, Hop. leopoldinus, Hop. Deshayesi, Corbis corrugata, Pinna robin- aldina, and Enallaster Fittoni. Hythe Beds. In Kent, between Hythe and Sevenoaks, these beds consist of greenish-yellow sand, often marly or argillaceous, and hard bluish-grey calcareous sandstone, the soft beds being locally known as hassock and the hard beds as rag ; the latter being known as Kentish rag. The beds of rag and hassock alternate in regular layers. From Hythe to Maidstone the thickness of these beds is only from 70 to 80 feet, but they increase rapidly westward and are 160 feet at Sevenoaks, and still thicker in Surrey. This increase of thickness is mainly due to the incoming of an upper set of sands with frequent layers of brown chert. These upper beds Fi<*. 160. SECTION THROUGH THE LOWER CRETACEOUS SERIES NEAR DORKING. Distance 8 miles. 8 Middle and Lower Chalk. 5. Folkestone Beds. 2. Atherfleld Clay. 7. Malmstone and sand. 4. Sandgate Beds. 1. Wealden Beds. 6. Qault marls and clays. 3. Hythe Beds. attain their maximum thickness about Reigate, Dorking, and Leith Hill, where the lower group of soft ferruginous sand is from 100 to 130 feet, and the cherty beds are also in places over 100 feet thick. The Hythe Beds maintain this facies all round the western end of the Wealden area, but thin rapidly eastward again as they are traced along the southern border, disappearing entirely near Lewes. The chief fossils of the Hythe Beds are Terebratula sella, Exogyra sinuata, Trigonia spinosa, Tr. ornata, Actinocamax bruns- vicensis, Nautilus pseudelegans, Ancyloceras BowerbanJci, Macro- scaphites gigas, Hoplites Deshayesi, and Holcostephanus Hambrovi. Sandgate Beds. These are a variable set of beds, but are generally more or less argillaceous. At and near Sandgate (see Fig. 161) they consist of dark-green clayey sand full of glauconite grains and about 70 feet thick. Fossils are not numerous, but at the base, resting on the Hythe Beds, there is frequently a layer of pebbles and phosphate nodules, with Brachiopods and other fossils in the state of casts. THE CRETACEOUS SYSTEM 479 In Surrey this division is represented by the Nutfield and Bargate Beds, which at Nutfield consist chiefly of soft sandstone and Fuller's Earth (a fine silty clay). 4 Between Reigate and Godalming the Bargate Beds consist of sands and hard calcareous grits (Bargate stone), with some thin layers of Fuller's Earth and coarse pebbly beds at the base, con- taining phosphate nodules and many fossils ; the whole thickness is about 50 feet. Avicula pectinata is a characteristic fossil of the Bargate stone ; Terebratula oblonga, T. depressa, Waldheimia tamarindus, Terebratella Fittoni, T. trifida, and T. Menardi are some of the fossils occurring in the pebble beds, and with them are many fossils derived from the Oxford Clay, which must at that time have formed part of the shore-line north of Godalming. In Hants and West Sussex the Sandgate Beds reappear in a more normal form, consisting of shaly clay at the top, with sand and sandy clay below. In the Isle of Wight the Hythe and Sandgate Beds are represented by the Ferruginous sands, which consist of alternating beds of sand and clay, the sands of grey, green, or yellow tints and the clays brown or blue. The lowest beds are known as " the crackers," and contain Holcostephanus Hambrovii, Gervillia anceps, Trigonia caudata, Pleuromya plicata, etc. The overlying beds contain Terebratula sella in great profusion with Douvilleiceras Martini, Macroscaphites gigas, M. Hillsi, and Grioceras BowerbanJci. These lower beds may be called the Walpen sands, and they corre- spond to the Hythe Beds. Above them are the equivalents of the Sandgate Beds, comprising two beds or bands, the lower of sand with ferruginous concretions about 20 feet thick, the upper a bed of clay without fossils and 40 feet thick. The concretions yield Rhynchonella sulcata, Avicula pectinata, Thetis Sowerbyi, Gyprina angulata, and Trigonia vectiana. Folkestone Beds. At Folkestone these consist of light-green and grey sand with beds of hard stone, partly siliceous and partly calcareous in composition, which have been described by Dr. G. J. Hinde as veritable sponge-banks, the sponges being of the Hex- actinellid order, and their fine needle-like spicules are clearly visible on weathered portions of the rock. The total thickness is about 80 feet. Fossils are not abundant in these beds, but they have yielded Avicula pectinata, Exogyra sinuata, Pecten (Neithea) atava, Neithea Morrisi, Ostrea frons, and Waldheimia pseudojurensis. For a few miles inland the composition of the Folkestone Beds is much the same, but west of Saltwood they begin to change their character, the cherty sponge-beds disappear, and the sands are chiefly white, yellow, and brown, current-bedded, and including LU *3 IS! e ^ CO SJ 2. . - * I HI 3 * st 3 THE CRETACEOUS SYSTEM 481 irregular layers of ironstone. This facies they preserve with little change all round the Wealden area, and consequently the name of Folkestone Beds is not a very happy one. In Sussex they include thin bands of hard ferruginous grit, which is locally known as car- stone. In Surrey and Hants they are from 130 to 160 feet thick, but like all other members of the Vectian, they thin eastward through Sussex, till near Eastbourne the whole stage appears to be absent, and the Gault rests directly on the Weald Clay. The Sand-rock Group appears to be the representative of the Folkestone Beds. It consists mainly of coherent sands with bands of laminated clay, but contains no fossils. The carstone is a coarse ferruginous grit which varies greatly in thickness, being 72 feet thick at Eedcliff, north of Sandown, but only from 6 to 12 feet on the western side of the island. It contains rolled pebbles and phosphatic nodules, and is considered by Mr. Strahan to be more closely connected with the Gault than with the Vectian. 5 The fossils found in it comprise Desmoceras Beudanti, Plicatula carteroniana, and Lima parallela. When followed across to Dorset the Vectian is found to be much reduced in thickness, being only 200 feet at Punfield Cove, near Swanage, 136 at Worbarrow Bay, and 66 in Mupe Bay, beyond which it is not seen. At Punfield the Atherfield Clay is 50 feet thick, and the upper part contains some remarkable fossils, especially species of Vicarya, Natica, Actceon, Gerithium, and Pleurotoma, which are common in the Rhodanian of Spain, but have not been found elsewhere in England. 2. Midland District The Wealden is not seen anywhere north of the Vale of Wardour, and the sands and sandstones which emerge from beneath the Gault in Berkshire, and thence at intervals northward to Cambridge and Ely, seem to belong to the higher part of the Vectian Group, nothing comparable to the Atherfield Clay or the Hythe Beds having been observed in situ. Near Faringdon 6 (see Fig. 163) the Vectian consists for the most part of soft yellow and brown sands, with some layers of sandstone and thin bands of ironstone, but at the base there are irregular beds of pebbly, fossiliferous gravel, compacted in places into a conglomerate. The pebbles are chiefly of quartz and porcellanous slate (lydianite). Among the fossils sponges, Brachiopods, and Echinoids are con- spicuous, the following being some of these : Terebratula tornacensis. Peltastes Wrighti. ,, depressa. Cidaris faringdonensis. Waldheimia tamarindus. Raphidonema porcatum. 2 I CO = 1=2 . P Mi S I sal THE CRETACEOUS SYSTEM 483 Terebratella Menardi. Rapliidonema faringdonensis. ,, Fittoni. Trematocystia anastomosans. Rhynchonella latissima. Peronidella ramosa. The summit of Shotover Hill, south-east of Oxford, consists of ferruginous sands with bands of ochre and cream-coloured loam, which contain remains of freshwater shells, Unio, Cyrena, and Viviparus. Similar beds occur at Wheatley, Brill, Crendon, and Thame, where they include beds of grey and lilac clay. At Crendon, Pecten aptiensis has been found in a layer of calcareous concretions which overlies the clays, and marine fossils have also been found at Stone near Aylesbury. Hence there can be little doubt that the Shotover Beds, which rest in some places on Portland, in others on the Purbeck Beds, are freshwater deposits of Aptiaii age. Their thickness is variable with a maximum of about 50 feet. 7 Yellow and white sands emerge from beneath the Gault near Leigh ton Buzzard and form a range of hills by Woburn and Ampthill as far as the neighbourhood of Shefford. Near their base are beds of phosphatic nodules, which were worked near Brickhill, and include casts of fossils derived from the under- lying Jurassic Clays, together with well-preserved contemporaneous fossils. Among the latter are Terebratula sella, T. depressa, T. microtrema, T. moutoniana, and others, Terebratella Menardi, T. Fittoni) Waldheimia celtica, W. tamarindus, and others, Peltastes Wrighti, and many of the Faringdon sponges. 8 The sands above are dug for glass-making purposes, and are believed to be about 200 feet thick near Woburn ; they also contain a band of Fuller's Earth, which has been worked for more than a century. These Woburn sands, passing beneath the valley of the Ivel, become conspicuous again near Sandy and Potton, with the same nodule beds near the base, in which derived reptilian bones are very abundant ; thence the sands run through Cambridgeshire to the Fens south of Ely, near Streatham and Upware. At Upware in the Fens of Cambridgeshire are sands and silts with two nodule beds similar to those of Brickhill and Potton, but containing a still larger number and variety of fossils. The beds are banked against the Corallian limestones, are less than 12 feet thick, and are overlain by the Gault, so that they seem on the point of thinning out. 3. Northern District This includes the Lower Cretaceous tracts in the counties of Norfolk, Lincoln, and York. Each of these exhibits a somewhat 484 STRATIGRAPHICAL GEOLOGY different facies, the deposits in Norfolk being mainly sands, while those in Yorkshire are clays and those in Lincolnshire are sands r clays, and limestones. In the two latter counties we have a more complete succession of marine deposits of Lower Cretaceous age than is found in any other part of England, and our detailed knowledge of them is mainly due to the labours of Professor J. W. Judd and Mr. G. "W. Lamplugh. 9 Unfortunately the Norfolk outcrop is separated from that of Lincolnshire by the breadth of the Wash, and in Yorkshire the exposed areas near Speeton are very small, their southerly continuation being concealed beneath the overlap of the Upper Cretaceous (see map, Fig. 165). Con- sequently correlation is rendered difficult and entirely dependent on the discovery of fossils. The probable correlation is shown in the following table : Yorkshire. Lincolnshire. Norfolk. Clays with Belemnites brunsvicensis Clays with Act. jaculum Clays with Belemnites lateralis {Carstone Tealby limestone and\ Roach ironstone J Tealby clay /Claxby ironstone \Spilsby sandstone Carstone. Snettisham clay. ^ Sandringham / sands. (Absent). It will be convenient to take Yorkshire first, and in that county the only good exposure is 011 the coast near Speeton Gap, and hence the whole series, which has a total thickness of about 330 feet, is sometimes termed the Speeton Clay. The base of this clay is a layer of black phosphatic nodules embedded in stiff clay. The fossils are badly preserved, but among them Professor Pavlow identifies Belemnites absolutus with Ammonites resembling Holcostephanus Panderi and Hole. scythicus T species characteristic of beds in Russia which are regarded a& Portlandian. It is a question, however, whether these fossils- are contemporaneous or derived. This bed is succeeded by clays which contain Belemnites lateralis, Holcostephanus subditus,. Hole. Lamplughi, Hoplites amblygonus, with Exogyra Couloni and Astarte senecta. These clays are about 35 feet thick and were considered to be of Purbeck age by Professor Pavlow, but the- continental zone of Hoplites Boissieri with which he correlated part of these beds is now regarded as the base of the Valenginian, and the natural base of the Cretaceous System both in England and Germany is at the base of this zone of Bel. lateralis. The overlying clays form the zone of Actinocamax jaculum and THE CRETACEOUS SYSTEM 485 PEETON Flamboro' Head gton Quay Bridlington Bay SKETCH-MAP OF THE YORKSHIRE AND LINCOLNSHIRE WOLDS, (based on the map in H.B. Woodwards 'Geology of England <2r Wales'.) Scale:! inch= about 18 miles. Railways *~*+- : The outcrop of the S PEE TON SERIES is oughly indicated by the horizontal shadinjf. THE WASH Typo. Etching Co.So, Fig. 165. MAP OF THE CRETACEOUS AREAS IN LINCOLNSHIRE AND YORKSHIRE. By G. W. Lamplugh (from Quart. Journ. Geol. Soc. vol. lii. p. 187). lie outcrops of the Lower Cretaceous Series are indicated by horizontal lines, and show up darker than the other parts of the map. 486 STRATIGRAPHICAL GEOLOGY are about 120 feet thick. They appear to be divisible into three sub-zones characterised by Ammonites which in succeeding order are Hoplites regalis, Holcostephanus speetonensis, and Hole. Decheni. Besides these Crioceras Duvali, Pecten cinctus, Meyeria ornata, and many other fossils occur. The whole of this zone appears to be older than the Atherfield Clay, and is therefore a marine repre- sentative of the Weald Clay. The highest part of the Speeton Clay consists of grey and black pyritous clays containing Belemnites brunsvicensis, Hoplites Deshayesi, a large Crioceras, Area securis, Isocardia anyulata, and others. These beds are clearly equivalent to the Vectian of Southern England ; their total thickness is more than 150 feet, and may be 180, and they are succeeded by gritty clays of Gault age containing Belemnites minimus. In Lincolnshire the Lower Cretaceous Beds emerge from beneath the Upper near Caistor, and extend thence in a gradually widening strip to the neighbourhood of Spilsby and Candlesby at the southern end of the Lincolnshire Wolds. The succession of beds found in this district is shown in Fig. 164, and the full thickness does not exceed 200 feet. The Spilsby sandstone has a bed of derived phosphatic nodules at the base resting on the Kimeridge Clay. Above are soft yellow and grey sands, including large masses or doggers of calcareous sandstone, which are rich in fossils and yield Belemnites lateralis, Holcosteph. subditus, Hole, plicomphalus, Trigonia ingens, T. robinal- dina, etc. The Claxby and Hundlesby ironstones also appear to belong to this zone, as they have yielded Bel. lateralis, Hole. Beani, Hole. Gravesiformis, Astarte senecta, Trigonia ingens, Rhynchonella multiformis, Terebratula sella. The Tealby Clays are from 40 to 100 feet thick, and represent the zone of Bel. jaculum, yielding that fossil with Hoplites speeton- ensis, Crioceras Duvali, and Perna Mulleti. The Tealby limestone contains Bel. brunsvicensis, Exogyra sinuata, Pecten cinctus, and appears to pass southward into a ferruginous oolitic marl, locally called "roach." Above these calcareous beds is the carstone, which has not yet yielded any fossils except at its junction with the red chalk, where Bel. minimus and Terebratula biplicata occur. In Norfolk the thickness of the series is rather less than in Lincolnshire. The Sandringham sands consist of light- coloured sands with some beds of brown flaggy sandstone in the upper part ; they have not yielded determinable fossils, and are not like the Spilsby sands ; hence it is more likely, as Mr. Lamplugh suggests, that they represent the Tealby Clay. The Snettisham Clay contains Bel brunsvicensis, and appears to be the equivalent of the THE CRETACEOUS SYSTEM 487 Tealby limestone, but it is nowhere more than 20 feet thick and thins out southward. The carstone of this country is a brown ferruginous grit, often cemented by oxide of iron into a sandstone capable of being used as a building stone. Its thickness is about 40 feet, and it is only at its base that fossils have been found. Some of these occur as phosphatic nodules which may or may not be derived, others occur in concretionary lumps of hard grit, and include Douvilleiceras cornuelianus, Trigonia scapha, Perna Mulleti, Isocardia angulata, and Cardium subhillanum species which prove that this part of the carstone is of Aptiaii age and comparable with the Hythe Beds. Hence we may infer that the carstone of Lincolnshire is of the same age. 4. Scotland Bocks of Lower Cretaceous age have been found in Caithness and Aberdeen. Those of Caithness occur at Leavad near Wick and appear to be in situ ; the exposure showing 20 feet of yellow sand with doggers of hard calcareous sandstone, which have yielded species of Oraspedites, Desmoceras, Crioceras, and Hamites. 1 Cras- pedites suggests a low horizon in the series. In Aberdeen near Cruden boulders of a glauconitic sandstone or gauze occur in boulder-clay and contain fossils which form a curious assemblage of Vectian and Selbornian species, but the presence of Hole, speetonensis, Grioceras Duvali, and Trigonia vectiana appears conclusive of their Vectian age. 11 C. CONTINENTAL EEPRESENTATIVES The Lower Cretaceous deposits of the European continent vary greatly in their lithic components, their thickness, and their fossil contents. They may be regarded as comprising three different facies of the series, (1) a southern deep-sea facies typically developed in the south-east of France - (Dauphin e and Provence), (2) a central or intermediate facies found in the Jura and Switzerland, (3) a northern facies found in Germany and Eussia. It will suffice to describe the typical succession in each of these regions. 1. Provence The finest sections in this area are those of the Montagne de Lure and Mount Ventoux near Sisteron and Simiane. The most recent studies of the strata forming these mountains are those of MM. Kilian and Leonhardt, and I shall here follow the capable THE CRETACEOUS SYSTEM 489 guidance of Professor Kilian with regard to the succession and the grouping of the zones. 12 The illustration, Fig. 166, is reproduced from one of his sections. In this part of France there is a complete passage from the marine (Tithonian) facies of the Jurassic to the Lower Cretaceous, and the exact line of separation has been a matter of dispute, but Professor Kilian has made it clear that the zone of Hoplites Boissieri should be regarded as the first term in the Cretaceous Series. The thickness of the several stages varies considerably in different parts of the area, but the total is about 2000 feet, and as it largely consists of limestones its formation must have occupied a long period of time. The following is a tabular view of the succession in the M. de Lure: ( Marls with Belemnites semicanalicutus, Oppelia nisus, . ... and Hoplites fur catus. 300 to^ 600 I Limestone of Graves with Toxaster Collegnoi, etc. Limestones with Hoplites Deshayesi passing into Orbito- [ Una limestone with Requienia ammonia. Barremian, / Limestones with recticostatus and Macro scaphites Yvani. 200 to 400 ( Limestones with Holcodiscus fallax and Crioceras Emeriti. Hauterivian, about 300 Valenginian, 600 to 1200 Limestones with Hoplites angulico status. Limestones with Crioceras Duvali, Hoplites radiatus, and Bel. dilatatus. Marly limestone with Hoplites neocomiensis and H. regalis. Marls with Hoplites neocomiensis and Belemnites Emeriti. Marls with Hoplites Boissieri, H. occitanicus, and Phyllo- ceras semisulcatum. The limestone with Requienia ammonina, classed above as Aptian, is part of the special facies developed at Orgon in Vaucluse, which was formerly regarded as the type of a stage called Urgonian by d'Orbigny but since found to pass laterally into Barremian and Lower Aptian. 2. The Jura In this area the whole Lower Cretaceous Series is still purely marine, but the thickness is much less, being only from 500 to 700 feet, and Ammonoids are much less common than they are in the southern region, so that the same zones cannot be recognised. According to the most recent authorities the following is the general succession in the Eastern Jura and in the basin of the Rhone. Aptian [Marls with Ostrea aquila and Plicatula placunca repre- 40 to 60 sented at Perte du Rhone by sandstones, underlain [ by marls with Orbitolina lenticularis. Barremian f White limestone with liequienia ammonia represented inn fppt ' 1 at Perte du Rhone by red and grey limestones with I Heter aster oblongus and Het. Couloni. 490 STEATIGRAPHICAL GEOLOGY I Yellow limestone of Neuchatel with Toxaster complanatus, Terebratula tamarindus, and Ehynchonella lata. Marls of Hauterive with Hoplites radiatus, H. leopoldinus, Hole. (Astieria) Astieri, and Belemnites dilatatus. Marls with Bryozoa and Ostrea Couloni. v i (Red oolitic limestone with Pygurus rostratus. oAA g f ?"' \ White limestone with Natica leviathan. 200 feet [Marl and rubbly limestone with Toxaster Campichei. 3. Germany The northern facies is well developed though not well exposed in Hanover. In the northern part of the province there appears to be a nearly complete marine series comparable with that of Speeton, and according to the recent observations of von Koenen the descending succession can be made out as follows : f Marls with Hoplites furcatus and Belemnites brunsvicensis. 4 Clay with Hoplites Deshayesi. [Clay with Ancyloceras gigos and Belemnites Grasi. g f Clays with several species of Crioceras and Ancyloceras with \ Belemnites brunsvicensis. o /Clays with Crioceras Tombecki and C. capricornus. \Clays with Hoplites radiatus (=noricus). _ / Clays with Polyptychites tercissus and P. psilostomus. \Clays with Belemnites lateralis and Oxynoticeras Gevrili. The figures 1 to 4 indicate a grouping which roughly corre- sponds with the Swiss and Jura stages, but the lowest marine clays are not probably equivalents of the lowest Valenginian, for in Southern Hanover the marine beds are underlain by a thick estuarine and freshwater formation comparable with the English Wealden and like it consisting of sandstones in the lower part and of clays in the upper part, containing Corbula inflexa, Melania strombiformis, Cyrena, and Viviparus. In some places these clays have been observed to alternate with the lower marine clays, and the German Wealden is probably a delta formation formed by a river discharging into the Northern Sea. 4. Russia The corresponding series in Eussia is best exposed on the right bank of the Volga between Simbirsk and Syrzan, but its total thickness is very small, being only from 50 to 60 feet, and it must therefore be regarded as a condensed representative. The divisions recognised by Professor Pavlow and other Russian geologists are : THE CRETACEOUS SYSTEM 491 7. Shales with Hoplites Deshayesi ...... 6. Black clay with Holcostephanus (Simbirskites) Decheni . 5. Black clay with Hole. (Simbirskites) versicolor 4. Sand and phosphatic nodules, Hole. Keyserlingi 3. Calcareous sandstone of Riasan ...... 2. Sands and phosphate nodules with Hoplites nodiger and Hop. kashpuricus .......... 1. Sands with Craspedites subditus ...... Feet 17 - 16 5 3 12 10 The four lower zones contain Belemnites lateralis and the two upper Bel. Jasekoivi, neither B. jaculum nor B. brunsvicensis occurring in Russia. II. UPPER CRETACEOUS SERIES A. STAGES AND ZONES As stated on p. 471, the members of this series have a far wider extension than those of the lower series ; they extend southward below the English Channel into France, they pass eastward beneath the eastern counties of England, and below the floor of the North Sea into Belgium and Holland. The following is a tabular view of the stages and zones into which the English Series is divisible, with the names of the corresponding French stages : Upper Chalk -< Zones. Ostrea lunata Belemnitella mucronata Actinocamax quadratus Marsupites testudinarius Micraster coranguinum Micraster corcestudinarium Holaster planus Middle Chalk f Terebratulina lata x \ Rhynchonella Cuvieri T pn i, /Holaster subglobosus Lower Chalk J Schlcenbachia varians Selbornian f Pecten asper | Schloenbachia rostrata 1 Hoplites interruptus 1 Douvilleiceras mammillatum French Divisions. Campanian. ! Senonian. rTuronian. j-Cenomanian. Ulbian. B. CHARACTERISTIC FOSSILS The following are some of the species which are characteristic of each of the four stages or subdivisions of the Upper Cretaceous series : 492 STRATIGEAPHICAL GEOLOGY Fossils of the Selbornian and Albian Porifera. Chenendopora Michelini, Doryderma Benettise, Siphonia tulipa, Barroisia Orbignyi. Hydrozoa. Parkeria sphserica. Fig. 167. GROUP OF SELBORNIAN FOSSILS (GAULT). a. Trochocyathus conulus. /. Bellerophina minuta. b. Inoceramus sulcatus. g. Hoplites splendens. c. Plicatula gurgitis. h. Hamites attenuatus. d. Aporrhais carinata. i. Nucula pectinata. e. Scalaria gaultina. k. Natica gaultina. I. Lima parallela. Actinozoa. Cyclocyathus Fittoni, Trochocyathus conulus. Echinoderma. Cardiaster fossarius, Cidaris gaultina, Echinospatagus murchisonianus, Cottaldia Benettise, Pentacrinus Fittoni. THE CRETACEOUS SYSTEM 493 Annelida. Serpula concava, Ditrupa difformis. Crustacea. Etyus Martini, Necrocarcinus Bechei, Palaeocorystes Stokesi, Hoploparia longimana. Brachiopoda. Terebratula biplicata, Terebratella pectita, Rhyncho- nella dimidiata, Lingula subovalis. Lartullibranchia. Grammatodon carinatas, Cucullsea glabra, Pectunculus umbonatus, Ostrea vesiculosa, Exogyra conica, Pli- catula gurgitis (-P. pectinoides), Lima gaultina, Fig. 168. GROUP OF SELBORNIAN FOSSILS (GREENSAND). a. Chenendopora expansa. b. Micrabacia coronula. c. Cottaldia Benettise. 7. Hoplites auritus. d. Terebratula biplicata. . Exogyra columba. /. Pectunculus sublsevis. Gastropoda. Scaphopoda. Cephalopoda. Inoceramus sulcatus, I. concentricus, Nucula pectinata, Panopea mandibula, Trigonia aliformis. Dicroloma (Anchura) carinata, D. (Perissoptera) Parkin- soni, Solarium ornatum, Bellerophina minuta, Natica gaultina, Scalaria gaultina, Turritella granulata. Dentalium decussatum. Douvilleiceras marnmillatum, Desmoceras Beudanti, Hoplites interruptus, Hoplites lautus, Hoplites splendens, Hoplites auritus, Schlcenbachia .rostrata, Schloanbachia varicosa, Hamites attenuatus, Nautilus clementinus, Belemnites minimus. 494 STftATIGRAPHICAL GEOLOGY The following are common species which occur both in the Selbornian and the Lower Chalk (Cenomanian) : Elasmostoma consobrinum, Catopygus columbarius, Holaster lcevis, Salenia petalifera, Peltastes clathratus, Pseudodiadema variolare, Ps. ornatum, Discoidea subuculus, Kingena lima, Rhynchonella grasiana, Lima globosa, Pecten asper, P. orbicularis, P. hispidus, Ostrea vesicularis, Aucellina gryphwoides, Nautilus elegans. Fig. 169. GROUP OF SELBORNIAN FOSSILS (GREENSANDS). a. Nautilus elegans. b. Salenia petalifera. e. Siphonia pyriformis. c. Pecten asper (young). d. Rhynchonella dimidiata. Fossils of the Lower Chalk and Cenomanian Porifera. Stauronema Carteri, Plocoscyphia labrosa, P. fenestralis. Actinozoa. Micrabacia coronua. Echinoderma. Discoidea cylindrica, Holaster subglobosus, H. trecensis, Cidaris vesiculosa. Annelida. Serpula umbonata, S. annulata. Crustacea. Enoploclytia brevimana. Brachiopoda. Rhynchonella Martini, R. Mantelliana, Terebratula semiglobosa, Terebratulina nodulosa. Lamellibranchia. Cucullaea mailleana, Inoceramus Crippsi, Lima aspera, Pecten elongatus, P. fissicosta, Plicatula inflata, Spondylus striatus, Unicardium ringmeriense. THE CRETACEOUS SYSTEM 495 Gastropoda. Dicroloma (Perissoptera) Mantelli, Avellana cassis, Solarium bicarinatum. Scaphopoda. Dentalium majus. Cephalopoda. Schloenbachia varians, Acanthoceras rotomagensis, Acanthoceras Mantelli, Turrilites costatus, Scaphites sequalis, Baculites baculoides, Actinocamax plenus. Fossils of the Middle Chalk Echinodermata. Galerites ( = Echinoconus) subrotundus, Cardiaster pygmsea, Discoidea Dixoni, Cidaris hirudo. Fig. 170. GROUP OF LOWER AND MIDDLE CHALK FOSSILS. a. Holaster subglobosus, f 6. Turrilites costatus, \. c. Inoceramus labiatus, |. d. Schloenbachia varians, e. Rhynchonella Cuvieri, nat. size. f. Mantelliana, f. g. Actinocamax plenus, f. h. Scaphites aequalis, |. Brachiopoda. Terebratula semiglobosa, Rhynchonella Cuvieri, Tere- bratulina lata ( = gracilis of most authors). Lamellibranchia. Inoceramus labiatus (mytiloides), I. Lamarchi var. Cuvieri. Cephalopoda. Mammites nodosoides, Pachydiscus peramplus, Priono- tropis Woolgari. Fossils of the Upper Chalk (lower zones) Porifera. Camerospongia campanulata, Doryderma ramosum, Ventriculites impressus, V. mammillaris. 496 STRATIGRAPHICAL GEOLOGY Fig. 171. GROUP OF UPPER CHALK FOSSILS. a. Belemnitella mucronata. b. Terebvatula carnea. c. Spondylus spinosus. d. Galerites conicus. e. Echinocorys vulgaris. /. Marsupites testudinarius. 7. Micraster coranguinum. h. Leskei (base). i. precursor (base). 7r. coranguini\m (base). THE CEETACEOUS SYSTEM 497 Echinoderma. Cidaris clavigera, C. perornata, Phymosoma ( = Cypho- soma) radiatum, Holaster planus, Micraster Leskei ( = breviporus), M. precursor, M. cortestudinarium, M. coranguinum, Epiaster gibbus. Lamellibranchia. Inoceraraus involutus, I. Lamarcki, and var. Brongni- arti, Lima Hoperi, L. divaricata, Plicatula Barroisi. Gastropoda. Solariella gemmata, Trochus Schliiteri, Cerithium Saundersi. Cephalopoda. Heteroceras reussianura, Scaphites Geinitzi, Baculites bohemicus. Fossils of the Upper Chalk (higher zones] For if era. Cceloptychium agaricioides, Plinthosella squammosa, Scytalia radiciformis, Siphonia Koenigi, Stichophyma tumidum. Anthozoa. Caryophyllia cylindracea, Ccelosmilia laxa, Parasmilia Fittoni. Echinoderma. Galerites ( = Echinoconus) abbreviates, OfFaster pilula, Marsupites testudinarius, Micraster coranguinum, Cardiaster ananchytis. Brachiopoda. Rhynchonella limbata, Magas pumilus, Crania costata, Terebratulina Rowei, T. gracilis. Lamellibranchia. Ostrea acutirostris. 0. curvirostris, 0. lunata, 0. wegmaniana, Pecten Nilssoni, P. cretosus, P. Mantellianus, Spondylus dutempleanus. Cephalopoda. Haploceras leptophyllum, Scaphites constrictus, Bacu- lites Faujasi, Hamites cylindraceus, Actinocamax quadratus, Act. granulatus, Belemnitella mucronata, B. lanceolata. Besides the above, the following are more or less common throughout the Upper Chalk : Porifera. Heterostinia obliqua, Ventriculites decurrens, V. radiatus. Anthozoa. Parasmilia centralis, Axogaster cretacea. Echinoderma. Echinocorys ( = Ananchytes) vulgaris, Galerites (Echino- conus) conicus, Cidaris sceptrifera, Phymosoma ( Cyphosoma) Kcenigi. Brachiopoda. Terebratula carnea, Rhynchonella plicatilis, Rh. reedensis. Lamellibranchia. Ostrea semiplana, Spondylus spinosus. C. THE UPPER CRETACEOUS OF BRITAIN 1. England In describing the Upper Cretaceous strata it will be more convenient to take each stage separately and trace its variations across England. Selbornian. This stage includes the deposits which have long been known as Gault and Upper Greensand, names which date from a time when a Middle Cretaceous Series was supposed to exist, comprising the Lower Greensand, the Gault, and the Upper Green- 2K 498 STRATIGRAPHICAL GEOLOGY sand, which were supposed to occur in regular upward succession. As the palaeontology of these deposits was more carefully studied it was seen that the " Lower Greensand " was a distinct stage, but that the Upper Greensand was inseparable from the Gault, and was to a large extent merely a sandy facies of the Gault. Hence it became necessary to combine these two facies under one name as a single stage. For this the name Selbornian has been chosen from the village of Selborne in Hampshire, though the names Gault and Greeusand are convenient terms for the argillaceous and the sandy facies respectively. 13 Where the stage is complete four zones can be distinguished. These are : 4. Zone of Pecten asper and Cardiaster fossarius. 3. Zone of Schloenbachia rostrata ( = inflaf,a). 2. Zone of Hoplites lautus and H. interruptus. 1. Zone of Douvilleiceras mammillatum. Of these zones the lowest is always a sand, the second is generally clay (i.e. Gault), the third may be clay or marl or malmstone or sand, the fourth is always a " greensand " or sandstone. The typical development of the argillaceous facies is in Kent, and the beds are well exposed at Folkestone, where the " Lower Gault" consists of grey clays with several bands of phosphatic nodules and a layer of such nodules at the base embedded in dark- green glauconitic sand. The characteristic fossils are Hoplites interruptus, H. splendens, H. lautus (upper part), Lima parallela, Pinna tetragona, Inoceramus concentricus. Below and forming the base of the stage are 5 or 6 feet of coarse green sand containing Douvilleiceras mammillatus. The Upper Gault or zone of Sch. rostrata consists of pale-grey marly clay about 78 feet thick ; this includes a bed of dark-green glauconitic sand in the upper part, and is separated from the Lower Gault by a layer of phosphatic nodules. These beds yield Schlcenbachia rostrata, Sch. Goodhalli, Pecten raulinianus, Tere- bratula biplicata in the upper part, Schlcenbachia varicosa and Inoceramus sulcatus in the lower part. Inland the thickness of the Gault becomes much greater and is about 200 feet near Maidstone. This increase appears to be chiefly in the Upper Gault, but it is not till we reach Westerham that the so-called " Upper Greensand " makes its appearance at the top of the Gault, and thence it thickens westward through Surrey. This deposit is known in Surrey as "firestone" and " hearth- stone," and in Hampshire as " malmstone " ; it is a greenish-grey siliceous stone consisting partly of fine sand and glauconite grains, partly of sponge spicules and globular colloid silica, and it passes down into sandy marls which are clearly part of the Upper Gault. THE CRETACEOUS SYSTEM 499 In Hampshire, near Alton and Selborne (see map, Fig. 157), the general succession is as follows : Feet. 5. Soft greenish sands without fossils . . . 2 to 15 4. Malms tone with beds of calcareous stone . . 80 to 100 3. Sandy marls with Sch. rostrata and Inoc. sulcatus 30 to 50 2. Dark clays with Hoplites interruptus . . . about 100 1. Green sands with phosphatic nodules . . . 3 to 15 About 240 From the malmstone Schlwnbachia rostrata, Hoplites auritus, Anisoceras armatum, Area carinata, and Pleuromya mandibula have been obtained, and there is no doubt that it represents the greater part of the Upper Gault of Folkestone. In Wiltshire and the Isle of Wight there is a similar succession, and the beds are well exposed in the Undercliff along the south side of the Isle of Wight, where the following are the main sub- divisions : Feet. P. asper zone Sands and chert-beds (few fossils) . . about 30 Sch. rostrata ( Soft sandstones and sandy malm . . ,,70 zone \ Dark-grey sandy clays . . . ,,45 Lower Gault Blue clays with Hoplites denarius, etc. . ,, 100 The "Gault" passes down into the "Carstone" (see p. 481), which Mr. A. Strahan is inclined to regard as representing the zone of Douv. mammillatum ; but that fossil has not yet been found, and the question remains undecided. In South Dorset the succession is similar, but the base of the Selbornian is sharply denned by a thin layer of quartz pebbles, while the Lower Gault becomes more and more sandy as it is traced westward. The zone of Sch. rostrata consists of green sands with beds of calcareous sandstone, and there is always a well-marked zone of Pecten asper at the top. In Devon the beds are well exposed in the cliffs between Lyme Regis and Sidmouth, the base of the Selbornian gradually passing over the Lower Lias and across the Keuper marls till it rests on the Keuper sandstones (see Fig. 131, p. 396). At the base near Lyme and Axmouth there are still some few feet of clay (Lower Gault), but westward this either thins out or passes into sand. At Beer and near Beer Head the succession is as follows : Feet. Zone of f Hard calcareous sandstone (few fossils) . . about 10 P. asper \ Sandstones with chert-nodules, Exogyra digitata 60 Zone of j Soft grey sands passing down into dark-green S. rostrata \ clayey sand, Dosiniopsis caperata ... 90 160 500 STRATIGRAPHICAL GEOLOGY In this area Pecten asper does not occur in the Selbornian, but is found in the overlying arenaceous representative of the Lower Chalk. At Sidmouth and in the Blackdown Hills the lower sands have yielded a rich fauna, the shells being preserved in chalcedony, and the following are some of the commonest species, Sch. varicosa, Aporrhais calcarata, Turritella granulata, Protocardia hillana, Cucullcea glabra, Cytherea (Callista) plana, Dosiniopsis caperata, Lucina orbicularis, Trigonia aliformis, and Tr. scabricola. An outlier of the Selbornian sands occurs on the Haldon Hills about 12 miles south-west of Sidmouth. Here they rest on the red Permian breccias, and are reduced to a thickness of 70 or 80 feet. Passing now to the main outcrop of the Selbornian from Wiltshire northward, we may briefly indicate the lithological variations which it exhibits in that direction. In Berkshire and Oxfordshire the facies is similar to that of Hampshire, but the total thickness is greater, being about 300 feet. The zone of Sch. rostrata is partly represented by marls and partly by malmstones, and the bed of unfossiliferous greensand at the top is a continuous band 15 or 20 feet thick. But both the greensand and the malm- stone die out in Buckingham, leaving a purely argillaceous facies like that of Folkestone, the Lower Gault being about 150 feet thick and the Upper about 80 feet, a notable fact being the occurrence of a layer of phosphatic nodules at the base of the Upper Gault, embedded in glauconitic clay and lying on an eroded surface of the lower clays. Traced to the north-east both subdivisions become thinner and the Upper Gault has also suffered erosion, with the result that a layer of nodules and fossils derived from it occurs at the base of the Lower Chalk. Near Cambridge the Gault is from 140 to 120 feet thick, including a variable thickness of the rostrata zone, but at Soham, where it passes below the level of the Fenland, it is only 90 feet. When it emerges at Stoke Ferry in Norfolk it is still 60 feet thick, but at Eoydon and Grimston it is less than 20 feet and consists of red and grey marls with two beds of grey limestone. 14 Farther north it passes into a red limestone, which is generally called the " Eed Chalk " and is so strikingly exposed in the cliffs at Hunstanton. This red rock is only 3^ feet thick, and consists of three layers, each passing into the other. The lowest is sandy and of a deep brick-red colour full of Bel. minimus and Terebratula Uplicata ; the central part is a red nodular limestone with the same fossils and many others, including Hoplites lautus, H. splendens, H. tuberculatus, Inoceramus sukatus, and Inoc. concentricus. THE CRETACEOUS SYSTEM 501 The highest layer is a pale-pink limestone with few fossils. Thus the Red Chalk appears to be a condensed representative of the whole Selbornian stage. It appears again near Welton and Willoughby in Lincolnshire, where it is 10 to 12 feet thick, but thins again to 4 or 5 near Caistor. In South Yorkshire the beds are about 7 feet thick, thinning to the north-east, till at Leavening and Wharram it is less than 2 feet, its base containing pebbles of quartz and ironstone. Eastward, however, it expands, and at Speeton it is represented by over 30 feet of material as below : Feet. Smooth and firm red marl ....... about 4 Red marl, enclosing lumps of harder and more calcareous material, with many fossils . . . . . . ,,16 Tough red marly chalk in irregular beds . . . . ,,10 Red, yellow, and green clays, shaly and gritty . . . ,-, l The lowest beds contain BeL minimus and Inoceramus sulcatus, the higher hold Terebratula biplicata, T. capillata, Kingena lima, etc. Lower Chalk. This stage as developed in the south of England is divisible into two parts or zones, the lower part forming the zone of Schlanbachia varians, and the higher part that of Holaster subglobosus. At the base, and included in the range of Sch. varians, is the sub-zone of Stauronema Carteri, consisting of sandy glauconitic chalk which is often called the " Chloritic marl." At the top is a band of soft grey marl which is the sub-zone of Actinocamax plenus, often called Belemnite marl from the presence of that Belemiioid. At Folkestone the beds composing the Lower Chalk are : 15 Feet. Zone of Hoi. ( 7. Belemnite marl (with Act. plenus] ... 6 subglobosus, -i 6. Whitish chalk in massive beds (few fossils) . . 60 126 feet (5. Grey chalk in massive beds ..... 60 (4. Grey marly chalk ....... 20 3. Alternating beds of grey marl and hard grey chalk with many fossils ...... 22 -s 68 feet 2. Grey chalk with reefs of sponges (Plocoscyphia] . 10 1 1. Soft green glauconitic chalk (Stauronema) . . 16 194 The Lower Chalk attains its greatest thickness in Wiltshire, where it is about 250 feet, the greater part belonging to the zone of Sch. varians, which includes beds of a peculiar siliceous chalk containing imperfect flints. In the Isle of Wight the best exposure of the Lower Chalk is in the Culver Cliffs north of Sandown (see map, Fig. 172), where 1 1 1 II I r THE CRETACEOUS SYSTEM 503 the total thickness is 2 1 2 feet, the zone of Hoi. subglobosus with the Belemnite marl being 92 feet and the zone of Sch. varians about 120 feet. The basement bed or Chloritic marl is from 6 to 8 feet thick along the Undercliff (Ventnor, etc.), and abounds in phosphatised casts of Schlwnbachia varians, Acanih. Mantelli, Turrilites Morrisi, and Cucullcea mailleana, together with shells of Pecten asper, Rhynchonella grasiana, and Stauronema Garteri. In Dorset there is a more decided break between the Selbornian and the Lower Chalk ; the basement bed is a glauconitic chalk with phosphatic nodules and fossils, but it does not seem to be the equivalent of the Chloritic marl, for it does not contain Stauronema, while it has other fossils which in the Isle of Wight are common from 10 to 20 feet above the Chloritic marl, such as Scaphites cequalis, Acanthoceras rotomagensis, Turrilites Wiesti, and Holaster subglobosus. Other common fossils are Cucullwa mailleana, Uni- cardium ringmeriense, and Galerites castanea. The thickness of the Lower Chalk is also less (140 feet near Swanage), diminishing west- ward till it is only about 60 feet in West Dorset. The Dorset facies of the Lower Chalk can be traced inland by Bridport, Beaminster, and Crewkerne to Chard in Somerset, where it is still about 60 feet thick with a highly fossiliferous bed at the base ; but when it reappears on the coast near Lyme and Seaton it has undergone such a change that it ceases to be chalk, and is reduced to a thin band of calcareous sandstone enclosing large grains of quartz. Near Lyme this bed is only 3 or 4 feet thick, but at Beer Head it swells out into two beds which have a combined thickness of 18 feet and are highly fossiliferous. 16 The fossils in these calcareous sandstones show that they represent the Chalk marl of more eastern counties, the commonest being Sch. varians, Acanth. Mantelli, Turr. costatus, Scaphites cequalis, Holaster subglobosus, Hoi. Icevis, and Galerites castanea, while species which connect them with the Cenomanian of France are Trigonia vicaryana, Pecten asper, P. puzosianus, P. subacutus, Lima tecta, Terebratula tornacensis, T. capillata, Rhynch. dimidiata, Gottaldia Benettice, and Gatopygus columbarius. The sponges Elasmostoma and Trematocystia are not uncommon. Passing to the Midland counties, we find a continuous outcrop of Lower Chalk in the lower part of the Chalk escarpment which runs from Wiltshire to Norfolk and is broken only by the valley of the Thames. North of this valley through Berks, Bucks, Bedford, Cambridge, and Suffolk the Lower Chalk has the following composition : 504 STRATIGRAPHICAL GEOLOGY Feet. 5. Soft grey shaly marl enclosing a band of bard white chalk, Actinocamax plenus . . . . . . 3 to 5 4. Tough blocky chalk without distinct bedding, Holaster subglobosus, Acanth. rotomagensis . . . . . 60 to 80 3. Totternhoe stone, hard grey sand chalk, Rhynch. Man- telliana, Lima echinata, Kingena lima, Pecten jissicosta 2 to 20 2. Soft marls with occasional layers of hard marly rock in Oxford and Bucks, Schlcenb. varians, Rhynch. Martini, Inoceramus latus . . . . . . . . 60 to 80 1. Marl with green grains, passing northward into the nodule bed known as Cambridge Greensand . . . . 1 to 2 The Cambridge Greensand (see Fig. 162) commences near Barton-le-Clay (north of Luton), to which locality the nodule bed in the Gault has been traced, and there can be little doubt that all the nodules and the phosphatised fossils which occur in the Cambridge Greensand have been derived from the erosion of the Upper Gault. This derived fauna comprises some 210 species of Invertebrata, and about 80 of these have not yet been recog- nised elsewhere in England; out of the remaining 130 no less than 114 occur in the Gault of Folkestone, nearly all the commoner fossils occurring in the Upper Gault. 17 The commonest species are Sch. rostrata, Hoplites auritus, H. raulinianus, Avicula gryphceoides, Terebratula biplicata, Plicatula gurgitis, and Rhyn- chonella sulcata, but it is noteworthy that the last is a very rare fossil in the Gault elsewhere. The Vertebrate fauna of the Cambridge Greensand is very remarkable, the remains of many species of Ichthyosaurs, Plesio- saurs, Dinosaurs, Chelonians, and Pterodactyles having been found, together with bones belonging to a bird (Enaliornis}. Most of these bones have doubtless been washed out of the Gault, but some may be Chalk Marl forms. The indigenous fauna of the Cambridge Greensand is found in the marly matrix of the bed, and some of the species are Vermi- cularia umbonata, Discoidea subucula, Micrabacia coronula, Ehyn- chonella lineolata, Kingena lima, and Terebratulina triangularis, most of which occur also in the Chalk Marl above. Another speciality of this district is the Totternhoe stone, so named from a place near Dunstable, where it is 20 feet thick, and is quarried for building purposes. It varies greatly, however, in thickness, and is sometimes represented by two beds of stone, with Chalk Marl between them. At its base there is usually a layer of green-coated phosphatic nodules, and small fragments of phosphate are common in the stone above. In Norfolk the Lower Chalk becomes greatly reduced in thickness and at the same time harder and more purely calcareous, both these changes being evidently due to a diminution in the THE CRETACEOUS SYSTEM 505 quantity of intermingled argillaceous matter. Thus at Heacham and Hunstanton the stage has the following composition : 18 Feet. Zone of [5. Hard whitish chalk in rather thin beds ... 35 Offaster -j 4. Rough dark-grey chalk with a layer of green-coated sphcericus ( nodules at the base (Totternhoe stone) . . 2 7 f f 3. Hard creamy white chalk ..... 13 2. Hard grey rough and shelly chalk with a layer of reen _ coate( i nodules at the base ... 4 nodular ii mes tone 1 The hard shelly chalk (No. 2) is often called the " Inoceramus Bed " because it is crowded with fragments of Inoceramus shells, probably of Inoc. latus. It also contains Holaster subglobosus and Acanth. rotomagensis ; but the former here ranges throughout the stage, so that it is convenient to take another urchin as index for the upper zone, and for this purpose Offaster sphcericus, common in the upper beds of Lincolnshire and Yorkshire, seems best. In Lincolnshire the only noticeable variations are (1) the red and pink coloration of two beds in the zone of Offaster sphcericus near Louth ; (2) the reappearance of the Belemnite marl, which is absent near Hunstanton, but is often 2 feet thick in Lincolnshire, and is sometimes stained dark red or purple. The Totternhoe stone is a constant bed about 2 feet thick throughout the county. 19 In Yorkshire the Lower Chalk maintains the Lincolnshire facies along the greater part of the escarpment, but at Speeton it shows some changes. The zone of Sch. varians expands to 77 feet, and the lower part consists of nodular chalk in alternating pink and white bands. The zone of Offaster sphcericus is 46 feet thick, and the lowest bed is in places stained pink. At the top the Belemnite marls are well marked, enclosing a medial bed of hard smooth white chalk. Middle Chalk. At the base of this stage there is always a certain thickness of hard nodular chalk which forms a great con- trast to the soft marl on which it rests ; this rocky chalk is called the Melbourn rock from a village in Cambridgeshire. Commencing as before in Kent, we find a good section of this stage in the cliffs near Dover, which show : Feet. Z n f f Rather rough lumpy white chalk .... 20 T a n i *,! I Smooth white chalk with thin layers of grey marl lata \ and several courses of flints .... 90 t White chalk with seams of marl but no flints . 60 Zone of /Hard chalk with frequent nodular layers . . 38 Rh. Cuvieri \ Hard and rough nodular chalk .... 32 ~240 506 STRATIGKAPHICAL GEOLOGY The fossils in the lower zone are Bhynchonella Cuvieri, Inoceramus mytiloides, Cardiaster pygmceus, Galerites subrotundus, Discoidea Dixoni, and its variety minima. In the upper zone are Terebratu- lina lata, Inoceramus Cuvieri, and Micraster corbovis ; here too Spondylus spinosus makes its first appearance, and Holaster planus comes in near the top. These beds can be traced all round the Wealden area, and another fine exposure of them can be seen in the cliffs of Beachy Head, where, however, there are no flints in the zone of Ter. lata. In the Isle of Wight, Culver Cliff and quarries near Yarbridge exhibit the following succession : Feet. /"Rough chalk with a seam of black clay in the Zone of I middle and a layer of green-coated nodules near Ter. lata 1 the base 19 1, Thick-bedded white chalk with seams of marl . 90 2 f (Hard chalk, becoming nodular below and containing 7?}> n ' \ many broken Inoceramus mytiloides ... 76 1 (.Rough nodular chalk with marly veins ... 8 193 Westward in Compton Bay the thickness is reduced to 150 feet. In Dorset, at Ballard Cliff, it is only 122 feet, but the same layer of green-coated nodules occurs about 20 feet from the top and has been mistaken for Chalk rock. At White Nothe Mr. A. W. Rowe estimates its thickness at 134 feet. 20 In Devon the Middle Chalk is well exposed near Beer, and is about 100 feet thick, the zone of Rh. Cuvieri here including some thickness of hard shelly chalk which is quarried for building purposes, and is known as Beer stone. In the Midland counties this division exhibits the same litho- logical and palceontological characters as in the south of England. In Berkshire it is not more than 150 feet, but north of the Thames it rapidly thickens to about 220 feet, and maintains this thickness into Suffolk. The Melbourn rock at the base is from 8 to 10 feet thick, and consists of hard nodular chalk ; it passes up into bedded chalk, in which Inoceramus mytiloides and Rhynchonella Cuvieri are generally abundant. The higher zone consists of soft white chalk with layers of grey shaly marl and occasional nodules of flint, which latter become more numerous towards the north, till in Cambridge and Suffolk flints are frequent in this zone and are often of elongate finger-like form. Fossils are seldom abundant, but Galerites subrotundus and Terebratulina lata occur, with Spondylus spinosus, Holaster planus, and Micraster corbovis in the higher part. THE CRETACEOUS SYSTEM 507 In the north-west of Norfolk the Middle Chalk appears to be only about 100 feet thick. The Melbourn rock continues at the base of the zone of Rhynchonella Cuvieri, but the beds above are less nodular than they are farther south, and the zone of Terebratu- lina lata consists of hard white chalk with many layers of flints. In Lincolnshire there is no definite Melbourn rock, and the zone of Ehynck. Cuvieri is reduced to a thickness of 10 or 15 feet, con- sisting of yellowish -grey chalk full of fragments of Inoceramus shell and containing Inoc. mytiloides, Rhynch. Cuvieri, and Terebratula semiglobosa. At the top is a layer of grey shale, and the succeeding zone of Ter. lata consists of firm white chalk with frequent layers of grey flints. Inoceramus Guvieri is the only common fossil, and the thickness of the zone may be 80 to 100 feet. In South Yorkshire both zones are exposed in some large quarries near Hessle Station and have yielded a larger number of fossils, but little is known of the Middle Chalk in its course through Yorkshire except that the whole of it becomes very hard. Near Speeton, on the coast, the Middle Chalk is exposed in a fine range of cliffs and has been described by Mr. A. W. Eowe. 21 The zone of Rh. Cuvieri is, as usual, in the north very thin, being less than 1 2 feet, but the overlying Ter. lata zone has expanded to a thickness of 210 feet, that fossil with small Rhynch. Cuvieri, Inoceramus Srongniarti, and Holaster planus being common throughout the whole of it. Upper Chalk. The chalk of this division has been divided into seven zones (see p. 491), but the higher zones are not every- where present, having been removed from large areas in the south of England during the upheaval which preceded the formation of the Eocene deposits. These seven zones may be grouped into two sub-stages, the lower of which, including the zones of Hoi. planus, Micraster cortestudinarium, and M. coranguinum, may be termed the Micraster chalk, while the higher zones may be called the Belemnite chalk until more definite names are proposed. In Kent the Upper Chalk forms the cliffs from St. Margaret's Bay near Dover to Walmer and Deal, and again from Kamsgate round the Isle of Thanet to Margate and Eeculvers, but the highest zone found in the county is that of Marsupites. Mr. W. Hill finds a convenient base for the zone of Holaster planus in a bed which is full of scattered flints, and in which Echinocorys vulgaris makes its appearance. 22 Measured from this the thickness of the zone is about 48 feet, and it consists of a succession of beds of rough lumpy chalk, the lumps being very hard but embedded in a softer matrix, and the top is marked by a coarse hard nodular rockyfchalk. In these beds Micraster Leskei, M. precursor, and Hoi. planus are common. 508 STRATIGRAPHICAL GEOLOGY The complete succession of the zones exposed between Dover and Margate in descending order is as follows : Feet. Soft chalk in two sub-zones, the upper characterised by Marsupites and Uintacrinus, the lower by Actinocamax granulatvs, and A. verus also occur . ........ 116 Zone of Micr. coranguinum. Smooth chalk with many layers of black flints ; Cidaris sceptrifera, C. perornata, Galeritcs conicus, and Cyphosoma Kcenigi . . . . . .250 Zone of Micr. cortestudinarium. Rough nodular chalk with many flints ; Micraster precursor and Hoi. placenta . . 75 Zone of Holaster planus (as above described) . . . . 45 486 On the coast of Sussex, however, higher zones come in. The Marsupite zone has been identified at Seaford Head, and above it is 170 feet of chalk containing Offaster pilula, Actinocamax granulatus, Ccelosmilia laxa, and other fossils of the quadratus zone. This chalk extends westward from Newhaven to Brighton, and thence to the valley of the Arim, where a succession of quarries between Houghton and Arundel shows parts of all the zones above mentioned (see Fig. 173). Still farther west chalk containing Belemnitella mucronata has been found, and this highest zone underlies the Eocene of Hampshire and is well exposed in quarries on Portsdown Hill, where an inlier of chalk is brought up within the Tertiary area by an anticlinal flexure. The total thickness of the Upper Chalk in West Sussex and Hampshire must be from 800 to 900 feet. In the Isle of Wight the Upper Chalk forms a long and narrow outcrop across the centre of the island in consequence of the steep dip imparted to it by the flexure which carries them below the Eocene of the Hampshire basin (see Fig. 172). Its thickness is from 1200 to 1300 feet, and the thicknesses of the several zones at the western end of the outcrop, as recently determined by Mr. Eowe's measurements, are as follows : 23 Feet. Zone of Bel. mucronata. White chalk with flints, Echinocorys (pyramidal form), Cidaris serrata, Magas pumilus . . . 475 Zone of Act. quadratus. White chalk with some marl seams but few fossils ./........ 342 Zone of Marsupites. Compact white chalk with flints ... 80 Zone of Micraster coranguinum. Chalk with many layers of flint nodules and the usual fossils . . . . . . .310 Zone of Micraster cortestudinarium. Rough nodular chalk with flints '53 Zone of Holaster planus. Hard greyish nodular chalk . . 60 1320 LU it 1 TT I i 1 I I. 1 2 2 J! g I 1 JS c8 o; ^ 1| eo e-5 !| ii 510 STKATIGRAPHICAL GEOLOGY The same zones with a total thickness of about 1000 feet are found in Dorset, where Mr. Howe has recently identified the zone of Marsupites and has been able to separate the zone of Act. quadratus from that of Bel. mucronata, assigning a thickness of 354 feet to the former and not less than 250 to the latter, 24 the characteristic fossils of the one being Actinocamax granulatus, Act. quadratus, and Offaster pilula, while those of the higher zone are Belemnitella mucronata, B. lanceolata, Cardiaster ananchytis, Magas pumilus, Crania costata, and Rhynchonella limbata. In Wiltshire and all along the inland outcrop from Salisbury Plain to Suffolk the Micraster zones are exposed in numerous chalk pits and the higher zones are all present in South Wiltshire, but north of the Vale of Pewsey the zone of Bel. mucronata does not occur, and that of Act. quadratus is gradually cut out by the pre- Tertiary planation, till the Eocene comes to lie on the Marsupites zone. In all these counties the lower part of the Hoi. planus zone consists of several beds of very hard compact chalky limestone, generally of a creamy colour, and each bed having a layer of calcareo-phosphatic nodules at the top. This limestone is known as " The Chalk Kock " ; above it there is from 15 to 20 feet of greyish nodular chalk consisting of hard lumps embedded in a soft matrix. The total thickness of the zone seldom exceeds 26 feet, so that it may be regarded as a condensed equivalent of the more expanded zone to the south. The Chalk rock frequently contains grains of glauconite and is often full of fossils, among which Gastropods are noticeable, for these are rarely found either in Middle or Upper Chalk. The most characteristic fossils are Pachydiscus peramplus, Prionocyclus Neptuni, Scaphites Geinitzi, Heteroceras reussianum, Turbo Geinitzi, and Solariella gemmata. Holaster planus, Micraster prcecursor, and M. Leskei are also common. The zones of M. cortestudinarium and M. coranguinum have the same characters and fossils as in the South Coast sections, as have also the higher zones as far as they can be traced, except in the case of the Marsupites zone which in Berkshire and the Thames Valley presents some peculiar features ; for it contains two beds of brown phosphatic chalk, one at the base of the Uintacrinus band and one at the base of the upper division. These beds are well exposed in the large quarry at Taplow near Maidenhead, where the upper bed is 8 feet thick. The phosphatic matter is in the form of grains, some of which are casts of Foraminifera and others are phosphatised portions of Inoceramus shells. In Suffolk and Norfolk the higher zones come in again from THE CRETACEOUS SYSTEM 511 beneath the Eocene and include the highest zone found in England, that of Ostrea lunata. A change also takes place in the zone of Holaster planus, for in Norfolk it passes into hard white chalk with dark-grey flints, without any rocky or nodular beds. The following is a tabular view of the Upper Chalk of Norfolk, but the thicknesses of the zones are only approximate : Feet. Zone of Ostrea lunata. White with beds and patches of grey chalk. Ter. gracilis, T/iecidium vermiculare, Trigonosemus elegans, and many Bryozoa . . . . . . . . .110 Zone of Bel. mucronata. Soft white chalk with many flints, some very large and cup-shaped ..... probably 250 Zones of Act. quadratus and Marsupites. Not yet separated from the zones above and below, Act. granulatus common . probably 400 Zones of M. coranguinum and M. cortestudinarium. Chalk with many flints, both nodules and courses ..... 340 Zone of Hoi. planus (as above described) . . . . . .50 1150 The Upper Chalk of Yorkshire, as exhibited in the cliffs at and near Flamborough Head, has been the subject of special study first by Mr. Lamplugh and more recently by Mr. Eowe, 25 from whose account the following resume is taken : Feet. Zone of Act. quadratus. Chalk without flints, partly seen in the cliffs near Sewerby and partly inland ..... 330 Zone of Marsupites, in two bands, without flints .... 208 Zone of M. coranguinum, the upper part flintless, the lower with many flints and marl seams ....... 260 Zone of M. cortestudinarium. Hard chalk with many flints . .120 Zone of Hoi. planus. Hard chalk with grey flints, both nodules and in continuous courses ....... 125 1043 With regard to its fossils the Chalk of Yorkshire differs con- siderably from that of Southern and Central England. Thus, Holaster planus being just as common in the zone of Ter. lata as in its own zone, one has to rely on its association with Echinocorys vulgaris for the recognition of the latter. The zone of M. cortestu- dinarium yields the usual assemblage, but in the overlying zone M. coranguinum is a rare fossil and its place is taken by Infulaster rostratus. Again no specimen of Actinocamax quadratus has been found in Yorkshire, and, 0/aster pilula being also rare, the zone finds a local index in Inoceramus lobatus ( = lingua auct., non Goldf.). It is also characterised by the abundance of fossil sponges belonging chiefly to the genera Bolospongia, Scytalia, Seliscothon, Siphonia, Heterostinia, Pachastrella, Phymatella and Verruculina, most of which do not occur in the south of England. 512 STRATIGRAPHICAL GEOLOGY 2. Ireland and Scotland Ireland. Upper Cretaceous rocks appear nearly everywhere round the edge of the great basaltic plateau of Antrim, Tyrone, and Londonderry, and we may therefore safely infer that they underlie the greater portion of it. It is also clear that they once extended much farther westward than the present edge of that plateau, as is proved by a small outlier near Draperstown. They usually rest 011 a planed down surface of Liassic or Triassic rocks, but sometimes overstep the limits of these, as in North-east Antrim, and rest upon the Palaeozoic rocks (see Fig. 186). These Irish Cretaceous rocks were described in 1865 by Pro- fessor Tate, and an attempt to establish a zonal classification was made by Dr. W. F. Hume in 1897, as shown in the following table : 26 Tate's Divisions. Hume's Zones. Feet. {( Zone of Belemnitella mucronata\ - Q Q 5. White limestone -I , , Actinocamax quadratus / [ ,, Actinocamax verus . . 3 to 5 4. Chloritic Chalk = ,, Echinocorys gibbus . . 3 to 4 {3. Chloritic sands and/Inoceramus zone (Upper Chalk) . 6 to 16 sandstones (Zone of Exogyra columba . . . 4 to 15 2. Yellow sandstones = , , Ostrea carinata . . . 4 to 30 1. Glauconitic sands = ,, Exogyra conica . . . 4 to 16 The glauconitic sands are regarded by Dr. Hume as in part equivalent to the Selbornian zone of Sch. rostratus, but the occur- rence of Sch. varians, Pecten asper, and an Actinocamax like verus (probably lanceolatus, Sow.) makes it difficult to accept this view. The yellow sandstones contain layers of chert, and their fauna is a curious mixture of Selbornian and Cenomanian forms, but as the beds are shallow -water deposits it is probable that Selbornian species here survived into Cenomanian time, and the occurrence of Acanth. rotomagensis marks them as equivalents of Lower Chalk. The so-called Chloritic sands appear to include two deposits of very different age, separated by a break and unconformity. The lowest beds are calcareous glauconitic sandstones with Exogyra columba, Pecten asper, and Trigonia crenulata, and are evidently of late Cenomanian age. The higher glauconitic sandstones, as seen north of Belfast and on the eastern coast, contain Spondylus spinosus, Galerites conicus, Rhynchonella limbata, Eh. plicatilis, and a species of Micraster with many broken Inocerami. These beds pass up into a glauconitic sandstone, often pink in colour, which yields Echinocorys vulgaris, var. gibbus, and a few other fossils. If the fossils are correctly identified these zones of Sp. spinosus and Ech. gibbus must be regarded as a condensed representative of THE CRETACEOUS SYSTEM 513 the Micraster zones of the English Chalk ; the whole of the Middle Chalk and perhaps that of Hoi. planus also being absent. Above the " Chloritic chalk " is the " White Limestone," which has a bed of nodular chalk at its base containing Act. verus and many sponges, and in the north of Antrim a similar nodular bed is succeeded by 5 feet of flintless chalk containing plates of Marsupites. Finally in conformable succession is white chalk with many flints, about 100 feet thick, having Act. quadratus near the base and Bel. mucronata in the higher part. Scotland. Cretaceous rocks are found again especially on the west coast of Scotland, in the island of Mull and in the district of Morvern on the adjacent mainland ; as in Ireland, they belong exclusively to the upper division of the system, and though they can be correlated generally with the Irish Series, they exhibit a still more abnormal facies and a more decided approach to littoral conditions. They are best exposed in Morvern, along the shores of Loch Aline, and beneath the singular outlying masses of Beinn-y- Hun and Beinn-y-Hattan. 27 The general succession here is : Feet. 4. Sandstones and white marls with plant remains (high Cretaceous or early Eocene) ...... 20 3. White indurated chalk with bands of flints, Belemnitella mucronata, and fragments of Inoceramus ; some beds of glauconitic chalk at the base ...... 10 2. White sandstones without fossils, but containing a thin seam of coal 30 to 100 1. Glauconite sands, passing sometimes into dark-green argil- laceous sand, sometimes into calcareous sandstone ; Pecten asper, P. orbicularis, Exogyra covtica, but no Ammonites 20 to 60 On comparing these beds with those of Antrim we may safely regard No. 1 as equivalent to Nos. 1 and 2 of the Irish sequence. The overlying white sandstones (No. 2) would appear to be homotaxial with the zone of Ostrea columba in Ireland, whatever the precise age of that may be. Above it there is a palaeontological break, though apparently no unconformity, and the hard white chalk corresponds closely to that of Ireland, except that there is no sign of the Micraster zone at its base. D. CONTINENTAL REPRESENTATIVES 1. Northern France In France the Cretaceous exhibits two very different facies, that of the southern provinces being very different from that of the northern region, which is often called the Paris basin. The deposits found in the latter closely resemble those of England, and 2L 514 STRATIGRAPHICAL GEOLOGY were evidently formed in the same sea, while those of Southern France belonged to a different marine province. Those of the Paris basin may first claim attention, and the following is a brief notice of them (see table on p. 491). Albian. The name of this stage is taken from the river Aube (or Alba), and in the department of that name it consists of the following members : Feet. Marls with Schlcenbachia rostrata and Ostrea vesiculosa . up to 200 Clays with Hoplites splendens and H. auritus . . . ., 30 Clays with Hoplites interruptus and Douvilleiccras mammillatum ,, 60 Green sands without fossils . ' .. % . ; . '_ ,, 30 Traced northward into the Departments of the Meuse and Ardenne the basal sands contain D. mammillatum, and pass into a siliceous rock (the " gaize de Draize "). The marls also pass into similar rock (the "gaize de 1'Argonne "), which closely resembles our malmstone, and is undoubtedly of the same age. Still farther north, near Mons, the zone of Sch. rostrata overlaps the lower beds, and is represented by sands with layers and nodules of chert, the fauna resembling that of the Blackdown sands in Devon. These Albian deposits pass completely under the Paris basin. They have been reached by deep borings at Paris and Dieppe. They appear again in the anticlinal uplift of the Pays de Bray, and in diminished thickness they are found in the cliffs of Normandy. Near Cap la Heve the clays are about 10 feet, and the gaize with ttchlwnbachia rostrata is about 26 feet thick. Oenomanian. The district where this stage was first estab- lished, that of Le Mans in the Sarthe, is unfortunately one where the beds have a sandy and shallow-water facies, and are difficult to compare with the chalky facies which prevails in the eastern part of the Paris basin as it does in England. In Normandy, however, there is an intermediate type which affords valuable assistance in the correlation, and a fine section of it is visible in the cliffs near Cap la Heve : - 8 Feet. 4. Greyish- white chalk with grey flints .... about 30 3. Grey chalk with cherts and phosphatic nodules . . , , 10 2. Yellowish-grey silty and glauconitic chalk with many layers of grey and black cherts, and some beds of bluish marly chalk near the bottom ....... ,,84 1. Bluish-grey sandy glauconitic marl with black phosphatic nodules and many fossils ...... ,, 6 The basement bed contains Stauronema Carteri and Schlwnbachia varians, and is the equivalent of our Chloritic marl. The beds above contain Schlcenbachia varians, Acanthoceras Mantelli, THE CRETACEOUS SYSTEM 515 Turrilites costatus, Pecten asper, P. Galliennei, and Holaster sub- globosus. Most of these also range into No. 3, but the whitish chalk above has few fossils. Turonian. The name of this stage is taken from Touraine, where it is divisible into two sub-stages called (1) the Ligerian and (2) the Angoumian. (Calcareous sands and sandy limestones with Callianassa, Ostrea columba, and Acanthoceras Deverice, 30 feet. Yellow sandy chalk with Acanthoceras deverioides. , Micaceous chalk with Prionotropis Woolgari, Pachydiscus I peramplus, and Mammites Revellieri (=Jiochbrunei), 60 to Ligerian -j 100 feet. I Marly chalk with RhynGh. Cuvieri, Inoceramus labiatus, and ^ Galerites subrotundus, 30 feet. Traced northward into Normandy these beds pass into pure chalk, and near Rouen the facies is the same as in England with a thickness of about 280 feet. At the 'base is hard nodular chalk like our Melbourn rock, succeeded by chalk with Mammites nodosoides, Prionotropis Woolgari, and Galerites subrotundus, and this by soft chalk with Terebratulina lata, the higher parts of which contain Holaster planus and Micraster Leskei. Senonian. For this stage I retain d'Orbigny's original name instead of the awkward one of Emscherian which has been adopted by de Lapparent or the Corbierian of de Grossouvre. The original name should not be dropped, and should be specially applied to that portion of the Chalk which is typically developed round the town of Sens and in the country of the ancient Senones. This comprises the Micraster zones, i.e. those of M. decipiens (or M. cortestudinarium}, M. coranguinum, and of Marsupites (or Micraster turonensis). In the Departments of the Yonne, Aube, and Marne, this stage consists entirely of white chalk, in which flints are often but not universally numerous, and it has a thickness of from 400 to 500 feet. Northward in the Aisne and in Picardie it passes into grey and dolomitic chalks with a less thickness, and on the western side of the Paris basin it is represented by sandy marls, hard limestones, and fine siliceous deposits (" tuffeaux "), the full thickness of which at Villedieu and Saumur is little more than 100 feet. According to M. de Grossouvre the " craie de Villedieu," containing Spondylus truncatus, Micraster turonensis, with species of Mortoniceras and Peroniceras, is the equivalent of the Marsupites zone. Campanian. This stage is typically developed in the district of Champagne, whence it takes its name, and where it has been divided into four local zones. It comprises the broader zones of Actinocamax quadratus and Belemnitella mucronata, which have a 516 STKATIGRAPHICAL GEOLOGY total thickness of about 250 feet, and its highest portion is found in the chalk of Mendon near Paris. In the Campanian should be included a still higher zone, some- times called Maestrichtian, which is found in Belgium near Ciply and Maestricht and in a small isolated patch in the Cotentin (Normandy). At Ciply there is a continuous series of chalks up to the top of a phosphatic and glauconitic chalk in which the little Brachiopod Thecidea papillata is abundant. Resting unconformably on this is a conglomerate of rolled phosphate nodules and fossils which forms the base of the " Tuff'eau de St. Syrnphorien," a soft yellowish rock composed largely of Bryozoa and containing occasional grey flints. Its fossils include Belemnitella vnucronata, Baculites Faujasi, Ostrea larva ( = lunata}, Thecidea papillatum, and Terebratula carnea. At Maestricht in Limbourg the series is carried a step higher, the tuffeau with flints being overlain by another set of limestones (120 feet thick), consisting of alternating soft and hard beds. These also yield Bel, mucronata and Ostrea larva, with Hemipneustes striatoradiatus, some Gastropoda, and remains of the marine reptile Mosasaurus Camperi. Some writers include the Maestrichtian in the Danian, but it is better to restrict the latter to deposits which are destitute of Ammonites and Belemnites. Thus restricted there is no Danian in the north of France or Belgium. 2. Denmark and Sweden The Upper Cretaceous deposits of these two countries are of special interest, because they carry the series continuously up to a higher horizon than is found in England or France, and include beds which seem to bridge the interval between Cretaceous and Eocene. Those of Scania also are transgressive and indicate the gradual northward extension of the Cretaceous Sea, which did not begin to encroach on Scandinavian land till the close of the M. coranguinum zone. In fact, nothing older than this zone is exposed in either country. The Danish Beds are seen at Faxe and Saltholm, where the succession is : 5. Chalky limestone with flints, Nautilus danicus, Holaster faxensis> Ananchytes sulcatus, Terebratula carnea. 4. Limestone with Cerithiumfaxense and Baculites vertebralis. 3. Clay with fish remains and Ostrea lateralis. 2. White chalk with Belemnitella mucronata. 1. Chalky limestone with Actinocamax granulatus. No. 2 is regarded as the top of the Campanian, and the higher beds as a separate stage under the name of Danian. THE CRETACEOUS SYSTEM 517 In Sweden the most complete succession is found near Malmo and in the basin of Ystad, the highest beds being found at the former and the lower at the latter locality. The total thickness appears to be more than 1500 feet, and the combined succession is as follows : s [Grey and white chalks without Belemnites or Ammonites, but con- taining Nautilus danicus, Ananchites sulcatus, two species of Cyprcea and one of Pleurotoma. White chalk with flints of Malmo, and sandy glauconitic marls of Kopinge with Belemnitella mucronata and Magas pumilus. Granular limestone with Actinocamax mammillatus. Marl with Act. quadratus, passing into a conglomerate at Tosterup with Ostrea vesicularis, 0. semiplana, and Inoc. latus. 3 Marly limestones with Actinocamax granulatus and A. verus with Inoceramus lingua and Scaphites binodosus. 5 I Marls with Actinocamax westphalicus. To the northward near Christianstad the Senoniaii is absent, and the quadratus beds rest directly on the Archaean platform with a basal conglomerate. By some authors the Danian is regarded as equivalent to the Montian of Belgium (see Chap, xv.), but its fauna is different and more closely allied to the Cretaceous. 3. The Mediterranean Region A different facies is found in Southern France, Spain, and through the whole of Southern Europe from the Alps and Italy to the shores of the Black Sea. The Albian of this region consists mainly of sands and sandstones, the Cenomanian of sandy limestones containing Orbitolina (Patellina) concava and many species of Caprina Caprotina and Caprinula. The Ttironian and Senonian are also represented by sandstones, sandy limestones, and marls abounding with other genera of irregular Lamellibranchs, such as Hippurites, Radiolites, and Sphcerulites, some of the beds being known as Hippurite limestones. These two divisions reach a thickness of 1500 feet in some places. The higher stages are more local, but often comprise a great thick- ness of strata ; thus in Provence they include some marly lime- stones of Campanian age, succeeded by a vast group of freshwater deposits (the Garumnian), having a thickness of 2500 feet and yielding a large number of freshwater and terrestrial shells. In character and thickness this group is analogous to the English Weal den. The most complete succession of marine deposits in the south of 518 STIIATIGKAPHICAL GEOLOGY France is that of Aquitaine, and may be given as an illustration of the lithological development of the Mediterranean facies. According to M. Coquand and M. de Grossouvre it is as follows : ' 29 H ( Limestones of Tercis, with Micraster tercensis, in two zones, separated by Miliolina limestones and lacustrine deposits, the whole about ^ I 600 feet thick. z fReddish- yellow limestone with Baculitcs anccps, Pachydiscus colligatus, Turrilites polyplocus, and Thecidea papillata. J White limestone with Bryozoa and Belemnitella mucronata. J" I White marly limestones with Actinocamax quadratics, Scaphites \ binodosus, Inoceramus latus, and Exogyrm. a f Sandy limestone Avith Spondylus truncatus, Hippurites, a,nd\fiadiolites. g I White and grey limestones with Mortoniceras texanum. g | Hard nodular limestones with Scaphites Meslei, S. Lamberti, J> I and Micraster turonensis. ^ fSoft limestones and marls with Sphcerulites. .2 I Soft beds with Radiolites, Hippurites, Sphwrulites, and Acantho- \ ceras Deverice. z I Nodular limestone with Prionotropis Woolgari. t* \ Marls with Ostrea columba. o I Sands, clays, and limestones with Ostrea, Caprina, Schlcenbachia < 1 varians, and Turrilites costatus. .= I Only found on the Pyrenean side of the basin, where it is repre- j sented by black marls and limestones. 4. Germany, Bohemia, and Silesia In Westphalia and Northern Germany there is an Upper Cretaceous Series comparable to that of England and the north of France, having Gault Clays and marls at the base, succeeded by Cenomanian marls and a thick mass of chalk which includes representatives of the Turonian, Senoniaii, and Campanian stages. Farther east, however, in Saxony, Bohemia, and Silesia, a different facies is found, where terrigenous sandy material prevails over that of organic origin, the chalky limestones of Western Europe being largely replaced by sandstones which the Germans call " Quaderstein," because they are much jointed, and split into square blocks. The following is the general succession in this area according to Schlonbach and Fritsch : Chlomek f Sandstones with Placenticeras Orbignyi, Peroniceras Beds ^ subtricarinatum, and P. loestphalicum. rClays with ironstone nodules, Placenticeras Orbignyi Priesen | and Scaphites Lamberti. Beds "I Marls with Pachydiscus peramplus, Prionotropis {, Neptuni, and Baculites bohemicus. j \ THE CRETACEOUS SYSTEM 519 Teplitz /Marls and limestones with Micrasters, Scaphites Beds \ Geinitzi, and Pachydiscus peramplus. /"Sandstones with Inoceramus Brongniarti. Malnutz I Glauconitic sandstones with Mammites nodosoides and Beds | Rhynchonella bohemica. vSandy limestones and marls with Spondylus spinosus. Weisemberg (Sandstones with Inoceramus labiatus, passing north- Beds \ west into marly limestones. ,,. f Sandstones and sandy limestones with Acanthoceras g f ^-OJTcan J rotomagense, Scaphites cequalis, and Baculites bacu- Perutz /Sandstones and shales with plant remains resting on Beds \ pre- Cretaceous rocks. The highest beds of this series represent the zone of Micraster coranguinum, nothing higher having been found in any part of the area. 5. Russia Strata of Upper Cretaceous age occupy large areas in Southern and Central Kussia, though the exposed portions are not of great extent, because so much is concealed beneath a covering of Tertiary Beds. The deposits are of great thickness, and must include repre- sentatives of the Cenomanian, Turonian, and Senonian stages, but though Greensands with Cenomanian fossils occur at the base, fossils are scarce in the overlying chalks and marls, so that no zones have been established ; moreover no definite Turonian division has been recognised, for there seems to be an intermingling of species which in Western Europe would be regarded as Turonian and Senonian. Useful information concerning the Eussian Chalk was collected in 1892 by Dr. W. F. Hume, 30 who describes the mass of the formation as consisting of grey and white chalks with a band of marly chalk or "Chalk Marl" between them. A boring at Kharkov has proved the great thickness of these beds, and the details given by Dr. Hume may be thus summarised : Feet. Tertiary deposits . . . . . .. .133 Soft white chalk without flints .... 967 Bluish and grey marly chalk ..... 391 Grey chalk ........ 312 White chalk with flints ...... 161 Grey and green sands and sandy clay . . 93 2057 The higher white chalks contain Bel. mucronata in abundance, and this species is said also to occur in the marly chalks. These deposits extend all across Southern Russia from Podolia and Volhynia on the west to Voronetz, Saratov, and Simbirsk on the 520 STEATIGRAPHICAL GEOLOGY east. From Poland and Silesia another band of chalk is believed to extend across the Moscow basin, but no deposits of Upper Cretaceous age have been found in Northern Russia, though it seems likely that they originally spread over those of the Lower Series. E. THE HISTORY OF THE PERIOD The history of Cretaceous time in Europe is the history of the gradual submergence of the large tracts of land which existed in the region at the beginning of the period. These tracts included (1) a large area in the north which may be called the Fenno- Scandian region ; (2) a western land including the British area, the whole of Western France, and the greater part of Spain ; this may be called the Atlantic region ; (3) a long tract extending from the north-east of France through Central Europe, Bohemia, and the southern part of Russia, which for want of a better name may be called the Hercynio-Sarmatian land. The Atlantic and Hercynian regions were connected by a broad isthmus across Belgium and the North Sea, and it is also possible that there was a land -connection between the Fenno-Scandian and the Sarmatian regions. There is fairly good evidence that some of this land rose to a considerable elevation above the sea-level of early Cretaceous time, because some parts of it were never entirely submerged even after the deposition of several thousand feet of strata and the formation of chalk below water which must have been over 3000 feet deep. We know also that there were both large rivers and large lakes 011 this continent, and we know the actual position of two of these lakes. The one, known as the Wealden lake, occupied a large part of Southern England with the adjacent parts of what is now the English Channel, The other occupied a large area in the north- west of Germany, and appears to have been connected with the estuary of a river which opened into a gulf of the Northern Sea. With regard to the Wealden lake, I think there is no doubt about its having been an enclosed inland lake with an excurrent river emerging from its southern side and running south-eastward through France. In spite of the views held by certain French geologists, there is no good reason for believing that the Southern Sea invaded the Pays de Bray or the Boulonnais before Aptian time. The few species of Lamellibranchs on which they rely for the presence of the Hauterivian Sea in the Pays de Bray are all found in the Vectian of the Isle of Wight. The northern termination of this sea appears to have been in the Haute Maine, where its characteristic fauna occurs in a limestone overlying a set of freshwater beds, some 60 feet thick. Such a conjunction seems THE CRETACEOUS SYSTEM 521 to indicate the estuary of a river, which was probably that flowing out of the Wealden lake. It is generally admitted that during the earlier part of this epoch, i.e. during Wealden and Neocomiaii times, land existed in the north-east of France and over Belgium and that it was con- nected with land over the east of England. It is my own opinion that this was a fairly broad tract of land enclosing the Wealden lake and connecting the western land with that of Central Europe. Further, I hold that this isthmus was never broken through till the Aptian epoch, when the Northern and Southern Seas met in a strait which passed obliquely across England from Wiltshire to Cambridge and Norfolk. We may now turn our attention to the general position and extensions of these Southern and Northern Seas. The former appears to have covered most of Eastern France (south of the Ardennes), and to have stretched eastward through Switzerland and Italy across the southern part of Austria and the Danubian provinces to the Black Sea and the Crimea. The Northern Sea occupied a large part of what is now the Arctic Ocean and extended southward in two great gulfs, one on each side of the Fenno- Scandian land. The western gulf touched the north of Scotland (Caithness and Aberdeen), and seems to have passed southward through the North Sea area into Yorkshire and Lincolnshire on the one hand and into Germany on the other. The eastern gulf occupied a large part of Northern Eussia, extending southward to the Moscow basin and south-east to the province of Simbirsk ; but near Moscow the sea was shallow, as the zone of Bel. Jasekowi is there represented by sands with plant remains but no marine fossils, so that they seem to be of brackish or freshwater origin. The question remains whether there was any water-connection between the extremities of the eastern and western gulf, or whether Fenno-Scandia was united to the Sarmatian land. So far as the similarity of the English and Eussian faunas is concerned, that both were derived from the Northern Ocean is sufficient explanation. The only evidence for the theory that the gulfs were otherwise connected is the existence of some deposits of Lower Cretaceous age at Brzezie on the borders of Posen and Poland, where fresh- water beds with Cyreiia and Cypris are succeeded by sands and conglomerates which are overlain by clays with Astarte and Exogyra Gouloni ; but these bivalves may be of any zonal age, and only indicate the line along which the sea eventually made its way across the space between Hanover and Moscow. Having regard to the presence of freshwater deposits of early 522 STRATIGRAPHICAL GEOLOGY THE CRETACEOUS SYSTEM 523 Cretaceous age in Hanover, Posen, and the Moscow area, it seems probable that Prussia and Poland were land surfaces at this time (see Fig. ] 7 5) and were not submerged till Aptian or Albian times. We know that it was in Aptian time that the subsidence carried the sea across the English isthmus, and it may well have been at the same epoch that a strait was formed across Prussia and that Fenno-Scandia was reduced to the condition of a large island. In Albian and Cenomanian times there was a more rapid subsidence, and the continental land was broken up into a number of islands. The Southern and Northern oceans were united across France and England, much of the Sarmatian land was submerged, and the central area was reduced to an island which, at any rate in Cenomanian time^ extended only from the Ardennes to Bohemia. Minor changes subsequently took place, with a temporary elevation of the Franco-British area, during the time of the Holaster planus chalk, but the renewed subsidence of Senonian time must have still farther reduced the land -areas till the only remaining large tract of land in the European region was that of Fenno- Scandia, for of this only a small portion was ever submerged. This subsidence appears to have reached its maximum during the Campanian epoch, for the zones of Actinocamax quadratus and Belemnitella mucronata seem to have had the widest extension both over Limbourg and over Scania. At this time it is probable that the British area was reduced to an archipelago of islands (see Building of the British Isles, 1911), and it is doubtful whether any part of England remained above water ; even of Wales only those parts continued as islands which are now more than 2000 feet above the sea. French geologists seem to have taken it for granted that Brittany and the Central Plateau of France still formed islands in the Senonian and Campanian Seas ; apparently because no trace of Cretaceous strata has been found upon them. But the existence of land cannot be postulated on negative evidence ; the question to be considered is how far the chalk of neighbouring areas is likely to have been prolonged over them. In the case of Brittany we have every reason to suppose that Turonian, Senonian, and higher chalks followed the Cenomanian over the Cotentin and down the Channel area; and as no part of Brittany rises to more than 1400 feet above sea-level, while Dartmoor rises to 2000, it seems probable that if the latter was entirely submerged, the former was also. With respect to the Central Plateau there is more to be said. Since the central and southern portions of it are over 5000 feet above existing sea-level, it might be thought that these must have remained above that of the Cretaceous Sea ; but we must remember 524 STRATIGRAPHICAL GEOLOGY that these parts were the scene of prolonged volcanic activity in Tertiary time, and that they were doubtless then raised to a higher relative level than they had previously occupied, for the north- west part of the plateau does not rise to more than 3000 feet. The facts which have chiefly to be considered are the characters of the Upper Cretaceous deposits which surround the Central Plateau. Those on the north and north-east sides are chalks which have the character of deep-sea deposits, being fine foramini- feral oozes in which Textularia and Globigerina are the prevalent forms. On the north-west the Senonian is largely a Bryozoan deposit formed in no great depth of water, and the Campanian is a siliceous rock analogous to "gaize," and mainly composed of sponge spicules ; this may have been accumulated in water of 200 to 300 fathoms (about 1500 feet). On the south-east side in Aquitaine we find the Senonian composed of marly limestones, which are often sandy and include banks of Hippurite and Eadiolite limestone, deposits which are not of deep-water origin. The Campanian of the same area, however, consists for the most part of fine white chalky limestones in alternating solid and crumbly beds with layers of black flints. M. de Grossouvre describes this material as similar to chalk, and as "denoting a deposit formed at a much greater distance from shores, and in much deeper water than the sediments " of the Senonian. 31 The successive stages of the series probably passed transgressively one beyond the other on to the flanks of the Central Plateau, and if the average depth of water under which the Campanian sediments was deposited was only 400 fathoms (i.e. 2400 feet), it is clear that large parts of the plateau must have been submerged ; how much would require a special study to determine. From the Campanian subsidence there was a gradual recovery, accompanied by a certain amount of lateral pressure which produced a warping or upward flexure of parts of the mass of deposit, so that a series of broad anticlines and syiiclines were formed, and the summits of the anticlines were planed off by the currents and waves of the shallowing seas. Hence, as we shall see, even the earliest Eocene deposits are more or less unconformable to the underlying Cretaceous strata. EEFERENCES 1 W. Whitaker, Quart. Journ. Geol. Soc. vol. xlii. p. 26 (1886). 2 See "Geology of the Weald," by W. Topley, Mem. Geol. Survey (1875). 3 See " Geology of the Isle of Purbeck," by A. Strahan, Mem. Geol. Survey (1898). THE CRETACEOUS SYSTEM 525 4 See F. Leighton, Quart. Journ. Geol. Soc. vol. li. p. 101. 5 See " Geology of the Isle of Wight," by A. Strahan, Mem. Geol. Survey (1889). 6 See Reports in Proc. Geol. Assoc. vol. xii. p. 227, and vol. xx. p. 115 (1907). 7 A. M. Davies, Proc. Geol. Assoc. vol. xvi. p. 15. 8 W. Keeping, "Fossils of Upwareand Brickhill," Sedgwick Prize Essay, 1883. 9 J. W. Judd, Quart. Journ. Geol. Soc. vol. xxiii. p. 227, and vol. xxiv. p. 218 ; G. W. Lamplugh, Quart. Journ. Geol. Soc. vol. xlv. p. 575. and vol. lii. p. 179 ; also Mem. Geol. Survey, Explanations of Sheets 83 and 84 (Lincoln and East Lincolnshire). 10 See Summary Prog. Geol. Survey for 1908, p. 75, and for 1910, p. 80. 11 See Jukes-Browne and Milne in Geol. Mag. for 1898, p. 21. 12 See Livret-guide, Congres geol. Internal, for 1900, "Excur. Montague de Lure." 13 See "Cretaceous Rocks of Britain," vol. i., Mem. Geol. Survey (1900). 14 See Jukes-Browne and W. Hill in Quart. Journ. Geol. Soc. vol. xliii. p. 544. 15 See W. Hill, Quart. Journ. Geol. Soc. vol. xliv. p. 320. 16 Consult "Cretaceous Rocks of Britain," vol.ii., Mem. Geol. Survey (1903). 17 Jukes-Browne, Quart. Journ. Geol. Soc. vol. xxxi. p. 256, and the "Geology of Cambridge, " Mem. Geol. Survey (1881). 18 Jukes-Browne and W. Hill, op. cit. p. 561. 19 W. Hill, Quart. Journ. Geol. Soc. vol. xliv. p. 325. 20 A. W. Rowe, Proc. Geol. Assoc. vol. xvii. p. 1 (1901). 21 A. W. Rowe, Proc. Geol. Assoc. vol. xviii. p. 193 (1904). 22 W. Hill in "Cretaceous Rocks of Britain," vol. iii., Mem. Geol. Survey (1904). 23 A. W. Rowe, Proc. Geol. Assoc. vol. xx. p. 209 (1906). 24 A. W. Rowe, Proc. Geol. Assoc. vol. xvii. p. 42. 25 See G. W. Lamplugh, Proc. Yorkshire Geol. and Pol. Soc. vol. xiii. p. 65 ; and A. W. Rowe, op. cit. (1904). 26 W. F. Hume, Quart. Journ. Geol. Soc. vol. liii. p. 540 (1897). 27 J. W. Judd, Quart. Journ. Geol. Soc. vol. xxxiv. p. 728. 28 W. Hill, Quart. Journ. Geol. Soc. vol. lii. p. 99. 29 A. de Grossouvre, "Recherches sur la Craie Superieur," Mem. Carle Geol. de la France, p. 351 (1901). 30 W. F. Hume, Geol. Mag. for 1892, p. 385. 31 A. de Grossouvre, op. cit. p. 378. CHAPTER XV TERTIARY OR C/ENOZOIC TIME c PALAEOGENE SYSTEM As explained on p. 11, the Tertiary deposits are usually divided into five or six groups with a nomenclature based on the pro- portional number of species of Mollusca found in each group that have survived to exist at the present day. But these groups cannot be regarded as systems of the same palseontological value as those into which the Mesozoic and Palaeozoic rocks have been divided. Only two such systems can be recognised in Caenozoic time, the Eocene and Oligocene divisions forming one system, the Miocene, Pliocene, and Pleistocene divisions forming a second. For these two systems the German geologist Hoernes proposed the names Palaeogene and Neogene, signifying the older and the newer kind of Tertiary life, and though not altogether satisfactory names, they have been widely adopted on the European continent. The older Tertiaries of North-western Europe occur in several basins or broad trough-shaped areas, separated by parallel anticlines or axes of elevation. The most northern of these is known as the London basin, which, however, is only the western part of a large geo-syncline extending from Belgium across the North Sea and terminating in Wiltshire. A second but much smaller basin is known as the Hampshire basin, extending through Sussex, Hampshire, and Dorset, but bounded on the south by the sharp anticline which runs through the Isle of Wight and the Isle of Purbeck. A third is the Paris basin, occupying a broad depression in the Chalk area of Northern France. South of this, again, are the two separate basins of Aquitaine and Provence. Some account of the Palaeogene deposits in all these areas will be given in the following pages, taking first the Eocene and afterwards the Oligocene Series. 526 THE EOCENE SERIES 527 I. THE EOCENE SEEIES This series has been divided into a number of stages for which different names are at present used in England, France, and Belgium, as shown in the following table : England. Belgium. France. Barton Beds. Wemmelian. Bartonian. Bracklesham Beds. Laekinian and \ Tuff Bagshot Beds. Brussellian Beds./ London Clay. Sables de Mons. \ v . Woolwich and \ Argile de Flandres. J Reading Beds. / U PP er IT inrlpnian Sparnacian. Thanet-Beds. Lower / Thanetian. Montian. Montian. A. LIFE OF THE PERIOD Eocene Life. The difference between the fauna of the Chalk and that of the Eocene is very great indeed, for only one species, and that is a Brachiopod (Terebratulina striata), seems to have survived the change of conditions in Britain, and not only so, but the general aspect of the Eocene fauna is quite different from that of the Cretaceous. The great reptiles of the Secondary rocks have disappeared, and with them the numerous Cephalopods which are so characteristic of the Cretaceous System the Ammonites, Turrilites, Baculites, Scaphites, Hamites, Belemnites, and many other genera ; vanished also are the Hippurites, Eadiolites, and Inocerami of the Chalk, most of the Cretaceous Echinoderms, and all the deep-sea siliceous sponges (Ventriculites, Cephalites, Ploco- spongia, etc.). In England the most abundant fossils are Gastropods and Lamellibranchs, many of them belonging to genera which now characterise much warmer seas than those around our islands. Brachiopods, Echinoderms, Corals, and Sponges are rare compared with their numbers in Mesozoic deposits. The Eocene flora is marked by the abundance of Dicotyledons, especially of genera which now occur in tropical and sub-tropical regions. It should be mentioned, however, that many of the genera w r hich now make their appearance in Britain occur in the higher Cretaceous deposits of the continent. The flora of the Woolwich and Reading Beds has a temperate aspect ; it includes leaves like those of the poplar, plane, and hazel (Populus, Platanus, and Corylus). The plants of the London Clay and Bagshot sand, on the other hand, are those of a much warmer climate ; they include leaves and fruits of the conifers Pinus and Sequoia, with 528 STRATIGRAPHICAL GEOLOGY the palms Nipa, Sabal, and Areca ; the Dicotyledons Ficus, Aralia, Dryandra, Cassia, Eucalyptus, Magnolia, Cactus, and many kinds of Leguminosce. In the Bournemouth Beds a still larger number occur, and form a different assemblage, but still one of sub-tropical aspect ; this includes ferns of the genera Osmunda, Chrysodium, Pteris, and Gleichenia, the palms Sabal, Iriartea, Flabellaria, Phcenicites, and a large number of Dicotyledonous leaf impressions which have not yet been properly determined. Of invertebrate animals the following are the principal genera which now make their appearance in Britain, though some of them (particularly the Gastropods) really date from the Upper Cretaceous Beds of Europe : Protozoa. Alveolina, Nummulites. Actinozoa. Balanophyllia, Dendrophyllia, Goniopora ( = Litharea), Madrepora, Oculina, Paracyathus, Solenastrea, Turbinolia. Echinoderma. Schizaster, Spatangus. Crustacea. Dromia, Mithracia, Portunites, Xanthopsis. Lamellibranchia. Corbicula, Diplodonta, Potamomya, Sanguinolaria, Verticordia. Gastropoda. Ancilla, Cassis, Conus, Conorbis, Cyprsea, Fasciolaria, Marginella, Oliva, Pyrula, Rostellaria, Rimella, Sycum ( = Leiostoma), Tritonium, Trophon, Typhis, Voluta, Xenophora. Cephalopoda. Aturia, Beloptera, Belosepia. Vertebrate animals are now abundant, the four great classes of fish, birds, reptiles, and mammals being all represented, though remains of birds are rare. The following are some of the genera found in the Eocene Series of England and France : Pisces. Myliobatis, JLtobatis, Pristis ; Lepidosteus, Phyllodus ; Coelopoma, Goniognathus, Sphyrsenodus ( = Dictyo- dus) ; Percostoma, Sciaenurus, and others. Reptilia. Gavialis (a crocodile) ; Lacerta (a lizard) ; Palseophis (a sea-snake) ; Trionyx, Lytoloma, Argillochelys (tortoises). Aves. Argillornis, Dasomis, Gastornis, Halcyornis, Eupter- ornis, Lithornis, Odontopteryx, all large birds. Odontopteryx had a large beak furnished with tooth-like processes. Mammalia. These deserve more special description. Mar- supials of the genus Didelphys were common, and with these are carnivores (Arctocyon and Palceonictis} with strong marsupial affinities, but believed to be really placentals ; a little later appear other carnivora, Pterodon, and Proviverra. The earliest known ungulate is Coryphodon, a primitive tapir from the Woolwich Be3s and London Clay. Still later other genera appeared, Lophiodou, THE EOCENE SERIES 529 Pachynolophus, Paloplotherium, all related to tapirs and rhinoceroses, with Hyracotherium, which combined characters now found in the coney (Hyrax) and in the pigs. Other hog-like animals (Hyopotamus) appear in the Upper Eocene and become common in Oligocene time. There were also primitive hornless forms of deer or antelope named Dichodon and Amphitragalus. Early forms of the Probos- cidea have been found in the Upper Eocene of Egypt. Sirenia are represented by Eotherium, etc. The rodents were represented by Fig. 176. FOSSILS OF THE THANET BEDS. a. Cuculltea decussata. b. Pectunculus terebratularis. c. Protocardium Laytoni. d. Cyprina planata. e. Pholadomya cuneata. /. Corbula regulbiensis. g. Aporrhais Sowerbyi. Theridomys and Sciuroides, and the primates by the Lemuroid Microchcerus. Characteristic Fossils. The following are some of the commoner and most characteristic species of invertebrates found in the marine and estuarine deposits of the Eocene Series : Fossils of the Thanet Beds Lamellibranchia. Protocardia Laytoni, Corbula regulbiensis, Cucullsea decussata ( = crassatina), Pectunculus terebratularis, Cyprina scutellaria ( planata), C. Morrisi, Astarte tenera, Pholadomya cuneata. Gastropoda. Aporrhais Sowerbyi, Trophon subnodosum, Scala Bowerbanki. 530 STRATIGRAPHICAL GEOLOGY Fossils of the Woolwich and Heading Beds Lamellibranchia. Corbicula cuneiformis, C. tellinella, Cyprina Morrisi, Ostrea bellovacina, 0. tenera, Unio subparallela. Gastropoda. Potamides funatus, Calyptrea aperta ( = trochiformis), Melanopsis buccinoidea, Melanatria inquinata, Neritina globulus, Pitharella Rickmani. Fig. 177.- a. Melania inquinata. b. Potamides funatus. c. Pitharella Bickmani. d. Ostrea bellovacina. OF THE WOOLWICH BEDS. e. Unio subparallela. /. Ostrea tenera. g. Corbicula cordata. h. Corbicula tellinella. Fossils of the London Clay Annelida. Ditrupa plana, Vermicularia bognoriensis. Crustacea. Hoploparia Belli, Xanthopsis Leachi. Brachiopoda. Lingula tenuis, Terebratulina striatula. THE EOCENE SERIES 531 Lamellibranchia. Cardita Brongniarti, Axinus angulatus, Dosiniopsis orbicularis, Pholadomya margaritacea, Pinna affinis, Modiola elegans, M. depressa, Panopsea intermedia, Pectunculus decussatus, P. brevirostris. Fig. 178. FOSSILS OF THE LONDON CLAY. a. Cardita Brongniarti. b. Panopea intermedia, r. Hoploparia Belli. d. Cassidaria ambigua e. Nipa elliptica (fruit). /. Terebratulina striatula. ff. Pleurotoma teretrium. h. Voluta Wetherelli. i. Nautilus imperialis. Gastropoda. Aporrhais Sowerbyi, Cassidaria^ ambigua, Xenophora extensa, Pisania (Tritonidea) Morrisi, Natica hantoniensis, Sycum (Leiostoma) globatum, Pleuro- toma (Surcula) teretrium, Pyrula Smithi, Voluta Wetherelli, Volutospina elevata. Cephalopoda. Nautilus imperialis, N. regalis, Aturia ziczac. 532 STRATIGRAPHICAL GEOLOGY Fossils of the Bracklesham Beds Protozoa. Nummulites laevigatus, N. variolarius. Actinozoa. Goniopora ( = Litharsea) Websteri. Lamellibranchia. Area biangula, Pecten corneus, Cardita (Venericardia), planicosta, Solen obliquus, Sanguinolaria Hollo- waysi, Pectunculus pulvinatus. Fig. 179. FOSSILS FROM THE BRACKLESHAM AND BARTON BEDS. a. Litharea Websteri. b. Nummulites laevigatus. c. Ostrea flabellula. d. Chama squamosa. g. Conorbis dormitor. e. Corbula pisum. h. Clavella longseva. /. Crassatellites sulcatus. i. Murex asper. Gastropoda. Clavella longseva, Sycum (Leiostoma) pyrus (range into Barton Beds), Cassidaria coronata, Cyprsea tuber- culosa (var. Coombii), Cerithium (Campanile) gigan- teum, Turritella imbricataria, Mesalia sulcit'era, Conus deperditus, Pleurotoma (Surcula) attenuata^ Melongena minax, Strepsidura turgida, Voluta nodosa, V. spinosa, Ancilla buccinoides. Fossils of the Barton Beds Protozoa. Nummulites elegans. Echinoderma. Schizaster d'Urbani. Lamellibranchs. Cardita sulcata, Crassatellites sulcatus, Cardium porulosum, Corbula pisum, C. ficus, Pecten (Chlamys) reconditus, Pectunculus deletus, Chama squamosa. THE EOCENE SERIES 533 Gastropoda. Turritella imbricataria, Rostellaria (Hippochrenes) ampla, Rimella rimosa, Conorbis dormitor, Murex asper, Typhis pungens, Fusus porrectus, Clavella longseva, Buccino- fusus regularis, Pleurotoma (Surcula) rostrata, Cassi- daria nodosa, Xenophora agglutinans, Olivella Branderi, Voluta athleta, V. luctatrix. B. BRITISH EOCENE DEPOSITS The present restricted extent of the Eocene deposits in England and their separation into two basins are due to the upheaval and Fig. 180. FOSSILS OF THE BRACKI.ESHAM AND BARTON BEDS. a. Voluta athleta. b. Conus deperditus. c. Xenophora agglutinans. d. Olivella Branderi. e. Ancilla buccinoides. /. Cardita planicosta. disturbance which took place in Miocene time, when lateral pressure from the south established a series of east and west flexures ; and as the Palaeogene deposits have been removed from the anti- clinals, the parts which remain occupy two synclinal areas, i.e. the London and the Hampshire basins. A glance at any geological map will show the extent of the Palaeogene Beds in these two areas. The London basin extends from Hungerford in Berkshire to Eeculvers in Kent, and its northern boundary passes from Eeading through the counties of 534 STEATIGRAPHICAL GEOLOGY Buckingham, Hertford, and Essex to near Framlingham in Suffolk, where the influence of the syncline dies out and the boundary line then runs northward to the west of Yarmouth. Its length from Hungerford to Keculvers is about 120 miles, and its width 'from Sudbury to Canterbury is nearly 60 miles, but this is probably less than half its actual extent, the rest of the basin being concealed by the waters of the North Sea. The Hampshire basin is much smaller, and lies in a trough between the anticlinal axis of Winchester on the north and the monoclinal of Purbeck and the Isle of Wight on the south, and its extreme width is, therefore, only 25 miles, while its length from Dorchester to near Worthing in Sussex is about 60 miles, with an outlier at Newhaveii nearly 20 miles farther east. There is evidence, however, of the original extension of the Eocene through Dorset and East Devon. In Ireland freshwater beds of Eocene age are interstratified with Norfolk. London. Fig. 181. DIAGRAM TO ILLUSTRATE THE PRE-TERTIARY EROSION OF THE CHALK. 5, b. Upper Chalk. L. London Clay, o, a. Base of Upper Chalk. W. Woolwich Beds. T. Thanet Beds. the great sheets of basaltic lava which cover so large an area in Antrim and Derry. In Scotland a similar set of lavas, also with intercalated freshwater beds, attains a great thickness in the islands of the Inner Hebrides (Mull, Skye, etc.). In Britain there is a decided unconformity between the Cretaceous and Palaeogene Systems, a break which indicates a considerable lapse of time and a great change of physical and geographical conditions. We have seen that there are deposits in France and Belgium which partially bridge over this gap, and similar deposits may also have been formed in the British area, but if so, they were destroyed before the deposition of our lowest Eocene strata. It is now certain that large tracts of the Upper Chalk have been removed from certain areas either before or during the formation of the earlier Eocenes, for we cannot doubt that the higher parts of the Chalk were originally continuous from Hampshire to Norfolk, yet they do not now exist under the London basin, and consequently the Upper Chalk, which in Sussex THE EOCENE SERIES 535 is at least 750 feet thick, and in Norfolk is probably 1000 feet, has only a thickness of from 250 to 300 feet beneath the valley of the Thames. Moreover, where the Upper Chalk is thinnest, the London Eocenes are thickest, and there can be no doubt that these Eocenes were deposited in a broad and deep trough, which had been excavated out of the Upper Chalk. The relation of this trough to the general surface of the Chalk is indicated in the diagram, Fig. 181. The following are the divisions of the Eocene Series now recog- nised in the London and Hampshire basins. They can be grouped into Lower and Upper Eocene, but the creation of a Middle Eocene is quite unnecessary. London Basin. Hampshire Basin (Isle of Wight). Feet. Feet. ? Wanting. Barton Beds . 340 to 360 Upper Lower Brackles- f Sands about 200 f Bracklesham and ham \Clays Bagshot Beds . London Clay . Woolwich and Reading Beds Thanet Beds . 60 \ Bournemouth Beds 555 to 583 150 Bagshot Beds . 243 to 98 400 London Clay . 230 to 320 80 Reading Beds . 80 to 160 80 Absent About 970 1450 to 1520 It will be seen that the Lower Eocene is more fully developed in the London basin than in Hampshire and includes the lowest division (Thanet Beds), which does not occur in Hampshire. On the other hand, the Upper Eocene is thin and incomplete near London, but is fully developed in the Hampshire basin. The Thanet Beds, as their name implies, are well developed in the Isle of Thanet. They consist mainly of pale-coloured sands, the lowest part being always argillaceous and full of dark -green glauconitic grains, while at the base, and resting on the Chalk, there is invariably a layer of unworn green-coated flints. These ilints show no signs of attrition, and are believed to have been derived directly from the underlying chalk, not by mechanical erosion, but by chemical solution; carbonated rain-water per- colating through the sands and dissolving the subjacent chalk, but leaving the insoluble flints behind ; so that a layer of these nodules has gradually accumulated at the base of the sand, but was not there when the sands were originally deposited. This supposition is confirmed by the fact that no fossils occur in this basement-bed, whereas if the flints had ever lain on a sea-bottom small oysters, serpulse, etc., would certainly have fixed themselves on to such favourable surfaces for attachment. In East Kent the greater part of this stage consists of greenish sandy marl (50 or 60 feet), overlain by sharp greenish-grey sand & I I* "2 So'* "SO w u t-c I II * -2 o o S,| I a ^^^ S cS >> o r ..7 i Q S X 1! os III r 5 S B S& C5 oi CO THE EOCENE SERIES 539 surface of the Woolwich Sands is a mass of pebbles in a sandy matrix, taking the place of the Oldhaven Beds and often called the Blackheath Beds. The Blackheath Beds do not extend west of Croydon, and are not found along the northern edge of the London basin. The pebbles composing them are all well-worn and rounded flints without any subangular pebbles such as occur on our shingle beaches ; hence Mr. Whitaker concludes that they were not laid down in beaches, but were deposited by strong currents at some little distance from the shore, and his view is confirmed by the frequency of current bedding. Beds of mottled plastic clay begin to appear in West Kent near the base, and these increase westward through Surrey, the shell- bearing beds gradually dying out, till in Hampshire and Berkshire they consist entirely of mottled clays, in tints of red, yellow, grey, purple, and green, associated with beds of red, yellow, and white sand. These are known as the Beading Beds, and marine fossils only occur in the basement-bed, the only fossils in the clays being leaves of plants. Near Beading the group consists of : Feet. Mottled clays 40 to 50 Yellow and white sands 20 to 30 Bottom bed with marine fossils . . . 5 to 8 Beds of the same (Reading) type range along the northern lip of the basin through Herts and Essex, but include some pebble beds which are occasionally concreted into the kind of conglomerate known as " plum-pudding stone." They recur in the Hampshire basin, where they consist almost entirely of bright red and white mottled clays with several layers of brown sand, one of these being at the base and resting on the Chalk. No fossils except fragments of plants have yet been found in these clays, and they do not therefore seem to be of marine origin. Their thickness in Whitecliff Bay is 163 feet, at Alum Bay 84 feet, and they are still from 70 to 80 feet thick in Dorset near Bere Regis. Near Dorchester they include beds of sand and of subangular gravel consisting of Cretaceous flints and cherts. The London Clay occupies broad areas both on the north and south side of the Thames Valley. It is of much more uniform composition than the lower group, the mass of it consist- ing of stiff bluish clay with layers of septarian nodules. The surface colour of the clay is generally brown, but this is due to the oxidation of the iron it contains by exposure to the weather ; its normal tint in deep wells and borings is a bluish-grey colour. The basement-bed is of a different character to the rest, con- sisting of brown, grey, or greenish sand and sandy clay, in which 540 STRATIGRAPHICAL GEOLOGY are scattered many rolled flint pebbles ; these pebbles are always black, and sometimes have the remarkable peculiarity of falling into fragments when tapped by a hammer, for though to the eye they seem compact, they are really traversed by a multitude of invisible cracks. This basement-bed is generally from 6 to 12 feet thick, and often contains fossils, of which the chief are Aporrhais Sowerbyi, Galyptrea aperta ( = trochiformis), Pitaria obliqua, Panopcea intermedia, Pectunculus brevirostris, and Ditrupa plana. The thickness of the London Clay near London is between 400 and 500 feet, but it thins gradually westward to about 270 at Wokingham, 50 near Newbury and Hungerford, and only 15 in the outlier of Great Bedwin in Wiltshire. Fossils are not common throughout the London Clay, but sometimes occur in great profusion, especially where beds of sandy clay occur. The cliffs of the Isle of Sheppey are noted for the fossils they yield ; plant remains are there especially abundant, and bones of birds, turtles, and snakes have also been found. In the Hampshire basin, with the exception of an outlier at Newhaven, the most easterly exposure of this division in this area is at Bognor in Sussex, where beds of clay and calcareous sandstone, highly fossiliferous, are exposed 011 the shore, and are supposed to belong to the lower part of the London Clay. Borings at Portsmouth Dockyard below the summit of the formation proved a thickness of 290 feet, and exposures made in extending that dockyard from 1868 to 1870 were described by Mr. Meyer, 2 who found the clay divisible into three parts, each commencing with a layer of pebbles and passing upward from stiff clays into sandy clays. The lowest zone is about 100 feet thick, the middle zone has a thickness of 134 feet, and the sandy beds in the upper part of this yielded many fossils. The upper- most zone consists mainly of clays containing Cyprina planata, Pholadomya margaritacea, Rostellaria lucida, and other species. In the Isle of Wight the London Clay is about 320 feet thick at Whitecliff Bay, but only 230 at Alum Bay. 3 At neither place are many fossils to be found, but- at Whitecliff the higher part contains Panopcea intermedia, Cyprina planata, Callista tenuistriata, and Pinna affinis. At 262 feet from the base there is a layer of black flint pebbles, and above this are sandy and laminated clays. In Dorset the London Clay consists chiefly of brown sandy loam with seams of sandy ironstone and a bed of flint pebbles at the base. Its thickness in the Purbeck district is from 70 to 80 feet, but it thins westward, and is finally overlapped by the Bagshot Beds near Dorchester. Bagshot Beds. In the London Basin these beds occupy a THE EOCENE SERIES 541 much smaller space than the London Clay because they have been removed from the greater part of its surface by subsequent detrition and erosion, but there is good reason to believe that they were not only coextensive with the London Clay but spread beyond its original borders. The chief remaining tract of Bagshot Sand is an area about 24 miles in length with a breadth of about 10 or 12 miles, and occupying parts of Berkshire, Hampshire, and Surrey. Aldershot, Farnborough, Bagshot, Wokingham, Ascot, and Cobham all lie within this area. Westward there are several outlying tracts, some large and some small ; and north of the Thames in Middlesex are three small outliers capping Harrow, Hampstead, and Highgate Hills (see Fig. 183). The extreme thickness of these sands in the Bagshot and Farnborough country is from 130 to 150 feet. Westward at Ramsdell, in the outlier between Basingstoke and the valley of the Kennet, there is a bed of brown clay very like London Clay, and 30 feet thick, in the lower part of the Bagshot Sand. In the Hampshire Basin the Bagshot Beds consist of bright yellow and white sands with bands of grey laminated sand and clay and (in the upper part) layers of white pipeclay, which contain leaves and stalks of plants belonging to the genera Aralia, Ccesalpinia, Gomptonia, Dryandra, Ficus, Laurus, Quercus, and other Dicotyledons. Ferns are rare in the Isle of Wight, but Ghrysodium lauzeanum abounds at Studland in Dorset. This flora indicates a warmer climate than that of the Eeading Beds. There is much difference of opinion as to the thickness assign- able to these beds in the Alum Bay section (Fig. 185) ; the Geological Survey classes all the sands and clays which do not yield marine fossils as Lower Bagshot, and these amount to 662 feet. Mr. J. S. Gardner, 4 however, has shown that the special Bagshot flora doe& not occur above the pipeclays, and that the higher beds at Bournemouth contain a very different flora. He also thought that the lower part of these sands replaced the upper part of the London Clay, but if the latter is restricted to 233 feet, the Bagshot Sands will have a thickness of 243 feet at Alum Bay, and decrease east- ward till they are only 98 feet in Whitecliff Bay. At Studland and Corfe in Dorset the Bagshot Beds have the same facies, but westward they pass into coarse subangular gravels,, which near Dorchester overlap both London Clay and Reading Beds so as to rest directly on the Chalk. These gravels contain pebbles of Cretaceous flint and chert, Purbeck marble, with many of quartz and of Palaeozoic rocks probably derived from the Permian breccias of Devonshire. Still farther west they overstep the Chalk and rest on the Selborniaii Sands. 542 STRATIGRAPHICAL GEOLOGY The Brackleshaxn Beds. In the London basin the lower part of this group consists of brown laminated clays, overlain by dark-green clayey sand from 12 to 20 feet thick, and succeeded by a variable set of greenish sands with bands of clay, the total thick- ness varying from 40 to 80 feet. These beds have yielded a few fossils, Fusus LongcevuSj Turritella sulcifera, Gardita planicosta, Pecten corneus, Ostrea flabellula, Gorbula striata, Nummulites Icevigatus, and many fish teeth, Lamna obliqua, Garcharodon auriculatus, etc. Above the clays is a mass of sands formerly called the Upper Bagshot Sand, but now referred to the Bracklesham Group, and possibly representing the Boscombe Sands mentioned below. At the base is a layer of small flint-pebbles, and the fine yellow and white sands which succeed are the highest Eocene Beds that occur in the London basin. Their thickness appears to be over 200 feet, for in a boring at the Albert Asylum, Bagshot, their depth proved to be 226 feet. 5 These sands contain in the upper part blocks of white saccharoid sandstone, which, when weathered out, are known as " grey- wethers " ; they are sometimes 10 to 14 feet long, and 3 or 4 deep, and they often enclose flint pebbles. Organic remains are rare, but here and there are casts of shells, and about forty- eight species were obtained by Mr. Monckton from the Tunnel Hill cutting near Pirbright on the South - Western Kailway. Among these the commonest were Rimella rimosa, Natica patula, Turritella imbricaturia, Ostrea flabellula, Pecten reconditus, Gardita sulcata, Lucina mitis, Tellina scalaroides, and Corbula pisum. These beds received their name from Bracklesham Bay in Sussex, where they contain marine shells throughout their extent along the shore, but there are no cliffs in which they can be measured. The section in Whitecliff Bay exhibits the same marine type, and has been described in detail by the Kev. 0. Fisher, 6 who divided the group into four portions. Mr. H. Keeping has since been able to fix the limit between the Bracklesham and Barton Beds more exactly, and finds it to be about 70 feet lower than where Mr. Fisher placed it. Making this alteration, the thick- nesses of the several parts are as follows : Feet. A. Green and blue clays with a bed of sand about 52 feet from the base, Nummulites variolarius, Pecten corneus, Corbula pisum, Pleurotoma plebeia, etc. ...... 182 B. Yellow sands and sandy clays, Num. variolarius ... 27 C. Sandy clays and green sands, Num. Icevigatus, Sanguinolaria Hollowaysi, Valuta spinosa, etc. . . . . . .123 D. Laminated clays and green sands with a bed of pebbles at the base, Num. Icevigatus and a few other fossils . . .251 583 THE EOCENE SEKIES 543 It will be noticed that C and D form a zone of Nummulites Icevigatus, while A and B are characterised by Num. variolarius. On reaching Alum Bay we find that the greater part of this group is represented by unfossiliferous sands and clays, probably estuarine beds, and that marine shells are only found in the upper- most 157 feet, which answer to Mr. Fisher's group A. The lower beds were formerly relegated to the Lower Bagshot, but their true position has been determined by Mr. Gardner, who has compared them bed by bed with those of the opposite cliffs of Hampshire. 7 Their thickness is 398 feet, making a total of 555 feet. The succession of the Bournemouth and Boscombe Beds has been carefully worked out by Mr. Gardner, who gives it as follows : 6. Dark sandy clays with pebbles at base .... 5. Highcliff Sands, white and without fossils 4. Hengistbury Clays, with septaria containing sharks' teeth 3. Boscombe Sands, of various colours containing pebbles 2. Bournemouth marine beds, with plants and marine shells 1. Bournemouth freshwater beds, with plant remains only . Feet. 14 31 57 140 50 ?400 About 590 The Freshwater Beds consist largely of sand with occasional bands of clay in which leaves are abundant, especially those of Salix (willow), Triartea (a palm), Sequoia Coutssce, Araucaria, Eucalyptus, and the ferns Osmunda, Polypodium and Acrostichum. The marine beds are a similar set of deposits, but among the plants are Palmacites (a cactus), Dryandra, and the fruits of the Nipa palm. Barton Beds. The most careful study of these beds is the memoir by Messrs. Gardner, Keeping, and Monckton, 8 on which the following account is based. They divide the stage into three parts, which they call Lower, Middle, and Upper Barton, but the Geological Survey divide it into the Barton Clay and the Barton Sands, which is in some ways a more convenient division. According to these authors the thicknesses of their divisions at the three principal sections are as follows : Barton Sands or Uer Barton Bavton Barton. Alum Bay. Whitecliff. 90 53 49 114 167 57 221 feet. 92 55 192 338 368 feet. The Lower Barton has at its base a thin zone of green sandy 544 STRATIGRAPHICAL GEOLOGY clay in which Nummulites elegans (var. prestivichiana) is common, and serves to define the Barton from the underlying Bracklesham Beds. Above are grey clays and sands containing Valuta athleta, Cassis ambigua, Rimella bartonensis, and Cominella canaliculata. The Middle Barton, or Barton Clay proper, consists mainly of grey and brown clays with bands of septaria, and its prevalent fossils are Valuta luctatrix, V. ambigua, Rostellaria ampla^ Murex minax, Glavella longceva, and Crassatella sulcata. At Alum Bay there is a remarkable increase in the thickness of this group, and the upper 70 feet consist of yellowish sandy clays with many small Mollusca, no such beds being seen at Whitecliff. The Barton Sands at Barton present the following succession : Feet. 4. White and yellow sands with Potamides concavus, Olivella JSratideri, and Melania hordeacea ...... 20 3. Dark-grey sandy clay with Olivella Branderi and other shells . 26 2. White sand without fossils . . . . . . 25 1. Bluish-grey clayey sand (the Chatna Bed) with Chama squam- mosa, Terebratula bisinuata, Terrebellum, sopitum, Volutct costata, V. humerosa, and Conus scabriculus .... 18 At Alum Bay the Chama Bed is rarely exposed, and the higher beds are represented by a mass of yellow and white sand from 90 to 100 feet thick, which has been largely dug for glass-making. These sands are sometimes called the Headon Hill Sands, and at the east end of the island they thicken to 206 feet, and the Chama Bed is seen below them with a thickness of 15 feet. C. THE KEGION OF VOLCANIC ACTIVITY The northern parts of the British Isles were a scene of remarkable volcanic activity throughout a large portion, if not the whole of Eocene time, and evidences of this activity are found not only in the vast lava -flows above mentioned but in lava -filled fissures of great length. The region over which such phenomena have been found embraces the north of England, the north of Ireland, and the whole of Southern and Western Scotland, and it must also have included the area of the adjoining seas as far south as Anglesey, as far west as St. Kilda, and as far north as the Faroe Islands a total area of about 50,000 square miles. Its history has been written by Sir Archibald Geikie 9 and Mr. A. Harker, 10 and the following is an epitome of it. This great development of volcanic action seems to have coincided with a general elevation of the region above indicated, as if it had been pushed upward by the accumulation of a vast sea THE EOCENE SERIES 545 of liquid lava beneath it. Whether as cause or coincidence, it seems certain that great horizontal tension arose which started a series of fissures, these running generally from north-west to south-east, and sometimes extending for 50, 60, or even 100 miles. Into these fissures welled up the liquid rock which has consolidated into basalt, and where they reached the surface the lava welled out in vast streams like those of modern Iceland, forming the great basalt plateaux of Antrim and the west of Scotland. This process of dyke-making' and the outpouring of lava -floods was repeated again and again, till on the great plain or plateau which bordered the west of Scotland the lavas accumulated to a depth of several thousand feet. In some places 3000 feet of them still remain, and we know not how great a thickness has been removed. Here and there they filled up and obliterated river-channels like those in the island of Mull. After the building up of these plateaus another episode took place. Large bodies of a basic magma were pushed into and through the plateau basalts and consolidated into huge domes of gabbro, troctolite, and other such coarse-grained rocks, but whether they ever extruded matter at the surface cannot be known, because the summits of the bosses are removed. The next phase was the protrusion of material from a different magma, which has consolidated into acid rocks of various kinds, from glassy obsidians and rhyolites to granophyres and a rock of granitic character. These now form conical hills which resemble in some respects the trachytic hills of Auvergxie. The last phase was a second formation of fissures, producing another system of basic dykes which traverse all the previously formed rocks, cutting not only the older dykes but the great masses of gabbro and granophyre. Whether they poured lava to the surface is not known, and whether the formation was the final episode is not quite certain, for there is also a series of acid dykes which traverse all the older volcanics and do not appear to be cut by the later basic dykes. The age of these later basic and acid dykes is also uncertain ; they may be Oligocene, or even Miocene ; but whenever vulcanism did cease the whole region began to sink * down again, and at the present time great parts lie below the sea. Ireland. The largest remaining portion of the great lava plains is the basaltic plateau of Ulster (Antrim and Derry) (see Fig. 186), which has an area of about 2000 square miles. This plateau presents steep escarpments on all sides, but the ground falls towards a central trough in which lie Lough Neagh and the valley of the river Bann. The component lavas are almost entirely basaltic, but they are divisible into a lower and upper series. The 2 N 546 STRATIGRAPHICAL GEOLOGY Io t -5 THE EOCENE SERIES 547 lower lavas rest on an irregular floor of older rock, the substratum being in most places Chalk, but between it and the lowest lava- flow there is always a bed of brown ferruginous clay full of flints. These lower lavas are mostly massive or amygdaloidal basalts, and their combined thickness is from 400 to 500 feet. Between the two series of lavas there is a break which appears to have coincided with the plutonic phase above mentioned, when the intrusive masses of gabbro and granite were formed. In Antrim this is marked by a band of stratified beds consisting of brown and red clays with layers of lignite, beds of pisolitic iron-ore (bauxite), of volcanic sand, and conglomerates composed of pebbles of rhyolite and basalt. Over these beds are piled thick and extensive sheets of basalt and dolerite, which at Sleamish have a thickness of about 600 feet. The well-known columnar basalt at the Giant's Causeway is one of these sheets and overlies a bed of bole or laterite. At Glenarm and at Ballypalady near Templepatrick there are thick deposits of iron-ore, the lower bed being a compact ferru- ginous earth known as lithomarge, which is sometimes 40 feet thick, and is overlain by pisolitic ore, and these occur some 600 feet from the base, and 400 feet from the top of the basaltic formation. At Glenarm they are associated with sandy clays and pebble beds, which enclose a seam containing well-preserved leaves and plant remains. Among these DapJinogene Kanei, Sequoia Coutssiw, Macclintockia Lyelli, and leaves of Platanus, Quercus, and Ehamnus have been identified by Baily. Mr. J. S. Gardner considers the flora to be of early Eocene age, not Miocene as previously supposed. 11 Scotland. The basaltic plateau of Ulster is only a portion of a very extensive tract of similar materials which has been largely broken up and destroyed, but must have originally stretched northward to the west coast of Scotland, where other large remnants of it are found in the Isles of Mull, Eigg, Rum, Skye, and Raasey. In the present connection the most important Scottish locality is the promontory of Ardtuii Head in the south- western part of Mull. Here beds of gravel sand and shaly clay are intercalated between two great sheets of basalt and have yielded many plant remains. The beds were first investigated by the late Duke of Argyll, but more recently by Mr. J. S. Gardner, 12 from whose account the following is taken, together with the illustrative section, Fig. 187. The sedimentary rocks appear along the seaward face of the headland for a distance of rather more than a mile, and the succession seen in one of the quarries is as follows : 548 STRATIGRAPHICAL GEOLOGY Feet. Rudely columnar basalt ....... 10 Bedded sandstone ........ 8 Indurated gravel of flints and lava-fragments ... 7 Indurated dark-coloured mud with ferns ... 1 Black crumbly shale full of leaves ..... 2 J Hard gravelly sand resting on basalt .... 2 At other points there are several beds of clay or shale which contain leaves, and similar beds also occur near Carsaig. The plant remains include a fern (Onoclea hebridica) and an Equisetum, many Gymnosperms, such as Ginkgo, Podocarpus, Taxus, and Sequoia. Dicotyledons are abundant and include leaves of Platanus hebridicus, Populus arctica, Cornus hyperborea, Boehmeria antiqua, and leaves like those of Corylus, Laurus, and Rhamnus. Mr. Gardner considered the flora to be more like a late Cretaceous than a Tertiary one, but accepting its probable connection with the Antrim area, was inclined to regard it as of very early Eocene age, and not newer than the time of the Thanet Beds. That the island of Arran was included in the area of Tertiary volcanic activity has long been suspected, but it was only in 1899 that absolute proof of the fact was obtained. 13 No remnants of basaltic lava-flows remain on the island, but all the southern part of it is seamed with basalt dykes, and it is very probable that many of these belong to the earlier dyke-system, and that surface eruptions proceeded from them ; all traces of such eruptions have, however, been destroyed. The most remarkable feature of the island is the large volcanic vent which occupies the central district around Glen Craigagh, between Brodick and Machrie Bays. The complex of rocks by which this vent is now rilled has an oval form about 3^ miles in length by 3 in breadth ; its central portion (over 2 miles in diameter) is composed of a breccia or agglomerate, consisting largely of fragments of sedimentary rocks mixed with a smaller quantity of igneous rock-fragments. This area is surrounded by a ring of intrusive igneous rocks, which are chiefly granites, grano- phyres, and felspathic rocks. Some of the rock-fragments in the central agglomerate are Old Eed Sandstone, and have been blown up from below; others are large blocks of Trias, Rhsetic, Lias, and Chalk, which must have fallen into the vent from above, and testify to the existence of a former covering of Jurassic and Cretaceous strata of which all other traces have been destroyed. As remarked by Sir A. Geikie in the Summary of Progress for 1900 (p. 124), "the explosion was thus certainly later than the Upper Cretaceous period and may consequently be referred to older Tertiary time. The fixing of the THE EOCENE SERIES 549 age of this vent carries with it the determination of the age of all the associated eruptive rocks. There can now be no room to doubt that the granites, quartz-porphyries, felsites, pitchstones, diorites, diabases, and basalts are all a connected series of eruptive masses, coeval with the similar assemblages in Antrim, the Inner Hebrides, and St. Kilda." D. CONTINENTAL EQUIVALENTS The Eocene area of Belgium may be regarded as a continuation of the London basin, for the two are doubtless in actual connection beneath the North Sea. Again, the deposits of the Paris basin present such a great resemblance to those of the Hampshire basin that they can be correlated without much difficulty, and were doubtless formed within the same area of deposition. In the south of France, however, and in the south of Europe generally, the Eocene Series exhibits a very different facies, consisting largely of hard foraminiferal limestones full of Nummulites, Operculina, and Alveolina, and generally known as Nummulite limestones ; with these are calcareous sandstones containing many large shells, reef- building corals, and Echinoderms of many species. The accompanying table show r s the divisions recognised in the Eocene Series of Belgium and France, and their correlation with those of England. England. Belgium. France. Barton Beds. Wemmelian, sands. Bartonian, sands 1 and limestones. \ Bracklesham Beds. Bruxellian, sands and sandstones. Lutetian ( (Calcaire Grossier), 1 limestones. Bagshot Sands, London Clay. Sables de Mons, Argile de Flandres. Ypresian, \ sands only. Woolwich and Reading Beds. Landenian (freshwater beds). Sparnacian (freshwater beds), f Thanet Beds. Heersian, sands and marls. Thanetian, Marine sands. Montian, limestones. Montian, limestone and marl. CO ! UT O O 3 550 STRATIGRAPHICAL GEOLOGY The name Montian is taken from Mons in Belgium, Thanetian from our Isle of Thanet, Sparnacian from Eparnay (Sparnacum), Ypresian from Ypres in Belgium, Lutetian from Lutetia the Roman name of Paris, and Wemmelian from Wemmel in Belgium. From the above table it will be seen that deposition began earlier both in Belgium and France than in the English area, the Montian Beds forming a kind of link between the Cretaceous and Eocene Series. 1. Belgium Montian. The beds referred to this stage have a thickness of from 200 to 300 feet, and the Geological Survey of Belgium now recognises three subdivisions which in descending order are : 3. Freshwater beds with Physa montensis. 2. The limestone of Mons (and Upper Tuffeau of Ciply). 1. The Cerithium limestone and basal conglomerate. The basal conglomerate lies unconformably on the higher Cretaceous strata, and the Cerithium limestone is so called from containing a large Cerithium (C. Corneti). The Mons limestone is a rough yellowish friable rock composed mainly of shell-fragments, Foraminifera, and calcareous algae. It contains a large fauna, consisting mainly of Gastropoda, with a fair number of Lamelli- branchs and a few Echinoderms ; Cerithium montense, Cer. in- opinatum, Turritella montensis, Mitra Dewalquei, Pseudoliva robusta, Cidaris distincta, Cassidulus elongatus, and Echinanthus Corneti are characteristic species, and there are a number of freshwater and terrestrial shells which have been washed into the deposit from neighbouring land. No Ammonites or true Belemnites occur, but the latter are represented by two species of Beloptera. The Heersian or Lower Landeniaii of Mourlon consists of glauconitic sands passing laterally into soft sandstone or sandy marl. Its fauna is marine and like that of our Thanet Beds, but near Gelinden the sands include a band of white marl containing the remains of an early Eocene flora (Osmunda eocenica, Laurus Omalii, Cinnamomum Sezannense, etc.). The total thickness of the Heersian is about 60 feet. Landenian. This group is of freshwater origin and resembles our Woolwich Beds, consisting of sands and marls with beds of shaly lignite. The thickness is only from 30 to 50 feet. Ypresian. This stage falls naturally into two parts, of which the lower (Argile de Flandres) is a continuation of the London Clay. In a boring at Ostende its thickness was found to be 460 feet, but it thins eastward just as the London Clay thins westward. Fossils are scarce in this clay, but it is overlain by sands and THE EOCENE SERIES 551 sandy clays which yield Nummulites planulatus, Turritella edita, Pectunculus decussatus and other fossils, and are from 100 to 150 feet thick. Of about the same age are the sands of Mount Panisel near Mons and those of Oeltre in Flanders, which contain Pinna margaritacea, Rostellaria fissurella, Turritella edita, and Gardita planicosta. Bruxellian, better latinised as Brussellian ; this name is restricted by Belgian geologists to the lower part of a mass of sand and sandstone which corresponds to our Bracklesham Beds. Fossils are abundant in these sands, but generally in a bad state of preservation, except sharks' teeth such as Lamna elegans, Otodus macrotus, Galocerdo minor, Enchodus Blecheri, and Myliobates toliapicus. The fruits of the Nipa palm (Nipadites Burtini) are also common. The higher sands are known as Laekiniaii and are characterised by Nummulites Heberti, N. variolaria, and Ditrupa strangulata. The total thickness is not more than 180 feet. The Wemmelian has a basement - bed of sandy gravel containing Nummulites variolarius, overlain by sands and a band of sandy glauconitic clay. In these beds Num. wemmelensis is a characteristic species, with Turritella sulcifera, Valuta elevata, Gardium porolosum, Tellina rostralis, and Pectunculus pulvinatus. The thickness of these beds varies from about 50 feet near Brussels to 170 at Malines, and over 200 in a boring at Anvers. 2. The Paris Basin Montian. This is not so well developed in France as it is in Belgium, because it has been deeply eroded in many places before the deposition of the succeeding Thanet sands. It varies, con- sequently, in thickness from 6 feet at Meudon to 162 at Mont Airne near Vertus. 14 Typically it is a yellow granular limestone, partly oolitic and partly composed of shell debris. It is probably some- what older than the limestone of Mons, for it contains Nautilus danicus, but with this are large Gerithia, Turritella montensis, Pseudoliva robusta, Fasciolaria prima, Lima Garoli, Gorbis multi- lamellosa, and other Montian species. At Meudon it is succeeded by white marls which are partly of marine and partly of freshwater origin. Limestones of the same age occur in the Pays de Bray at Vigny, and at Laversine near Beauvais. Thanetian. Glauconitic sands which appear to be about the age of our Thanet Beds are found all along the northern and eastern borders of the Paris basin, and as a rule they rest directly on the Chalk. Such are the sands of Watten and St. Qmer near Calais, which are characterised by Cyprina planata and are often indurated by a siliceous cement into a stone known as tuffeau. 552 STRATIGRAPHICAL GEOLOGY resembling gaize or malmstone arid containing diatoms and sponge spicules. Farther south in the Aisne are the sands of La Fere and of Chalons-sur-Vesle, which seem to be of the same age, while westward near Beauvais are the sands of Bracheux, containing Cyprina planata, Ostrea bellovacina, Pectunculus terebratularis, etc., a fauna which is classed as Thanetian by the majority of French geologists, but as Sparnacian by Professor Prestwich. 15 None of these sands extend so far as Paris. Sparnacian. This group corresponds very closely to our Woolwich and Heading Beds. Near Paris it consists principally of plastic clays, but at the base is a conglomerate with bones of mammals and birds, overlain by laminated clay containing Physa, Viviparus, and Unio, and near the top is a bed of sand with seams of lignite. This sand and lignite thickens eastward, while the plastic clays die out, till in the Soissonnais the group is represented by beds of lignite and black clays with associated marls and sands (30 to 50 feet). These contain Cyrena cuneiformis, Melania inquinata, and other Woolwich species. The highest beds are found at Sinceny on the Oise. They contain the same freshwater shells and consist of sands and clays with rolled flint pebbles. Ypresian. In France the London Clay is only found near Calais and Orchies ; evidently it thinned out southward and does not appear to be represented elsewhere unless by the lower part of the sands of the Soissonnais, which are not more than 160 feet thick. These are divisible into three portions : Sables de Guise with Cyrena Gravesi and freshwater shells. Sables de Pierrefonds, Turritella edita and Nummulites planulatus. Sables d'Aizy with Rostellaria Geoffroyi. Lutetian. In the Paris basin this stage consists of the beds generally known as the Calcaire Grossier. It is the most fossili- ferous formation in the Paris basin, and the middle portion yields the light-yellow calcareous freestone which is the chief building stone of Paris. The group is divisible into three sub-stages as follows : Feet. 3. Upper, thin-bedded fissile limestones with white and green marls and some beds of siliceous and dolomitic limestone, Cardium obliquum, Cerithium denticulatum, Potamides lapidum, Lucina saxorum . . . . . . 30 to 79 2. Soft thick-bedded limestones, some beds largely composed of Miliola and Orbitolites : Cardium granulosum, Chama calcarata, Fusus bulbiformis, Turritella imbricataria . 25 to 40 1. Glauconitic limestones and sands with a layer of pebbles at the base, Cerithium giganteum, Cardium porulosum, Ostrea flabellula, Nummulites Icevigatus . . . . . 25 to 70 80 to 145 THE EOCENE SERIES 553 554 STRATIGRAPHICAL GEOLOGY The Calcaire Grossier is succeeded by a set of sands known as the Sables Moyens or Sables de Beauchamp, which are about 30 feet thick. They form the zone of Num. variolarius and N. Heberti, and are also found at Anvers on the Oise, whence Mr. Dollfus has called them Anversian. 16 Near Paris they are sur- mounted by a freshwater limestone, the Calcaire de Ducy, the fauna of which is different from that of the higher limestones. Bartonian. The argillaceous facies of this stage does not appear to exist in the Paris basin, where it is represented by sands alternating with freshwater limestones in the following succession, and having a total thickness of from 100 to 120 feet : 4. Calcaire de Noisy with Limncea longiscata. 3. Sables de Marines with Corbula pisum, Natica ambulacrum, and Voluta athleta. 2. Calcaire de St. Ouen with Limncea longiscata. 1. Sables de Mortefontaine with Avicula, Potamides tricarinatus, and Pot. concavus (=pleurotomoides). 3. Southern Europe After the general uplift which took place at the close of Cretaceous time (see p. 524) the south of Europe seems to have remained for a time above sea-level, but it gradually sank again and was covered by a sea which occupied the whole of the Medi- terranean region. The mountain ranges of the Pyrenees, the Alps, and the Apennines were not yet in existence, and the Eocene or Nummulitic Sea extended from the north of Spain and south of France through the Alpine region to Bavaria and the Carpathian Mountains, and through Italy to the Balkan States, Greece, and Asia Minor. It also occupied parts of Northern Africa. The northern parts of this sea were shallow, and the deposits are mainly arenaceous ; the central parts were deep, and the preva- lent deposits are limestones crowded with species of Nummulites. Thus in Bavaria and near Vienna the Eocene consists of 3. The Vienna Sandstone (upper part), which in Hungary contains Num. striatus and at Haring in the Tyrol includes a thick seam of coal. 2. Vienna Sandstone (middle part) and Kressenberg Beds of Bavaria, sands and marls with a bed of limestone, Num. complanatus, and N. perforatus. 1. Burberg Beds ; greensands with small Nummulites, Orbitoides and Operculina. Probably of London Clay age. Nothing comparable with the lowest Eocene deposits of England and France has yet been found in Central Europe, large parts of which must have been land, for a special feature in the Vienna Sandstone Series is the occurrence of beds of coarse conglomerate THE OLIGOCENE SERIES 555 and of large isolated blocks of granite and gneiss which appear to have come from Bohemia. In Provence the Eocene is represented only by freshwater and terrestrial deposits, but south of it the Nummulitic Sea passed westward through the north of Spain and into Aquitaine. The Provencal deposits may be tabulated as follows, and are specially interesting from the mammalian remains which occur in them : Bartonian. Freshwater clays, sands, and conglomerates. Lutetian. Limestones and marls with Planorbis -pseudo - ammonia, Limncea aquensis, and Bulimus Hopei, with bones of Lophiodon Pachynolophus, Paloplotherium, and Hyracotherium. Ypresian. Sands and mottled clays (freshwater). Sparnacian. Limestone with Physa Draparnaudi and Limncea obliqua. Tlianetian. Limestone with Physa prisca and Melania. Montian. Limestones with Physa montensis. In the Southern Alps, Vicentin and Venetian districts, there is a more varied and complete series, and the succession according to von Hauer, but slightly modified by more recent researches, is as follows : Ba. Lu. Fine-grained sandstone or "macigno." Limestone with Cerithium diaboli. Volcanic tuffs of Ronca. Limestones with Num. complanatus and N. perforatus. Limestones with Alveolina, and fish beds of Bolca. Limestones with Num. ataicus and Lithothamnium. Limestones with flints, Brachiopods, and Nummulites. y . _ Cosina Beds with Melania and Chara and leaves of Dryandra and IBanksia. ^Miliolina limestone, with Cerithium and Anomia. The letters to the brackets stand for Liburnian, Lutetian, and Bartonian, the Liburnian being apparently conformable to the underlying Cretaceous strata, and consequently representing nearly the whole of the Lower Eocene. In the Northern Apennines, according to Professor Sacco, 17 the Eocene Series attains a thickness of 5500 feet. At the base is a group which he classes as Suessonian (about 300 feet), succeeded by a great thickness of sandy and marly beds with marine fossils (Lutetian), surmounted by sandstones, shales, and shaly limestones with few fossils (about 5000 feet). At the top are 300 feet of grey marls with many fossils, including Num. striatus and Orbitoides radians (Bartonian). II. THE OLIGOCENE SEEIES A. NOMENCLATURE AND DIVISIONS The name Oligocene was proposed for this series by Professor Beyrich in 1854, who showed that it contained an important 556 STRATIGRAPHICAL GEOLOGY fauna, differing considerably from that of the typical Eocene strata of England, France, and Belgium, and that it attained a considerable development in Northern Germany. The older classification arranged these beds partly in the Eocene and partly in the Miocene System, but this rendered it necessary to draw a line of separation between the two so-called systems in every area of deposition, and in the midst of a perfectly continuous series. The difficulty of making this separation, and the close connection of the faunas of the groups which were formerly called " Upper Eocene " and " Lower Miocene " respectively, has induced geologists to accept Beyrich's proposal to unite these groups under the name of Oligocene. The following table shows the divisions which have been made in the Oligocene Series of England, Belgium, and France : England. Belgium. France. Strata in Paris Basin. Hamstead Beds. Rupelian. Stampian. Calcaire d'Etampes. Sands of Ormoy and Morigny, etc. Bembridge Beds. Upper Tongrian. Sannoisian. Calcaire de Brie. Marls and Clays. Upper Gypsum. Headon Beds. Lower Tongrian. Ludian. Lower Gypsum Beds. Marine Marls of Ludes. B. LIFE OF THE PERIOD The Oligocene. flora does not differ in any great degree from that of Upper Eocene time ; species of Sequoia are still abundant ; palms are common ; leaves of cinnamons, laurels, figs, camphor- trees, acacias, chestnuts, and other trees occur. The invertebrate fauna is not large and is a direct development from the Eocene fauna, the genera being all survivors from the Eocene. Echiiioderms were not abundant in our northern seas, but were common in the Mediterranean area, especially forms of the genera Echinolampas, Echinocyamus, and Coelopleurus. Of Vertebrates the Mammalia deserve special attention. A large number of new genera make their appearance in the gypsum beds of Paris, which correspond to our Headon Beds and form the Lower Oligocene of English and German geologists. Prominent THE OLIGOCENE SERIES 557 among these are Palceotherium and Anoplotherium, the former allied both to tapirs and rhinoceroses, the latter an ungulate of compre- hensive type, combining characters now found in the tapir, pig, and horse. There were several species of Anoplotherium, the largest being about the size of a donkey and having a long, thick, and strong tail which may have assisted him in swimming. Anchi- therium, regarded as an ancestor of the horse, was no larger than a sheep, and had three hoofs on each foot. Hyopotamus (see p. 529) was a common hog-like animal, and Chceropotamus is supposed to have been a river hog. Dichodon, Xiphodon, and Ccenotherium were hornless forms of antelopes, and Hycenodon one of the early car- Fig. 190. RESTORATION OF PALCEOTHERIUM MAGNUM. nivores resembling the marsupial wolf of Tasmania. In Southern France the jaw of a lemur (Adapis) has been found. In higher Oligocene deposits corresponding to our Bembridge and Hampstead Beds other genera have been discovered, such as Anthracotherium, an ungulate remarkable for its strong canine teeth, and presenting several species which vary from the size of a hog to that of a hippopotamus. This beast and species of Hyopotamus are very abundant in the higher Oligocene. Other genera now appearing are Dacrytherium, Dremotherium, Elotherium (Entelodon}, all ruminants ; the small carnivore Plesictis, and toward the close of the period the genus Protapirus with Ehinoceroses. Proboscidea are represented by Palceomastodon and other genera ; and the Sirenia by HaMtherium. The following are some of the fossils which are characteristic of the British Oligocene Beds, and many of them also occur in France and Belgium. Species restricted to or most abundant in one of 558 STRATIGRAPHICAL GEOLOGY the stages are given under its heading, but the following are more or less common in all three groups : Lamellibranchia. Corbula pisum. Gastropoda. Potamides elegans, Tomichia Duchasteli, Bayania fasciata, Melania acuta, Melanopsis carinata, Natica labellata, Neritina concava, Viviparus lentus, V. angulosus, Plan- orbis obtusus, P. platystoma. Reptilia. Trionyx incrassatus (a freshwater tortoise). Fossils of the Headon Beds and Lower Tongrian Actinozoa. Solenastrea cellulosa, Goniopora ( = Litharea) brockenhursti, Madrepora anglica. Lamellibranchia. In the marine beds are : Cardita deltoidea, Corbula cuspidata, Dosinia ( = Cy there*) incrassata, Pecten belli- costatus, Ostrea velata, 0. prora, Trinacria ( = Trigono- ccelia) deltoidea. In the freshwater beds are : Corbicula obovata, C. cycladiformis, Erodona ( = Potamomya) plana, E. gregaria, Unio Solandri. Gastropoda. In the marine beds are : Ancilla buccinoides, Cancellaria elongata, C. muricata, C. evulsa, Neritina aperta, N. concava, Pisania labiata, Voluta spinosa, V. geminata, V. decora. In the freshwater beds : Potamides concavus, P. ventricosus, P. vagus ( = pseudocinctus), Limnaea longiscata, Melanopsis fusiformis, Stenothyra parvula, Planorbis euomphalus, P. lens, Helix occlusus. Mammalia. Dacrytherium ovinum, Dichodon cervinus, Hysenodon minor, Palseotherium crassum. Fossils of the Osborne and Bembridge Beds Plantce. Chara Lyelli, C. medicaginula, Flabellaria lemanonis, Cinnamomum sp. Lamellibranchia. Corbicula semistriata, C. pulchra, C. obtusa, Ostrea vectensis, Area Websteri. Gastropoda. Glandina costellata, Amphidromus ellipticus, Cyclotus cinctus, Helix d'Urbani, H. pseudo-globosa, Viviparus lentus, Planorbis discus, Melania acuta, M. muricata, Potomaclis turritissima, Potamides mutabilis. Mammalia. Anoplotherium commune, Choeropotamus gypsorum, Palseo- therium magnum, P. medium, Xiphodon gracile. Fossils of the\Hamstead and Rupelian Stage Plantce. Chara tuberculata, Andromeda reticulata, Arthrotaxis sp. Lamellibranchia. Corbula subpisum, C. vectensis, Dosinia (Sinodia) Lyelli, Ostrea callifera, 0. cyathula, Pectunculus obovatus, Lucina Thierense, Leda Deshayesiana. Gastropoda. Potamides plicatus, P. elegans, Tomichia Duchasteli, Bayania fasciata, Stenothyra ( = Nematura) pupa, Voluta Rathieri, Tritonium flandricum. Mammalia. Hyopotamus bovinus, Anthracotherium minus. THE OLIGOCENE SERIES 559 Fig. 191. GROUP OF OLIGOCENE FOSSILS. a. Chara medicaginula. b. Cyrena pulchra. c. Glandina costellata. d. Viviparus angulosus. e. Amphidromus ellipticus. /. Helix d'Urbani. g. Tomichia Duchasteli. h. Potamides elegans. i. Limntea longiscata. k. Planorbis euomphalus. I. Dosinia incrassata. m. Voluta Rathieri. C. THE ENGLISH SERIES Headon Beds. This group is divisible into three portions, Lower, Middle, and Upper, the Middle portion containing marine 560 STRATIGRAPHICAL GEOLOGY and brackish -water shells, while the lower and upper are of fresh- water origin. These beds underlie all the northern part of the Isle of Wight, but except at the western end their outcrop is very narrow. They are well exposed in the cliffs which extend from Headon Hill to Cliff End, and also in Whitecliff Bay at the east end of the island (see map, Fig. 172). In Hampshire they underlie a large part of the New Forest, and the Lower Headon is exposed on the coast in Hordwell Cliff. Recently they have also been found in Dorset (Isle of Purbeck). 18 The section at Headon Hill may be regarded as the typical one, although the total thickness is less there than elsewhere, being only 146 feet. The succession at this place is as follows: 19 Feet. I Variegated clays with Erodona, (Potamomya) gregaria . 21 Limestone with Limncea longiscata and Planorbis euom- phalus .......... 8 Blue clays with the same Limncea and Planorbis . . 5 Limestone with the same fossils and 2 feet of sand below 12 M'rlrll fClays with Potamides concavus, Cyrena obovata, etc. . 7 Miacue I Limestone with Limncea and Planorbis .... 1 HeaaotM Sandy clays with mar i ne fossils 15 ' iSand and clays with Neritina, Cyrena, and Cerithium . 10 I Limestone with Limncea and Planorbis .... 3 Sands and clays with carbonaceous layers and lignite . 20 Clays and sands with two beds of limestone, Vimparus angulosus, Nematura parvula, Limncea, and Planorbis . 25 Sands and clays with firodona plana, etc. . . .19 In Colwell Bay the Middle Headon contains many marine shells, such as Ostrea velata, Dosinia (Sinodia) incrassata, Corbula cuspidata, Nucula headonensis. In Whitecliff Bay the marine beds have expanded to 126 feet, and contain many of the Brocklehurst shells mentioned below, while the Lower Headon freshwater beds are reduced to 28 feet, a fact which suggests that part of these lower beds may be replaced by marine deposits. The marine beds are found also at Brockenhurst and Lyndhurst in the New Forest, where they yield a rich fauna ; 150 species of shells and corals have been found there, about half of which range up from the Barton Beds, but the fauna as a whole closely resembles that of the Tongrian in Belgium and Germany. Some of the commonest species are Rimella rimosa, Cancellaria muricata, Pisania labiata, Pleurotomaria transversaria, Valuta decora, V. geminata, Cardita deltoidea, Dosinia incrassata, Corbula cuspidata, with Solenastrea cellulosa and many other corals. At Hordwell only the Lower Headon is exposed, and the beds are freshwater with a thickness of 83 feet. 20 Many reptilian and mammalian remains have been found at this locality. THE OLIGOCENE SERIES 561 Bembridge Group. The lower beds of this group were separated by Professor Forbes in 1853, under the name of the St. Helens Beds, afterwards changed to " Osborne Beds." In the Memoirs of the Geological Survey they are described as a separate group, but as the fossils they contain are not distinctive, and are all of freshwater species, they are certainly not entitled to rank as a stage. Professor Judd and Mr. H. Keeping group them with the overlying Bembridge Beds under the name of Lower Bembridge marls, 21 and if this is adopted the Bembridge Group will consist of three members, as below : Hamstead whitpMiff andHeadon. Whltechff - 3. Upper marls 75 106 2. Bembridge Limestone ... 15 25 1. Lower marls or Osborne Beds 74 80 164 211 The Lower or Osborne Marls at Headon Hill and Colwell Bay consist of white, blue, and red marls with calcareous concretions which contain Limncea longiscata, Planorbis discus, Viviparus lentus, and other freshwater shells. At Whitecliff Bay they consist of dark-green clays and sands, but are much concealed by landslips, and the best sections occur along the north-eastern coast from Osborne to Nettlestone Point ; here, however, they assume a very different facies, the lower 20 feet consisting of marls, with bands of hard yellowish sandy ragstone and shell limestone full of Viviparus lentus and Melania excavata, the upper 50 feet consisting of green and yellow sands alternating with marls and clays, containing Melania excavata, Planorbis obtusus, Limncea longiscata, and Cyrence. The Bembridge Limestone is a cream-coloured stone with inter- stratified marls ; sometimes it is compact, sometimes quite tufaceous and concretionary ; it is quarried at Headon Hill, at Sconce near Cliff End, and at Binstead near Eyde, and it is also exposed at Bembridge Point and in Whitecliff Bay. Its contents are terrestrial and freshwater, many species of Helix, Amphidromus, Achatina, Glandina, Cyclotus, Clausilia, Limncea, Planorbis, and Viviparus occurring, with oviform bodies which are supposed to be the eggs of the large Amphidromus ; stems and seeds of Chara are common, and mammalian remains are found in the Binstead quarries. The Upper Bembridge Marls extend from the river Yar to Bembridge and Foreland Point, and occupy some space round Osborne, Hyde, and St. Helens. At Hamstead they consist entirely of freshwater beds, chiefly blue and green clays contain- ing Viviparus lentus, Melania acuta, Potamaclis turritissima, Melanopsis carinata, Cyrena semistriata, and C. pulchra. 2 o 562 STRATIGRAPHICAL GEOLOGY In Whitecliff Bay there is a marine bed near tlie base of these marls containing Ostrea vectensis, Dosinia incrassata, Nucula similis, and Mytilus affinis. About 50 feet up is a grey sandy limestone (4 feet thick) containing Amphidromus and Glandina, and the highest beds contain Potamaclis turritissima in abundance. The recent discovery of the Bembridge Limestone and Osborne Marls on Creechbarrow Hill in Dorset is important as showing that the Oligocene Series extended some distance westward, and that the Bembridge lake must have been a fairly large one. The Hamstead Beds cover a large area extending from Yarmouth and Hamstead through Parkhurst Forest and across the Medina Kiver to Brading Harbour. Throughout this tract they maintain a thickness of over 200 feet, and their base is everywhere marked by a band of carbonaceous clay. At Hamstead the succession is given by Mr. Reid as follows : Feet. Marine Beds with Corbula vectensis, C. pisum, Ostrea callifera, Valuta Rathieri, Potamides plicatus, P. elegans . . .31 Leaf and Seed Beds. Red and green clays with carbonaceous layers containing leaves of palms and water-lilies and seeds . .153 White Band. Green clay with white shelly marls ... 6 Nematura Beds. Green and black clays with Stenothyra ( = Nema- tura) pupa, Assiminea conica, Sphcerium Bristovi, Hydrobia (Tomichia) Duchasteli 64 Black Band. Black clay with Fiviparus lenta and Unio . . 2 256 The " white band " is only traceable as far as Parkhurst Forest, but east of the Medina it appears to be represented by a bed of fine sand which, commencing near Newport, thickens eastward till it is about 40 feet and forms a marked feature on the surface. D. CONTINENTAL REPRESENTATIVES As in the case of the Eocene Series, the Oligocene deposits which are found in Belgium and Northern France can be correlated with- out much difficulty with those of our own country, but those of Trance include some beds which appear to be higher and newer than any which occur in England or Belgium. Again, in the central part of France there are lacustrine deposits of great thick- ness containing many fossils of great interest and associated with -volcanic phenomena. 1. Belgium The Oligocene occupies a large area in Belgium, and though the beds differ considerably in lithological character from those of the THE OLIGOCENE SERIES 563 Hampshire basin, the general succession corresponds very closely with that of our country. Belgian geologists have divided them into two stages : (1) the Tongrian, from Tongres near Limbourg, and (2) the Rupelian, from the river Eupel south of Antwerp, on which is situate the town of Boom. Tongrian. There is in most parts of Belgium a break between the Tongrian and the Eocene Series, the Oligocene having a layer of quartz-pebbles at the base which rests indifferently on various members of the lower series. 22 It is divisible into two parts, a Lower and an Upper Tongrian. The Lower Tongrian is of small thickness, from 20 to 60 feet only, and consists of fine sands and sandy clays with Ostrea ventilabrum ( = 0. prona, Wood), Astarte Omaliusi, Cardium porolosum, Dosinia incrassata, Pecten bellicostatus, Valuta decora, Clavella longceva, Cancellaria evulsa, a marine fauna which closely resembles that of Brocklehurst (Middle Headon). The Upper Tongrian is an estuarine or fluvio-marine group, equivalent to the Bembridge Beds, and comprises three members, as below, but the total thickness is only 20 to 30 feet. 3. Sands of Vieux Jones with Potamides plicatus and P. 2. Green clays of Henis with Dosinia incrassata and Mya angustata. 1. Sands of Bauterem with Cyrena semisfriata, Ceritliium Lamarcki, Melania muricata, Bithinia helicella, and B. plicata. Rupelian. This is a purely marine formation, and appears to be the equivalent of our Hamstead Beds. It commences with the sands of Berg, characterised by the abundance of Pectunculus obovatus, with Valuta Rathieri, Cominella Gossardi, Tritonium flandricum, and locally it includes a band of clay with Nucula compta. The Upper Rupelian consists of a thick mass of sandy clay, the Argile de Boom, which contains many fossils, some of the more important being Valuta Rathieri, Fusus elongatus, Triton flandricum, Corbula pisum, and Leda deshayesiana ; it has also yielded many remains of vertebrates, including two Sirenians (Halitherium), several birds, turtles, and a number of fish, chiefly Elasmobranch (Lamna, Odontaspis, Oxyrhina, Carcharodon). 2. Germany Since the name Oligocene is based upon the deposits of this age found in the north of Germany they really form the type of the series, but although they occupy a large area in that country they are to a great extent concealed beneath Miocene and newer deposits. Here and there, however, the Oligocene Beds are 564 STRATIGRAPHICAL GEOLOGY exposed at the surface, as at Latdorf near Bernberg, Egeln near Magdeburg, at Neu Brandenburg, and other places. They are the oldest Tertiary deposits in Germany, and the lowest beds are the sandy clays of Latdorf and Egeln, which contain Ostrea ventilabrum and the other fossils of the Lower Tongrian of Belgium, but the equivalents of the Upper Tongrian are not yet known at those places. On the Prussian coast, near Konigsburg, are other deposits of Tongrian age, the lowest being sands with 0. ventilabrum and a bed containing pieces of amber, which is the resin of Pinus succinifera and other conifers. This amber encloses an immense number of insects and arachnids (about 2000 species), with leaves of Pinus, Sequoia, and other conifer ae Quercus, Acer, Castanea, Magnolia, Ginnamomum, Palms, and other trees. The overlying beds are sands with layers of lignite. Above the Tongrian is the Septarian Clay or Eupelian, which contains the same fossils as that of Belgium, and was clearly formed in a continuation of the same sea. It reaches a thick- ness of nearly 500 feet, but passes eastward into the Stettin sands. To the south-west around Mayence on the Rhine is a small basin of Oligocene deposits which must have been formed in a gulf of the Northern Sea. The lower beds are similar to those of Germany and Belgium, but the Septaria Clay passes up into marl and sand with Gyrena semistriata and Potamides plicatus, above which are other sandy beds with P. plicatus, Mytilus Faujasi, and other brackish-water fossils, with some land-shells (Helix, etc.). 3. Switzerland The northern part of Switzerland was also occupied by a sea during Oligocene time which seems to have been connected both with the Southern Sea and with the Gulf of Mayence. The area occupied by Oligocene and Miocene deposits is that of the great lakes between the Alps and the Jura, from Geneva to the Chiem See in Bavaria, a distance of 300 miles. The deposits are like those of Mayence but are much thicker, the upper or Rupelian part alone having a thickness of nearly 1000 feet and presenting the following succession : Feet. Freshwater beds with Helix Ramondi and PlanorUs dedivis ~\ Marls with Cyrena semistriata and Ostrea cyathula . . 60 Sandstone with Potamides plicatus and plants . . J Septaria Clay (few fossils except Foraminifera) .... 600 Sand with Ostrea longirostris, Pectunculus obovatus, Cyrena semi- striata, Potamides trochlearis, etc. ...... 300 THE OLIGOCENE SERIES 565 Southward toward the Alps these beds pass into a great mass of shales, sandstones, and conglomerates which are called Molasse by the Swiss geologists and Flysch by the Germans and Austrians. Of these sandy deposits about 5000 feet belong to the Oligocene, and the lower part of them is probably of Tongrian age, but they contain very few fossils. " Nothing," says Jukes, " is more calculated to strike the geological traveller on his first visit to Switzerland than the vast deposit of the ' Molasse,' occupying the central region between the Alps and the Jura. This is the country of the great lakes, extending from that of Geneva to that of Constance. The hills by which these lakes are environed have all the rugged and broken character of mountains, and rise in peaks of various altitudes up to that of 6050 feet, which is the height of Ehigi Kulm. These hills, which, if they were not overshadowed by the still loftier Alps, would themselves be celebrated mountains, are composed from top to bottom of beds of sand and gravel, occasionally compacted into sandstones and conglomerate." 4. The Paris Basin Ludian. French geologists are not agreed as to the precise line of division between the Eocene and Oligocene Series, the fact being that there is a gradual transition from one to the other, 23 but it is generally conceded that the Lower Gypsum beds of Montmartre may be correlated with the Headon Beds of Hampshire, and it is convenient to retain the name Ludian (from Ludes, near Eheims) as a designation for these beds. Near Paris they comprise three beds of gypsum, separated by beds of yellow marl which contain a few marine fossils, but in the eastern part of the basin the marls predominate and are more fossiliferous, containing Pholadomya ludensis, Gardium granulosum, Psammobia neglecta, Potamides concavus, etc. The total thickness varies from 60 to 120 feet. Sannoisian. This includes the highest bed of gypsum, which is 65 feet thick at Montmartre and has yielded the bones of fifty- different species of mammals, including Palceotherium magnum, P. medium, Anoplotherium commune, and Xiphodon gracile. This mass of gypsuni may be regarded as contemporaneous with the Lower Bembridge or Osborne Beds and the Bembridge Limestone. Above are a set of marls known as the Supra-gypseous Beds ; the lowest are white and blue marls with a few freshwater fossils, overlain by green marls containing Cyrena semistriata, Dosinia incrassata, Psammobia plana, Potamides plicatus, etc. Above there is another lacustrine deposit, the Calcaire de Brie, consisting 566 STRATIGRAPHICAL GEOLOGY of white marls with beds of siliceous limestone which have been used for making millstones (meulieres) ; total thickness about 100 feet. Stampian. This part of the series consists chiefly of sands which were formerly known as the " Sables de Fontainebleau," but being more fossiliferous near Etampes to the south of Paris are now named after that village. The lowest bed is a marl containing marine fossils, Ostrea cyathula, 0. longirostris, and Corbula subpisum. The overlying sands of Morigny and of Ormoy yield many fossils, including Pectunculus obovatus, Lucina Thierensi, Valuta Rathieri, Tritonium flandricum, and Potamides plicatus, a fauna which correlates them with the Rupelian of Belgium. At the top is a lacustrine limestone with Helix Eamondi and Cyclostoma antiquum, formerly known as the " Calcaire de la Beauce inferieur," but now termed the Calcaire d'Etampes. The thickness of the Stampian Group near Paris is about 250 feet. 5. Central France Extensive lakes were also formed in Central France during the Sannoisian epoch, and continued to exist through the Rupelian (or Stampian) and the greater part of Miocene time, leaving their records in the lacustrine deposits of the Auvergne, Cantal, and Velay. French geologists believe that there were three separate lakes, the largest and most northern occupying the valley plain of the Allier known as the Limagne d' Auvergne, the second being in Velay, and the third in the Cantal ; but the latter is now for the most part buried below the volcanic mass of that district. 24 The deposits of the Limagne are the most complete and accessible ; this plain has an average breadth of 20 miles, and its length from Nevers to Brionde is nearly 100 miles. It consists of nearly horizontal strata of sandstones, marls, and limestones, which often rise into low hills capped by fragments of ancient lava-streams. This tract is bounded both on the east and the west by lofty hills of granite and gneiss which rise to a height of 1600 feet above the plain and 3000 feet above the sea ; and the western plateau supports a chain of extinct volcanoes from which in Miocene and Pliocene times lava-streams flowed into the valley below. The succession of the Oligocene portion of these deposits is as follows (see Fig. 194) : Feet. ( Limestones and marls with Helix Ramondi .... 200 St. -I Marls with Cyprisfaba and Limncea pachygaster ... 60 ( Limestones with Potamides Lamarcki ..... 160 ( Sands and clays with marly partings and a limestone at the Sa. I base, Melania, Limncea, and Cyrena semistriata , . . 200 ( Sandy clays and coarse sands (arkoses) ... . . .150 THE OLIGOCENE SERIES 567 From the Samioisian of the Velay bones of Palceotherium, Paloplotherium, and Entelodon have been obtained. 6. Italy and Dalmatia Parts of Southern Europe were covered by a more open sea than, that of the northern region ; it was, in fact, the Eocene Sea, shallowed and made less extensive by a slow movement of upheaval. Oligocene deposits reach a great thickness in Northern Italy, being found in Liguria, around Turin, in the Vicentin, and along the range of the Apennines, and they have been classified by Professor Sacco in the following manner : 25 Feet. Stampian. Grey sandy marls with few fossils .... 2000 /"Sandy marls and sands with beds of conglomerate, j, . I Nummulites Fichteli, N. intermedius, Natica crassa- tina, Cyrena semistriata, and bones of Anthraco- { therium 6000 SestidM. Sandy marls with Num. Fichteli, Num. vascus, etc. . 60 The lowest beds are conformable to the highest of the Eocene, and the Stampian marls are succeeded by the Miocene. III. GEOGRAPHY AND SEDIMENTATION We have seen (pp. 524 and 535) that the Cretaceous period closed with a general uplift of the whole European region, so that for a long time sedimentation was restricted to a few bays or gulfs, some dependent on the Southern Sea, and one reaching in from the far north to Belgium and the Paris basin. There is some reason for thinking that the Atlantic Ocean began to assume its present form in Eocene time, or at any rate that its eastern trough between Europe and the medial sub -oceanic ridge, from which the Azores rise as mountain-summits, came into existence at this epoch ; and that the depression of this trough was correlative with the uplift of the Western Franco-British region, which was accompanied by such wonderful volcanic eruptions. After these early changes, however, a slow subsidence ensued, and the seas gradually spread their waters over larger spaces till about the middle of the Eocene period, or in Lutetian time, these seas had acquired their greatest extension. There was still land over the greater part of France, Germany, Poland, Russia, and Scandi- navia, but there was a large southern sea covering the Mediter- ranean area, together with the most northern parts of Africa, and stretching eastward through Asia Minor and Syria far into Asia 568 STRATIGRAPHICAL GEOLOGY The northern border of this sea passed through Switzerland, the south of Bavaria, Austria, and Southern Russia. Westward this Nummulitic Sea passed by two channels into the western ocean, which occupied the Central Atlantic area ; one of these channels lying through the south of Spain and Morocco, the other through the north of Spain, the Pyrenees, and Aquitaine, leaving Portugal and Central Spain to form a large island. Moreover, the Atlantic Sea reached northward as far as the latitude of Cornwall, and a narrow channel of communication was established between it and the Northern Sea along the line of the English Channel. The proofs of this are found in a small outlier of Lutetian limestone lying directly on the Chalk near Orglandes in the Cotentin, in the fact of Lutetian fossils being sometimes washed up on the shore at St. Malo, and in a block of Miliolina limestone dredged up from the floor of the Channel 39 miles south of the Eddystone Rock. 26 This early English Channel, however, did not remain open very long, for its closure may be inferred from the great contrast between the faunas of the Bracklesham and the Barton Beds. The Bracklesham or Lutetian fauna has a semi-tropical aspect, and a large number of species are restricted to it, while the Barton ian fauna has a greater resemblance to that of the London Clay, the southern types having become extinct. The movement which cut off the connection between the Western and Northern Seas seems to have been an uplift of the whole western area, for the Aquitanian channel was also closed and the eastern end of the Aquitanian basin converted into an area of marshes and lagoons. Farther east, however, in the Alpine region and in Northern Italy, the sea seems to have maintained its position. With the commencement of Oligocene time a further change took place, consisting principally in the subsidence of Germany, Poland, and Central Russia, thus opening up communication between the Anglo-Gallic Sea and the Asiatic Sea above mentioned, so that many of the same species of shells occur both in Belgium and near Simbirsk in Russia, such as Valuta suturalis, Ostrea flabella, and Astarte Bosqueti. Various minor changes also took place, the sea penetrating into the Central European land by several channels and inlets, in which a variable series of marine, brackish, and freshwater deposits were laid down, while in France some altera- tion of the drainage system led to the formation of several large lakes. It seems to have been about the beginning of Oligocene time that certain important physical changes took place in the British THE OLIGOCENE SERIES 569 region. During the Eocene period all the north-western part of our region formed a tract of high plateau-land on which a remarkable series of volcanic eruptions took place, as briefly described on pp. 544 et seq. The southern part of this western plateau is believed to have drained eastward or south-eastward across England, but after the cessation of volcanic activity crust-move- ments occurred, which resulted in the elevation of the Pennine range and the dome of the Lake District. Dr. Marr 27 has shown that the final uplifts of both of these areas took place in Tertiary times, and I have given reasons for assigning the date to late Eocene or early Oligocene time. 28 It is probable that these uplifts were correlative with the subsidences which formed the trough of the Hebridean Sea west of Scotland, and that of the Irish Channel between Ireland and Wales ; these depressions then becoming broad valleys which drained southward, and thus formed new features in the geography of the British region. Moreover, the Pennine uplift was probably continued southward down the centre of England, forming a watershed which threw oft' streams to the east and the west, and thus initiating the valleys of the Severn and the Thames. Recurring now to the area in which the varied succession of Eocene and Oligocene deposits took place, some remarks may be made on the source of the materials of which they consist. Of course, the sands and clays were carried down by the rivers which drained into the seas from the surrounding land, and the marine limestones are mainly of organic origin. Two kinds of sediment, however, are especially conspicuous ; the one being clays of white, green, or variegated red and white tints, the other consisting of pale-coloured foraminiferal limestones. The clays occur on many horizons, and are generally of a plastic nature ; in England they are found in the Reading Beds, in the Bagshot sands, in the Bournemouth facies of the Brack- lesham Beds, while in the Headon and Bembridge Beds green clays predominate, though white, red, and variegated clays are also of frequent occurrence. Similar clays occur at successive horizons in the deposits of the Paris basin. The white clays have engaged the attention of several observers, the earliest writers attributing them to the detrition of granitic rocks and the decomposition of felspars. Mr. G. Maw, 29 however, in 1867 dissented from this view, regarding them as more probably derived from argillaceous chalks ; Dr. A. Irving 30 also attributed the Bagshot pipeclays to the " decomposition of chalk and the leaching out of the iron by the action of peaty acids." There are, however, strong objections to both these views ; to 570 STRATIGRAPHICAL GEOLOGY the first because the white clays occur at a considerable distance- from the nearest areas of granite. It is true that there may have been granite masses not far south of the Isle of Wight, as Mr. 0. Fisher has pointed out, and there may have been others between Normandy and Devonshire ; still none of them are likely to have been large enough to provide so much pure argillaceous material. Again, with respect to the Chalk, the Eocene deposits rest nearly everywhere on Upper Chalk, which contains a very small percentage of clay. The Chalk-marl could have supplied a larger amount, but the contemporaneous outcrops must also have- been some distance from the areas in which the Eocene Clays were being deposited, and a very large mass of Chalk-marl would be required to furnish the extensive beds of white clay that seem to have been formed. Much of the insoluble residue obtained from marly chalks is colloid silica, and the amount of alumina is generally small, the amount of argillaceous matter varying from 6 to 20 per cent, and much of this is glauconite, not clay. Lastly, there seems no good reason for supposing that the white clays had a different source from the variegated clays of the Reading Beds or of the Oligocene Series. So far as is known, they only differ in colour, and this, as Dr. Irving points out, may be due entirely to the leaching action of peaty acids which would operate on clay derived from any source. He has also shown that the green colouring matter of certain Eocene sands and earths is due to decomposed vegetable matter, while we know that variegated white and red clays are common among lacustrine deposits of many ages all over the world. It seems probable, therefore, that these fine Tertiary Clays are merely the fine muds brought down by rivers and streams from all parts of the country which surrounded the areas of deposition, and that their colour or want of colour is the result of the chemical actions to which they were exposed during their transportation and during the process of deposition, the red tints being due to the oxidation of iron or manganese, the green to the presence of decomposed vegetation, and the white to leaching out of all colouring matter. With respect to the other prevalent kind of deposit mentioned on p. 569, that is more easily accounted for. The foraminiferal limestones are specially characteristic of the Mediterranean area of deposition, where they constitute masses of limestone many hundred feet in thickness. The ordinary " Nummulite limestone " has a pale buff or grey matrix, composed mainly of the debris of decayed and disintegrated Nummulites, in which are embedded more perfect tests of one or more species. The genera Nummulites r THE OLIGOCENE SERIES 571 Orbitoides (sub-genus Orthopragmina), and Operculina, are the largest and most abundant forms ; next to these Alveolina, is perhaps the most important as a rock-builder, and in some districts the Miliolidce occur in great profusion, as in the Miliolite limestone of the Paris basin, of which rock the sub-genera Triloculina and Quinqueloculina are the chief components. KEFERENCES 1 See "Geology of the London Basin," Mem. Geol. Survey, vol. iv. (1872), and " Geology of London " (1889), both by Mr. W. Whitaker ; also the smaller Guide to the Geology of London, by the same author (7th ed. ). 2 C. J. A. Meyer in Quart. Journ. Geol. Soc. vol. xxvii. p. 74. 3 See " Geology of the Isle of Wight, " Mem. Geol. Survey, 2nd ed. (1889). 4 J. S. Gardner in Quart. Journ. Geol. Soc. vol. xxxviii. p. 1, and Proc. Geol. Assoc. vol vi. p. 88. 5 "Bagshot Beds of the London Basin," by H. W. Monckton, Quart. Journ. Geol. Soc. vol. xxxix. p. 348 (1883). 6 Rev. 0. Fisher in Quart. Journ. Geol. Soc. vol. xviii. p. 65. 7 J. S. Gardner in Quart. Journ. Geol. Soc. vol xxxv. p. 209, and Geol. Mag. for 1879, p. 153. 8 Gardner, Keeping, and Moncktou on the Upper Eocene Beds in Quart. Journ. Geol. Soc. vol. xliv. p. 578. 9 The Volcanoes of Great Britain, by Sir Archibald Geikie. 10 A. Marker in Geol. Mag. Dec. 4, vol. viii. p. 506. 11 J. S. Gardner, Quart. Journ. Geol. Soc. vol. xli. p. 84. 12 J. S. Gardner, Quart. Journ. Geol. Soc. vol. xliii. p. 270. 13 Geol. Survey Summary of Progress for 1900. 14 Traitede Geologic, by A. de Lapparent, 5th ed. 1906, p. 1471. 15 Prestwich on the "Correlation of the Eocene Strata in England, Belgium, and France," Quart. Journ. Geol. Soc. vol. xliv. p. 88. 16 See Harris and Burrows, "The Eocene and Oligocene Beds of the Paris Basin " (Geol. Assoc. 1891), and G. Dollfus, Proc. Geol. Assoc. vol. xxi. p. 101. 17 Sacco in Bull. Soc. Geol. France, for 1889, p. 212, and papers in Boll. Soc. Geol. Ital. 1892, 1893, 1896. 18 W. H. Huddleston, Geol. May. for 1903, p. 197, and H. Keeping, Geol. Mag. for 1910, p. 436. 19 See "Geology of the Isle of Wight," Mem. Geol. Survey (1889), and Keeping and Tavvuey in Quart. Journ. Geol. Soc. vol. xxxvii. p. 85. 20 See Keeping and Tawney in Quart. Journ. Geol. Soc. vol. xxxix. p. 566. 21 J. W. Judd, Quart. Journ. Geol. Soc. vol. xxxvi. p. 169, and E. Keeping, Geol. Mag. for 1887, p. 48. 22 See Van der Broeck, Ann. Soc. Geol. Beige, vol. vii. p. 208 (1893). 23 Harris and Burrows (op. cit.} ; also G. Dollfus (op. cit.). - 4 T. G. Bonney, "Volcanic Region of Auvergne," Proc. Geol. Assoc. vol. xvii. p. 191 (1901), and " Livret Guide," Auvergne Geol. Congress, Paris, 1900. 25 Sacco, op. cit. supra. 26 R. H. Worth, Journ. Marine Biol. Assoc. Plymouth, 1908, p. 118. 27 J. E. Marr, Pres. Address, Geol. Soc. (1906), p. Ixxxvi. ^ 8 Building of the British Isles, 3rd ed. p. 378 (1910). 29 G. Maw, Quart. Journ. Geol. Soc. vol. xxiii. p. 387 (1867). 30 A. Irving, Quart. Journ. Geol. Soc. vol. xliii. p. 378 (1887). CHAPTEK XVI NEOGENE SYSTEM MIOCENE AND PLIOCENE SERIES THE Newer Tertiary strata, i.e. those which are newer than the Oligocene Series, have so much in common, both as regards their mode of occurrence and the species of fossils which they contain, that they can hardly be considered to form more than one system, and even when so grouped together it is only in the south and south-east of Europe that they constitute a series of deposits comparable in thickness and extent to those of older systems. As explained on p. 526, the name Neogene will be adopted for this system, and it will be regarded as including the three groups or series which were termed Miocene, Pliocene, and Pleistocene by Lyell. It is true that many geologists, both in this country and on the Continent, exclude the Pleistocene from the Neogene System and place it in what they call a Quaternary or Post- Tertiary System, but the only essential difference between the Neogene and the Quaternary is that the remains of man have hitherto only been found in the latter. It would, however, be rash to say that man did not exist during the later part of Pliocene time, and thus the distinction on which a Quarternary System is based must be regarded as an assumption. No hard-and-fast line can be drawn between deposits of Pliocene and Pleistocene age, and from a purely geological point of view it is more logical and correct to say that Tertiary time merges into the Recent or Historic period than to imagine that Tertiary time ended when man first appeared. When the Miocene, Pliocene, and Pleistocene Groups were established by Lyell the percentage of living species of Mollusca in each fauna was taken as a guide and as a criterion of age, but subse- quent researches have shown that the percentages originally adopted by Lyell cannot be sustained, and that if the princinle is to be retained the proportional numbers must be altered, and further, that much latitude be allowed in certain cases. It may be considered, however, that the Miocene molluscan fauna has only from ten to forty 572 THE MIOCENE SERIES 573 living species, the Pliocene from forty to ninety, and the Pleistocene from ninety to a fauna without species that are more than locally extinct, and finally to one that does not differ from that of the surrounding country or neighbouring sea. Some geologists, however, prefer to take the mammalian faunas as a guide instead of the molluscan, and to take note of the extinction of genera rather than of the introduction of new species. Agree- ment on this question has not yet been reached, and there are differences of opinion as to the respective limits of the three series. I. THE MIOCENE SERIES A. NOMENCLATURE AND DIVISIONS Although Continental geologists have differed on the question of including the stage known as Aquitanian in the Oligocene or in the Miocene, the latter view is now the prevalent opinion in France. It is adopted not only by de Lapparent in the last edition of his Traite de Geologic (1906), but its propriety has been placed almost beyond doubt by Mr. G. F. Dollfus in his recent revision of the typical Aquitanian fauna. 1 His examination shows that the number of species of marine Mollusca common to the Stampian and the Aquitanian of the Bordeaux district is only thirty-four, or about 10 per cent of the total ; while the number of those common to the Aquitanian and the overlying Burdigalian is 164 or 55 per cent, so that the affinity of the Aquitanian fauna with that of the Miocene is five times as great as with the Oligocene. Even when compared with the still later (Helvetian) fauna of Tour- aine a proportion of 38 per cent is found to be common to the two. There is also much difference of opinion with respect to the upper limit of the series, some placing both Sarmatian and Pontian in the Miocene Series, but Fuchs and others regard the Pontian Beds, which include what is known as the Pikermi fauna, as belonging to the Pliocene. In the present state of our knowledge it seems best to draw the line of division between these two groups,, and to consider the Sarmatian as the highest member of the Miocene. Thus defined, the Miocene Series is divisible into five stages : the Aquitanian and Burdigalian having their typical development in the Bordeaux district, the Helvetian in Switzer- land and in Touraine, the Tortonian at Tortona in Italy, the Sarmatian in Austria and Southern Russia. The following table shows the succession of deposits which are attributed to the Miocene in each of these districts. The only representative of the series in the British Isles are the Bovey Beds in Devonshire, which were correctly referred to the Aquitanian 574 STRATIGRAPHICAL GEOLOGY by Heer in 1867, but were supposed to be Eocene by Starkie Gardner in 1879. 2 Mr. C. Keid, 3 however, lias recently shown that the flora is identical with that of the lignites of Wetteran near Bonn on the Ehine. B. LIFE OF THE PERIOD Flora. The Miocene flora does not differ very greatly from that of the Oligocene period, and indicates a climate of sub-tropical warmth, but that of the highest stage furnishes evidence of a certain diminution of temperature and probably of an increase of the winter's cold. It was during this period that palms ceased to be abundant in Central Europe. The earliest Miocene flora is that of the Aquitanian, the Bovey Beds and their equivalents in Northern Germany. This flora includes an abundance of the Conifers Sequoia, Taxodium, and Glyptostrobus, associated with palms of the genera Phcenicites, Flabellaria, and Sabal, many deciduous trees, such as Quercus, Laurus, Vitis, Magnolia, and Liquidambar, and several ferns, Osmunda lignitum, Lastrea styriaca, and Lygodium Gaudini. A fauna of a similar character occurs in the Swiss Burdigalian. Another well-known flora is that of the (Eninghen Beds of .Switzerland, from which no fewer than 475 species of plants were obtained by Professor Heer. In this assemblage palms are rare, ,and the commonest trees are maples, planes, poplars, and elms, with species of Cinnamon, Camphor, Vine (Vitis teutonica), Fig- tree, Cypress (Taxodium) and Glyptostrobus, with a few Proteacece. Looking to the Miocene flora as a whole, Professor Heer remarked that "increased prominence is given to the Japanese types by the abundance of camphor - trees and Glyptostrobi ; to the Atlantic element by the laurels ; to the American types by the numerous evergreen -oaks, maples, poplars, planes, liquidambars, Robinice, Sequoice, Taxodia, and ternate-leaved pines ; and to the types of Asia Minor by the Planerce and a species of poplar (Populus mutabilis). The greatest number and the most important of the types of the Swiss Miocene flora belong, therefore, to a belt lying between the isothermal lines of 59 and 77 Fahr., and in this zone America is now the region which is most correspondent to the natural character of the Swiss Miocene land." Fauna. Among invertebrate animals the Echinoderms are perhaps the most noteworthy, and they are conspicuous members of the marine fauna in the French, Italian, and Austrian regions ; the principal genera are Clypeaster, Echinodiscus (Amphiope), ficutella, Echinolampas, Spatangus, and Arbacia. Nearly all the marine Mollusca belong to living genera, but form 576 STRATIGRAPHICAL GEOLOGY an assemblage which resembles that now found in the Carribean Sea more than that of the existing Lusitanian and Mediterranean provinces, and the same is the case with the Echinoderm fauna. Among Lamellibranchs the prevalent genera were Area, Lucina, Tivela (Grateloupia), Tellina, Cardita, and Ostrea. Of Gastropods Fig. 192. GROUP OF MIOCENE FOSSILS (natural size). a. Cassidea saburon. c. Area barbata. b. Ancilla glandiformis. d. Ostrea crassissima (young). e. Clypeaster altecostatus. there is a great variety, the chief genera being Conus, Fusus, Melongena, Ranella, Murex, Cassidaria, Pleurotoma, Oliva, Ancilla, Turritella, Potamides, Cerithium. The only known representatives of the tetrabranchiate Cephalopoda are the generas Nautilus and Aturia ; the dibranchiate order is represented by Spirulirostra. THE MIOCENE SERIES 577 It is, however, the vertebrates, and especially the Mammalia, which form the most characteristic and distinctive features of the Miocene fauna. When this fauna is contrasted with that of the Oligocene the principal changes are seen to be as follows : (1) the appearance of Proboscidea in Europe, genera Dinotherium and Mastodon ; (2) the replacement of the Palaeotheres and Anthra- cotheres by Tapirs and by true Rhinoceroses, both of the hornless (Aceratherium) and horned type (Rhinoceros) ; (3) the occurrence of Anchitherium, an ancestor of the horse ; (4) the appearance of large cats (Machcerodus) comparable to modern panthers and tigers ; (5) the appearance of monkeys and apes ; (6) appearance of horned deer, Dicroceras ( = Procervulus) (allied to the Muntjac of India) ; (7) the disappearance of marsupials from Europe. The assemblage of animals was thus very different from that of Oligocene times, Aceratherium, Anthracotherium, and Amphicyon being almost the only mammals that survived from the one period to the other. The pig tribe were represented by Palceochcerus, which first appears in the highest Oligocene of the Allier, and by Hyotherium, a closely allied genus. Listriodon was another two- toed and pig-like beast. The carnivores include Machcerodus (the sabre-toothed tiger), Amphicyon (a progenitor of the dogs and bears), Hycenarctos (a bear-like animal), with otters (Lutra), cats, and foxes. Beavers made their dams across the streams ; a gigantic ant-eater (Macro therium) tore up the ant-hills with his great claws ; lastly, true apes made their appearance, several forms being known (Plio- piihecus, Dryopithecus), these being anthropoid apes of large size. The Dinotherium was remarkable for having two large tusks in the lower jaw which curve downwards and are supposed to have been used by the animal to moor himself to the banks of the lakes in which .he lived while he fed upon the herbage. The Mastodon had long tusks in both jaws, and was more like an elephant in build, and probably also in habits. C. THE MIOCENE IN ENGLAND The Bovey Beds rest on Devonian and Carboniferous rocks, and occupy a basin-shaped depression between Dartmoor and the Haldon Hills, extending for about 9 miles from Pullabrook near Bovey Tracey to Kingskerswell near Newton Abbot. They consist in the lower part of thick beds of lignite and of brown clay, and in the upper part of alternating beds of sand and clay, with some layers of lignite. Their total thickness has been estimated at about 600 feet, a boring at Heathfield having been carried to a depth of 526 feet without reaching their base, while west of the place it is believed that higher beds come in ; where the base is exposed at 2P 578 STRATIGRAPHICAL GEOLOGY Staple Hill there is a bed of pebbles and stones overlain by some thickness of sand dipping steeply toward the east (see Fig. 193). 4 The lignites are described by Pengelly and Heer as consisting mainly of Sequoia wood and the compressed fronds of Osmunda lignitum. The leaves and seeds of the following plants are also common : Nyssa, a tree now living in the southern United States ; Quercus Lyelli (an evergreen oak), Oinnamomum (two species), Vitis teutonica and several other species, Palmacites (the prickly cactus), and the water plants Stratiotes Websteri and Potamogeton tenuispina. D. CONTINENTAL MIOCENE 1. France (Aquitaine) Aquitanian. The typical Aquitanian is found in the near Bordeaux, 5 and the typical marine fauna of this stage occurs in the " faluns " or calcareous shelly sands of Bazas and Lariez, some of the commoner species being Ostrea aginensis, Area cardiiformis, Turritella Desmaresti, Protoma Basteroti, Neritina Ferussaci, Cerithium calculosum, C. fallax, Melongena Lainei, M. cornuta, Oliva clavula, and Nassa aquitanica. These beds are overlain by a freshwater limestone containing Helix Morognesi and Planorbis Mantelli ; this " calcaire gris de 1'Agenais " must not be confounded with the " calcaire blanc de 1'Agenais," which forms the top of the Oligocene, and is character- ised by Helix Ramondi and Cyclostoma antiquum (see p. 566). Burdigalian. The typical Burdigalian fauna is that of Leogiian ; some of the characteristic species are Turritella terebralis, Protoma cathedralis, Tudicla rusticula, Pecten burdigalensis, Clypeaster margi- natus, and Scutella subrotunda. On the Sancats horizon Agaronia ( = Oliva) Basteroti and Dorsanum baccatum are common shells. Helvetian. The faluns of Salles furnish another fauna in which figure Valuta (Scaphella) Lamberti, Cassidea saburon, Pecten solarium, P. scabrellus, Gardita Jouanneti, and Panopcea Menardi. Eastward these marine sands pass into a freshwater formation, the molasse of Armagnac, consisting of marls of various colours (yellow, green, and red), limestones and calcareous sandstones, the whole said to be 1000 feet thick. The lower beds yield bones of Mastodon angustidens, Rhinoceros sansannensis, and Amphicyon major-, the upper beds those of Mastodon tapiroides, Dinotherium giganteum, and Anchitherium aurelianense. 2. Orleanais and Touraine Aquitanian. The deposits of this age in the north of France are entirely of freshwater origin, and comprise the larger part of THE MIOCENE SERIES 579 the formation which was formerly known as the Calcaire de la Beauce, from which the Calcaire d'Etampes has been separated (see p. 566). The Miocene part is now called the Calcaire de 1'Orleanais, and in the Gatiiiais district the two are divided by a band of greenish marl or clay, which is taken at the base of the Aquitanian. The thickness of these marls and limestones is about 100 feet, and with the Calcaire d'Etampes they overstep all the older beds both southwards and westwards so as to lie directly on the Cretaceous and Jurassic rocks. Their characteristic fossils are Helix aurelian- ense, H. Morognesi, Limncea pachygaster, and Planorbis solidus, with remains of Anchitherium aurelianense. Burdigalian. The Orleanais Limestone is overlain by an extensive deposit of granitic sands, the Sables de 1'Orleanais and the Sables de la Sologne, which are divided by a band of white and green marls with Melania Escheri. The sands have been derived from the contemporaneous erosion of the Central Plateau, and out- lying patches of them extend westward to Blois and Poitiers and northward to Rouen and Havre. The thickness of the group is about 90 feet, and the lower sands contain remains of Mastodon angustidens and two other species, Dinotherium Cuvieri, Rhinoceros aurelianense, and Amphicyon. Helvetian. This does not occur within the limits of the Paris basin, but consists of a number of separate tracts and patches of shelly calcareous sand around Blois, Tours, and Angers, with others to the north-west near Rennes and Dinan in Brittany. These beds were termed Falunian by d'Orbigny from the peasant's name of falun for this kind of material, which resembles that known as crag in the east of England. The beds are marine, and mark an invasion of the western ocean. The most typical fauna is found near Pontlevoz and Savigne, and some of the commoner fossils are Area turonica, Ostrea crassissima, Pecten scabrellus, Lucina Dujardini, Trochus miocenicus, Potamides pictus, Conus (Conospira) Dujardini, and Nassa limatula. Tortonian. In this western area (Maine and Anjou) there is a still newer set of deposits, distinguished by the abundance of Bryozoa with fragments of Lithothamnium (a calcareous alga), many Echinoderms of the genera Amphiope, Scutella, and Echinolampas, and the teeth of large sharks, Gar char odon and Oxyrhina. 3. Auvergne and Velay Aquitanian. Since the limestones with Helix Ramondi are now referred to the Oligocene, there is little to represent the Aquitanian in this part of France, and M. Dollfus believes that there is a gap (lacuna) or want of continuity in the series of lacustrine deposits. 6 580 STRATIGBAPHICAL GEOLOGY ; -HI' I - tJ m R I ll III III- I.S" II -^ \\ =0 & !yi o gS M i .! ll< S so ^ 2S > = v n THE MIOCENE SERIES 581 In many places the Helix limestone is directly overlain by marls with Melania Escheri, but on Gergovia hill near Mardogne there is a sandstone containing the remains of plants which seem to be an Aquitanian assemblage. Burdigalian. This is represented by rnarls and sands containing plant remains and a few shells such as Melania Escheri, Melanopsis callosa, Cyrena and Unio. Above these beds are basaltic lava flows and thick beds of volcanic ash. 7 The sands of Givreul near Moulins, which contain bones of Dinoiherium, also belong to this stage. Helvetian. No beds of this age have yet been recognised. Tortonian. The deposits referable to this stage are also scanty, but beds of coarse sand and clay near Aurillac have yielded remains of Dinotherium giganteum, Hipparion gracile, and Machairodus cultridens. In the Cantal also are extensive deposits of shaly clays and lignites associated with volcanic ash and basaltic lavas which contain the same Mammalian fauna and plant-remains analogous to those of (Eninghen mentioned hereafter. 8 4. Switzerland Aquitanian. The only deposits in the northern plain which can be attributed to this stage (as now restricted) are the gypsiferous beds near Lausanne which overlie those containing Helix Ramondi, and in which are found other species of Helix with Planorbis, Neritina, Potamides, and Cyrena, and bones of Anthracotherium. Part of the " red molasse " is probably of the same age. Burdigalian. Near Lausanne the above are succeeded by the "grey molasse," a set of sandstones and marls no less than 1000 feet thick, mainly of freshwater origin and containing the remains of a flora in which palms (Sabal, Phcenicites, and Flabellaria) are associated with figs, laurels, acacias, maples, cinnamon, and some Proteacece. Helvetian. East of Lausanne the highest freshwater beds appear to pass into estuarine deposits with Ostrea crassissima and Potamides lignitarum, and these are covered by the shelly marine molasse which is more fully developed in Freiborg and Berne. In St. Gall and Appenzell there are beds of an intermediate character consisting of red marly sands with bands of conglomerate, the sands containing Cardita Jouanneti, Panopcea Menardi, Pectunculus pilosus, Conus ventricosus, etc. To this. stage also belong a great part of the conglomerates of the Nagelfluh, which contain pebbles of various rocks not now occurring in the Swiss Alps. Tortonian. The (Eninghen Beds, consisting mainly of shaly limestones, occur on the northern border of the Lake of Constance, and have yielded an enormous number of plants, insects, fish, and 582 STRATIGRAPHICAL GEOLOGY reptiles. Heer described nearly 500 species of plants, and the assemblage indicates a cooler climate than that of earlier Miocene time. It contains a mixture of generic forms which are now characteristic of widely separated countries (see ante, p. 574). Among insects wood -eating beetles are particularly abundant. With these are remains of fish and freshwater tortoises, a huge salamander (Andrias Scheuchzeri), which was taken for a human skeleton when first discovered ; also bones of Dinotherium giganteum, Mastodon tapiroides, Palceomeryx eminens (a ruminant), Galecynus (Eningensis (a fox having affinities with the civet cat), hares, beavers, and many small carnivores. 5. Austria The Miocene has a wide extension in Austria, and consists there mainly of marine deposits which occupy the Vienna basin and extend thence eastward to Moravia and the Carpathian Mountains. Their thickness varies in different places, but the average may be estimated at about 1500 feet. Aquitanian. The lowest beds are the sands and clays of Horn and Molt containing Potamides plicatus, Pectunculus Fichteli, and Mytilus Haidingeri, which seem to represent only the higher Aquitanian. Burdigalian. To this stage may be referred the sands of Gaudernsdorf containing Turritella turris, T. cathedralis, and other fossils of the Leognan fauna, also the molasse of Eggensburg con- sisting of sand and Bryozoan limestone. Of about the same age are the gypsiferous marls known as the " Schlier " containing Pecten denudatus, Cryptodon angulatus, Solemya Doderleini, and Aturia Aturi. Helvetian. The Grund Beds are sandy marls which contain a fauna like that of the faunas of Touraine, and the Baden Tegel are marly clays of about the same age but containing rather a different assemblage of shells, species of Pleurotoma being more abundant, with Cardita Jouanneti, Ancilla glandiformis, Cassidea saburon, Pecten cristatus, and Corbula gibba. Tortonian. This seems to be represented by the Leithakalk, a limestone consisting largely of Lithothamnion and Foraminifera (A.mphistegina and Triloculina), but in some beds Bryozoa are very abundant and in others Corals with many bivalve shells (Pecten latissimus, etc.), large Echinoderms (Clypeaster altecostatus, C. marginatus\ and teeth of sharks (Carcharodon, etc.). Sarmatian. This consists mainly of sand and soft sandstone, but includes some beds of sandy limestone and marl, the whole being about 800 feet thick, but reaching 1200 feet in Galicia. THE MIOCENE SERIES 583 Shells are exceedingly abundant, but the number of species is not large, and the fauna is that of an enclosed sea, with much cooler water than that of the older Mediterranean. The Cones, Olives, Cypreas, and Ancillas have all disappeared, and the commonest shells are species of Cerithidce Potamides pictus, Cerithium rubiginosum, Potamides disjunctus with the bivalves Tapes gregarius, Mactra podolica, Ervilia podolica, and Cardium obsoletum The flora is similar to that of (Eninghen, and the Mammalian remains include Mastodon angustidens, Anchitherium aurelianense, and Listriodon splendens. 6. Italy The Miocene Series of Italy is thicker than that of Austria and must have been found in a deeper sea, the floor of which was probably a subsiding area, for the beds attain a thickness of about 6500 feet. The Aquitanian is represented by the Schio Beds of the Vicentin district, which are marls from 700 to 1000 feet thick containing the Foraminifer Lepidocyclus with Scutella subrotunda and many kinds of fish. In Liguria its equivalent probably exists in the sandy marls at the base of the so-called Laiighian Group. Surdigalian. The typical equivalent of this is found near Langhe in Liguria, where sandy marls with Pectens (solarium and burdigalensis) pass up into blue marls with Aturia Aturi, Solemya Doderleini, and many Pteropod shells ; the whole having a thickness of 2000 feet (Mayer). Helvetian. This stage is described as having a thickness of over 3000 feet in Liguria and as divisible into three groups, a lower consisting of whitish marls full of Pteropoda (1300 feet), a middle of sandy molasse with Bryozoa and Pecten ventilabrum (650 feet), and an upper of nullipore beds with Pectens and Oysters (300 feet). The Superga Beds of Turin belong to the same stage, and contain the fauna of the Touraine faluns. 9 Tortonian. Succeeding the above are about 650 feet of blue marls with an abundant shell fauna, rich in species of Pleurotoma and Conus with Valuta rarispina, Ancilla glandiformis, Apollon marginatus, Trochus patulus, and corals of the genera Turbinolia and Stephanophyllia. The assemblage indicates a sea that was rather warmer than the Mediterranean of the present time. Sarmatian. In Liguria the Tortonian is overlain by about 60 feet of marls and sands containing a limited fauna like that of the Austrian beds, Potamides pictus, Cerithium rubiginosum, Pecten cristatus, etc. In other parts of Italy, on both sides of the Apennines, in Calabria and in Sicily, there are extensive beds of 584 STRATIGRAPHICAL GEOLOGY siliceous earths (tripoli), composed largely of Diatoms and Radiolaria, with a fauna containing some Helvetian survivals, such as Cardita Jouanneti and Pecten aduncus, mixed with the characteristic Sarmatian species of Tapes and Ervilia. 7. Belgium and Germany Our survey of the European Miocene would not be complete without a brief notice of the Belgian and German, equivalents. These are of small thickness, and comprise only two stages, the Bolderian, which probably represents the Helvetian, and the Anversian, which corresponds with the Tortonian or Sarmatian. The Bolderian, or sands of the Bolderberg, about 30 feet thick, and contain a fairly large fauna, including Conus Dujardini, Murex Nysti, Ancilla obsoleta, Terebra Basteroti, Panopcea Menardi, and Venus multilamella. They rest on the Eupelian Clay, and have at their base a pebble -bed containing blocks of septaria derived from that clay and often pierced with P/io/as-borings. The sands of Anvers (Anversian} are dark glauconitic sands about 20 feet thick, and contain layers almost entirely composed of the shells of Pectunculus pilosus ; other common species are Pirula condita, Turritella subangulata, and Corbula striata. A gulf of the Bolderian Sea seems to have extended into the north-west of Germany through Schleswig-Holstein, Mecklenburg, and Hanover, as far as the Teutoburg and Oldenburg. The deposits consist of sand at the base overlain by micaceous clay, and the commoner fossils are Terebratula grandis, Area diluvii, Pectunculus pilosus, Isocardia cor, Pecten Janus, P. decussatus, Conus Dujardini, and C. antediluvianus. E. HISTORY OF THE MIOCENE PERIOD During the course of this period a succession of uplifts and terrestrial disturbances took place by which great changes were produced in the geography of Europe. The first change, i.e. that from Oligocene to Miocene conditions, seems to have been a general uplift, causing a general retreat of the seas, both northern and southern, and raising the greater part of Northern Europe into dry land. At the same time it is clear that the elevation of this land was not great, and that its lower levels in France, Switzerland, and Germany were occupied by large lakes. The deposit formed in these lakes preserves the remains of a flora which indicates a mild semi-tropical climate, inducing a rich vegetation which supported a large mammalian fauna. These conditions seem to have continued through Burdigalian time, at any rate in Northern Europe, but in the southern region, THE MIOCENE SERIES 585 and more especially in the south-east, subsidence occurred which carried the sea northward through Provence into Dauphin e and Savoy, through parts of Switzerland, and through Austria and Hungary to the Carpathian mountains. The history of this epoch has not yet been very clearly deciphered, and there may have been too great a desire on the part of some continental geologists to find an equivalent of each stage in every region. In other words, it is not certain whether all the stages recognised in France are repre- sented in Switzerland and Austria ; and in some districts there are indications of a break at the base of the Helvetian stage. Suess divided the Miocene Series of Central and Eastern Europe into two groups, which he called the first and the second " Medi- terranean Stages," and he came to the conclusion that the main upheaval of the Alpine chain took place in the interval between them. Subsequent researches have tended to confirm this view, and it is generally believed that it was at the close of the Burdi- galian or the beginning of Helvetian time that the greatest orogeiietic movements took place. By these movements the rocks of the Alpine ranges were bent into sharp flexures and overfolds, and whole tracts of strata were thrust northward over and along planes of dislocation. These pressures ridged up the older rocks of Switzerland and Austria into mountains upon which rain and running water came into active operation, and some of the results are exhibited in the Nagelfluh conglomerates of Northern Switzer- land (see p. 581). According to Suess, while the Alps were being ridged up in the manner above mentioned, " in other places, as in parts of Tuscany, Austria, Western Hungary, and Styria, great in -sinkings were formed, and in these lie leaf -bearing lignitiferous beds which are im- mediately subsequent in age to the Schlier. It is about this time that the Alpine basin of Vienna was broken in, that the Alps were separated from the Carpathians, that the eastern downbreak of the Alps with the gulf of Gratz was formed, that the way was paved for the subsequent discharge of the Danube to the east, and that the Tuscan part of the inner Apennine depression was formed or at least indicated. It is probable that most of the inner Carpathian subsidences belong to this period." It seems more likely, however, that these subsidences were of subsequent date, and led to the extension of the Helvetian Sea over these areas. At the same time in Northern Europe the effect of these crust- pressures showed itself along the old lines of Armoricaii and Hercynian folding, a new series of flexures was produced along these lines across the south of England and through the north- east of France, Belgium, and Germany. It was by these movements 586 STRATIGRAPHICAL GEOLOGY that the London, Hampshire, and Paris basins were formed and separated from one another, and it is specially noticeable that the sharpest flexures are along the southern sides of the Hampshire and London basins. The well-known monocline of the Isles of Wight and Purbeck is one of these flexures, and that of the Hog's Back ridge west of Guildford is another. The general result was to elevate the whole of Northern Europe into high ground 011 which no large areas of deposition seem to have existed. After these disturbances there seems to have been a partial sub- sidence ; the Atlantic Sea of Helvetian time invaded the west of France, and occupied not only part of the Aquitanian area but also that which is now the valley of the Loire, forming a gulf which reached as far up as Blois. The sea also passed northward by Rennes to Gahard and Dinan, so that the central and western parts of Brittany were probably reduced to the condition of an island. The Southern Sea also spread northward through Provence and Dauphinq to the Jura, and thence round the northern side of the Alps to Austria, through parts of Bavaria to Bohemia and Poland, and eastward through Galicia and across the south of Russia. There must also have been some connection between the French or Swiss gulfs and the Belgian Sea, though by what route has not yet been definitely ascertained ; it is certain, however, that such shells as Conus Dujardini, 0. antediluvianus, and Terebra Basteroti must have come from the south. The concluding phase of this period seems to have been pro- duced by a general uplift of the southern region, which raised parts of the Mediterranean area into dry land and led to the formation of the shallow Sarmatian Sea, a large enclosed sea which extended from the Danubian basin through the Pontic area to the Caspian region, with a gulf that stretched southward into the jEgean area. The two main basins of the Mediterranean seem also to have been occupied by shallow seas which covered parts of Italy and Sicily and reached westward to the neighbourhood of Barcelona in Spain, where sands and marls with Cerithium pictum and Mactra podolica are found, but the rest of Spain appears to have been part of the Continental area, and the westward opening of the Mediterranean into the Atlantic lay through Morocco. II. THE PLIOCENE SERIES A. SUBDIVISIONS AND NOMENCLATURE In most parts of Europe the Pliocene deposits are of small thickness and of no great extent, and were mostly accumulated in THE PLIOCENE SERIES 587 shallow seas not far removed from the existing coast-lines of the Continent. Of such deposits good examples exist in England and Belgium. The only areas where marine deposits of Pliocene age attain any great thickness are Holland, Italy, and the south-east of France, but there is also a large area in the south-east of Europe, where the Sarmatian Beds (see p. 573) are succeeded by another set of deposits formed in a similar inland sea. These Pontian Beds are only represented in Western Europe by a few isolated lacustrine and terrestrial accumulations which yield a similar mammalian fauna. The great extent and thickness of the Italian Series (over 2000 feet), and the abundant marine fauna which it contains at suc- cessive horizons, have caused it to be regarded as the type of the Pliocene in the south of Europe. It has been divided into three stages named after Italian localities, and those names have been adopted in France. In the northern region, on the other hand, names have been given to small subdivisions which have no higher stratigraphical value than zones, and are only useful in the detailed correlation of the sequence found in one area with that of another. In a former edition of this book I grouped these zones into Lower, Middle, and Upper Stages, and as these correspond very closely with the major divisions of the Mediterranean Series, the same nomenclature will be employed in the present edition. With regard to the deposits of the eastern inland sea, there is no reason to suppose that they are older than the lowest marine Pliocene Beds of Italy, i.e. those to which the name Messiniaii was given by Professor Sacco ; they may therefore be included in the Lower Pliocene group or stage, and the divisions which have been made in different countries may be correlated as follows : TABULAE, CORRELATION OF THE PLIOCENE SERIES Stages. England. Belgium and. Holland. Provence. Italy. Austria and Greece. 53 ( Cromer Beds ? Absent. Sands \ 1 and with Elephas ! Sicilian. Sicilian. t>l Upper Crag. meridionalis J Middle Red Crag, Walton Crag. Amstelian. Scaldisian. Sands with Mastodon. Astian. \ If Coralline Crag. Lenham Beds. Casterlian. Diestian. Congeria Beds. Plaisancian. Paludina Beds. 31 Absent. Absent. Luberon Beds. Messinian. Pontian. 588 STRATIGRAPHICAL GEOLOGY B. LIFE OF THE PERIOD Plants. The flora of the Pliocene is transitional between the luxuriant, more or less, evergreen vegetation of the Miocene period and the temperate flora of modern Europe. In Southern Europe the plant-assemblage was similar to that of CEninghen, but without palms or cactuses. Later plant -beds give evidence of a gradual refrigeration of the climate, and toward the close of the period, if we may judge from the plants of the Cromer Beds, the climate of Northern Europe cannot have been very different from that of the present day. Fauna. Of the Invertebrata the majority of the species are either identical with, or merely varieties of, species now living, and the proportion of extinct species gradually diminishes from a percentage of 36 to 40 in the lowest beds to one of 10 or 11 in the highest beds. Very few of the fossils belong to extinct genera, but Congeria may be mentioned as a genus of Dreissensiidoe which is not now living. Among the Vertebrata, however, many in the Lower Pliocene belong to extinct genera, and the majority of the species even in the higher stages are extinct. With the advent of the Pontian epoch and its southern land-connections with Africa and Asia, a new mammalian fauna invaded the plains of Europe. This included large herds of antelopes and gazelles, varying in size and form, and for the most part allied to those of Africa. With these was an animal of intermediate form between the giraffe and antelopes (Helladotherium\ allied to the short -necked giraffe or okapi recently discovered in Africa. Herds of small wild asses (Hipparion), intermediate in structure between Anchitherium and the horse, also appeared in great numbers. Numerous carnivores followed and preyed on the herbivorous animals, large cats and panthers, including Machcerodus (a tiger), Hycena, Ictitherium (allied to the civet cats), Hycenictis (combining characters of cats and hyaenas), and several species of true cats (Felis). The higher Pliocene is distinguished by the appearance of elephants (Elephas), oxen (Bovidce), true horses (Equus), bears (Ursus), and wolves (Ganis lupus), and by the abundance of deer (Gervidce). In France and Italy the Pliocene deposits are characterised by Mastodon (Tetralophodon) arvernensis, Rhinoceros (Diceros) mega- rhinus, R. (Cwlodonta) etruscus, Elephas meridionalis, Hippopotamus major, Capreolus cusanus, Elaphus perrieri, Axis pardinensis, with Machairodus cultridens, Hycena striata, bears, and wolves. The following are some of the marine fossils which are charac- teristic of the several divisions of the Pliocene Series, but as the THE PLIOCENE SERIES 589 faunas of the Northern and Southern Seas were somewhat different, those restricted to the southern region are indicated by an asterisk : Fig. 195. GROUP OF CORALLINE CRAG FOSSILS. a. Echinus Woodwardi. e. Cardita senilis. b. Temnechinus Woodi. /. Voluta Lamberti. c. Terebratula grandis. g. Brissus unicolor ( = Scillab). d. Astarte Omalii. Ji. Fascicularia aurantium. Fossils of the Lower Pliocene Ecliinodermata. Echinus Woodwardi, Temnechinus Woodi, Brissus unicolor. Bryozoa. Cupularia canariensis, C. porosa, Theonoa ( Fascicu- laria) aurantia, Alveolaria semiovata, Salicornaria crassa. 590 STRATIGRAPHICAL GEOLOGY Brachiopoda. Terebratula spondyloides ( = T. grandis). Lamellibranchia. Area (Anadara) diluvii, Gardium papillosum, Tellina Benedeni, Cardita senilis, Astarte mutabilis, A. Omalii, Cyprina rustica, Isocardia cor, Congeria simplex,* C. subglobosa,* Pecten benedictus,* Ostrea cochlear,* 0. princeps. Gastropoda. Conus (Conorbis) Dujardini, Pleurotoma (Clathurella) consobrina, Terebra acuminata, Tritonium (Ranularia) heptagonum, Voluta (Scaphella) Lamberti, Pyrula reticulata, Fusus lamellosus, Nassa (Amycla) serni- striata,* Turritella subangulata.* Fig. 196. GROUP OF RED CRAG FOSSILS. a. Neptunea contraria. b. Cassidaria bicatenata. e. Buccmopsis Dalei. d. Cardium Parkinsoni. e. Pectunculus glycimeris. /. Carcharodon megalodon. Fossils of the Middle Pliocene Actinozoa. Balariopliyllia calyculus. Echinoderma. Echinocyamus pusillus. Lamellibranchia. Astarte obliquata, Dosinia exoleta (lentiformis), Cardimn Parkinsoni, C. angustatum, Lucina borealis, Mactra constricta, M. (Spisula) ovalis, Tellina (Macoma) obliqua, Pecten jacobaeus.* Gastropoda. Nassa (Uzita) reticosa, Neptunea contraria, N. antiqna, N. despecta, Purpura lapillus, P. tetragona, Coluni- bella (Anachis) sulcata, Natica hemiclausa, Scala groenlandica, Tritonofusus gracilis. Fish teeth. Carcharodon megalodon. THE PLIOCENE SERIES 591 Besides the above the following are equally common in the Coralline and Red Crags Astarte Burtini, Mactra (Spisula) arcuata, Pecten opercularis, Pectunculus glycimeris, Venus casina, Cassidaria bicatenata, Buccinopsis Dalei, Trochus Adansoni, Turritella in- crassata, Natica millepunctata, Nassa labiosa. Fig. 197. FOSSILS OF THE RED CRAG. o. Balanophyllia calyculus. b. Echinocyamus pusillus. c. Astarte obliquata. d. Cardium angustatum. Nassa reticosa. e. Mactra constricta. /. Columbella sulcata. g. Natica hemiclausa. h. Purpura tetragona. Fossils of the Upper Pliocene La'niellibrancMa. Astarte borealis, A. compressa, Cardiura edule, Cyprina islandica, Yoldia oblongoides, Nucula Cobboldise, Mya truncata, Mactra (Spisula) subtruncata, Tellina (Macoma) calcaria, T. (Macoma) obliqua, T. (Macoma) prsetenuis, Scrobicularia plana. Gastropoda. Potamides tricintus, Buccinum groenlandicura, Melam- pus pyramidalis, Littorina littorea, Natica catena, Purpura lapillus, Turritella comraunis. 592 STRATIGRAPHICAL GEOLOGY Most of these species occur also in the Chillesford Beds, with the addition of Cardium grcenlandicum and Yoldia lanceolata, and most of them are found in the Weybourn Crag, which latter is specially characterised by the incoming of Tellina (Macoma} balthica ( = T. solidula). The Norwich Crag has also yielded the following mammalia : Elephas antiquus, Cervus carnutorum, Equus stenonis, Trogontherium Guvieri (an extinct beaver), Gazella anglica, Arvicola intermedius. C. BRITISH PLIOCENE DEPOSITS The English Pliocene was formerly divided into two groups, a lower and an upper, it being supposed that there was a decided break between the Coralline Crag and the Red Crag, both faunally and physically ; but recent researches have shown that most of the common fossils of the Coralline Crag occur also in the more southern part of the Red Crag, while it also appears that the higher crags to the north of Aldeburgh are quite as distinct from the Red Crag as that is from the Coralline. Hence the series seems to fall naturally into three groups or stages, which may be tabulated as below : (Cromer Beds ( = Forest Bedn UPP SSdBel '. \ -bout 200 feet. ^Norwich Crag . . J ( ( Butley Crag ) Middle-! Red Crag^ Newbourn Crag ^combined up to 70 feet. ( Walton Crag J Lo er /Coralline Crag . . . about 70 feet. (Lenham Beds . . . thickness unknown. The Lenham Beds are only found in Kent, and the Coralline Crag only in Suffolk, though in Cornwall there are sands which are probably of about the same age as the latter. The newer crags occupy parts of Essex, Suffolk, and Norfolk, lying unconform- ably either on the London Clay or the Chalk. The only other traces of the Pliocene period in England are a deposit of sand in Dorset containing bones of Elephas meridionalis^ and a cave in Derbyshire which has yielded bones of that and other Pliocene animals. 11 In Scotland shells of Upper Pliocene species occur in the Glacial deposits of the eastern coast, and indicate a northern extension of these Crags. Lenham Beds. The oldest Pliocene deposits in England are certain patches of ferruginous sand which occur at intervals along the Chalk downs of Kent from the heights above Folkestone to those above Maidstone, most of them being about 600 feet above THE PLIOCENE SERIES 593 the sea. They were first described by Professor Prestwich in 1857, and more fully explored by Mr. C. Reid in 1886. 13 The Lenham Beds are mere remnants of a deposit which must originally have had a wide extension, not only in England but eastward through Belgium. These remnants have been preserved because they have subsided into deep pipes and hollows caused by the solution of the underlying Chalk. The hollows are lined with brown clay full of unworn flints, and these clay walls enclose a set of sandy deposits, yellow and red sands, reddish fossiliferous sandy ironstone, and greenish sand with scattered flint pebbles. From these beds, and chiefly from the ironstone, Mr. C. Reid obtained sixty-seven species of Mollusca, and all but fifteen have been found in the Coralline Crag, so that there cannot be much difference in their relative age in. spite of the great difference in relative level. There are, however, some species which do not occur in the Coralline Crag, but are found in the older Pliocene or later Miocene Beds of Europe ; these are Area diluvii, Gardium papillosum, Terebra acuminata, Pleurotoma consobrina, P. Jouanneti, the first two shells being very abundant. On the whole, therefore, the Lenham Crag is probably older than the Coralline, and Mr. F. W. Harmer thinks that it was considerably older. 13 Some patches of similar sand occur farther on the Surrey downs, but no recognisable fossils have yet been found in them, and it is consequently uncertain whether they are of the same age. . The nodule bed at the base of the Suffolk Crags contains many rounded lumps of tough brownish sandstone which are known as boxstones, and have evidently been derived from an older deposit. Many of these contain fossils, and some of them are species which occur in the Coralline Crag, but two are only known from the older Pliocene of the continent ; these are Gonus Dujardini and Valuta auris-leporis. Coralline Crag. This crag is only found over a small area in Suffolk between Aldeburgh and Boyton, and in three other small isolated patches at Sutton, Ramsholt, and Tattingstone (see map, Fig. 198). It rests on an eroded surface of the London Clay, and its basement -bed (as seen at Sutton) contains a remarkable assemblage of materials derived from older deposits, and chiefly from the London Clay ; the pebbles are chiefly phosphate nodules of a dark-brown, colour, but with these are fragments of septaria, a few small pebbles of quartz and flint, and the boxstones which have been mentioned above ; there are also Crustacea and fish teeth derived from the London Clay, reptilian vertebrae from the Oxford Clay, mammalian teeth of doubtful age, many Cetacean 2Q THE PLIOCENE SERIES 595 bones and teeth, and some sharks' teeth which are probably of Pliocene age, such as Odontaspis hastalis and Carcharodon Rondeleti. Above the nodule bed there are yellowish marly sands and fine quartz sands, with layers of comminuted shells, from 30 to 40 feet thick, and these form the lower portion of the Crag ; above them .at Sutton, Gedgrave, and Aldeburgh come about 36 feet of a soft porous yellowish-brown calcareous rock, consisting largely of com- minuted shells and fragments of Polyzoa, which are often arranged in oblique lamina?, while the bedding is of that irregular kind due to the action of rapid and changeable currents. This soft limestone has probably been formed by the infiltration of carbonate of lime from the overlying Red Crag. Fossils are abundant in both portions. The Mollusca of the Coralline Crag (if they are not drifted) indicate a sea of some depth (40-60 fathoms), with warmer water than that of the existing North Sea. According to S. V. Wood's summary in 1874 the total number of species was 391, of which 142 were extinct, while 249 are identical with recent forms. Of these living species 205 occur in the Mediterranean, while only 174 occur in British seas, 50 being Mediterranean species which .are not British, and 20 British species which are not Mediterranean. Wood has also pointed out " that the most abundant and therefore the most characteristic species of the Coralline Crag are southern : species unknown to British seas, and that among the 154 which occur both in British and Mediterranean waters there are many which are really characteristic Mediterranean shells, and are only marked as British in consequence of some rare occurrence." St. Erth Beds. At St. Erth, near Hayle in Cornwall, and at ; an elevation of about 100 feet above the sea, there is an interesting deposit of sand and clay, the latter containing shells which prove it to be of Pliocene age. A pit near the vicarage exposed the following beds : Feet. Loamy clay full of angular stones . . . . . . 2 to 6 Fine gravel and coarse ferruginous sand . . . . . 4 to 6 Clay, yellow above, blue beneath, and containing many fossils 6 Fine foundry sand, base not exposed . . . . . . ? 10 The number of Mollusca which have been found in the clay is between eighty and ninety, and the greater part of them are such as pccur in the lower part of the Red Crag, but many species of a southern character are present, such as Fusus corneus, Nassa recti- .costata, Nassa mutabilis, Cardium papillosum, and Cardita aculeata, which do not occur in the eastern crags, while the Boreal and Arctic forms found so abundantly in the higher part of the Red rag are absent at .St. Erth. 596 STRATIGRAPHICAL GEOLOGY Mr. C. Reid has pointed out that the proportion of extinct forms, so far as the fauna has been worked out, is 41 per cent, which exactly corresponds with the proportion in the Coralline Crag ; 14 and in view of this and of the preponderance of southern Mollusca y he concludes that the St. Erth Beds are of older Pliocene age. Mr. Robert Bell, Professor Kendall, and Mr. Harmer are, however,, all opposed to this view, and consider that the St. Erth Beds corre- spond more nearly with the older part of the Red Crag. From the' nature of the deposit, from its geographical position, and from the occurrence of beach deposits round St. Agnes Beacon up to 375 feet above the sea-level Mr. Reid infers that the fossiliferous clay was- formed at a depth of 40 or 50 fathoms. Red Crag. Returning now to the eastern counties we come to the deposit which is generally known as the Red Crag, this name denoting the usual reddish-brown colour of the material, which differs considerably from that of the Coralline Crag. The latter (when not decalcified) is a light-yellow calcareous deposit consisting chiefly of organic debris with some fine sand, while the Red Crag is a ferruginous shelly sand, its unaltered portions consisting partly of quartz-sand and partly of shells (either perfect or broken), the whole stained by a variable proportion of peroxide of iron (from 5 to 16 per cent). This formation covers a much larger area than the older (Coralline) crag, and is apparently continuous over a space of about 300 square miles in Suffolk, but is so often concealed by the sands and clays of the Glacial Series, that generally it is only on the slopes of the valleys which intersect the district that the crag actually comes to the surface. The best exposures occur in the cliff-sections of Waltoii-on-Naze, Felixstow, Bawdsey, and in the valleys of the rivers Orwell and Deben. The Red Crag rests partly on the London Clay and partly on the Coralline Crag, wrapping round the isolated reefs of the latter, and filling up the hollows between them. At its base there is generally a bed of phosphatic nodules with " boxstones " and other derived pebbles, together with rolled bones and teeth of mammalia and of sharks. The manner in which the Red Crag is banked against the Coralline Crag is illustrated in the section, Fig. 199. In the pits near Sutton there is evidence of two shore-levels, an upper and a lower cliff-line. The upper cliff is about 12 feet higli r the lower shore-line is about 9 feet below the upper, and these two shore-lines are traceable all round the small mass of Coralline Crag which evidently formed a reef in the Red Crag Sea. From the researches of Mr. S. V. Wood 15 and Mr. F. W. Harmer 16 it appears that the crag in the southern part of this area is THE PLIOCENE SERIES 597 decidedly older than that in the northern part. The beds at Walton-on-Naze yield a fauna that Mr. Harmer considers to be -closely allied to that of the Coralline Crag, for it includes many southern species and very few of northern origin, while at Butley in the northern part of the area northern species are abundant and southern forms are comparatively rare. Thus it seems as if these Red Crags were deposited while the sea was gradually retreating northwards, the slowly receding waters leaving a succession of beach and shallow- water deposits behind them. This shelly crag is seldom more than 20 feet deep at any one place, and its bedding is generally oblique, so that it may be described as consisting of a number of shelly sand-banks, each composed of highly inclined laminae, one bank often being sharply truncated by another. Mr. Harmer has recently divided the Red Crag into three parts on the basis of the changing proportion of northern and southern species, and these divisions may be termed the Walton Crag, the Newbourn Crag, and the Butley Crag. The relations of these crags is best expressed by the following table of percentages, in the compilation of which Mr. Harmer has taken into account only the more abundant and characteristic species. Extinct Species. Southern Species. Northern Species. Species of Northern and Southern Range. Butley Crag . . 13 13 23 47 Newbourn Crag . 32 16 11 36 Walton Crag . . 36 20 5 35 Coralline Crag . 38 26 1 31 The Walton Crag is characterised by the abundance of Neptunea contraria, Columbella sulcata, Nassa labiosa, N. elegans, Trochus Adansoni, Buccinopsis Dalei, and Dosinia exoleta ; the New - bourn Crag by abundance of Cardium angustatum, Mactra ovalis, M. constricta, and Tellina obliqua ; the Butley Crag by Neptunea antiqua, Buccinum grcenlandicum, Cardium grcenlandicum, Nucula Cobboldice, Tellina obliqua, and T. prcetenuis, Norwich Crag. This group has a more extended range than either of the older crags, since it is found more or less continuously through Eastern Suffolk from Aldeburgh to Bungay and Beccles, and thence through Norfolk to Brundall and Norwich (on the river Yare), and to Coltishall and Burgh in the Bure Valley, a distance of about 40 miles. It is well exposed at several localities in the neighbourhood of Norwich, whence it takes its name ; it is a variable group of sands, laminated clays, and pebbly gravels, THE PLIOCENE SERIES 599 with occasional seams or patches of shells, and different local names were given to different portions of the group before it was ascertained that they all belonged to one division of the Crag. Mr. F. W. Harmer remarks that " the Norwich Crag Beds are separated by a considerable interval [of time] from any part of the Red Crag. Their molluscan fauna has a much more recent character ; they never exhibit the highly inclined bedding so characteristic of the Red Crag, and they attain a much greater thickness than the latter ; they occupy an entirely different area, and appear to have originated under somewhat different conditions." 17 The Norwich Crag appears to cut off the Coralline Crag abruptly near Aldeburgh (see Fig. 199), and it probably also truncates the Red Crag, for the latter has never been recognised north of that place, while the newer crag thickens northward very rapidly. In a boring at Leiston, only 2 miles from the border of the Coralline Crag, it was found to be 134 feet thick, at Southwold 147 feet, and at Lowestoft 180 feet were traversed without reaching its base. Norwich Crag is not well exposed in any cliff-section, but a few feet of it are seen at the base of the cliffs near Southwold and Easton Bavent, where the succession is as follows : Feet. Glacial Beds Bedded sands with flint pebbles . . up to 7 Chillesford Beds Bedded sand and clay . . . ,,12 Norwich Crag Sand with three layers of shells . ,,4 At Aldeby, on the north side of the river Waveney, a brickyard exposes a similar succession, and about sixty-six species of Mollusca have been obtained from the shelly sands. A special feature of this locality is that many of the bivalves occur with united valves in the position of life ; they include Mya arenaria, M. truncata, Yoldia oblongoides, Y. lanceolata,Lucina borealis, Mactra ovalis, Scrobicularia plana, Tellina lata, and T. obliqua. Farther west along the same valley this crag is exposed in many pits near Norwich, Bramerton, and Brundall. The beds vary from 15 to 20 feet in thickness and consist mainly of sand with occasional seams of brown clay, and have generally a bed of rolled flints and pebbles at the base, in which mammalian bones and teeth have been found. Ohillesford Beds. There has been much difference of opinion about these beds and their relation to the Norwich Crag. The geological surveyors (Messrs. W. Whitaker and C. Reid 18 ) distinguish between Chillesford Crag and Chillesford Clay, regarding the former as part of the Norwich Crag and the latter as a separate and higher horizon. Mr. Harnier agrees with these 600 STRATIGRAPHICAL GEOLOGY authors that what has been called Chillesford Crag or Aldeby Beds is in most localities indistinguishable from Norwich Crag, but he differs from them in regard to the sand below the clay at Chillesford, which he regards as belonging to the Chillesford Beds, believing it to be an exceptional bed without a parallel elsewhere. The Chillesford Beds consist of finely laminated clay and sand, always micaceous, lying horizontally, and often in alternating layers of clay and sand. They are often from 18 to 20 feet thick ; they rest indifferently on the Coralline, the Eed, and the Norwich Crags, and have been traced from Waltoii-on-Naze to Mundesley in Norfolk, a distance of about 90 miles. Mr. Harmer has pointed out that when the exposures of these beds are plotted on a map they arrange themselves in the form of a sinuous band of gradually increasing width as it is traced northward, a form which suggests that they were formed in the estuarine portion of a large river flowing from south to north, and if this was so, the river was probably a continuation of the Rhine. 19 Weybourn Crag This is only found in the north of Norfolk, and takes its name from Weybourn on the coast near Cromer, where it consists of from 1 to 1 2 feet of loam and sand with a band of blue clay in the lower part and a basement bed of unworn flints resting on the Chalk. Its characteristic shell is Tellina balthica, which is very abundant and is a northern species that does not occur in the older crags. Out of fifty-three species of Mollusca only five are extinct, and all the rest are such as now live in the North Sea, nine being regarded as Arctic forms, which is a larger proportion than occur in the Norwich Crag. The fauna, therefore, seems to indicate that the Weybourn Crag is newer than the Norwich Crag, and Mr. Harmer believes it to be newer than the Chillesford Beds. Cromer Beds. These beds are exposed at intervals along the Norfolk coast from Happisburgh to Weybourn. They have been described by many writers, and were carefully explored for the Geological Survey by Mr. C. Eeid, who found the general succession in these clift's to be as follows : 20 Feet. Pleistocene / Glacial deposits ... . \Leda myalls Bed 4 to 15 {Upper freshwater bed "I Forest bed (estuarine) > Cromer Forest Bed . 10 to 20 Lower freshwater bed] Weybourn Crag 1 to 12 The Lower freshwater bed is a local and discontinuous deposit, its relation to the central estuarine portion resembling that of more recent " submerged forests " to the mud of modern estuaries. The THE PLIOCENE SERIES 601 middle division consists of clay, sand, and gravel in irregular layers, and contains a mixture of marine and freshwater shells with many mammalian bones ; it received its name of " Forest Bed " from the frequent occurrence of stumps, logs, and matted roots of trees, but these are not in the position of growth, as formerly supposed ; they have all been drifted, though probably not from any great distance, and Mr. Reid concludes that they grew on the side of a steep river- bank which was continually being undermined and washed away by the current, " and settling generally in an upright position, as we should expect from the greater density of the roots and from the weight of the adhering soil, they have formed ' snags ' in the river, such as are constantly met with in streams flowing through a forest- clad country." The upper surface of this estuarine deposit is often weathered into a soil and penetrated by small roots (hence termed the Rootlet Bed), and is in some places covered by lacustrine deposits containing freshwater shells, with remains of Fish, Amphibia, and small Mammalia ; this is the Upper freshwater bed. The marine Mollusca, of which there are only nineteen species, all occur in the Weybourn Crag, and include some extinct species, as Melampus pyramidalis, Tellina obliqua, and Nucula Cobboldia. With respect to the land and freshwater Mollusca these number sixty-two, of which thirty-three also occur in the Norwich, Chilles- ford, and Weybourn Crags. Of the others, which do not occur in older beds, some are extinct, e.g. Viviparus gibbus, Limax modioli- formis, Nematura runtoniana, and Pisidium astartoides, and some no longer live in Britain, e.g. Hydrobia marginata, H. Steini, Valvata fluviatilis, and Corbicula fluminalis. The mammalian remains are also numerous and important, and when the species are arranged in three groups, as below, it is seen that they prove the beds to have been formed just at the epoch of transition from Pliocene to Pleistocene time, a certain number being survivals from the Pliocene (either of Britain or France), and about an equal number being new-comers which range to Pleistocene or to modern times, while a few have not yet been found elsewhere. (1) Survivals from Pliocene^ 19 Elephas meridionalis. Cervus poligniacus. ,, antiquus. ,, Etueriarum. Rhinoceros etruscus. ,, elaphus (stag). ,, megarhinus. ,, verticornis ( = C. belgrandi). Hippopotamus major. Machaerodus latidens. Equus stenonis. Trogontheriura Cuvieri. ,, caballus (horse). Castor fiber (beaver). 602 STRATIGRAPHICAL GEOLOGY Ursus arvernensis. Microtus ( = Arvicola) intermedius, Canis lupus (wolf). Lutra vulgaris (otter). Cervus Sedgwicki. (2) New-comers ranging to Newer Beds, 19 Bison bonasus (var. priscus). Sorex vulgaris (shrew). Ovibos moschatus (musk ox). ,, pygmseus (pigmy shrew). Ursus spelseus (cave bear). Talpa europaea (mole). ,, ferox (grizzly bear). Mus sylvaticus (mouse). Gulo luscus (glutton). Microtus arvalis (field vole). Canis vulpes (fox). ,, gregalis. Mustela martes (marten). ,, amphibius (water vole). Hyaena crocuta. Microtus glareolus (red vole). Sus scrofa (wild boar). Sciurus vulgaris ? (squirrel). Myogale raoschata (musk shrew). D. CONTINENTAL PLIOCENE DEPOSITS 1. The Pliocene in Belgium and Holland In these countries a succession of beds has been described which appears to correspond with the lower and middle portions of the British Series, but does not include any representative of the Norwich Crag or of the later Pliocene deposits in England. This succession, as interpreted by the researches of Messrs. Van den Broeck, Dr. Lorie, and Mr. Harmer, 21 is as follows : 4. Amstelian. 3. Scaldisian or zone of Neptunea antiqua. 2. Casterlian or zone of Isocardia cor. 1. Diestian or zone of Terebratula grandis. The Diestian Sands occupy a position similar to that of the Lenham Beds, and are only found on the high ground along the southern side of the Belgian basin, occupying a considerable area near Louvain and Diest, but not extending far to the northward. Near Antwerp are beds which contain a fauna similar to that of our Coralline Crag. At the base is a pebble bed covered by a blackish sand yielding sharks' teeth, Cyprina rustica, and Gardita senilis ; this passes up into pale greenish-grey sand, from which above sixty species of shells have been obtained, the most abundant being Isocardia cor, Astarte Omalii, A. corbuloides, Lucina borealis, Ringicula buccinea, and Turritella incrassata, with bones of Balcena and Balcenoptera. This zone is succeeded by the Scaldisian, which consists of red and grey sands similar to our Red Crag, and containing many of the same shells, such as Tritonofusus gracilis, Neptunea contraria, Valuta THE PLIOCENE SERIES 603 Lamberti, Nassa reticosa, N. labiosa, Purpura tetragona, Pecten maxi- mus, and P. Gerardi. These beds are regarded by Mr. Harmer as the equivalents of the Walton Crag ; he points out that Neptunea contraria is as abundant as in the English Beds, while no specimen of the dextral form N. antiqua has been found in the Scaldisian. No newer beds are found near Antwerp, but borings in Holland, and especially at Utrecht and Amsterdam, have traversed a great thickness of Pliocene, and prove that all the zones thicken greatly toward the north. Thus the Utrecht boring was carried to a depth of 1198 feet without reaching the base of the Pliocene, and this depth is divided as follows by Mr. Harmer : Feet. Recent and Pleistocene deposits . . . . 513 I" Amstelian, yellow sands .... 262 Pliocene^ Scaldisian, grey glauconitic sands . . 123 \Casterlian, grey glauconitic sands . . 300 1198 Here, therefore, the Pliocene Series is probably at least 700 feet thick. The shells obtained from the Amstelian include Nucula Cobboldice, Yoldia lanceolata, Cardium edule, Cardium grcen- landicum, Tellina prcetenuis, Mactra subtruncata, Mya arenaria, and Littorina littorea, an assemblage which corresponds to that of the later parts of the Red Crag. 2. France In the north-west of France there are a number of small patches of reddish sand and blue clay containing a marine fauna which is claimed to be Upper Miocene by Mr. Dollfus, but which has generally been regarded as Pliocene, and the latter view is supported by Mr. C. Reid. These tracts occur, some in the Cotentin, near Carenton and Valognes ; some in He et Vilaine, near Rennes and Redon ; and some still farther south, both north and south of the valley of the Loire. According to Mr. Dollfus the fauna is substantially the same at all localities. Cardita striatistrina, C. aculeata, Venus fallax, Nassa prismatica, N. mutabilis, and Turritella incrassata are some of the characteristic shells. In the Cotentin the basement-bed is a layer of rolled pebbles with teeth and bones of Miocene and Eocene animals, together with shells of Terebratula grandis. This bed is overlain at Gourbesville by sands, and at St. Martin d'Aubigny by greenish clays, and the faunas of these two deposits are not quite the same. Mr. Reid states 22 that of 143 species from Gourbesville 91 604 STRATIGRAPHICAL GEOLOGY occur in the Coralline Crag, and that 85 per cent of the whole are living species, which is about the same proportion as in the Red Crag. The clays of d'Aubigny, on the other hand, have only 50 per cent of living species, and contain a larger number of southern species which give them a different and older aspect ; but the difference may be due to the different conditions under which the two deposits were formed rather than to difference of age. It seems probable, therefore, that all these deposits are of about the same age as the Lenham Beds and the Diestian of Belgium, and that it was by connection between the Atlantic and the Belgian Seas across the north of France that the southern species of Mollusca made their way eastward. In Central and Eastern France the Pliocene deposits are mainly of lacustrine and fluviatile origin, but these yield assem- blages of mammalian remains by means of which they can be referred to one or other of the three stages of the series. Thus the earliest of these assemblages is that found in the ossiferous breccias of Mt. Luberon and Cucuron in Vaucluse ; these beds contain many of the same species as occur in the Poiitian of Eastern Europe, especially Hipparion gracile, Rhinoceros Scheier- macheri, Palceoryx Cordieri, Helladotherium, and Dinotherium. In the Auvergne district the general succession may be tabulated as follows : 23 Plateau basalts with tuff's and breccias containing bones of Elephas meridionalis. /Stratified deposits of Perrier and Le Puy. \ Upper phonolite and hanyne andesites. T /Andesitic trachytes and porphyritic basalts with intercalated ^ breccias. The stratified beds of Perrier overlie the Miocene (see Fig. 194), and have a conglomerate at the base containing bones of Mastodon arvernensis and Machcerodus cultridens, succeeded by cenerites and sands, the former embedding leaves of plants which comprise maple (Acer polymorphum), beech (Fagus phocenica), and a bamboo (Bambusa lugdunensis). These beds are overlain by a breccia of trachyte fragments, among which are bones of Elephas meridionalis, Equus stenonis, and Gazella Julieni ; while in Le Puy pebbly sands beneath the basalts have yielded Rhinoceros etruscus and Hippopotamus major. Farther south we find a more complete succession of stratified deposits, without volcanic intercalations, and including a thick marine Placentian stage. These beds extend from near Lyons southward along the valley of the Rhone and spread out in Provence. The general succession is as follows : TT / u PP er \ THE PLIOCENE SEKIES 605 Sicilian Alluvial pebbly sands with Elephas meridionalis. Astian Yellow sands with Mastodon arvernensis. p] ,. /Sands and marls with Potamides Basteroti (160 feet). 111 \ Marls with Nassa semistriata, Area diluvii, etc. (500 feet). p ,. (Marls of Bollene, with Congeria simplex and Melania in l Matheroni, lying imconformably on the Miocene (100 feet). 3. Italy When the Pliocene is followed eastward into Liguria, the marine deposits are still thicker, presenting a series of marls and limestones over 600 feet thick, and representing the Pontian y Placentian, and Astian, and abounding in fossils. These beds are overlain by 300 feet of yellow sands with few fossils, but these also are of Astian age, and no marine deposits of Sicilian age occur in Northern Italy ; near Asti, however, and other places there are fluviatile sands and gravels containing bones of Elephas meridi- onalis and other Upper Pliocene species. Pliocene deposits occupy large tracts on both sides of the Apennines and are well displayed in and near Rome. They cover about half the island of Sicily, and are largely developed in Calabria. In these southern areas they are over 2000 feet thick and rise to a height of 4000 feet above the sea, consisting almost entirely of marine deposits, the succession being as follows : Sicilian f Calcareous sands of Palermo with many fossils, including 300 f t northern species, such as uccinum grwnlandicum, Mya truncata, and Cyprina islandica. Astian, 700 feet Placentian. 900 feet ' Yellow sands with Pecten opercularis, P. varius, Pectunculus pilosus, and other shells. Limestone and calcareous sand with Amphistegina. Limestones with Blue marls with Nassa semistriala. Aporrhais pespelicani and Pecten jacolceus. -\T -~t ,/ Sands with Bryozoa and Brachiopods (Terebratulina}. Messinian, (Foraminiferal marls and limestones with Ostrea cochlear 300 feet \ and Clypeaster pliocenicus. In the Val d'Arno there is an extensive lacustrine deposit over 700 feet thick, the lower part including beds of lignite with plant-remains and the bones of a tapir ; the upper part yielding the bones of many mammalia of Upper Pliocene age, such a& Elephas meridionalis^ Rhinoceros etruscus, Equus stenonis, with antelopes, deer, a bear (Ursus etruscus), an ape (Macacus florentinus\ and three species of Machcerodus (sabre-toothed tigers). 24 4. Austria and the Balkans Pontian. In Austria the Sarmatian Clays are overkin by the Gongeria Beds or Pontian stage, consisting of sandy marls and 606 STRATIGRAPHICAL GEOLOGY clays which have a thickness of more than 300 feet. The prevalent shells are species of Congeria, C. conglobosa occurring in the lower part, with G. triangularis and C. rhomfioidea in higher beds, with species of the peculiar Cardiidse Adacna and Monodacna like those of the Caspian Sea. These Congeria Beds extend eastward through Hungary, Croatia, Roumania, and Southern Russia to the Caspian Sea and Lake Aral, which lakes, together with the Black Sea, may be regarded as remnants of the ancient and much larger Pontian Sea. In Greece the Pikermi Beds are a lacustrine deposit of the same age ; they consist chiefly of red marly clays with land and fresh- water shells, varied by layers of sandy marl and lenticular beds of pebbles ; in the clays are several layers of mammalian bones. 25 These beds are part of a formation which has a wide extension in Attica and is regarded by Fuchs as of Pliocene (not Miocene) age, as the freshwater beds overlie deposits which contain marine shells of Pliocene species (Pecten benedictus, Ostrea undata, and 0. lamellosa). The bones are mostly separate and some are broken, but many .complete limbs occur with the bones in their natural position, as if they had been washed down from higher ground during heavy floods. The principal members of this fauna have been mentioned on p. 588. Similar deposits have been found in Euboea and in the island of Samos. Placentian. In Austria the Pontian Clays are succeeded by some lacustrine beds with land and freshwater shells, but these are more fully developed in Croatia, Roumania, and Bulgaria, where they are known as the Paludina Beds. The lower part of these contain smooth rounded forms of Viviparus ( = Paludina), but in the higher beds they are caririated. Mastodon arvernensis has been found in the lower beds. Similar beds extend through Southern Russia to the Caspian Sea and are well exposed near Odessa, where beds containing Prosodacna and Dreissensia overlie the Congeria Beds and are themselves succeeded by the " Paludina Beds." Astian and Sicilian. Newer Pliocene deposits, both of marine and lacustrine origin, occur at different places all over the same eastern region, and marine Pliocene Beds occur also in Cyprus and Asia Minor, but their stratigraphy has not yet been sufficiently investigated for the recognition of stages or zones. 5. Germany No marine Pliocene deposits occur in Germany, but at Eppelsheim near Worms there are sands and gravels, overlying Miocene Beds, which have yielded a Pliocene mammalian fauna THE PLIOCENE SERIES 607 resembling that of Pikermi, but without Helladotherium or the antelopes. E. HISTORY OF THE PLIOCENE PERIOD At the beginning of Pliocene time all the mountain ranges of Central and Western Europe had been formed, and the greater part of the European region was in the condition of dry land, though the coast-lines were not the same as those of the existing con- tinent. The British area was still a compact and extensive tract of land, extending northward beyond Ireland and Scotland, but it seems to have been separated from France and Belgium and must therefore have been a large island. We have seen that the southern aspect of the fauna of the Diestian and Coralline Crag Seas makes it almost certain that there must have been a connec- tion with more southern seas, and this connection most probably lay along the line of the English Channel and through Normandy to the country between the Vilaine and the Loire, where it opened into the Atlantic. This matter is more fully discussed in my Building of the British Isles (1911). In the south of Europe the Mediterranean Sea still existed, but probably opened southward through Egypt, for there is 110 evidence of any opening across Spain or by the Straits of Gibraltar until the time of the Sicilian Beds, when northern species of Mollusca first entered the Mediterranean area. This Mediterranean only covered small parts of Southern France, but seems in Placentian time to have extended up the valley of the Rhone, converting it into a sort of fiord. Thence its northern shore passed along the feet of the Maritime Alps and round the great basin of Northern Italy, the whole of that country except the ridge of the Apennines being covered by the Lower Pliocene Sea, together with nearly the whole of Sicily and parts of Northern Africa, especially in Tunis, Tripoli, and Egypt. East of this sea, and separated from it by the ridge of the Dinaric Alps and the Highlands of Montenegro and Albania, lay the large inland sea of the Ponto-Caspian region, which extended from the Vienna basin 011 the west to and beyond the Sea of Aral on the east. Whether the deposits of Pontian age in Greece and the jEgean area were formed in a gulf of this sea or in separate lake-basins is at present uncertain, but the probabilities are in favour of the latter view. These conditions continued to prevail in the Pontian area through Placentian time, but the waters of the great lake gradually became more and more fresh, so that the Congerias died out and their place was taken by Unio, Viviparus, and Melanopsis. 608 STRATIGRAPHICAL GEOLOGY The change from the conditions of Lower Pliocene to those of Middle Pliocene (or Astian) time seems to have been produced by a general uplift of what may be termed the central axis of Europe, in other words, across the whole of that part of Europe which lies between the parallels of 40 and 51 ; while further subsidence took place both in the southern Mediterranean and also in the northern region. Thus the Italian area continued to sink, permitting a great thickness of marine deposits to be accumulated, and the Pontian lakes continued to exist, but probably they had shrunk in size. The whole of France and the English Channel became dry land, and all direct communication between the Atlantic and the Belgian Sea was thus cut off; but farther north subsidence was taking place, forming and widening a channel between Scotland and Norway which opened into the Arctic Ocean. The facts which relate to this gradual formation of the North Sea Basin and to the extension of the great delta of the Rhine may be mentioned more fully. Having regard to the present position of the Lenham Beds, and the progressive overlap of the newer crags toward the north, as well as the great thickness of the Pliocene deposits in Holland, we may infer that the crust move- ment in the Anglo -Belgian area had the effect of a tilt, which caused a subsidence on its northern side, while its southern Border was raised through at least 780 feet. Thus all connection with more southern seas was severed, On the other hand, the subsidence to the northward allowed the sea to spread farther north than in the earlier part of the period, and before the epoch of the Newbourn and Butley Crags it would seem that part of the land which had hitherto united Scotland and Scandinavia was submerged, so that an opening was formed into the Arctic Ocean. The Red Crag is generally regarded as a shore deposit, and by Mr. Harmer it is compared with the shelly banks and beaches which are now being formed all along the coast of Holland. These accumulation* of sand and shells are attributed by Dutch geologists to the prevalence of gales from the west, and as no such shell-beds exist now on our eastern coasts this seems to explain the facts, but if such was the origin of the Red Crag we must suppose that in Pliocene time the prevalent gales were from the eastward and not from the westward, as is now the case. 26 As the sea extended its area northward and the land rose on its southern border, the valley of the Rhine must have prolonged itself northward over the southern part of the North Sea plain, and we have seen (p. 600) that the Chillesford Beds may possibly THE PLIOCENE SERIES 609 be deposits formed in the estuary of that river. Other indications of a large river flowing from the south present themselves in the Croiner (Forest) Beds, and Mr. C. Eeid has carefully examined the stones which compose the gravelly portion of the deposit. He found the collection of pebbles to be such that if the river had flowed from the south, west, or north, it must have brought a very different assemblage. " It seems, therefore, that only from the south-east and east could the stones be derived, and that the river must have been very large is shown by the uniformity of the composition of the gravels at considerable distances apart ; if this be the correct reading, the river can be no other than the Rhine, a view held by various writers ever since the Forest Bed was first known." 2 7 With respect to the change in the climate, the fauna of the Diestian Sand and Coralline Crag is that of a rather warm sea, and the mean annual temperature of the surrounding country must have been decidedly higher than it is now in the same latitude. The subsequent exclusion of the warm southerly current doubtless caused a general diminution of temperature, but there were other meteorologic and cosmic causes which operated in rendering the climate more and more frigid, till at length in the Weybourn Crag we find a fauna indicative of a climate somewhat colder than that which at present prevails in Norfolk. This prepares us for the great change which took place very soon after the formation of that bed. In the Mediterranean area the closing scenes of Pliocene time were connected with the subterranean disturbances which resulted in the formation of the great volcanoes of Etna and Vesuvius and the smaller vents of the Latian Hills. The earliest volcanic eruptions appear to have been submarine, for the Astian and Sicilian deposits of Sicily, Calabria, and Rome include inter- stratified beds of volcanic ash and sand. It was not, however, till the close of Pliocene time that the greatest discharges began, for the building up of Etna and the elevation of the Pliocene deposits to the levels at which they are now found, i.e. more than 3000 feet above the sea, must have taken place in early Pleistocene time. REFERENCES 1 G. F. Dollfus, "Essai sur 1'etage Aquitanien," Bull. Service Carte Geol. France (1909). 2 S. Gardner, Geol. Mag. for 1879, p. 152, and Monograph. Pal. Soc. on the Eocene Flora, pp. 18, 19. 3 C. Reid, Phil. Trans. R.S. vol. 201, B. p. 161 (1910). 4 See Pengelly in Phil. Trans, vol. cliii. p. 1019 (1863), and Jukes- Browne in Geol. Mag. for 1909, p. 257. 2R 610 STRATIGRAPHICAL GEOLOGY 5 See Dollfus, op. cit., and G. F. Harris, Geol. Mag. for 1890, p. 22. 6 See Dollfus, op. cit. table on p. 19. 7 See T. G. Bonney in Proc. Geol. Assoc. vol. xvii. p. 191. 8 See M. Boule, Bull. Soc. G6ol. France, Ser. 3, vol. xx. p. 104. 9 See C. Mayer, Bull. Soc. Gtol. Fr. Ser. 3, vol. v. p. 282, and F. Sacco, II Bacino Terziario del Piemonte, Turin, 1889. 10 O. Fisher, Quart. Journ. Geol. Soc. vol. Ixi. p. 35. 11 W. B. Dawkins, Quart. Journ. Geol. Soc. vol. lix. p. 105. 12 C. Reid, "Pliocene Deposits of Britain," Mem. Geol. Survey (1890). 13 F. W. Harmer, Quart. Journ. Geol. Soc. vol. liv. p. 308. 14 C. Reid, Pliocene Deposits, p. 59. 15 S. V. Wood, "Monograph of the Crag Mollusca," Pal. Soc., 1872 to 1874, with Supplements in 1879 and 1884. 16 F. W. Harmer, Quart. Journ. Geol. Soc. vol. Ivi. p. 705 (1900). 17 F. W. Harmer, op. cit. p. 737. 18 W. Whitaker, "Geology of South wold," Mem. Geol. Surv. (1887), and C. Reid, Pliocene Deposits of Britain (1890). 19 F. W. Harmer, Proc. Geol. Assoc. vol. xvii. p. 446. 20 C. Reid, op. cit. p. 146. 21 F. W. Harmer, " Pliocene Deposits of Holland," Quart. Journ. Geol. .Soc. vol. Hi. p. 748. 22 C. Reid, op. cit. p. 68. 23 See M. Boule in Proc. Geol. Assoc. vol. xvii. p. 204. ;8 * C. Reid, op. cit. p. 219. " 25 See A. S. Woodward, Geol. Mag. for 1901, p. 482. ' 26 F. W. Harmer, Quart. Journ. Geol. Soc. vol. Ivi. p. 732. - 27 C. Reid, op. cit. p. 188. CHAPTER XVII THE PLEISTOCENE SERIES A. GENERAL CONSIDERATIONS THE term Pleistocene was used by Lyell to denote the deposits which are newer than the Pliocene and yet not modern enough to be called Recent. But different writers have different ideas as to what deposits should be called Recent. It seems best to consider the Pleistocene period as including all deposits in Europe which are older than those containing metal implements of human manufacture. Some of these deposits have obviously been formed during a time when the climate became so cold that the higher parts of the country were covered with ice and snow and the North Sea was more or less filled with ice. Such deposits are called Glacial, and the time during which they were formed is known as the Glacial period, or epoch, and sometimes as the " Great Ice Age," while all the subsequent deposits can be classed as " Post-Glacial." This, however, is not a satisfactory division of Pleistocene time, because the Glacial deposits do not extend all over Europe, being limited to its northern parts and to a certain area round the Alps. In other parts of the region fluviatile, alluvial, and subaerial deposits were being formed, which are of a more normal character, though some of them exhibit signs of severe climatic conditions. The Pleistocene deposits cannot therefore be treated as one consecutive series, but must be regarded as including two different sets of deposits, which may be called Glacial and Non-Glacial, but are only distinct from a regional point of view, some of the one set being contemporaneous with the other set. Again the Glacial deposits, even those of a single country like England, are not easily classified ; for they vary so much in different districts and are so irregularly stratified that correlation is difficult. There is of course a general succession in time, and elaborate classifications have been formulated, but these are based upon assumptions and 611 612 STRATIGRAPHICAL GEOLOGY theoretical views, and must be received with great caution ; for it is obvious that, owing to the special conditions under which Glacial deposits have been accumulated, ordinary stratigraphical principles cannot be applied to them. Neither are the organic remains which occur in the Glacial deposits to be trusted as implicitly as those of other series of strata, because portions of an older deposit have sometimes been trans- ported and mixed with newer accumulations. Bones of Mammalia are occasionally found in the Glacial Series and more frequently in Non-Glacial deposits, and they do afford a certain criterion of age, as some of them belong to extinct species and they occur in different assemblages which seem to have succeeded one another. In this way three divisions of Pleistocene time have been distinguished, (1) a time when Elephas antiquus, Rhinoceros Mercki ( = R. lepto- rhinus), and Machcerodus were common animals ; (2) a time when Elephas primigenius (the Mammoth), Rhinoceros tichorhinus, and Hyceana spelcea took their places ; and (3) a period when the Rein- deer (Cervus tarandus) ranged over the greater part of Europe > accompanied by many other existing species of animals which live now within the Arctic Circle. Lastly, it was during the Pleistocene period that man appeared in Europe, and one of the special characteristics of man is that he is a manufacturer of implements. Accordingly implements fashioned by the hands of men are found here and there even in early Pleistocene deposits ; the older tools being always made of stone and roughly manufactured, but those of the later phases of the period showing more and more careful shaping and finishing. Stone implements thus afford another means of correlating the Non- Glacial deposits of different districts and countries, but they are rarely found in or under the Glacial deposits. The division into Glacial and Non- Glacial is therefore the only arrangement under which the Pleistocene deposits can be conveniently described, and this will be adopted in the following pages ; but the reader must remember that the two sets of deposits are more or less contemporaneous. With respect to the life of the period, it will be convenient to- defer giving a list of the Pleistocene Mammalia till we deal with the Non-Glacial deposits in which they are most frequently found, but mention may be made of the marine shells which often occur in Glacial deposits and in beds associated with them. Those which actually occur in boulder - clays are generally broken, but those found in the interstratified sands and gravels are often perfect. The marine mollusca of Pleistocene times all belong to living species, but those found in the Glacial deposits are such as live THE PLEISTOCENE SERIES 613 in the seas round the northern parts of our islands, the species of our southern coasts being generally absent. The following are some of the commoner species in Glacial Beds, and some of them are shown in Fig. 201 : Gastropoda. Natica affinis (clausa), N. pallida ( = groenlandica), Trophon clathratus, T. Fabricii, Buccinuni undatum, B. groen- landicum, Bela turricula, Turritella comraunis, Scalaria groenlandica. a Fig. 201. GROUP OF PLEISTOCENE MARINE SHELLS. a. Pecten islandicus. b. Astarte borealis. c. Corbicula fluminalis. d. Trophon Fabricii. e. Naticaaffinis ( = clausa). /. Trophon clathratus. g. Scala groenlandica h. Puncturel la noachina. Lamellibranchia. Pecten islandicus, Leda truncata, Yoldia arctica, Y. lanceolata, Tellina lata, T. balthica, Astarte borealis, A. compressa, Pauopsea norvegica, Saxicava rugosa, Cyprina islandica. B. GLACIATION AND GLACIAL DEPOSITS Glacial Deposits. These are the relics of a time when the conditions which are now confined to the Arctic and Antarctic regions prevailed over the whole of Northern Europe. They are 614 STRATIGRAPHICAL GEOLOGY principally such as result from the action of moving masses of ice (boulder-clays), or from the action of strong currents of water (coarse gravels), but interstratified with these there are sometimes sands, brick-earths, and laminated loams which have been deposited in quiet water and in a more gentle manner. The chief product of the Glacial epoch is the material known as boulder-clay. There are several varieties of such clay, and it is by no means certain that they have all been formed in the same manner. Some boulder-clays are stiff compact masses of unstratified clay containing numerous stones and boulders, and they seem to have been compacted by the passage of heavy masses of ice over them. It does not follow, however, that they were dragged along beneath the ice in the form of what has been called "ground- moraine." The prevalent view at the present day is that the materials of boulder-clay were enclosed in the moving ice, princi- pally no doubt in the lower portion of it, but embodied in and moving with the ice. Wherever the glacier or ice-sheet melted this " englacial " material thawed out, and was dropped to form an unstratified clay full of stones and boulders. It might even accumulate by the thawing of the under surface of stationary ice during a temporary amelioration of climate, and if climate conditions again became severe a renewed movement of ice might take place in such a manner as to pass over such boulder - clays without removing them. Other boulder - clays are of a looser texture, and are more frequently associated with sands and gravels which contain marine shells, and such clays must either have been formed where ice invaded the sea, or must be the product of an ice-sheet which had passed over a sea floor, or was moving from the sea over the retreat- ing shore of a sinking land. These shells occur at various heights above the present sea-level from 100 to over 1300 feet, and also far inland, often many miles away from the present coast -line. The question naturally arises whether they are in situ or whether they have been transported to their present positions by the agency of ice. When they were first discovered and described they were regarded as proofs of a great submergence, during which the British Isles sank to an extent of more than 1300 feet. Later it was perceived that this was highly improbable, because no evidence of such a subsidence exists in the south of England nor in the north of France ; marine deposits in these areas being restricted to coastal belts, and not occurring at more than 130 feet above sea-level. Latterly glacialists have tried to explain the occurrence of all marine relics at high levels on the supposition that they were THE PLEISTOCENE SERIES 615 frozen into the base of enormous ice-sheets, which not only filled the basins of the North Sea and the Irish Sea, but were able to take up portions of the sea floor, and carry them up to high levels in England and Ireland. Further it has been maintained that instead of the land being lower and the seas deeper than they are now, the whole of Northern Europe had been raised to a greater elevation than that at which it now stands, and that land-ice was the sole agent concerned in the transport of marine shells as well as of large stones and boulders. These are extreme views, and the truth probably lies in a modification of them. The idea of the land being at a higher level than it is at present, either at the beginning of the Glacial Period or at the time of maximum glaciation, is a mistake. The evidence only proves that the land had previously been at a higher elevation, i.e. in Pliocene and Miocene times, and it favours the conclusion, that the land sank as the thickness of ice upon it increased. At the same time it must be admitted that the shells at very high levels may not be exactly in situ, and that perfect shells have not been found in sufficient number to prove a sub- mergence of more than 500 or 600 feet. Thus at Clava, near Inverness, there is a shell - bearing deposit with Arctic species at about 500 feet, and overlain by boulder-clay. A British Associa- tion Committee, appointed to investigate this and other localities, reported that the molluscs at Clava had probably lived on the spot where their shells are now found. It is probable, however, that the subsidence was greater in the north than it was farther south, and that while it may have been 600 feet in the latitude of Inverness it was not more than 130 or perhaps 160 feet in the south of England. Even such a moderate subsidence, however, would greatly facilitate the movement of ice- sheets over tracts which are now dry land, and would account for the transport of many rock-fragments from lower to higher levels under 500 feet. In spite of this we are still obliged to admit that there was sufficient vis a tergo to force the ice not only over the sea floors but out again and up to considerable heights on the other side. It has also been pointed out that under such conditions shear- planes may have been produced, and portions of the ice thrust forward and upward over parts in front of them. Such upward thrusts have been observed in Spitzbergen at the present day. 1 Glaciation of Rock Surfaces. In nearly all districts where boulder-clays occur the rock -surfaces which underlie the lowest clays are glaciated, that is to say they are grooved, scratched, and smoothed as they would be by the passage over them of heavy ice, which, when its sole is full of stones and boulders, becomes a gigantic 616 STRATIGRAPHICAL GEOLOGY rasping, graving, and moulding tool. In most cases it is found that the grooves run parallel to the main valleys of a district, and in many cases when the directions are marked on a map it is seen that they radiate outward in all directions from some central mountain tract ; but occasionally they run in one direction across some tract of country without any reference to its valley system, Fig. 202. MAP SHOWING THE GLACIATION OF SCOTLAND AND THE NORTH OF IRELAND (J. R. Kilroe). (Reproduced by permission of the author and the Council of the Geological Society.) as if the ice had been locally deflected from its natural course. This system of grooves and striations is often called the glaciation of a district or country. Thus the plotting of the principal lines of striation in Scotland has shown that country to have been an independent centre of ice- dispersal, and that at the time of maximum glaciation it was wholly THE PLEISTOCENE SERIES 617 covered by snow and ice which moved outward from the principal water-sheds ; the ice-fields of the Central and Northern Highlands forming a confluent ice-sheet which spread out in all directions and descended into the neighbouring seas (see Fig. 202). The south of Scotland formed another centre of dispersion, the ice flowing off its uplands, moving chiefly in western, southern, and eastern directions, because free movement to the north was blocked by the more northern ice-sheet. There are, however, exceptions to the coincidence between the direction of the striae and that of the main valleys ; thus the dis- trict known as Knapdale in Argyllshire, which rises to 700 or 800 feet above the sea, is glaciated obliquely from north-east to south- west, as if the ice had filled up the valley of Loch Fyne, and had flowed over the ridge to the Sound of Jura. There seems also to have been a time when the ice from the Perthshire highlands went across the Ochil Hills, some of which reach a height of 2300 feet, and these are important facts to remember when the mode of glaciation is considered. In the north of Ireland again the Glacial stria tion is not of a simple radiating character. According to Mr. J. R. Kilroe, the striae can be resolved into two sets, the one set having a general direction from east-north-east to west-south-west across the whole area with local variations to west and south-west in Mayo and Galway, while another set show lines of movement northward, southward, and south-eastward. These facts he interprets to mean that there was first a local snow-field covering all the high ground between Mayo and Antrim and shedding its ice both to the north and the south ; but that at the time of maximum glaciation the Irish ice was so dominated by that flowing off the Scottish Highlands that it was forced to move westward (see Fig. 202). The Lake district and the Welsh mountains were also independ- ent centres, and the glaciers of the former seem to have maintained their radiating onward movement throughout the whole of the Glacial epoch, but their terminal parts were merged into the ice flowing off the south-west of Scotland. In Wales the striae also have a general radiating arrangement from two centres, the Siiowdonian and Merioneth mountains forming one, the Plynlimmon Group and the highlands of Central Wales forming the other. Movement to the north and north-west was, however, blocked by the northern ice, so that Anglesey is glaciated from north-east to south-west, Denbigh, Flint, and South Lancashire from north-west to south-east. Consequently the ice of North Wales moved chiefly to the east and south-east, and boulders from the Arenig mountains have been carried eastward into Shropshire and Staffordshire. 618 STRATIGRAPHICAL GEOLOGY The ice from the central ice-field of Wales was also free to flow eastward, southward, and westward, but it seems to have been partially turned aside by the mass of the Brecknock Beacons and REFERENCE Welsh Ice ^ZAuorthern Ice m Irish Sed Ice ^^fl/orth Sea Ice Aw?//^ Ice I \nrifrJm Areas Scale of Miles Fig. 203. MAP SHOWING THE AREAS OCCUPIED BY THE DIFFERENT ICE-SHEETS IN ENGLAND. to have flowed chiefly in souther]y and westerly directions. Thus it overran the whole of the South Wales coalfield, the striae on the rocks of the northern borders of that area all indicating flow THE PLEISTOCENE SERIES 619 from the north. Pembrokeshire on the other hand has been glaciated by ice coming in from the Irish Channel (see map, Fig. 203). C. THE GLACIAL DEPOSITS OF BRITAIN No classification of the Glacial drifts has yet been proposed which is applicable to all parts of the British Isles. In some areas it is possible to establish a local succession of deposits, but all attempts to correlate those of one area with a similar series in another area have hitherto broken down. In some districts, such as Aberdeenshire, Norfolk, North Wales, and the south-east of Ireland, there are certainly older and newer boulder-clays, differing in colour and contents, and in some places they are separated by stratified sands and gravels. We may safely infer that such different deposits were formed during different successive phases of the Glacial Period. In Germany the Glacial deposits appear to exhibit a greater uniformity, and it is generally held that there are two boulder -clays, a lower and an upper, and that they indicate two epochs of advance and retreat of the northern ice. The great variation observable in the Glacial deposits of Britain is doubtless due to their being the product of several distinct ice- sheets coming from different directions. This is especially the case in England,' where the influence of 110 fewer than five ice-sheets can be distinguished (see map, Fig. 203). At present, therefore, the British deposits can only be described in a piecemeal fashion, and brief accounts will be given of those found in the following areas : 1. Central and Northern Scotland. 2. South Scotland and the north of England. 3. Yorkshire and Lincolnshire. 4. East Anglia. 5. The Midlands. 6. Wales and the West of England. 7. Ireland. 1. Central and Northern Scotland The Glacial deposits of Scotland may be classified as follows, but it is seldom that all the members are found in one locality, and it is probable that the moraines were more or less contempora- neous with the later shell-bearing clays : 4. Moraines of local glaciers. 3. Shell-bearing clays near the coast. 2. Upper boulder-clays with stratified sands and gravels. 1. Lower boulder-clay or till. 620 STRATIGRAPHICAL GEOLOGY The till is described as a very stiff clay full of stones and boulders, the whole being so hard and compact that it often weathers like a conglomerate. Boulder-clay of this kind occurs up to heights of 1600 feet, and is sometimes over 100 feet thick, being always thickest on the low grounds and thinning away on the highland slopes. The stones and boulders found at any given locality are principally derived from local rocks, and in the low- land clays there are not many stones which have travelled more than a few miles from their parent site ; most of the fragments are smoothed and striated, and many are scratched all over, often exhibiting five or six definitely striated surfaces. The local character of the clay is also shown by its variations in colour. Hugh Miller, sen., long ago remarked that red sandstones were covered by red boulder -clay, the grey flags of Caithness by a lead -coloured clay, and the coal-measures of the Lothians by a bluish -black clay ; subsequent observations have shown that the clays of these colours do not exactly coincide with the limits of the formations from which they derive their tint, but slightly overlap them in an outward direction. Thus, though the materials of this inland boulder -clay afford evidence of a certain amount of transport, still its local character is so strongly marked that "whatever may have been its mode of production, it cannot have been carried from a distance, but must have been formed on the whole in the districts in which it is found." In many places the hard till or boulder -clay above described is covered by deposits of a more varied character, consisting of stratified sand, gravel, and clay ; such deposits occupy large areas in the central lowlands between the firths of Forth and Clyde. Associated with these beds or stranded on bare hill -sides are numerous large erratics or isolated boulders, which have usually come from points to the north or north -west of the positions they now occupy, and some have travelled distances of 50 or 60 miles ; it is to be noted that such boulders rarely occur in the boulder-clay, though they often lie on it, the travelled stones in the boulder-clay being always of small size. Along the east coast, from Stonehaven to Peterhead, a very different kind of boulder-clay is found; this is a reddish -brown clay which contains broken marine shells, and often includes seams of fine sand or laminated silt ; moreover, its materials appear to have been derived from the southward, from the Old Red Sand- stone districts in Kincardine and Forfar, a conclusion which is confirmed by the occurrence of striee on the rocks below, pointing from south-south-west to north-north-east. ia Near Collieston and Slains this clay rests on stratified sands and gravels which are more THE PLEISTOCENE SERIES 621 than 50 feet thick and contain numerous broken marine shells, which seem to have been derived from the destruction of Pliocene Beds, comparable to the English Crags. The red clay varies from 10 to 30 feet in thickness and reaches inland up to heights of 300 feet, while pebbly gravels with broken shells range up to 470 feet. In the basin of the Clyde there are certain deposits which were evidently formed during the last stage of the Glacial epoch. The usual succession of these beds is as follows : Clyde 3. Sandy clay or sandy gravel. r>J-, -i 2. Fine grey clay with many perfect shells. n \l. Laminated reddish clay without shells. Till. Hard tough boulder-clay with striated stones. There is always a layer of laminated, stoneless, and shell-less clay between the till and the shell-bearing clay (No. 2), but its thickness varies from 1 to 18 feet. The clay above is a consolidated mud through which many stones, both rounded and angular, are scattered, but its consistency is not that of boulder-clay, and the stones are not glaciated ; the shells which occur in it are generally perfect, and are of northern and Arctic species. These clays fringe the Firth of Clyde from Glasgow westward, and pass up the Kyles of Bute to Loch Fyne ; they are also found on the shore of Loch Lomond, but do not anywhere extend beyond 50 feet above the sea. Inland the last phase of the Glacial epoch is marked by the occurrence of terminal moraines in many of the Highland valleys-; sometimes several such moraines occur one above another, and huge blocks of rock are often perched on the top of them as well as on the neighbouring slopes, while the course of the glen above the highest moraine is swept clear of all such detritus. 2. Southern Scotland and North of England In this area there were three principal centres of dispersion, one in Galloway, whence confluent glaciers moved outwards from the central granitic hills, another in the Lake District already mentioned (p. 617), and a third on the Hartfell, Ettrick, and Cheviot Hills, from which ice flowed chiefly eastward and south- ward (see Fig. 202). In each area the northward flow was short, and the main lines of transport were south and south-east. Thus Eskdale granite is found all over Lancashire and Cheshire, and boulders from the fells near Shap have been carried not only down the valley of the Lune, but over the Ingleborough district into Ribblesdale, showing that at one time the ice over-rode hills up to 1500 feet above present sea-level. Another stream of boulders has passed through the Stainmoor pass and through Teesdale to the east coast of Northumberland. 2 622 STKATIGRAPHICAL GEOLOGY The whole of Southern Scotland and Northern England below the level of 1600 feet is more or less covered by boulder-clay, which in many places has the same character as the Scottish till. The larger boulders embedded in this till are striated in situ in the same general direction and with the same firmness and evenness as the strioe on the underlying rocks ; thus the till of Northumberland may be likened to one vast irregular " striated pavement." The smaller boulders and stones frequently have their longer axes drawn into the line of glaciation, as if by a heavy dragging force moving contemporaneously with the arrangement of the material. At the surface it is often shaped into sow-backs or drumlins stretched in the same direction. All these features of the till are remarkably developed on the watershed among the Bewcastle Wastes, north of Gilsland in Cumberland. 3 The composition of the till, again, shows that the outcrops of rock lying in the path of movement- all the way from the red rocks of Cumberland, for instance, to the coaly shales of Newcastle contributed one after the other their quota of material to the mass of the clay. The texture and colour of the till thus constantly varies it is bright Indian red on the red rocks ; leaden-coloured and clayey on the carbonaceous rocks ; sandy where the pale Carboniferous sandstones predominate ; brown on the Cheviot porphyry ; and when it rests on composite formations such as the Carboniferous limestone series, it never remains for half an acre the same. Among the slopes of the higher hills it is thin, loose, local, angular, and almost morainic-looking ; in the valleys and low grounds it becomes dense and well-kneaded up, and contains a larger assortment of boulders ; in the lower grounds it is often gravelly and much intercalated with seams of sand, gravel, and laminated clay ; but it is the same deposit throughout. In the south-west of Scotland, and especially in Ayrshire, broken marine shells are found in the boulder-clay up to heights of 1000 feet above the sea, and the precise manner of their con- veyance against the outflow of the local ice has not yet been explained. On the coast of Northumberland and Durham, however, there is a newer or upper boulder-clay limited to the maritime districts, and to elevations under 350 feet. It is divided from the till sometimes by stratified sand and gravel, and sometimes by an eroded junction line. It seems to be the same deposit, very uniform in character, that is found along the sea -board from Aberdeenshire southward, differing from the till by its uniformly reddish colour and the fewness and smallness of its stones. THE PLEISTOCENE SERIES 623 3. Yorkshire and Lincolnshire When this northern ice reached the Cleveland Hills in East Yorkshire the mass of it was diverted into the Vale of York and, reinforced by contributions from the west, it spread out a confused mass of clays and gravels over the Triassic plain as far south as York and Escrick, where it has left some remarkable terminal moraines. Another lobe of this ice passed down the eastern side of the Cleveland Hills and along the coast to Flamborough Head. South of York there is a long and fairly broad area which is almost free of boulder-clay, a fact which is not yet fully understood, for there are boulder-clays to the west of it about Wakefield, Barnsley, and Doncaster, and one would have expected these to extend into the plains west of the river Trent. Boulder-clays occur in East Yorkshire near the coast and over the greater part of Lincolnshire, but they are of two different types ; those found on the eastern side of the Chalk Wolds being reddish and purple clays with intercalated marine deposits, while those on the western side are grey and more or less chalky clays. The Chalky Boulder-clay covers the central parts of Lincolnshire, spreading over the Jurassic Clays east of Lincoln by Tattershall, Horncastle, Wragby, and Market Rasen ; and the increase in the amount of chalky matter in the direction of the Chalk Wolds is particularly noticeable, till near these hills it becomes a stony Chalk-marl, and has been burnt for lime in some places. There is, however, very little of it on the Wolds themselves, though its occurrence at Kelstern, near Louth, proves that it originally spread over all levels below 400 feet. 4 It occurs also in South Lincolnshire, not only on the border of the Fenland district, but on the higher ground to the west, and appears to have formed a continuous mantle which covered even the summit ridges of the Jurassic escarpment and passed southward through Eutland, Northampton, and Huntingdon. The boulder- clay of the high ground between Stamford and Grantham is remarkable for containing huge masses of Lias Marlstone and Lincolnshire Limestone, which have been carried many miles from the spots whence they were quarried by the ice. These blocks appear to have come from the west and north-west. Carboniferous rocks came from the north-west, and Chalk from the north-east, so that several ice- currents seem to have met in this area. The boulder-clays of the eastern type occupy the area in Yorkshire known as Holderness, and underlie the whole of the eastern lowland of Lincolnshire. In both cases they and the associated sands and gravels are banked up against an ancient UJ *K<> I ;i I . TJ-^J I? II fc& m T5 ^ fl 61 _2 OT d s 11 6 g ?' i! to w P< o & S W . & 2 ^ M in THE PLEISTOCENE SERIES 625 coast-line which is in many places a line of buried cliffs. Such cliffs are intersected by the present coast -line at Sewerby in Yorkshire, and though they are not seen in Lincolnshire their existence is proved by the depth of Glacial deposits found at many places on the east of the chalk-surface. These deposits also enter the valleys of the Wolds, and appear to have mantled the slopes of the hills up to a height of at least 380 feet. The succession of beds seen in the cliffs near Bridlington and Sewerby is as follows, and is from 60 to 100 feet thick : 5 5. The Sewerby gravels. 4. The Hessle Boulder-clay, 3. The Purple Boulder-clay. 2. The Basement Boulder- clay. 1. Chalky gravel and Buried-cliff Beds. The buried -cliff beds are only seen at Sewerby, where they are banked against the ancient coast -line above mentioned (see Fig. Fig. 206. CLIFF-SECTION AT SEWERBY (after G. W. Lamplugh). 5. Sewerby gravel. 7. Purple boulder-clay. 6. Basement boulder-clay. 5. Chalky gravel. 4. Blown sand ^ r> n _. Vf , _,:<* 3. Rainwash I Buried cliff 2. Pebble beach J beds ' 1. Chalk 206). From the old beach and the overlying rain wash bones and teeth of the following Mammals were obtained : Elephas antiquus, Rhinoceros meroni, Hippopotamus amphibius and Hycena crocuta. Elsewhere the chalky gravel appears to form a nearly continuous basal bed. The Basement Clay is of a greenish-grey colour, is sometimes 30 feet thick, and includes twisted layers or patches of sand which contain marine shells ; no fewer than 110 species have been found, and among them are some which no longer live in British waters, such as Nuculce Cobboldice, Tellina obliqua, Astarte borealis, Fusus spitzbergensis, and F. Sarsi. The Purple and the Hessle boulder - clays have the same characters as in Lincolnshire (see below). The commonest stones in these and the lower clay have been derived from the Carboniferous rocks to the northward, but some have come from the Cheviot Hills, 2S 626 STRATIGRAPHICAL GEOLOGY others from Teesdale and from Cumberland, while a few are identical with rocks in the south and west of Norway. In Lincolnshire the basement -beds are nowhere visible, but borings prove that there is always a bed of chalk-rubble at the bottom, from 2 to 8 feet thick, and that this rests on a floor of Chalk which is about 80 feet below sea-level along the present coast- line. The thickness of the overlying deposits is from 60 to 100 feet, and they are of two kinds, (1) a lower clay of a purplish brown colour (the " Purple Clay "), containing stones and boulders both large and small ; (2) an upper clay (the " Hessle Clay ") of dull red colour mottled with ashen grey, containing many small stones, pellets of chalk, and bits of coal, but few large boulders. Near the border of the Wolds patches of sand and gravel are frequent, often resembling beach deposits, and containing many marine shells, some of which are perfect, but most are broken. They include Mya truncata, Oorbula gibba, Mactra subtruncata, Tellina balthica, T. lata, Cardium edule, Gyprina islandica, Venus gallina, and (at Croxton) Corbicula fluminalis, which is also abundant at Kelsey in Yorkshire. At Kirmington, and about 80 feet above sea- level, an interesting deposit of laminated silk or warp occurs con- taining Foraminifera, Rissoa ulvce, Scrobicularia plana, and other bivalves ; it resembles the warp formed between tide marks on the present coast, and is overlain by 10 or 12 feet of beach shingle. The occurrence of such undisturbed marine silt in association with the boulder-clays and marine gravels is of much importance. 6 The correlation of the beds above described with those of more southern counties and the relation of the eastern to the western deposits are debated questions, but two opinions have been ex- pressed, and a third view is also possible. (1) The eastern series may be newer than the western (Chalky Boulder-clay), and there- fore later than any deposit in East Anglia ; (2) the eastern series may be the older of the two and equivalent to the Lower Glacial of Norfolk ; (3) the eastern and western clays may be of nearly the same age but formed by different ice-streams, the one coming from the north and north-east, the other from the north-west. 4. East Anglia In the north-east of Norfolk, where the most complete sections of the Lower Glacial Beds are found along the coast, the following succession of deposits can be recognised, but there is 110 place where they can all be seen in vertical succession : The Upper or Chalky Boulder-clay. {Contorted Drifts (loams and sands). Brown Boulder-clay (Cromer Till). Arctic plant bed and Leda myalis bed. THE PLEISTOCENE SERIES 627 Above the Cromer freshwater beds and below the Glacial deposits there are two discontinuous beds which appear to be the oldest Pleistocene Beds in Norfolk. The Leda Myalis Bed is only found between Old Hythe and Cromer, for south of Cromer it either thins out or is cut off by the boulder-clay. The bed consists of fine current-bedded loamy sand, with thin seams of loam and gravel. The fossils are marine, and indicate a depth of 5 or 10 fathoms, as some of the bivalves occur with the valves united in the position of life. All the Mollusca yet found (nineteen species) belong to living species, and they include two characteristic crag shells, Neptunea antiqua and Tellina obliqua. One of the best exposures is at Runton Gap, near Cromer, where Leda myalis, Mya arenaria, and Astarte borealis may be found in position of life, and the bed is 15 feet thick. The Arctic Freshwater Bed was first described in 1880 7 by Mr. C. Reid, who then distinguished it from the Forest Bed, and placed it in the Pleistocene Series, because the plants it contains indicate a great change from the climate of the earlier bed. Trees seem to have entirely disappeared, and the plants include the dwarf Arctic birch (Betula nana], the Arctic willow (Salix polaris), and mosses which now only live within the Arctic circle. It consists of a layer of clay and loam about 2 feet thick, but is not by any means continuous, having only at present been found at three localities on the coast, viz. Beeston, Mundesley, and Ostend, near Bacton. At Mundesley it has yielded freshwater shells, elytra of beetles, and bones of Spermophilus (a squirrel-like creature). The Cromer Till, where thickest near Happisburgh, consists of two beds of tough grey unstratified boulder-clay, with an inter- mediate band of finely laminated and ripple -marked clay. The boulder-clays contain fragments of marine shells, with pebbles and boulders of many kinds of rocks, including gneiss, mica- schist, basalt, red and grey granites, " rhombynporphyr " from Norway, pieces of Carboniferous limestone, and pebbles of hard chalk which have been bored by marine annelids, and subsequently striated. The Contorted Drift is usually a yellowish marly and sandy clay full of stones, so that it is really only a variety of boulder-clay, but it exhibits a rude kind of stratification, and includes knots, patches, and seams of sand and gravel, and the whole has a twisted, contorted, and kneaded appearance. Another remarkable feature of the deposit is the occurrence of large tabular masses of chalk, some of which are 20 to 30 feet thick, and from 100 to 180 yards long. This drift contains boulders similar to those in the till .below and many broken shells, the latter being frequent in the 628 STRATIGRAPHICAL GEOLOGY nests and patches of sand, Tellina balthica, Cardium edule, Cyprina islandica, and Mya arenaria being the commonest species. Near Cromer the thickness of the Contorted Drift is sometimes as much as 170 feet, and the contortions are very marked, but southward and south-westward the deposit passes into a set of loams, sands, and fine gravels. Moreover, the underlying Cromer Till appears to thin out, so that over large parts of Norfolk, Suffolk, and Essex the Lower Glacial Group consists only of sands and gravels, with occasional intercalations of laminated loam or brick-earth. These beds were termed "Middle Glacial" by Mr. S. V. Wood. As they are followed to the higher ground in the western parts of these counties, and to the Chalk-escarpment in Cambridgeshire, they thin out and are overlapped by the Chalky Boulder-clay, which then rests directly on the Chalk. The Chalky Boulder-clay is an extension of that already described, and has been formed partly from the detrition of the Chalk and partly from that of the Jurassic rocks, materials from the Oxford and Kimeridge Clays being numerous, and derived fossils, such as Gryphcea arcuata, Gryph&a dilatata, Cardinia Listeri, and Belemnites abbreviatus, are not uncommon. In this respect it contrasts with the boulder -clays of the Lower Glacial Group. This boulder -clay exhibits a fairly uniform character over Norfolk, Suffolk, Essex, Cambridge, Herts, and Middlesex, the most southerly outlier being that on Muswell Hill, north of Highgate ; here the clay is of the usual pale colour, and contains occasional fragments of red chalk, which must have journeyed a distance of some 100 miles, for the nearest place whence they could have been derived is Hunstanton in Norfolk. The thick- ness of this great sheet of boulder-clay is very variable ; sometimes only a few feet can be seen, and in other places borings have gone through 100 or 150 feet of it. Another fact of some importance in connection with the formation of this clay is the occurrence in it of pebbles which have been carried up to a level much higher than that of their source ; thus phosphatic nodules derived from the Cambridge greensand have been found at a level of 500 feet above the sea, whereas the outcrop of the parent stratum to the north and west is everywhere under 150 feet. 5. Tlie Midland Area In the counties of Derby, Stafford, Warwick, and Leicester we find the meeting ground of four different ice -lobes (see map, Fig. 203). Around Derby and Nottingham, and thence south- THE PLEISTOCENE SERIES 629 east to Leicester, there are two very different kinds of boulder - clay ; the one a reddish clay containing stones of Pennine origin, the other a chalky clay with Chalk-flints and Jurassic debris, derived from places to the north-west. This Chalky Boulder-clay is clearly the same as that of Lincolnshire, and it is stated to overlie the Pennine Clay, but according to Mr. R. M. Deeley there is also in the Trent valley a still newer reddish clay with Pennine material, which he calls the Newer Pennine Clay. 8 There seems also to have been a dispersion of boulders in different directions from Charnwood Forest before the incidence of the ice from the north-west, for such boulders occur at places in Nottingham and Lincoln, as well as near Leicester and Coventry, that is to the north-east and east as well as the south-east and south-west. In Warwickshire, from near Broinsgrove to the north-east of Birmingham for a distance of some 20 miles, there is a line of boulders which have been transported from Wales, often accompanied by blocks of basalt from the Rowley Hills near Dudley. Lastly, in Staffordshire, round Wolverhampton, Cannock, and Rugeley, and even as far east as Lichfield, there are many boulders which have come from the Lake District and the south of Scotland. 6. Wales and the West of England The Glacial deposits of this district are also divisible into two sets, the one set consisting of detritus which has come from the north and has been transported by ice through or over the Irish Sea, while the other set consists entirely of Welsh material and has been carried by the glaciers which issued from the Welsh mountains (see p. 617). The following account of the Glacial Drifts of Lancashire and Cheshire has been compiled mainly from the writings of Mr. Mellard Reade 9 and Dr. A. Strahan. 10 The whole of the lower parts of Lancashire, Westmoreland, Cheshire, and North Shropshire, from Coniston and Windermere southward and inland up to heights of 500, and in some places 600 feet above sea -level, is covered by a nearly continuous mantle of boulder - clay, which is only broken in places where Triassic or Carboniferous rise up through it. This boulder-clay, however, is not a solid mass like the till of Scotland and Northumberland, but consists of irregular masses or sheets of stony clay with local intercalations of stratified sand and gravel. The deposits are in fact very like those of East Yorkshire and Lincolnshire, for the lower clays are generally purplish brown, and the upper clays resemble the Hessle Clay in being streaked 630 STEATIGRAPHICAL GEOLOGY with grey and in containing fewer stones. A good section of these drifts is exposed in the cliffs near Blackpool and is represented in Fig. 205). Both the boulder-clays and the sands fre- quently contain marine shells, generally broken, but sometimes perfect, and also abundance of Foraminifera. Moreover, the Gastropod shells are seldom filled with clay or red sand, but generally with a fine white sand or silt which is full of Foraminifera, showing that they have been picked up by the ice from silty shores and sea-bottoms, and these silt-filled shells are just as common in the boulder-clay near Macclesfield, at 600 feet, as they are near the coast. Near Macclesfield, too, there are sands with the same species of shells at a height of 1200 feet above the sea. The stones and boulders in these clays have come principally from the Lake District and the south of Scotland. They include granite from Eskdale and granophyr from Buttermere, other granites from Galloway, rocks from the Borrow- dale Volcanic Series, Silurian Grits, and Carboni- ferous limestones. Mixed with these in smaller numbers are blocks of Triassic sandstone and flints, which latter must have come from Antrim or from the west of Scotland. The finer material has, however, been largely derived from the Triassic sandstones and clays which underlie the Glacial deposits, and doubtless extend far to the west and north- west below the floor of the Irish Sea. These deposits fill up and conceal the in- equalities of a deeply sculptured pre-Glacial surface. The lakes of Coniston and Windermere are merely the upper ends of pre-Glacial valleys, the lower reaches of which are filled with Glacial drift, and where these old valleys open into Morecambe Bay the valley bottom is probably 200 or 300 feet below the present sea- level, for in Furness one such valley descends to 460 feet below sea-level and is filled with drift. Southwards these deposits sweep over the THE PLEISTOCENE SERIES 631 Cheshire plain into Shropshire and over the lower parts of North Wales up to heights of 400 and 500 feet above the sea ; while accumulations of sand and gravel with marine shells occur at still higher elevations, up to heights of more than 1 200 feet on both sides of Wales. Thus, on the summit of Moel Tryfaen, a hill overlooking Carnarvon Bay, with a height of 1350 feet, there is an extensive deposit of stratified sand and gravel containing marine shells, more than sixty species having been obtained. Among the most abund- ant of these are Tellina balthica, Cardium edule, Astarte borealis, Gyprina islandica, and Turritella communis. It is irregularly overlain by boulder-clay, and both deposits contain erratics from Cumberland and Scotland, as well as Chalk -flints which came probably from Ireland. These flints, as Mr. Mackintosh observes, have travelled to a higher level than their source, for the highest parts of the Irish chalk are between 900 and 1000 feet. 11 Similar shell-bearing gravels have been found by Mr. Mackintosh at heights of 1000 to 1230 feet on the east side of the hills near Minera in Denbighshire ; here also Eskdale granite, Cumberland felstones, and Chalk-flints are mixed with the debris of local rocks, and all are well rolled and rounded as if they had formed part of a sea beach. Another similar deposit of shelly sand and gravel occurs at Gloppa, near Oswestry. Within this Welsh area, how- ever, we find tracts and patches of another kind of drift, which is clearly the product of Welsh ice, and wherever the two are in juxtaposition the local drift always lies underneath the deposits left by the northern ice. This is well shown in the cliffs of Colwyn Bay, where the following succession is visible : 12 A' CO V f 4. Brown boulder-clay with broken shells and some -q- , I northern erratics . . . . . . 8 to 12 1 3. Obliquely stratified sands resting on an eroded I surface 6 to 12 {2. Cream-coloured till in discontinuous patches . to 2 1 . Hard bluish-grey till packed full of stones derived from the Carboniferous and Silurian rocks of Wales seen for 6 Most of Central and Southern Wales was covered by another ice-field, its centre of accumulation being the high ground which extends from Plyiilimmon southward and eastward through parts of Cardigan, Brecknock, and Carmarthen, and rises in many places to over 1800 feet. The thickness of the ice generated 011 this area was such that the ice -flow was only partially turned aside by the great mountain masses of the Brecknock Beacons and the Black Mountains, and was actually able to override the northern 'escarpment of the South Wales Coalfield. Proof has 632 STRATIGRAPHICAL GEOLOGY recently been found 13 that a great mass of ice passed down the valley of the Towy, but that only its lower part followed that valley westward, the upper ice -layers escaping southward over the escarpment on the southern side. Much boulder - clay occurs in the track of this ice - sheet from heights of 1500 feet downward, that around Plynlimmoii being described as a stiff blue clay or till crowded with rock-fragments of all sizes up to blocks of several tons weight. In South Wales much of it is coloured red by the debris of the Old Red Sandstone. In Glamorgan it passes southward into coarse angular gravel, and finally near the coast into roughly stratified gravels with rounded but striated stones. The interest of the last is increased by the fact that they overlie certain raised beaches so that the beach - material with its marine shells and the marine sand of the caves with the mammalian bones are thus proved to be of Glacial age. 14 The shells are of recent species, but the mammals include Elephas antiquus, Rhinoceros mercki, Bison prisus, Hycena spelcea, and Ursus spelceus. In Pembrokeshire, however, there is quite a different set of Glacial deposits. Those of Northern Pembroke between Cardigan and St. Bride's Bay have been examined by Dr. T. J. Jehu, 15 who found them divisible into (1) a lower boulder - clay, (2) inter- mediate sands and gravels, (3) an upper boulder - clay. The first is a tough bluish - grey clay containing boulders of Carboniferous Limestone, of rocks from the south-west of Scotland and many Chalk- flints, together with many broken marine shells. It occurs in patches up to 600 feet above sea-level. The sands and gravels frequently contain, marine shells which are generally broken, but some are perfect. Most of the species are such as now live in the Irish Sea, but some are Arctic forms, e.g. Trophon clathratus (common), T. scalariformis, Astarte compressa, and A. arctica ( = borealis) ; while the commonest shell is Pectun- culus glycimeris. These marine sands occur up to 640 feet. The upper clay is of a yellowish-brown colour and is best seen in the bays on the coast. The included stones are well scratched and are mostly from northern sources. It is sometimes 40 feet thick, but in places it passes into stony rubble. In South Pembroke the deposits are similar, 16 and near Nolton on St. Bride's Bay a mass of brown boulder-clay fills up a valley and forms cliffs which are 50 to 60 feet high ; it includes flints and many broken marine shells. Isolated transported boulders also occur all over the country, and their disposition shows that the ice-movement was from the north-west ; many come from distant sources. THE PLEISTOCENE SERIES 633 7. Ireland Ireland is largely covered by Glacial deposits, about which much has been written, but no comprehensive account of them has yet been published. Enough is known, however, to make it obvious that they were accumulated under the same conditions as those of England ; those of Northern and Eastern Ireland indeed greatly resemble the drifts of Western England, but those of the central and western districts constitute rather a different type. The central plain of Ireland " is largely covered by a widely spread mass of drift, consisting of dark sandy clay with pebbles and blocks and occasional beds of sand and gravel, which are sometimes very regularly stratified. The great majority of the pebbles are rounded fragments of Carboniferous limestone, whence the deposit usually goes by the name of the Limestone Drift. This deposit rests not only on the limestone, but sweeps up on to the flanks of all the hills which rise from the central plain. In. such case the Limestone gravel (often) becomes largely mingled with the detritus of the rocks of which the hills are made, . . . but gravel almost entirely composed of limestone pebbles is found up to heights of 1200 feet on the granite mountains south of Dublin. It spreads across the lower part of this granite range, and runs down by Bray into the county Wicklow, where it is covered by beds of sand and marl that spread through Wicklow and Wexford over all the low grounds between the mountains and the sea-coast " (Jukes). In some cases this Limestone Drift overlies a deposit composed entirely of local rocks, as in Glenbarrow, on the northern flank of the Slieve Bloom. Here at a height of 800 feet the river-cliff, 120 feet high, exhibits the following succession : Feet. Coarse drift with limestone boulders 50 Fine laminated sand ......... 20 Rubbly sand, with angular blocks of Old Red Sandstone, the same rock lying below .....'... 50 On the north-east coast there are deposits similar in all respects to the low-level drifts of Lancashire, consisting of red and brown boulder -clays, with frequent intercalations of sand containing marine shells. Cliff -sections of such bed are frequent along the coasts of Antrim and Down, and their marine origin is proved not only by the shells in the sands, but by the presence of Leda pernula and Leda minuta with attached valves in boulder-clay at Woodburn Glen, near Belfast. 17 Southwards these reddish clays appear to overlie the Limestone Drift, as seen in the cliff-sections at Killiney and Ballybrack, south 634 STRATIGEAPHICAL GEOLOGY of Dublin, and they attain a great thickness near Carlow, where a colliery shaft traversed 89 feet of stiff red clay with stones of various kinds overlying 33 feet of Limestone Drift. Drifts with marine shells occur again in Wicklow and Wexford, chiefly below 600 feet, but on Three Rock Mountain in Wexford they occur at a height of 1300 feet, and include some more southern forms, as Dosinia lincta, Venus striatula, and Venus casina. No notice of Irish drifts would be complete without a reference to the long winding banks or ridges of gravel which are known as eskers. These ridges often rest on the surface of the other Glacial deposits, and from these the materials composing them have been chiefly derived. They often extend in sinuous lines for many miles, but are confined to the plains and low grounds, not extending above the 400 feet contour. They may therefore be regarded as among the latest of the Pleistocene deposits, but the occurrence of large boulders on these banks shows that ice was still present. Morainic mounds occur in some of the valleys of the more mountainous regions, and testify to the existence of small valley glaciers during the last stage of the Glacial period. D. THE GLACIAL DEPOSITS OF EUROPE Ice-extension. The principal centre of ice -dispersion in Europe was the Scandinavian region. The striated rock-surfaces of Norway and Sweden show that a great snow-field accumulated on the highlands of those countries, forming ice -sheets which radiated outwards in all directions from the central axis of the Scandinavian chain. It has also been shown that the ice passed over mountain ridges which are 2700 feet high, so that the ice- sheet in such places must have been at least 3000 feet thick, and some authorities think 5000 feet a probable estimate of its maximum thickness. To the west and north-west this ice-sheet must have advanced far into the Atlantic before terminating in ice-cliffs like those of the Antarctic ice -sheet of the present day. South-westward it passed into the basin of the North Sea, and is believed to have crossed it so as to impinge upon the eastern coast of England. Southward it certainly crossed the Baltic, which is a shallow sea, and invaded the plains of Holland, Germany, and Russia. Eastward it not only crossed the Gulf of Bothnia, but passed up the Gulf of Finland and over Northern Russia to the Ural Mountains, and south-east into the very centre of the Russian plain (see Fig. 208). In the North Sea it was met by the independent and smaller ice -sheet formed by the eastern part of the Scottish ice; and the THE PLEISTOCENE SERIES 635 southern limit of tlie confluent and continuous ice at the time of its maximum extension, whether floating or resting on sea-floors, is shown in Fig. 208. This southern limit passes through the Bristol Channel and up the valley of the Severn, then across England to that of the Thames, through Belgium by Brussels and Dusseldorf to the foot of the Westphalian Hills ; round the Hartz Mountains (which are glaciated to a height of 1470 feet) to the Erz Mountains and the Kieseii and Sudeten ranges, passing by Cracow and Vladimir into the basin of the Dnieper and out again in an irregular line through Central Russia. Glacial Deposits. Many Swedish and German geologists have endeavoured to educe some definite order of succession out of the congeries of Glacial deposits which covers such a large part of their respective countries. Most of the Norwegian and Swedish geologists recognise two boulder -clays, a lower bluish clay and an upper clay which is generally of a yellow colour, but in many parts of the region it seems difficult to distinguish them. In Scania, however, the distinction is more clearly marked by the frequent interposition of stratified sands and gravels, and sometimes of laminated clay containing remains of the dwarf birch and other plants. There are also long continuous ridges of moraiiiic material, one behind the other, which are regarded as marking successive pauses in the recession of the ice-sheet. In Germany the evidence for two distinct boulder-clays seems to be stronger, and some find reason to believe in a third. The Lower " Diluvium " is the most extensive, and is described as a bluish or brownish clay containing many boulders of Scandinavian origin, and it is the southern border of this clay which is shown on the map (Fig. 208). Above it in many places are stratified deposits, generally sands and clays with occasionally beds of peat ; these contain plant remains and also the bones of Elephas antiquus, E. primigenius, Rhinoceros tichorhinus, Cervus megaceros, and C. elaphus. The plants in these beds are specially remarkable because they include not only pines and firs, but leaves of the aspen, poplar, oak, hornbeam, ilex, Trapa natans, and a water-lily (Oratopleura), the assemblage indicating quite a temperate climate no colder than that of Germany at the present time. The Upper Diluvium (or boulder -clay) covers Denmark, Schleswick-Holstein, Northern Germany, and the Baltic provinces of Russia, and is associated with mounds and ridges of terminal moraines which mark the final stages in the retreat of the ice- sheet. One such ridge has been traced from near Neustrelitz in Mecklenburg, south-eastward to Oderberg, a distance of 35 miles, and probably reaches still farther. Borings have shown that the 636 STRATIGRAPHICAL GEOLOGY THE PLEISTOCENE SERIES 637 combined thickness of Glacial deposits on the German plain is variable, as might be expected, but that it is sometimes 300 feet and in a few places even more. The Swiss Glaciers. The refrigeration of climate which brought Arctic conditions into Northern Europe was also felt in Switzerland and caused a great extension of the Swiss glaciers, till they not only filled their valleys, but coalesced with one another, and spread out over the plains which lie both to the north and south of the Alps. Thus the glacier of the Rhone spread across the plains of Geneva and Neuchatel onto the flanks of the Jura Mountains, where it left a long line or belt of transported boulders at an average height of 800 feet above the lakes. Similarly the glacier of the Rhine spread out far beyond the Boden See (Lake of Constance), passing northward and eastward into the drainage area of the Danube. As in Germany so in. Switzerland there is evidence of two distinct Glacial epochs or advances of the glaciers, separated by an interval of temperate climate. The older deposits 'are coarse gravels and a kind of boulder-clay or argillaceous moraine. The inter-glacial deposits are fluviatile gravels and sands with beds of lignite at Diirnten, Utznach, and other places, which yield leaves of plants and trees of the same species as now live at these localities, together with bones of Elephas primigenius, Rhinoceros etruscus, Ursus spelceus, and Bos primigenius. These beds rest in places on the older moraines, and are overlain by newer moraine-stuff. Of still later date than the last advance of the glaciers are the terraces of lacustrine deposit at 150 feet above the Lake of Geneva, which prove that the lake has been lowered to that extent since the retreat of the ice during the erosion of its valley by the ex- current river (the Rhone). E. NON-GLACIAL DEPOSITS 1. Classification and Fauna Relative Age. Under this head we group those Pleistocene deposits which have not been accumulated by moving ice and are not interstratified with boulder - clays. The principal deposits of this kind may be catalogued as follows : 1. Clay with Flints. 4. Lacustrine Deposits. 2. Plateau Gravels. 5. Cave Deposits. 3. River Gravels. 6. Raised Beaches. 7. Alluvial Levels. These deposits are chiefly found outside the glaciated areas, 638 STEATIGRAPHICAL GEOLOGY but some of those classed as Plateau gravels occur within these areas and overlie the higher boulder-clays ; moreover it has been proved that some of the raised beaches and some of the Cave deposits are of Glacial age. Lastly, the assemblage of mammals found in the older river deposits is the same as that which is regarded as early Pleistocene (see p. 612). Again, the older river gravels and the older cavern deposits .appear to be of about the same age, and in both the stone weapons of Palaeolithic man are associated with the bones of many extinct animals. In some places it seems possible to distinguish two Palaeolithic horizons or ages, an older of roughly made implements associated with Elephas antiquus, and a newer of more finished weapons for which the name Mesolithic has been proposed. Some authorities recognise a third age, that in which remains of the reindeer are abundant, associated with a special type of flint implements. If the latter is in real succession and not a mere local variation the Pleistocene (non-Glacial), succession may be tabulated as follows : 4. Neolithic, the age of domesticated animals. "^'^ M 1'th' /th e a S e f the reindeer (Rangifer tarand-us). 2. / \ the age of Elephas primigenius and Rhinoceros tichorhinus. 1. Palaeolithic, the age of Elephas antiquus. In France another scheme of classification is employed, all the rougher implements being called Palaeolithic, but classed according to type and finish in three groups, the Chellean or Acheulian, the Mousterian, and the Magdalenian. Of these the first corresponds with the restricted Palaeolithic in the above table, and the two latter with the Mesolithic. Fauna. The mammalian fauna consists first of a number of species which are common to all or most of the phases of the period, .and secondly of certain groups of species which are either restricted to one epoch or phase, or are much more abundant in the deposits of that phase. Again, many of the Pleistocene animals are survivals of the later Pliocene fauna, and of these some became extinct before Neolithic time, and some still continue to exist either in Europe or Asia. The oldest assemblage (Palaeolithic) includes Elephas antiquus, Rhinoceros mercki, Eh. leptorhinus, Hippopotamus major, Ursus spelceus, Machcerodus latidens, and Trogontherium Cuvieri. The Mesolithic implements are associated with Elephas primi- yenius, Rhinoceros tichorhinus, Cervus giganteus ( = megaceros], C. elaphus, Felis spelcea, Ganis lagopus (Arctic fox), Equus caballus, Saiga tartarica (an antelope), Arctomys marmota (the Marmot), and species of Spermophilus. THE PLEISTOCENE SERIES 639 Tlie Neolithic fauna is very different from the above, and includes only one extinct species (Cervus giganteus), though there are, of course, many animals which have subsequently become extinct in Britain, such as the reindeer, brown bear, wolf, wild boar, and beaver. Palaeolithic man was a hunter only; Neolithic man was a herdsman and agriculturist, and he introduced several domesticated animals, such as the dog, sheep, goat, pig, and short-horn cow. Fig. 209. PLEISTOCENE LAND AND FRESHWATER SHELLS WHICH ARE NOW EXTINCT IN BRITAIN. (Reproduced by permission of Mr. B. B. Woodward and the Council of the Geologists' Association.) 1, 2. Unio littoralis. 3, 4. Corbicula fluminalis. 5. Pisidium astartoides. 0. Pisidium amnicum. 7. Helix (Eulota) fruticum. S. Hydrobia inarginata. The molluscan fauna of the older Pleistocene deposits, i.e. the fauna of the freshwater deposits, is also interesting, for it includes a few extinct species and a certain number which no longer live in Britain, though most of them still exist in Europe or Asia. Prominent among these is Corbicula fluminalis, a species which is extinct in Europe but still lives in the rivers of North Africa ; another is Vivipara diluviana, believed to be entirely extinct, as is 640 STRATIGRAPHICAL GEOLOGY also the allied species V. clactonensis ; Neritina grateloupiana, found in the Thames valley, is also extinct though allied to the modern N. danubialis. A common shell in the older river gravels is Unio littoralis, which is extinct in Britain but still living in Europe. Among land shells there is Helix fruticum, now only found alive in Central Europe, and Olausilia pumila, found at Barn well, near Cambridge, but now only living in parts of Northern Europe and in Silesia. Hydrobia marginata and Pisidium astartoides are other species found in British Pleistocene deposits but not now living in Britain. These and some of the above are shown in Fig. 209, for the use of which I am indebted to Mr. B. B. Woodward. 2. British Deposits As it would be impossible within the limits of this volume to give anything like a comprehensive account of the clays, gravels, cave -earths, raised beaches, and other non-Glacial deposits in Britain, only a few examples will be described. Clay-with-Flints. This is a deposit or accumulation of indefinite age, occurring at high relative levels and mainly composed of materials derived from the Eocene clays and sands ; hence it only occurs in the south of England and the north of France. Its typical form is that of a stiff reddish-brown clay containing a large number of flints, the majority of those in the lowest layer being entire unworn Chalk-flints, while those in the higher part are angular pieces of broken flints, with some quartz pebbles and fragments of ironstone. In some districts there is also a large percentage of sand in the clay. The whole deposit lies irregularly on the Chalk, from which the unworn flints have been derived, and by the solution of which it has been let down into hollows, funnels, and pipes. This clay is often overlain by, and sometimes it seems to pass laterally into, thick deposits of loam, brick-earth, sand, and gravel, all of which may have been derived from the destruction of the Eocene Beds. 18 Plateau Gravels. This name has been given to various unstratified gravels in the southern, eastern, and midland parts of England, because they occur on the plateaus out of which the modern river-valleys have been excavated. In Norfolk and Suffolk they generally rest on the Chalky Boulder-clay, and consist mainly of large rolled and battered flints whence they have been called " Cannon-shot Gravels." In the Midlands they rest on Jurassic rocks and contain a mixed collection of stones, most of which have been derived from more or less distant sources. Thus around Oxford the Plateau drift contains quartzite -pebbles from the Bunter, Carboniferous THE PLEISTOCENE SEKIES 641 sandstones and cherts, Liassic stones, Red Chalk, and various igneous rocks. The assemblage is in fact that found in the Glacial deposits to the north, and Mr. Pocock is of opinion that the gravel is of Glacial age and is in part a decalcified Glacial deposit. 19 He admits, however, that at lower levels, 130 to 200 feet above the rivers, it is sometimes rudely terraced as if it had been redistributed by fluviatile action. In Berkshire and East Oxfordshire the constituents are chiefly subangular flints and materials derived from the Eocene Beds such as flint-pebbles, quartz-pebbles, sarsen-stone, and ironstone. Mr. Osborne White also regards the gravel as of Glacial age, but as belonging to more than one Glacial epoch, and he remarks that " the Plateau and Valley gravels are very often inseparable ; their more level spreads appear to be but the higher and lower members of a single graduated series, the steps and half-landings of a ruined stairway that was never complete." 20 River Gravels. The manner .in which disconnected tracts and terraces of gravel, sand, and loam have been formed and left at various levels in a river-valley is explained in all text-books of Physical Geology. They represent different stages in the excava- tion of a valley, those at the highest level above the existing stream being of course the oldest, and the others newer in proportion as they are nearer to the level of the modern alluvium. The deposits in the valley of the Thames may be taken as an example of such accumulations. In the higher parts of the valley tracts of gravel are found at various heights, now on one side of the river, now on the other ; three stages or terraces can often be distinguished, the highest being about 40 feet above the level of the river. From this highest terrace few bones have been obtained, but at the lower levels land and river shells, with remains of Elephas primigeniuSj Elephas antiquus, Rhinoceros tichorhinus, Bison priscus, Ranyifer tarandus, Equus caballus, Canis lupus, and Ursus arctos are frequently found. Between Windsor and London the outspreads of gravel become wider and more continuous, but the surface of the older parts still rises to 30, and even 45 feet above the river, and the shells are still of fluviatile species. Hence we may conclude that when these gravels were deposited the Thames had a much longer course than it has at present, and that this part of the valley was then far inland, otherwise the beds would be at a lower level and the shells would be estuarine. Below London, at Ilford, Grays, Crayford, and Erith, there are a series of gravels, sands, and loams which rise to 25 or 30 feet above the river, but are nearly 60 feet above the base of the alluvium at Erith. These beds have yielded the mammalian remains 2 T 642 STKATIGRAPHICAL GEOLOGY mentioned above, and it is only at these places that Rhinoceros leptorhinus and R. tichorhinus have been found together. 21 The presence of Rhinoceros megarhinus (a Pliocene form) and the absence of the reindeer (Rangifer tarandus) are regarded as indications of the antiquity of the deposits. Between fifty and sixty species of land and freshwater shells have been found, several of which 110 longer live in Britain, e.g. Cyrena (Corbicula) fluminalis, Unio littoralis, Pisidium amnicum, Hydrobia marginata, and Helix fruticum. A worked flake proving the presence of Man has also been found in the lower brick-earth. Many flint implements of Palaeolithic types have also been found in high-level gravels at Acton and Stoke Newington on the north side of London. The district now drained by the river Cam contains an interesting series of river deposits which bear testimony to the successive changes that have taken place in the drainage system of this area. 22 The oldest line of gravels has no connection with the modern river - valley, but crosses it nearly at right angles near Cambridge. Patches of gravel in the hill-valleys near Balsham testify to the action of streams where none now exist, and these seem to have united to form a river which ran by Wilbraham, Quy, Cambridge, Girton, Oakington, Long Stanton, and Willingham. The deposits it has left now form long gravel-capped ridges, the base of the gravels where intersected by the Cam being about 40, and their surface sometimes 60 feet above the modern alluvium. These fluviatile gravels appear to terminate abruptly near the fens of the Bedford Level, but were probably originally continuous with the gravels at Chatteris, Doddington, and March, which form similar ridges rising above the general level of the fens which surround them, and evidently mark the estuarine channel of a river which traversed the country when it stood at a somewhat lower level, and before the Fen-land existed. The shells found at March are chiefly marine, but include Corbicula fluminalis, Of deposits which bear a definite relation to the modern river Cam, remains of three terraces can be recognised. The highest of them runs by Trumpington and Barnwell, about 25 feet above the alluvium, and has yielded a number of mammalian bones and teeth, with a still larger number of shells, among which Corbicula fluminalis, Unio littoralis, and Hydrobia marginata occur. These three shells do not occur in the lower terraces, which are respec- tively about 15 and 8 feet above the alluvium near Cambridge. Lacustrine Deposits. Two notable instances of such deposits occur in the east of England. That of Hoxne, near Diss in Suffolk, is perhaps the best known, pits at that place exposing over 30 feet of lacustrine beds without reaching the bottom ; the lower beds are THE PLEISTOCENE SERIES 643 marls containing bones of mammalia, freshwater shells, and remains of plants ; these are overlain by brownish clay or brick-earth enclosing flint implements of a Palaeolithic type. The plants include leaves of Salix polaris, S. myrtinites, and Betula nana (all Arctic forms), with others of more southern character, but the flora somewhat resembles that of modern Iceland. 23 Similar deposits occur in a broad depression south of Hitchin in Herts, and consist of alluvial marls and silts resting on boulder- clay and overlain by brick-earth with implements of Mesolithic type. 24 The flora, however, does not include the Arctic species, and indicates a more temperate climate. Cave Deposits. -The origin of caves and the formation of the deposits found in them is explained in most books on Physical Geology. Many such caves have now been explored in England and Wales ; some of them have served as dens for bears or for Fig. 210. SECTION ACROSS THE TRUMPINGTON GRAVELS. Distance 2 miles. a. Lowest terrace. d. Chalk marl. . ft. Second terrace. e. Gault. ' c. Highest terrace. /. Alluvium. hysenas, and the bones of other animals belong to carcases which have been dragged in by these creatures. At other times they have served as shelters for man, whose occupation of them is proved by the frequent occurrence of stone implements Palaeolithic in the lower beds, and sometimes Neolithic in the higher. The caves of Creswell Crags in the Magnesian limestone on the borders of Derby and Notts afford good examples of such deposits. 25 In Robin Hood Cave the beds found were : Feet. 5. Stalagmite united to the roof 1 to 2 4. Breccia with bones and implements . . . . . to 3 3. Cave earth with bones and implements . . . . 2 to 4 2. Red clayey sand with bones . . . . . . 2 to 4 1. Light- coloured sand with blocks of limestone fallen from the roof 2 The lowest sand yielded no organic remains, but the red sand contained bones of many animals, all of them scored with the teeth- 644 STRATIGRAPHICAL GEOLOGY marks of the creatures which had dragged them into the cave, and that these were hyaenas is proved by the abundance of hyaena bones and teeth belonging to individuals of all ages from cubhood upwards. Flint implements are rare in the sand, but abound in the overlying cave earth, some consisting of flint and others of quartzite, but all of Palaeolithic form ; bones of the same animals occur in this layer. Lastly, in the breccia above are flint and bone implements of the Neolithic type. Some caves in the Vale of Clwyd (North Wales) have been explored by the late Dr. Hicks, and are important from their position in relation to the local Glacial deposits. 26 They are known as the Ffynnon Beuno and Cae Gwyn caves, and are situate in a small valley or ravine on the east side of the Vale at a level of nearly 400 feet above the sea. The valley has apparently been filled with Glacial drift of the low-level Lancashire type (see p. 629), and one of the entrances to the Cae Gwyn cave was found to be buried beneath a bank of stratified sand, gravel, and boulder-clay, the lower layers of which entered the cave and overlay the bone-bear- ing cave-earth. The section at this entrance was as follows : Feet. Ins. Reddish boulder-clay with sandy seams . . . . 64 Sand with boulders and seams of clay ..... 3 9 Gravelly sand with tine laminated sand below . ..35 Red laminated clay ........ 1 Cave earth with bones and flint implements . . . 2 to 5 Dr. Hicks inferred that the contents of the cave were pre-Glacial, but this was not proved by the evidence, and the assemblage of mammalia is that of the later Palaeolithic time, the commonest species being Rhinoceros tichorhinus, Elephas primigenius (the mammoth), and Eangifer tarandus (reindeer). They must, there- fore, be regarded as inter-glacial, but anterior to the submergence which enabled the northern ice to travel so far southward. Another important cavern is that known as Kent's Cavern, near Torquay in Devon. This was explored by Mr. Pengelly with assistance from the British Association, and classified collections of its contents may be seen in the museums at Torquay, South Kensington, and Oxford. Where the succession was complete it read as follows : 5. Black mould. 4. Granular stalagmite (5 feet) with some bones. 3. Cave earth with many bones and Palaeolithic implements. 2. Hard crystalline stalagmite (6 to 10 feet) with bones of bears. 1. Breccia with bones of bears and rude stone implements. The cave seems to have been first occupied by bears and THE PLEISTOCENE SERIES 645 occasionally by men of early Palaeolithic race. In the cave earth (No. 3) hyaena bones are the most abundant, but remains of twenty- five other mammals were found, including Elephas primigenius, Rhinoceros tichorhinus, and Rangifer tarandus. With these were associated flint implements of two recognised Mesolithic types and a few tools made of bone. This deposit evidently represents a long period of time, and is covered by the upper stalagmite containing remains of hysena, mammoth, fox, and horse. Raised Beaches. Raised beaches and sea -margins occur along many parts of the British coast-line. The beaches consist of sand and gravel, irregularly stratified, and often containing drifted shells, such as may be found on modern shores. Deposits of this kind occur at Brighton and westward along the southern part of Sussex as far as Portsdown Hill, the highest level at which they are found being about 130 feet above the sea. Inland they are banked against an old line of cliffs, the position of which is shown in Fig. 173. Eaised beaches exist also at Portland Bill, where the highest part of the deposit is 50 feet above sea-level, and at many places round the coasts of Devon and Cornwall. 27 All these are probably contemporaneous with the raised beaches of South Wales (see p. 632), which are older than the local Glacial deposits, and in fact of meso-Glacial age. On the Scottish coasts raised beaches and lines of cliff are found at various successive levels between 25 and 100 feet above high-water mark, four or five sometimes occurring one above another. These are of more recent age, the highest being contempor- aneous with latest moraines. Alluvial Levels. These are marshy flats formed of the deposits left by the flood-waters of a river ; they border the actual channel of the river and widen out in the lower part of its valley, till near the river's mouth they often form wide levels or marshes, which are partly of fluviatile and partly of marine origin. The depth of alluvial matter which occupies the ancient valleys of our rivers near their mouths is sometimes very great ; thus at Sheerness, and also at Upchurch, near Sheppey Island, borings have proved 77 feet of such alluvium. In the Essex marshes there is sometimes 50 feet of alluvial matter. At Yarmouth there is said to be 170 feet, and in the valleys of the Yare and Wensum there appears to be sometimes 60 or 70 feet of alluvium between the present surface and the chalk which lies at the bottom of the valley. Besides the deposits above mentioned which lie within river- valleys, there are other tracts round our shores which have once been bays and inlets of the sea, and are now filled principally with marine silts and clays, though layers of peat with trunks of 646 STRATIGRAPHICAL GEOLOGY trees are frequently intercalated with the silts, and mark pauses in the process of silting up. Many such tracts pass beneath the present beaches, and were evidently once much more extensive than they are now. Such are the marshes near Fleetwood, in Lancashire, the Bridge water Levels in Somerset which pass beneath the Bristol Channel and must formerly have extended far down this estuary, the Pevensy Levels in Sussex, the Romney marshes in Kent, the Fens of Cambridge and Lincoln, and the marshes of the East Lincolnshire coast. 3. Continental Deposits The cave earths and river deposits of France and other con- tinental countries do not differ in any essential respect from those above described, and the French classification of them according to types of flint implements has already been mentioned (p. 638). Here, therefore, it is only necessary to describe such deposits as differ from anything existing in Britain. Of these the two chief are (1) the Loess, (2) the post-Glacial deposits of the Baltic area. The Loess. This deposit is found in the north-east of France, and in the valley of the Rhine as far south as Basle, over parts of Southern Germany, the lower parts of Bohemia, Moravia, Hungary, and Galicia, up into the Carpathian Mountains, and also through Poland and Silesia over the plains of Southern Russia. Though thickest in the valleys, it also extends onto the slopes of the hills and reaches up to levels of 2000 and even 3000 feet, but is there mingled with local detritus. Loess is a fine yellowish or brownish argillaceous material, similar to some of the yellow loams which occur in fluviatile deposits, but it is unstratified and has a marked tendency to break along vertical planes and thus to form vertical walls. It is always more or less calcareous, and often contains calcareous concre- tions, but the upper part is generally decalcified. It appears to be a terrestrial deposit, for the fossils it contains are almost entirely land-shells and mammalian bones ; freshwater shells being of rare occurrence. The commonest shells are Helix (Hygromia) hispida, Succinea oblonga, and Pupa muscorum. Diverse views have been held with regard to the origin and the age of the loess ; some have thought that it resulted from the blocking of the great rivers flowing from Central Europe, their waters being ponded back by the northern ice-sheet and large lakes thus formed. Others with more reason regard it as an seolian deposit formed during an epoch of dry cold ; that it was, in fact, an accumulation of dust carried by the wind and deposited on grassy steppes like those of Southern Siberia. This has been THE PLEISTOCENE SERIES 647 confirmed by the discovery of a steppe-fauna in the loess at Thiede in Brunswick and at Westeregeln in Prussia. This assemblage includes some northern forms, such as Elephas primigenius, Cervus tarandus, and Ovibos moschatus, associated with species that now live on the Siberian tundras such as Saiga tartarica, Myodes torquatus (a Lemming), Arctomys bobac (a Marmot), Lagomys pusillus (the tailless hare), Alactatus jaculus (a Jerboa), and several species of Spermophilus. With regard to age it has been considered to be entirely post- Glacial or late Glacial, but the evidence tends to favour the view that its formation extended over a long period of time, that its lower parts are what may be termed meso-Glacial and contain flint implements of- Mesolithic types, while its higher parts are late Glacial and locally post-Glacial. Deposits of the Baltic Region. During the final retreat of the Scandinavian ice-sheet, the sea gradually reoccupied the trough of the Baltic, together with parts of Southern Sweden, a large part of Finland, and a small part of Kussia between the Gulf of Finland and the White Sea. The Swedish geologists Munthe and de Geer assume that this was due to a concomitant subsidence, but it is quite as likely that the ice-sheet had occupied the place of the sea, which simply returned into the area as the ice retired. The deposits of this sea are laminated clays containing Yoldia arctica, Y. hyperborea, Pecten islandicus, Sipho togatus, Trophon truncatus, Natica affinis, and other boreal forms which are only now found in the Arctic Ocean ; hence they have been called the deposits of the Yoldia Sea. They do not occur in Germany, but are found in the south of Sweden, and thicken northwards till they attain a thickness of over 800 feet in Angermanland, so the conditions must have lasted a considerable time. This geographical phase was brought to an end by an uplift of the region which not only raised the area north of Lake Onega, but also united Sweden to Denmark and converted the Baltic basin into an enclosed sea which quickly became a freshwater lake. This has been called the "Ancylus Lake" because Ancylus fluviatalis is a common fossil in its deposits, associated with Limncea ovata, Planorbis contortus, Bithinia tentaculata, and other freshwater shells ; these deposits rest unconformably on the Yoldia Beds. This phase was terminated by a subsidence which was greater on the western than on the eastern side, allowing the waters of the .North Sea to flow freely into the Baltic area, so that the freshwater lake was quickly converted into a gulf of the sea. Moreover the opening must have been wider than it is at the 648 STKATIGRAPHICAL GEOLOGY present time, for the gulf was completely filled with salt water, and from the abundance of Littorina littorea and L. rudis in the grey clays left by these waters it has been called the Littorina Sea. The present conditions of brackish water with a limited and stunted fauna have been produced by a second elevation of the area. The Ancylus Lake did not extend so far west as Christiania, where the succession of shelly clays has been described by Sars, Kjerulf, (5yen, and Brogger 28 ; the last-named recognising the following series : Post Glacial Late Glacial 6. Scrobicularia Clays with S. piper -ata. 5. Tapes Beds with T. pullastra. 4. Cardium and Ostrea Beds with C. edule. 3. Mya Beds with Mya truncata, Mytilus edulis, Cyprina islandica, Tellina baltica, etc. 2. Area Clays with Area glacialis and Yoldia lenticula. 1. Yoldia Clays with Y. arctica, Leda pcrnula, etc. Of these numbers 1 and 2 were formed during subsidence and correspond with the Yoldia Clays of Sweden. The Mya Beds were formed during the subsequent elevation and occur at a little over 200 metres above the sea (660 feet). The newer deposits range from 80 metres (260 feet) to sea-level, and the land appears to have been rising all the time. F. SOME PHASES OF PLEISTOCENE GEOGRAPHY The history of the Glacial epoch has yet to be written, for geologists are not yet agreed as to the interpretation of the records on which such a history must be based, or as to the causes which produced such a great change in the climate of Europe. So far as we know there are only three sets of causes which could have wrought such a change of climate ; these are (1) cosmical causes affecting the whole globe ; (2) geographical changes involving the connection or disconnection of continents, and the consequent divergence or introduction of marine currents ; (3) alterations in the arrangement of the areas of high and low barometric pressure, affecting the direction of prevalent winds. The cosmical causes have been advocated by the late James Croll, who published a book on the Cause of the Ice-age in 1891, but his views have been controverted by Professor Newcombe and Mr. E. P. Culverwell. The possible geographical causes were discussed by Lyell, who argued that the elevation of land in the Arctic regions would tend to refrigerate the northern hemisphere, and that if continuous land existed between Labrador and . the British Isles the climate of Northern Europe would probably become colder. Others again have suggested geographical changes THE PLEISTOCENE SERIES 649 which would alter the course of the Gulf Stream, but all these ideas are unproved hypotheses. The alteration of meteorological conditions as a cause of climatal change has recently been discussed by Mr. F. W. Harmer, 29 who points out that the winds are an important factor in determin- ing the distribution of climatal zones, and that existing deviations of the monthly or yearly isotherms from the normal are coincident generally with the direction of the prevalent winds. " The winds blow in a direction more or less parallel to the isobars ; the latter group themselves round centres of high and low pressure. . . . Anomalous weather is due to some unusual arrangement of these areas ; similarly former cases of anomalous climate can only have occurred when meteorological conditions were favourable. " During the Glacial period the regions covered by ice might have been, to a greater or less extent, anticyclonic at all seasons, low-pressure systems prevailing at the same time over the warmer regions immediately to the south of them and over the adjoining oceans. The relative positions of areas of high and low barometric pressure, the direction of the prevalent winds, and the consequent distribution of climatic zones would in such a case have differed from those of the present time." The prevalence of south-westerly winds over the British area at the present time is due to the fact that the centres of cyclonic storms generally pass from the Atlantic to the north or north-west of our islands. Under the conditions above imagined they would pass farther to the south, bringing oceanic winds over the Sahara and the Mediterranean Sea, while easterly winds would prevail over Britain. It has already been mentioned that the shell-banks of the Suffolk and Norfolk Crags afford actual evidence of such easterly winds during the later part of the Pliocene period. Further, as Mr. Harmer points out, " the diversion of the prevalent winds of the northern part of the north Atlantic from a south- westerly to a south-easterly direction would have tended to divert the Gulf Stream, or what remained of it, towards the American coast." This abstraction of the Gulf Stream would have an enormous effect on the climate of Western Europe, equivalent, according to Dr. Croll, to a lowering of the mean annual temperature of London by 40 F. With regard to the actual difference in the climate of Northern Europe at the time of the greatest extension of the ice, Mr. C. Eeid has estimated that the average temperature of the air along the southern margin of the ice -sheet was not higher than freezing point, and that the mean annual isotherm of 38 F. must then have passed through the south of England and across Central 650 STRATIGEAPHICAL GEOLOGY Europe in an oblique direction to the northern shores of the Black Sea ; in other words that the mean annual temperature was 20 lower than it is now along this line. 30 He adds that " the Arctic plants and animals found fossil in Britain point to a similar conclusion, for the same species now nourish in a climate fully 20 colder than that of the lowlands of Britain." Of the geographical changes which took place during the Pleistocene period within the European area some have been indicated in the preceding pages, but a brief summary of the probable sequence of events may here be given, the reader being warned that I only express the views of those who seek to avoid extremes and to reconcile the different inferences which have been drawn from the facts. At the beginning of the period, i.e. at the close of Pliocene time, the British Isles were broadly united to the continent across the English Channel and across the southern part of the North Sea to Holland and Belgium. The Rhine discharged itself into the North Sea Gulf near the coast of Norfolk, and the height of Eastern England above sea-level was about the same as at present. The western parts of the British region were, however, at a much higher level than they are now, and Ireland may have been broadly united to Scotland and England, though possibly a deep narrow estuary, receiving a river from the north, lay between Ireland and Wales. The first formation of glaciers in Ireland, Scotland, and Wales may not have been accompanied by any terrestrial movement, and the same may have been the case in Scandinavia. But of this early glacial episode few traces remain, for most of the glacial products of it would have been swept away by the subsequent advance of the ice. There is good reason to believe that the great extension of the ice and the consequent formation of ice-sheets coincided with a general subsidence of Northern Europe, which enabled the ice to spread much farther south than it otherwise would have done, and to rise over the land in such a manner as to carry material from lower to higher levels. The extent of this submergence in Britain is at present uncertain, but it seems to have been greater in the north, where the ice was thickest, and less in the south where the ice-sheets terminated. Thus as stated on p. 615, Scotland seems to have been about 600 feet lower than at present, while in the south of England the difference was not more than 150 or 160 feet. But although we may infer a subsidence of that amount in the area of the English Channel, we must not suppose that the Straits THE PLEISTOCENE SERIES 651 of Dover were formed by that subsidence. England was still united to France, but by a much narrower tract of land, possibly only by an isthmus joining the Wealden area to the Boulonnais, the invasion of the low-lying country being protected by the continuation of the South Downs across the Channel from near Beachy Head to the French coast north of the mouth of the Somme, and by the North Downs between Dover and Calais. For an account of the subsequent elevation of the British Isles and of the final subsidence by which Ireland was separated from England and the latter from France, the reader is referred to the author's Building of the British Isles (3rd ed.). The later geographical changes in the Baltic region have already been mentioned, and the history of the Neolithic time belongs rather to the antiquary than to the geologist. For a good account of what was known up to 1880 the reader is referred to Professor B. Dawkins's book on Early Man in Britain. EEFERENCES 1 See E. J. Garwood, Quart. Journ. Geol. Soc. vol. Iv. p. 681. la See T. F. Jamieson, Quart. Journ. Geol. Soc. vol. Ixii. p. 13 (1906). 2 See Goodchild, "Glacial Phenomena of the Eden Valley," Quart. Journ. Geol. Soc. vol. xxxi. p. 55 (1875). 3 See H. Miller on "Boulder Glaciation," Proc. Roy. Phys. Soc., Edinburgh, vol. viii. p. 156 (1884). 4 For Lincolnshire consult the Memoirs of the Geological Survey, and see Quart. Journ. Geol. Soc. vol. xli. p. 118. 5 On Holderness consult "Geology of Holderness" by C. Keid, Mem. Geol. Survey, and G. W. Lamplugh in Quart. Journ. Geol. Soc. vol. xlvii. p. 384. 6 See Rep. Brit. Assoc. for 1904, p. 272. 7 See C. Reid, Geol. Mag. for 1880, p. 548, and "Geology of Cromer," Mem. Geol. Survey. 8 R. M. Deeley, Quart. Journ. Geol. Soc. vol. xlii. p. 437 (1886). See also Sum. Prog. Geol. Survey for 1905. 9 M. Reade, " Drift Deposits of the north-west of England," Quart. Journ. Geol. Soc. vol. xxxix. p. 83, and "Eskdale Drift," Geol. Mag. for 1893, p. 9. 10 A. Strahan, "Geology of the Country round Chester," Mem. Geol. Survey and Quart. Journ. Geol. Soc. vol. xlii. p. 486 (1886). 11 D. Mackintosh, Quart. Journ. Geol. Soc. vol. xxxvii. p. 366. 12 M. Reade, " Drift Deposits of Colwyn Bay," Quart. Journ. Geol. Soc. vol. xli. p. 102. 13 See "Geology of Aramanford," Mem. Geol. Survey, p. 188 (1907). 14 See " Geology of Swansea," Mem. Geol. Survey, p. 118 (1907). 15 "The Glacial Deposits of N. Pembrokeshire," T. J. Jehu, Trans. Roy. Soc. Edin. vol xli. p. 53 (1904). 16 Summary Prog. Geol. Survey for 19Q9, p. 25. 17 M. Reade in Proc. Liverpool Geol. Soc. for 1878-9. 18 See Jukes-Browne, Quart. Journ. Geol. Soc. vol. Ixii. p. 132. 19 T. T. Pocock, " Geology of the Country around Oxford," Mem. Geol. Survey, p. 96 (1908). 652 STRATIGRAPHICAL GEOLOGY 20 H. J. 0. White, "Geology of the Country around Hungerford," Mem. Geol. Survey, p. 85 (1907). 21 "The Geology of London," by W. Whitaker, Mem. Geol. Survey (1889). 22 "The Geology of Cambridge," Mem. Geol. Survey, p. 82 (1881). 23 C. Reid, Rep. Brit. Assoc. for 1896, p. 400. 24 C. Reid, Geol. Mag. for 1897, p. 229. 25 See J. M. Mellow, Quart. Journ. Geol. Soc. vol. xxxii. p. 240 and vol. xxxiii. p. 589. 26 H. Hicks on "Bone Caves in North Wales," Quart. Journ. Geol. Soc. vol. xlii. p. 3 and xliv. p. 561. 27 See W. A. E. Ussher in Geol. Mag. for 1879, p. 166, and A. R. Hunt, Trans. Dev. Assoc. vol. xx. p. 225 (1888). Also "Geology of Torquay," Mem. Geol. Survey (1903). 28 W. C. Brogger in Geol. Mag. for 1902, p. 317. 29 F. W. Harmer, Quart. Journ. Geol. Soc. vol. Ivii. p. 405 (1901). 30 C. Reid, "Climate of Europe during the Glacial Period," Natural Science, vol. i. p. 427 (1892). INDEX Generic names are printed in Italics Abberley Hills, 314, 330 Abbey Hill Shales, 267 Abbotsbury, 442 Aberdaron, 119 Aberdeen, 487, 620 Aberdo Limestone, 255 Acanthaspis, 238 Acanthoceras, 466 Acanthodadia, 327, 343 Acanthodes, 226, 317, 327 Acanthopholis, 465 Acer, 464 Aceratherium, 577 Acervularia, 156, 159, 160, 195 Achanarras, 236 Acheulian, 638 Acidaspis, 110, 113, 156, 162, 195 Acrocidaris, 436 Acrodus, 350, 363, 386 Acrolepis, 327 Acrosalenia, 385, 413, 435 Acrothde, 93 Acrotreta, 73 Actceonella, 466 Actceonina, 385 Actinocamax, 466, 495 Actinoceras, 158, 162 Actinocrinus, 247, 249 Actinodesma, 195 Actinodon, 327 Actinostroma, 195 Acton, 642 Adapis, 557 Adelastrea, 410 Adiantites, 245 ^%er, 385 ^goceras, 386, 389 Albert, 99, 113 jEtobatis, 528 Aetosaurus, 371 Agathiceras, 319, 344 Agenais, limestones of, 578 ^grnosfrus, 75, 77, 114 Agoniatites, 176 Aisne, the, 515 Aizy, Sables d', 552 ^IZcma, 386 Albertia, 350 Albert!, von, 348 Albian stage, 463, 491, 514, 523 Aldborough, 593, 597, 599 Aldeby, 599 Alethopteris, 288, 290 Allorisma, 327 Alluvial Levels, 645 Almaden, 144 Alpine Region, 62, 188, 320, 373, 407, 431, 454, 517. 554, 555, 565 Alsace, 376, 405, 406 Alston, 263 Alton, 499 Alum Bay, 540, 541, 543, 544 Alveolina, 528, 570 Alveolites, 156, 195 Amaltheus, 386, 389 Amberleya, 385 Amblotherium, 434 Amblypterus, 317, 327 Ambonychia, 158, 195 Amphibia, 292, 317, 340, 352 Amphicyon, 577 Amphidromus, 559 Amphilestes, 415 Amphion, 104, 110 Amphiope, 574, 579 Amphipora, 195 Amphitherium, 384, 414, 415 Amphitragalus, 529 Amphitylw, 415 Amphoracrinus, 247 Amplexus, 247 Ampthill Clay, 443 Ampyx, 110 Amstelian, 587, 603 ^4na6acia, 410 Ananchytes, 466 ^4?i,arcesies, 196 Anatina, 434 Ancenis, 214, 280 Anchitherium, 557 Ancilla, 528, 533, 576 Ancistroceras, 147 Ancyloceras, 436 Ancylus Lake, 647 Andalusia, 313 Andraruin Limestone, 103 653 Andrews, A. R., 80 Andrews, W. R., 449 Andrias, 582 Angelina, 76, 78 Angers, basin of, 214 Anglesey, rocks of, 50, 122, 257 Angoumian, 515 Anisoceras, 466 Anjou, 141 Annularia, 286, 315 Anodon, 350, 415 Anomopteris, 350, 370 Anomozamites, 414, 471 Anoplophora, 350, 371 Anoplosaurus, 465 Anoplotherum, 557 Anopoltmis, 82 Anor Sandstone, 216 Antholites, 290 Anthracite, 321 Anthracomya, 291 Anthracosaurus, 292 ^ntftracosia, 291, 315, 321 Anthracotherium, 577 Antrim, 237, 402, 512, 545, 547, 633 Antwerp, 602 Anversian, 554, 584 Apatorius, 465 Apennines, the, 432, 555, 557, 583, 605 Apiocrinus, 410, 413, 425 Aplin quartzite, 43 Aplin phyllites, 43 Aporrhais, 466, 492, 529 Appenzell, 580 Appleby, 236, 337 Applecross, 45, 403 Applecross Group, 45 Applethwaite, 132 Aptian stage, 463, 487, 489 Aquitaine, 518, 524, 568 Aquitanian, 573, 575, 578, 579, 580, 582, 583 Arachnophyllum, 156, 159, 160, 195 Aralia, 464, 528 Aramaria, 385 Araucarites, 414 654 STRATIGRAPHICAL GEOLOGY Arbacia, 574 Arber, E. A. N., 301, 308, 309, 320, 322, 363 Arbroath Sandstones, 232 Area, 469, 576 Area Clays, 648 Arcestes, 349, 351 Archrean rocks, 5, 34 Archceocalamites, 245 Archceocidaris, 247 Archceocyathus, 72, 97 ArchcKoniscus, 438 Archceopteryx, 384, 435 Archceoscyphia, 97 Archanodon, 212, 225, 226, 238, 264 Archegosaurus, 327, 341 Archidesmus, 226 Archimedes, 319 Arctic Freshwater Bed, 627 Arctic Region, 238 Arctocyon, 528 Arctomys, 538 Ardara, granite of, 50 Ardennes, the, 101, 215, 430 Ardmillan Group, 135, 136 Ardnamurchan, 368 Ardtun Head, 547 Are schists, 68 Arenig Mountains, 117, 118, 123 Arenig Series, 109, 110, 116, 119, 128, 136 Argile de Boom, 563 Argile de Flandres, 550 Argillornis, 528 Argyllshire, 43, 403 Arietites, 386 Arionellus, 82, 104 Alley mine coal, 297 Armagnac, 578 Armorican flexures, 314, 345 Armorican Sandstone, 141 Arno, val d', 605 Arragonite shells, 19 Arran, Isle of, 269, 359, 368, 402, 548 Arthur's Seat, 267 Artinskian, 325, 343, 344 Arundel, 508, 509 Asaphellus, 76, 78 Asaphus, 110, 113 Ascoceras, 158 Ascoceras Limestone, 182 Ashburton, 207 Ashdown Sand, 475 Ashgillian, 110, 124, 132 Aspidoceras, 435, 436 Aspidorhynchus, 410, 435, 438 Astarte, 385, 436, 469, 589 Astartian, 451, 452 Asterolepis, 226 Asterophyllites, 286 Asteroplax, 238 Astian stage, 587, 605, 606 Antroccenia, 385 Asturias, 100, 214, 281, 313 Ateleaspis, 178 Atherfield Clay, 475, 476, 477 Athyris, 214, 248 Atlantic Region, 107, 239, 283, 320, 586 Atlahtosaurus, 434 Atractites, 349 Atrypa, 110, 156, 157, 159, 160, 162 Aturia, 528, 576 Aube, the, 514, 515 Auchenaspis, 158, 162 Auchmithie conglomerate, 232 Audierne, 60 Aulacoceras, 349 Aulopliyllum, 247 Aulosteges, 327 Aultbea Beds, 45 Aurillac, 580 Austria, Archaean of, 62 Carboniferous of, 313 Cretaceous of, 521 Jurassic of, 455 Miocene of, 582 Pliocene of, 602 Silurian of, 188 Trias of, 373, 375, 379 Autun basin, 340 Autunian, 325, 340, 345 Auvergne, 566, 579, 60* Avellana, 466 Aveyron, 340, 341 Avicula, 161, 162 Aviculopecten, 195, 246, 248 Avonian Series, 244, 262, 278 Axis, 588 Axmouth, 368, 499 Aylesbury, 444, 447, 449, 483 Aymestry Limestone, 165, 167 Ayrshire, rocks of, 134, 234, 268, 303, 338, 339, 622 Azeca, 528 Azoic era, 5 Babbacombe, 207, 333 Backwell Hill, 366 Bactrites, 196 Baculites, 466 Baggy Beds, 209 Bagshot Beds, 527, 549, 551 Baiera, 325, 350, 414 Bailey, B. B., 43 Bairdia, 248, 291 Bajocian, 383, 409, 431, 45S Bakevellia, 327 Bala Series, 110, 124, 128, 132, 138 Bala Limestone, 125 Balachulish Limestone, 43 Balanophyllia, 528, 591 Balclatchie Beds, 136 Balkans, the, 320, 605 Ballagan Beds, 267 Ballard Cliff, 506 Ballypalady, 547 Baltic Region, 103, 183, 223, 516, 647 Bamboo, 604 Banbury, 400 Bangor, 53, 88 Bangor Series, 53 Bannisdale Slates, 175 Bantry Bay, 275 Baphetes, 292 Bardon Hill, 58 Bargany Beds, 177 Bargate Beds, 479 Barmouth Grits, 82 Barnstaple, 276 Barr Series, 135, 136 Barrande, Dr. J., 104, 145, 184, 221 Barrandeiceras, 162 Barremian, 463, 489 Barrois, Prof. C., 60, 100, 144, 213, 214, 281 Barroisia, 466 Barrow, G., 97 Barrule Slates, 94 Earth, 94 Barton Beds, 527, 535, 543 Bartonian, 549, 554, 555 Basse Loire, 279 Basses Alpes, 408 Bath, 398, 416 Bathonian, 383, 409, 422, 428, 431, 458 Battle, 440, 450 Bauterem, 563 Bauxite, 547 Bavaria, 373, 375, 408. 554, 564 Bazas, 578 Beaches, raised, 645 Beaminster, 422 Bear Island, 238 Beauce, Calcaire de la, 566, 579 Beauchamp, Sables de, 554 Bedfordshire, 443, 483, 503 Beer Head, 499, 503 Belemnite Marl, 501, 503, 505 Belemnitella, 460, 496 Belemnites, 385, 390, 434, 435 Belemnoteuthis, 436 Belfast, 359, 365 Belgium, 101, 144, 186, 215, 278, 310, 550, 562, 602 Belhaven, 267 Belinurus, 292 Bell, R., 596 Bellerophina, 492 Bellerophon, 76, 78, 115, 158, 162, 196, 249, 327 Beloceras, 196 Belodon, 352, 371, 434 Bdoptera, 528 Belostpia, 528 Bernbridge Beds, 556, 561 Benan conglomerate, 135 Bencliff Grits, 442 Bencraft Shales, 140 Beneckeia, 354, 371 Berg, Sables de, 563 Bergeron, J., 61, 281 Bernberg, 564 Bernician Series, 244, 263 Beroun, 104, 222 Berriedale Beds, 236 Berwickshire, 235 Berwyn Mountains, 126 INDEX 655 Betton Shales, 128 Brent Tor, 276 Beyrichia, 93, 195, 291 Brest, 99 Bideford, 309 Brickhill. 483 Binstead, 561 Bridgend, 397 Birkenia, 178 Bridgenorth, 356 Birkhill Shales, 176, 178 Bridlington, 625 Black Forest, the, 370, 431 Bridport, 397 Black Mountains, 228, 229, Brie, Calcaire de, 556, 565 631 Brill, 447, 449 Blackband Group, 297 Brioverian System, 60 Blackbrook Beds, 58 Brissus, 589 Blackdown Hills, 500 Bristol district, 252, 304, 305, Blackhcath Beds, 537, 539 307 Blake, J. F., 52, 57, 442, 451 Brittany, Archaean of, 59 Blea Wyke, 401, 402 Cambrian of, 99 Boghead coal, 321 Carboniferous of, 279 Bognor, 540 Cretaceous in, 523 Bohemia, Archaean of, 63 Ordovician of, 141 Cambrian of, 104 Silurian of, 186, 213 Carboniferous of, 312 Brive, 61, 340 Cretaceous of, 518 Brixham, 206, 346 Devonian of, 221 Brocastle, 397 Orclovician of, 145 Brockenhurst, 560 Silurian of, 184 Brockwell coal, 300 Bolderian, 584 Brogger, 648 Bolospongia, 511 Bromsgrove, 363 Bolton, H., 307 Brongniart, 382 Bonney, Professor T. G., on Bronteus, 195 Archaean Rocks, 35, 53, Brontosaurus, 434 58, 60, 62 Brook Point, 476 ,, on Permian, 332, 345 Broquies, 340 ,, on Trias, 361 Brora, 429, 450 Bononian, 452 Browgill Beds, 175 Boom, Argile de, 563 Brown, H. T., 332 Bordeaux, 578 Brownstone Series, 227, 228 Borkholm Beds, 148 Brunswick, 407 Bornia, 238, 245 Brussellian, 527, 549, 551 Borrowdale Series, 132 Brussels, 145 Boscombe Sands, 542, 543 Bruton, 416, 422 Bothriolepis, 226 Bryograptus, 76, 78, 94 Boulder-clay, 614, 620 Bryozoan reefs, 191, 343 Boulonnais, 311 Buccinopsis, 590 Bourbon, 340 Buch, von, 382 Bourgelicrinus, 466 Buchenstein Beds, 373, 374 Bournemouth Beds, 535 Buchiceras, 466 Bovey Beds, 577 Buckinghamshire, 443, 447, Bovidae, 588 500 Bowland Shales, 259 Buckman, S. S., 387, 399, 406, Bowmore Grits, 46 416 Boxstones, 593 Budleigh Salterton, 361 Brabant, 145 Buildwas Beds, 165 Brabourne, 362, 450 Builth, 171 Brachycyathus, 4G6 Bulla, 410 Brachymetopus, 248 Bunter Series, 340, 355, 370 Brachypliyllum, 414 Bupreston, 438 Brachytrema, 410 Burdiehouse Limestone, 268 Bracklesham Beds, 527, 535, Burdigalian, 573, 575, 578, 542 579, 580, 582 Bradford Clay, 425 Branchiosaurus, 327 Burford, 362, 393 Burren Limestones, 274 Brancoceras, 196, 249 Burton Bradstock, 416, 422 Brand Beds, 58 Busaco, 100 Brandenburg, 564 Butley Crag, 592 Branscombe, 368 Buxiere, 340 Brasil, 406 Brathay Flags, 175 Caban Beds, 168 Bray Head Series, 95 Cadnant Shales, 126 Brecknockshire, 227, 631 Cadomilla, 385 Brecon Vans, 228, 631 Cae Gwyn Cave, 644 Brendon Hills, 207 Caen, 100 Caerfai Beds, 72, 79 Cairconnon Beds, 232 Caistor, 486, 501 Caithness, 230, 236, 487 Caithness Flags, 236 Calabria, 605 Calaminker, 299 Calamites, 226, 286, 288, 325 Calamodendron, 286, 315 Calamophyllia, 350 Calamostaehys, 286 Calathium. 97 Calcaire Grossier, 552 Calcareous shells, 19 Calcareous Grit, 442, 443 Calceola, 105, 217 Shales, 217 Caledonian area, 231, 235 Caledonian System, 48 Callaway, Ch., 51, 57, 89, 92 Callicrinus, 156 Callipteridium, 315 Callipteris, 325 Callonema, 196 Callovian, 432, 451, 452 Calp, the, 272 Calymene, 110, 156, 162 Cam, valley of the, 641 Camarophoria, 195, 248, 319, 327 Camarospongia, 466 Camarotcechia, 156, 162 Cambrian System, 8, 12, 70 Cambridgeshire, 443, 500, 506 Cambridge Greensand, 504 Campanian, 491, 515, 516, 517, 523, 524 Campbell, R., 97 Camphor-tree, 574 Campil Beds, 374 Caninia, 247 Canis, 588, 638 Cannel coal, 321 Cannock Chase, 356 Cannon-shot gravel, 640 Canonbie area, 301 Cantal, the, 566, 581 Cantrill, T. C., 117, 328 Cap la Hkve, 514 Capreoliis, 588 Capulus, 249 Caradoc Hills, 89, 128 Caradocian, 110, 124, 129 Caratomus, 466 Carbonicola, 291 Carboniferous System, 6, 12, 244, 285 Carboniferous Slate, 275 Carcharodon, 579, 590 ' Cardiaster, 466 Cardinia, 385 Cardiocarpus, 290 Cardioceras, 434, 436 Cardiola, 162, 195 Shales, 182 Cardiomorpha, 249 Cardiopteris, 245, 290 Cardita, 350, 385, 531, 533, 576, 589 Cardium, 385, 590, 591 656 STRATIGRAPHICAL GEOLOGY Carinthia, 313, 373 Carinthian stage, 374, 375 Carlisle, 359, 364, 365, 402 Carmarthen, 84, 117, 121, 123 Carmyllie Beds, 232 Cam Chuinnach, 42 Carnarvonshire, 52, 82, 87, 119 Carnic Alps, 320, 373 Carpathian Mountains, 432, 455, 582 Carrickfergus, 365 Carstone, 4S1, 484, 487, 499 Caryocaris, 112 Caryophyllia, 466 Cassia, 528 Cassiandla, 350 Cassidaria, 466, 576, 590 Cassis ( = Cassideal 528, 531 Casterlian, 587, 602, 603 Castor, 601 Catagma, 466 Catalonia, 187 Catopygus, 466 Cave deposits, 643 Cefn Slates, 82 Cement-stone Group, 267 Cenomanian, 463, 491, 514, 523 Centronella, 195 Cephalaspis, 158, 196, 224, 226 Cephalites, 466 Cephalopoda Bed, 399, 417 Cephalopteris, 238 Ceratites, 349, 351, 371 Ceratodus, 351, 371, 388 Ceratopyge Limestone, 102 Ceritella, 385 Cerithium, 350, 385, 435, 437, 576 Ceromya, 385 Cervidse, 588, 601, 638 Ceteosaurus, 384, 415, 434 Chcetetes, 245 Chalk, Lower, 463, 491, 501 Middle, 463, 491, 505 ,, Upper, 463, 491, 507 Chalk Marl, 503, 504 Rock, 510 Chalky Boulder-clay, 623, 628 Chama, 532 Chara, 471, 559 Chard, 503 Charmouthian, 383, 405, 406, 408, 457 Charmouth, 394, 397 Charnwood Forest, 57, 58 Chasmops Limestone, 147 Chateaulin basin, 279 Chatelier Sandstone, 141 Chatham, 440 Ckeiracanthus, 226 Cheirodus, 292 Cheirostrobus, 245 Cheirurus, 92, 156, 195 Chelidosaurus, 317 Chellean epoch, 638 Chelone, 471 Chemnitzia, 351 Chemnitz area, 342 Chenendopora, 466, 493 Chert Beds, 51 Cherwell Valley, 419 Cheshire, 294, 355, 363 Cheviot area, 231, 235, 621 Chideock, 416 Chillesforcl Beds, 592, 599 Chilocychis, 350 Chipping Norton, 419 Chiton, 156 Chloritic Chalk, 512 Marl, 501, 503 Sands, 512 Cheer opotamus, 557 Chonetes, 156, 162, 248 Chonograptus, 94 Chonophyllum, 199 Christiania, 147, 181 Chrysodomus, 525 Chudleigh, 208 Cidaris, 350, 351, 385, 435 Cimolestes, 466 Cimoliosaurus, 434 Cimolomys, 466 Cinnamon, 464, 574, 578 Ciply, 516 Cirrus, 385, 410 Cladiscites, 349 Cladophlebis, 350, 414, 471 Cladophyllia, 350 Cladyodon, 352, 363 Clare, 273 Clathrodictyon, 150 Clausthal, 221 Clava, 615 Clavella, 532 Clavinian Series, 382, 458 Claxby Ironstone, 484, 486 Clay-with-flints, 640 Cleidophorus, 158, 195 Cleistopora zone, 252 Clent Hills, 330 Cleveland Ironstone, 401 Cleveland Hills, 633 Clifton, 252 Clim'Mograptus, 110, 113 Climatius, 226 Clisiophyllum, 247 Clitheroe, 258, 259, 336 Clogan Beds, 82 Clogher Head, 180 Clonocrinus, 156 Clough, C. T., 42, 45, 97 Clunian, 154 Clwyd, Vale of, 644 Clyde Beds, 621 Clydoniceras, 413 Clymenia, 196 Clypeaster, 574 Clypeus, 410 Clypeus Grit, 415 Coal, formation of, 320 Coalbrookdale, 298 Coalfields, 294 Coal-Measures, 285, 294 Cobbold, G. S., 89 Coblentzian, 216, 220 Coccosteus, 196, 212, 224, 226 Coccoteuthis, 436 Cadaster, 247 Coddon Hill Beds, 276 Ccelacanthus, 250, 292, 327 Cceloceras, 386 Ccelolepis, 158 Ccelonautilus, 249 Coelopleurus, 556 Ccelopoma, 528 Codoptyckium, 466 Ccelosmilia, 466 Ccelostylina, 350 Cc&nograptus, 110 Ccenotherium, 557 Coldwell Beds, 175 Cole, G. A. C., 49, 50 Coleoloides, 73 Collins, F. G., 309 Collyrites, 410 Colly weston, 419 Colwell Bays, 560 Columbella, 591 Columbellina, 466 Combemartin, 207 Cominella, 466 Comley Sandstone, 89 Commentry, 317 Compton Bay, 506 Congeria Beds, 587, 606 Congleton Edge, 258, 295 Coniferse, 325, 350, 385, 574 Coniopteris, 350, 414 Coniosaurus, 465 Coniston Grits, 175 Conocardium, 158, 159, 246, 249, 319 Conoceras, 112 Conocoryphe, 75, 77 Conorbis, 528, 532 Contorted Drift, 627 Conularia, 76, 78 Conus, 528, 533, 576 Conwa) , 126 Coomhola Beds, 211 Coral Rag, 442 Coral Reefs, 189, 190 Corallian Beds, 383, 432, 441, 442, 443, 451, 452, 469 Coralline Crag, 587, 592, 593 Corax, 464 Corbicdla, 410 Corbicula, 528, 639 Corbieres, 281 Corbierian, 515 Corbis, 410, 469 Corbula, 529, 532 Cordaites, 290, 315, 325 Cordioxylon, 325 Cork, 210, 211, 274 Cornbrash, 415, 422, 424, 425, 426, 428 Cornstone Series, 227 Cornwall, 144, 200, 202, 208, 275, 308, 334, 595 Corona Beds, 132 Coroniceras, 389 Corwen Grits, 170 Corynella, 410 Coryphodon, 528 Cosmoceras, 410, 434 Cotentin, 516, 568, 603 Cottaldia, 466, 493 INDEX 657 Cotteswold Hills, 415, 417, 423 Cotteswold Sands, 399 Cottian Alps, 63, 408 Cracoe Limestone, 259 Cracow, 431 Crampton, 42 Craspedites, 436 Crassatellites, 466, 532 Craticularia, 410 Craven fault, 258 Crediton, 333 Credneria, 464 Creech barrow Hill, 562 Crendon, 483 Crenipecten, 248 Crepidula, 466 Creswell Crags, 643 Cretaceous System, 6, 12, 462 Crinoids, growth of, 189 Crioceras, 466, 467 Croagmartin Beds, 180 Croatia, 606 Croll, J., 648, 649 Cromer Beds, 587, 592, 600 Cromer Till, 627 Cross Fell, 259, 261 Crossfield, Miss, 84 Crotalocrinus, 156, 162 Croyde Bay, 209 Cruden, 487 Cruziana, 75, 77 Cryptcenia, 386 Cryptoccenia, 410 Cryptograptus, 110, 111 Cryptosaurus, 434 Ctenis, 414 Ctenocrinus, 195 Ctenodonta, 76, 114, 158, 160, 195 Ctenoptychms, 292 Ctenopyge, 92, 93 Cucullcea, 195, 198, 385, 464, 529 Cucuron, 604 Cuise, Sables de, 552 Culford, 362, 393 Cullion, 272 Culm, the, 280 Culm-measures, 308 Cultra, 327 Culver Cliffs, 501, 506 Culverwell, E. P., 648 Cumberland, 93, 129, 130, 300, 301, 337, 359, 364, 402 Cummingstone, 339 Cupar, 236 Cupressocrinus, 195 Cupressus, 574 Curtonotus, 195, 198 Ousel Beds, 341 Cyathaxonia, 247 Cyathocrinus, 156, 195, 247 Cyathophora, 410 Cyathophyllum, 159, 160, 195, 247 Cycadites, 471 Cycads, 325, 350, 385 Cyclocyathiis, 466 Cyclolobus, 326 Cyclonema, 466, 469 Cydophyllum, 247 Diabeg Group, 45 Cyclopteris, 226 Dibunophyllum, 247 Cylindrites, 386 Dicellograptus, 110, 113 Cyphaspis, 195, 248 Dichodon, 529, 537 Cyphosoma, 436 Dichograptus, 110 Cyprcea, 528 Dicranograptus, 110, 113 Cypress, 374 Dicroceras, 577 Cypricardiella, 249 Dicroloma, 386, 434 Cypridina, 248 Dictyograptus, 76 Cypridinella, 248 Dictyonema, 76, 77 Cyprimeria, 466 Dictyoneura, 315 Cyprina, 410, 529 Dicynodon, 353 Cyrena, 415, 438, 530 Didelphys, 528 Cyrtia, 156, 162 Cyrtina, 156 Didymograptus, 110, 111, lia Dielasma, 195, 248, 319, 327 Cyrtoceras, 76, 115, 157, 158, Diestian, 587, 602 160, 249 Digne, 408, 432 Cyrtodonta, 195 Diluvium, 635 Cyrtograptus, 155, 156, 158, 182 Dimerocrinus, 156 Cyrtotheca, 75 Dimetian granite, 53 Cystiphyllum, 156, 195, 197 Dimodosaurus, 371 Dimorphoceras, 249 Dachstein, 373, 375 Dimorphodon, 386 Dacrytherium, 557 Dinant, 215, 217, 241, 279 Dactylioceras, 386, 389, 390 Dinantian Series, 244, 278 Dodocrinus,l350, 351 Dinarian stage, 374 Dadoxylon, 290 Dingle Promontory, 178, 179, Dakosaurus, 434 210 Dalarne, 146, 148 Dinotherium, 577 Dalmanites, 195 Dalmatia, 373, 409, 567 Dalradian Series, 43 Dionites, 471 Diphya Limestone, 454 Diplacanthus, 226 Dalsland, 67 Diplodonta, 528 Danian stage, 463, 516 Diplograptus, 110, 113 Daonella, 350 Diplopora, 350, 379 Dapedius, 386 Diplopterus, 226 Dartington, 207 Dipteronotus, 351, 363 Dartmouth Slates, 202 Dipterus, 226 Darwin, Ch., 11, 15, 16, 20, 21 Discites, 249 Dasornis, 528 Discoidea, 466 Dasyceps, 327 Dithyrocaris, 248 Davidsonia, 195 Divesian, 452 Davies, D. C., 126 Dogger, the, 387, 420 Davis, J. W., 262 Dolgelly Beds, 86 Dawkins, W. B., 651 Dolichosaurus, 465 Dawlish, 333 Dolioli-na, 319 Dayia, 156, 161, 162 Dollfus, G. F., 573, 579, 603 Dean, Forest of, 253, 308 Dolomite Mountains, 373, 375 Dechenella, 195 Dolomitic conglomerate, 366 Deeley, R. M., 629 Dolwen, 80 Deganwy mudstones, 129 Deiphon, 156 Donegal, rocks of, 49, 237 Donetz basin, 281, 288, 409 Delamere Forest, 357 Delepine, G., 278 D'Orbigny, A., 382 Dordolot, Prof, de, 278 Delphinula, 410 Dorset, Cretaceous of, 470, Denbighshire, 170, 174 481, 499, 503, 506, 510 Dendrerpeton, 292, 317 Eocene of, 540 Dendrocrinus , 76 Jurassic of, 393, 396, 416, Dendrophyllia, 528 422, 442, 444, 445 Denmark, 516 Doryderma, 466 Dentalium, 371 Dosinia, 559 Derbyshire, 257, 359 Deroceras, 386- Douglas, J. A., 273 Doulough Slates, 140 Derry, rocks of, 272, 545 Doulting, 422 Desert/martin, 272 Dover, 440, 475, 477, 505, 508 Destruction of deposits, 20 Downtonian, 154, 167, 171, Devillian, 101 175 Devonian System, 6, 12, 193 Dremotherium, 557 Devonshire, 200, 275, 308, 332, Drepanaspis, 196 361, 367, 499, 506 Dryandra, 528 2u 658 STRATIGRAPHICAL GEOLOGY Dryopithecus, 577 Dublin, 274 Ducy, Calcaire de, 554 Eotherium, 529 Eozoic, 5 Eparchaean rocks, 35 Dudley, 167, 299 Epiaster, 466 Dufton Shales, 132 Epidote Grits, 46 Dumfries, 338 Eppelsheim, 600 Dunottar conglomerate, 231, Equisetites, 350, 363, 369 232 Equus, 601, 638 Danrobin Castle, 369, 403 Erbray, 186, 187 Dunstable, 504 Eretmosaurus, 386 Dura Den, 236 Eriboll area, 96 Durham, 300, 334 Erinnys, 75, 82, 104 Durlstone Head, 445, 448 Erpetosiichus, 352, 369 Durness Limestone, 96 Eryma, 385 Dusseldorf, 279 Dyas, the, 324 Eryon, 385 Erzgebirge, 63, 242 Dykes, Archaean, 38 Eskers, 634 ,, Eocene, 545 Estheria, 224, 226, 315, 350, 363, 371 Earthquake tissures, 346 Easdale Slates, 43 Estuarine Beds, 419, 420, 426, 428 East Anglia, 626 Etampes, Calcaire d', 556, Echinoconus, 466, 496 566, 579 JSchinocorys, 466, 496 Etheridge, R., 206 Echinocyamus, 556, 591 Etna, 609 Echinocystis, 156, 162 Etruria Marls, 297, 298 Echinodiscus, 574 Etyus, 466 Echinolampas, 556, 574, 579 Eucalyptocrinus, 156 Echinospatagus, 466, 496 Eucalyptus, 464, 528 Echinosphcera, 114, 115 Euchasma, 97 Echinus, 589 Eucorystes, 466 Edaphodon, 464 Eugnathus, 386 Eden, Vale of, 336, 369, 429 Eukeraspis, 158 Edge Coal Group, 267, 269 Eulima, 466 Edinburgh, 267, 268, 302 Euloma, 92 Edmondia, 195, 249 Eunema, 158, 162 Edzell Shales, 232 Euomphalus, 112, 196, 246, Egeln, 564 249, 349 Eggensburg, 582 Eupterornis, 528 Eggesford Grits, 308 Eurynotus, 250 Eifel district, 219, 370 Eurypterus, 156, 162, 226 Eifelian stage, 217 Exelissa, 386 Eigg, island of, 428, 547 Exeter, 309, 333 Eilde Flags, 42, 43 Exmoor, 207 Elaphus, 588 Exogyra, 436, 493 Elbingerode, 221 Elephas, 588, 601, 638 Elgin, 339, 369, 429 Fairlight Clay, 475 Fallowfield, 336 Elginia, 327, 329 Falmouth Beds, 144 Ellergill Beds, 131 Faluns, 578, 579 Biles, Miss G. L., 94, 119, Famennian stage, 219 126, 128, 131, 167, 170, Faou, greywacke of, 213 . 172, 174 Farewell Rock, 305 Ellesmere, 357 Faringdon, 442, 481 Ellipsocephalus, 104 Farrington Beds, 304, 308 Elotherium, 557 Fasciculites, 589 Els worth Rock, 443 Fasciolaria, 528 Ely, 444 Faunas, succession of, 21 Emscherian, 515 Favosponqia, 162 Enaliornis, 465 Faxe, 516 Enallaster, 466 Fearnsides, W. G., 87, 117, Enchodus, 464 118, 123 Encrinurus, 110, 156, 162 Felis, 588, 638 Encrinus, 350, 371 Fell Sandstones, 264 Endothyra, 245 Fenestella, 25^, 273, 327, 343 Entelodon, 5^7 Fenno-Scandia, 65, 66 Entomis, 195, 208, 209 Ferriters Cove Beds, 180 Enville Marls, 330, 331 Ffestiniog Beds, 86 EoUattina, 315 Ffynnon-Beuno Cave, 644 Eocene Series, 11, 526, 527 Fichtelgebirge, 280 Ficus, 528 Fifeshire, 236, 268, 302 Finisterre, 213 Finland, 64 Firestone, 498 Fisher, O., 542, 570 Fissurella, 410 Flabellaria, 528, 574 Flagstone Group, 304 Flandres, Argile de, 550 Flett, J. S., 41 Flintshire, 257, 297 Klotzeere Sandstones, 310 Flysch, 565 Folkestone, 475, 498, 501 Folkestone Beds, 475, 479 Fontainebleu, tables de, 566 Foreland Sandstones, 264 Forest of Dean, 253, 308 Forest Bed, 600, 601, 609 Forest Marble, 415, 422, 424, 425, 426 Forfar, 231 Fossils, illustrations of, 31 preservation of, 16, 17 Fox-Strangways, 443 France, Archaean of, 59 Cambrian of, 97 Carboniferous of, 278, 310, 315 Cretaceous of, 487, 513 Devonian of, 212 Ordovician of, 141 Permian of, 340 Silurian of, 186 Tertiaries of, 552, 566, 603 Trias of, 376 Franconia, 407, 431, 433 Frasnian stage, 217, 219 Freestones, the, 415 Fremington Beds, 276 Fret, greywacke of, 213 Fritsch, Dr., 317 Frodingham, 400 Fuller's Earth, 415, 422, 424, 425 Fundamental Gneiss, 49 Furada Grits, 215 Fusidina, 245, 312, 319, 320, 343 Fusus, 466, 576 Gahard Sandstone, 213 Gairloch, 38 Gaize, 514 Gala Beds, 176, 178 Galerites, 466, 496 Galloway, 521 Galway, 49, 139, 273, 274 Gamlan Shales, 82 Gampsonyx, 327, 341 Gannister Beds, 295 Ganodus, 410 Gardiner, C. T., 138, 139, 180 Gardner, J. S., 541, 543, 547, 548, 574 Garumnian, 517 Garwood, E. T., 262, 263, 204, 345 Gastornis, 528 INDEX 659 Gastrioceras, 249, 319 Gaudernsdorf, 582 Gault, 498 Gavialis, 528 Gedinnian stage, 216 Geikia, 327, 339 Geikie, Sir A., on Archtean rocks, 48, 49, 51, 53, 55 ,, on Eocene volcanoes, 548, 554 ,, on Old Eed Sandstone, 230 ,, on Silurian lavas, 180 Gelinden, 550 Gembloux, 145 Geneva, lake of, 637 Geographical restorations, 33 Geological Magazine, 30 ,, Record, 30 Society, 28, 30 ,, Survey, 30 Geologists' Association, 30 Gephyroceras, 196 Germany, Cambrian of, 104 Carboniferous of, 280, 317 Cretaceous of, 490, 518 Devonian of, 219 Jurassic of, 406, 431, 453 Neogene of, 584, 606 Palaeogene of, 563 Permian of, 341 Pleistocene of, 635 Silurian of, 185 Trias of, 369, 378 Gervillia, 350, 371, 464, 469 Giant's Causeway, 547 Gibson, W., 297, 298 Gigantosaurus, 434 Ginkgo, 325, 415 Girvan district, 134, 176, 177 Givetian stage, 217 Glacial deposits, 611, 613 Glaciation, 615 Glamorgan, 632 Glandina, 559 Glasgow district, 267 Glassia, 156 Glastonbury, 398 Gleichenia, 528 Glenarm, 547 Glencoe quartzite, 43 Glengarriff Beds, 210 Glenkiln Shales, 135, 139 Glevanian Series, 382, 457 Gloppa, 631 Glossograptus, 110, 113 Glossopteris, 325 Gloucestershire, 398, 399, 424 Glyn Ceiriog, 126 Glyphcea, 385 Glyphioceras, 246, 249 Glyptarca, 76 Glypticus, 436 Glyptolepis, 212, 226 Glyptostrobus, 574 Gneissic rocks, 36, 38 Gomphoceras, 158, 162, 196 Goniognathus, 528 Goniomya, 350, 435 Goniopholis, 434, 438 Goniophyllum, 156 Go7iiopora, 528 Goniopygus, 460 Goodchild, J. G., 235, 267, 337 Gordonia, 327, 339 Gosselet, J., 219, 406 Gottland, island of, 181 Gourbesville, 603 Gouriu, phyllades de, 60 Gower, 228, 230, 253 Grammatodon, 385 Grammoceras, 3S6 Grammysia, 158 Grampian Series, 43 Grampound Grits, 203 Granatocrinns, 247, 248 Grantham, 400 Granton Sandstone, 267 Gratdoupia, 576 Gravels of rivers, 641 Great Oolite, 415, 425, 426 Greece, 606 Green, J. F. N., 55, 84 Greenland, 239, 459 Greenly, E., 51 Greensand, see Lower and Upper Gregory, J. W., 48, 49, 63 Gres armoricain, 100, 141 felspathique, 99, 141 de May, 142 Gresslya, 385, 413 Griffith, Sir R., 274 Griffithides, 248 Grimston, 500 Gristhorpe Bay, 422, 428, 441 Groix, schists of, 60 Groom, T. T., 89, 92, 126 Gross Venediger Alp, 62 Grossouvre, A. de, 515, 518, 524 Ground-Moraine, 614 Gruinard Bay, 368, 369 Gryphcea, 385, 390, 434 Gryphite Grits, 415 Guettardia, 466 Guttenstein Limestone, 373, 374 Gwastaden Beds, 168 Gyffin Shales, 170 Gyracanthus, 292 Gyrodus, 432 Gyrolepis, 351 Gyroporella, 379 Gyroptychius, 226 Hafothy Shales, 82 Hailes Sandstone, 267 Halcyornis, 528 Haldon Hills, 329, 333, 500 Halirhoa, 466 Halitherium, 557 Hallstadt Limestone, 375 Halobia, 350, 351 Halonia, 286 Haly sites, 115, 156 Ham Hill stone, 399 Hamites, 492 Hampshire, 498, 508, 534, 539, 541, 560 Hamstead, 561, 562 Beds, 556, 562 Hamulina, 466 Hangman Grits, 204 Hanover, 407, 431, 453, 490 Hapalocrinus, 156 Haploceras, 410 Haplocrinus, 195 Hardman, E. T., 304 Barker, A., 544 Harkness, Professor, 338 Harlech Beds, 79 Harmer, F. W., 593, 596, 597, 599, 600, 602, 608, 649 Harnage Shales, 129 Harpes, 110, 156, 195, 197 Harpoceras, 386, 389 Hartfell Shales, 135, 139 Hartshill quartzite, 91 Hartwell Shales, 444 Hartz Mts., 185, 220, 241, 280, 342 Harwich, 102 Haslemere, 476, 477 Hastings Beds, 475 Hatherleigh, 333 Haugia, 286 Hauterivian, 463, 489 Haxey, boring at, 335 Headon Beds, 556, 560 Hearthstone, 498 Heathfleld, 577, 581 Hebridean gneiss, 36 Hedera, 464 Heer, O., 574 Heersian, 549, 550 Heliolites, 156, 160, 195, 197 Heliophyllum, 195 Helix, 559 Helladotherium, 558 Helvetian stage, 573, 575, 578, 579, 580, 582 Hemiaspis, 156, 162 Hemiaster, 466 Hemicidaris, 350, 410, 412 Hemicosmites, 115 Hemipedina, 385 Hengistbury Beds, 543 Henis, 563 Hercynian Flexures, 314, 345 Hercynian Land, 520 Hercynian Series, 185, 220 Hermsdorf, 407 Herpetocrinus, 156 Hesperornis, 465 Hesse, 343 Hessle, 507 Hessle Clay, 625, 626 Heterangiutn, 285 Heterustrea, 385 Heteroceras, 466 Heterostinia, 511 Hettangian, 383, 405, 408 Hexacrinus, 195 Hickling, G., 231, 338 Hicks, H., 53, 79, 84, 109, 117, 121, 172, 644 Highcliff Sands, 543 660 STRATIGRAPHICAL GEOLOGY Highlands of Scotland, 42, 44 Hildoceras, 386 Hill, W., 507 Hilton Shales, 337 Hind, W., 255, 259, 309 Hinde, G. J., 479 Hipparion, 588 Hippopodium, 385, 390 Hippopotamus, 588, 601, 638 Hippurite Limestone, 517 Hippurites, 466 Histioderma, 95 Hitchin, 643 Hoar Edge Grits, 129 Hoernesia, 327, 350 Hogbom, A. G., 68 Holaspis, 226 Holaster, 466, 495 Holcospongia, 410 Holcostephanus, 386, 436, 464 Holectypus, 410, 464 Holland, 602 Hollybush Sandstone, 89 Holoeystis, 466 Holopea, 97, 114, 115, 158 Holopella, 158, 160, 162 Holoptychius, 225, 226, 250 Holt, 356 Holywell Shales, 255, 257 Homalonotus, 110, 156, 161, 162, 195 Homomya, 350, 385 Homosteus, 19(5, 226 Homotaxis, 23 Hope Shales, 128 Hoplites, 466, 467, 470, 492, 493 Hoploparia, 531 Hoplopteryx, 464, 466 Hordwell Beds, 560 Horiostoma, 158, 162 Home, J., 48 Hoxne, 642 Hudleston, W. H., 420, 442, 443 Hughes, T. M'K., 53, 81, 87, 172 Hull, E., 211, 367 Hume, W. F., 512, 519 Hundsruch Slates, 220 Hunstanton, 500, 505 Huntingdon, 441 Hurlet coal, 268 Huy, 145 Hycena, 588, 602 Hycenarctos, 577 Hycenictis, 588 Hycenodon, 557 Hyalonomus, 317 Hybodypus, 410 Hybodus, 351, 386, 435 Hydrobia, 639 Hylceosaurus, 464 Hymenocaris, 75, 77 Hyolithes, 73, 75, 76 Hyopotamus, 528, 557 Hyotherium, 577 Hyperodapedon, 353, 363, 369 Hypothyris, 248 Hypsilophodon, 464 Hyracotherium, 528 Hythe Beds, 475, 477, 478, 479 Iberg Limestone, 221 Ice, extension of, 634 Ichthyocrinus, 156, 162 Ichthyornis, 405 Ichthyosaurus, 386, 392, 434 Ictitherium, 588 Iguanodon, 434, 464 Ilfracombe Beds, 207 lllcenus, 110, 114, 156, 160, 162 Imperfection of records, 16 Inferior Oolite, 383, 387, 409 Infulaster, 466 Ingleborough, 259, 260, 261 Inoceramus Bed, 505 Inoceramus, 385, 492, 495 Instow, 309 Inverness, rocks of, 39, 41 Ireland, Archaean of, 48 Cambrian of, 94 Carboniferous of, 269, 303 Cretaceous of, 512 Lias of, 402 Old Red Sandstone, 209, 237, 239 Ordovician of, 137 Pleistocene of, 617, 633 Silurian of, 178 Trias of, 365 Iriartea, 528 Irish Channel, 569 Irving, Rev. A., 367, 569,570 Isastrea, 350, 385 Ischadites, 162 Ischyodus, 410 Islay, 39, 46, 47 Isle of Man, 94, 365 Isle of Wight, 476, 499, 501, 502, 506, 508, 540 Isocardia, 410 Isocrinus, 350, 385 Italy, Tertiary Beds of, 567, 533, 605, 609 Jarrow coalfield, 304 Jasper, 51, 97 Jatulian System, 65, 67 Jehu, T. J., 632 John o' Groat's Sandstone, 236 Johns, Cosmo, 260, 263 Jotnian System, 65, 67 Judd, J. W., 368, 403, 428, 441, 484, 561 Jukes, J. B., 204, 210, 211, 274, 275, 565 Jura Mts., 382, 405, 406, 430, 452 Juvavian, 373, 374, 375 Kalevian System, 65 Kammin, 407 Kampecaris, 226 Karpinsky, 344 Keele Beds, 297, 298, 328 Keeping, H., 542, 543, 561 Keisley Limestone, 132 Kellaways Beds, 432, 441 Kellerwald, 220 Kelsey, 626 Kendall, P., 596 Kendall, Professor P. F., 19 Kenmare, 211 Kent, 475, 498, 505, 507, 535 Kent's Cavern, 644 Keraderpeton, 317 Kerforne, F., 186 Kerry, 178, 210, 211 Keuper Series, 348, 362, 370 Kharkov, 519 Kidston, D., 285, 290, 296, 302, 303, 307 Kilian, W., 487, 489 Killary Harbour, 139, 178 Kilrose, J. R., 617 Kiltorkan Beds, 211, 212 Kimberley, 335 Kimeridge Clay, 432, 444, 445 Kimeridgian, 383, 459 Kinaham, G., 273 Kinderscout, 295 Kingena, 466 King, W. W., 330, 331 Kinlock Beds, 45 Kirkby Moor Flags, 175 Kirklington Sandstone, 365 Kirmington, 626 Knorria, 212, 226 Koenen, von, 490 Kohlen Keuper, 371 Konigsburg, 564 KonincJdna, 350 Kossen Beds, 408 Kostromian, 325, 343 Krone, 320 Kupferschiefer, 342 Kutorgina, 73, 74 Labechia, 156 Labyrinthodon, 352 Labyrinthodonts, 338 Lacerta, 520 Lacopteris, 350, 414 Lacustrine deposits, 642 Ladoga, Lake, 148 Laekinian, 527 Lagena, 245 Lagenostoma, 290 Lake District, 93, 129, 130, 569, 617, 622 " Lake, Ph., 172 Lamarck, 14 Lamna, 464 Lamplugh, G. W., 94, 484, 486, 511 Lanark, 177, 234, 302 Lanarkia, 178 Lancashire, 258, 295, 296 Lancastrian flexures, 314 Landenian, 527, 549, 550 Langhe, 583 Languedoc, 187 Lapparent, A. de, 99, 573 Lapworth, C., 8, 71, 80, 82 86, 89, 91, 93, 128, 134, 154, 170, 176, 177, 299 Lapworth, H., 168 Lapworthura, 162 INDEX 661 Lasanius, 178 Lasiograptus, 110 Lastrea, 574 Latdorf, 564 IMUTUS, 574 Lausanne, 580 Laval basin, 280 La Vega Beds, 100, 101 Leaia, 315 Leavening, 501 Lebach Beds, 341, 342 Lebour, G. A., 263 Leda, 390 Leda Myalis Bed, 627 Ledbury Shales, 168 Leenane Grits, 140 Leicester, 296, 332, 399 Leinster coalfield, 303 Leintwardine Flags, 167 Leiston, 599 Leithakalk, 582 Leitrim, 272 Lenham Beds, 587, 592 Lepadocrinus, 156, 162 Leper ditia,, 73, 291 Lepidodendron, 226, 245, 286, 288 Lepidophloios, 245, 286 Lepidoxteus, 628 Lepidostrobus, 286 Lepidotosaurus, 327 Lepidotus, 386, 435 Leptcena, 110, 156, 159, 162 Lepta-na Limestone, 147, 148 Leptite, 65 Leptodomus, 195, 249 Leptograptus. 110, 115 Leptolepis, 386, 434, 435 Leptophragma, 466 Leptoplastus, 93 Leptopleuron, 353 Lesmahagow, 234 Letterewe Limestone, 39 Leven schists, 43 Lewisian gneiss, 36 Leyburn, 262 Lias, the, 382, 383, 386 Liburnian, 555 Lichas, 110, 114, 115, 156, 162, 195 Life-forms, succession of, 21 Ligerian, 515 Liguria, 583 Lima, 350, 390, 412, 464, 492 Limagne, the, 566 Limerick, 274 Limestone Drift, 633 Limnwa, 559 Limnerpeton, 317 Liinulus, 292 Lincolnshire Limestone, 419, 420 Lincolnshire, rocks of, 399, 401, 419, 426, 445, 486, 501, 505, 507, 623, 626 Lincombe Hill, 204 Lindstrcemia, 159 Lingula, 156, 162 Lingula Flags, 72, 84, 85, 87, Lingulella, 73, 75, 76, 77 Linnarsonia, 99, 106 Linopteris, 290 Lioceras, 410 Liostracus, 105 Liperoceras, 386 Liquidamber, 574 Listriodon, 577 Literature of Historical Geo- logy, 26 Litharea, 528, 532 Lithomarge, 547 Lithornis, 528 Lithostrotion, 246, 247 Lithothamnion, 579 Littorina, 386 Littry, 317, 341, 377 Lituites, 110, 114 Liverpool, 356 Livonia, 183, 223 Lizard, the, 55, 59 Llanberis, 81, 87 Llandeiniolen, 88 Llandilo Beds, 109, 110, 122, 128, 131, 140 Llandovery Beds, 129, 153, 163, 168, 176 Llandudno, 257 Llanfaelog, 122 Llanvirn Beds, 109, 110, 121, 128, 131, 140 Lleyn promontory, 52, 119, 126 Llyn Peris, 87 Lobites, 349 Lochmaben, 338 Loch Maree, 38, 39 Loch-na-dal Beds, 46 Lodeve, 341 Loess, 646 Loganograptus, 110 Lonan Flags, 94 London Basin, 533 London Clay, 527, 535, 539 London, gravels near, 642 Longmynd, the, 55, 128 Longmyndian Series, 56 Longville Flags, 129 Lonsdaleia, 247 J.ophiodon, 528 Lome, '234 Lorraine, 404, 406, 430 Lossiemouth, 369 Lower Greensand, 462, 498 Lowestoft, 599 Loxonema, 196, 249, 292, 327, 349 Luarca Slates, 144 Luberon, 604 Luberon Beds, 587 Lucina, 350, 437, 576 Ludian, 556, 565 Ludlow Beds, 154, 167, 177, 184 Ludwigia, 410 Lumulicardium, 158 Lure, Mont de, 487, 488, 489 Lutetian, 527, 549, 552, 555, 568 Lutra, 577, 602 Lybster Flagstones, 236 Lyckholm Limestone, 148 Lyell, Sir C., 572, 648 Lygodium,, 574 Lyme Regis, 395, 503 Lymnorella, 410 Lynton Beds, 204 Lyons, 376 Lyopora, 115 Lyra, 466 Lytoceras, 386 Lytoloma, 528 Macclesfield, 630 Machcerodus, 577, 588, 601, 638 MacHenry, A., 49, 50, 139, 210 Mackintosh, D., 631 Maclurea, 97 Macrocephalites, 410, 413 Macrochilina, 196 Macrochilus, 249 Macroscaphites, 466, 470 Macrotherium, 577 Mactra, 591 Madrepora, 528 Mceneceras, 196 Maentwrog Beds, 86 Maastricht, 516 Magdalenian age, 638 Magnesian Limestone, 334, 335 Magnolia, 464, 528, 574 Magnosia, 436 Malaise, C., 101, 186 Malmo, 517 Malmstone, 498 Malvern, 57, 88, 92, 164, 314 Shales, 92 Mammites, 466 Manchester, 336 Mansfield, 359 Mantellia, 438 Maplewell Beds, 58 Maps (geological), 27 Marchlyn-mawr, 87 Marcou, J., 324 Margate, 507, 508 Marginella, 528 Marines, Sables de, 554 Mariopteris, 290 Market Rasen, 445 Marl Slate, 334, 335 Marlstone Rock, 397, 398, 400 Marr, J. E., 94, 104, 110, 124, 174, 569 Marsipiocrinus, 156, 162 Marsupites, 466, 496 Marsupite zone, 507, 508 Mar wood, 209 Mastodon, 577, 588 Mastodontosaurus, 352, 363 Matagne, slates of, 219 Matley, Ch., 51, 274 Matonidium, 414 Mauchline, 339 Maw, G., 569 Maw, W., 165 May Hill, 164 Mayence 564 662 STEATIGRAPHICAL GEOLOGY Mayenne, 99 Mayo, 49, 139, 178, 272 Meadfoot Beds, 202 Meadowtown Beds, 128 Mediterranean Region, 313, 517, 607, 609 Medlicottia, 327, 344 Meekella, 319 Megalanteris, 195 Megalichthys, 250, 292 Megalodon, 195, 350 Megalosaums, 384, 415, 464 Megaphyllites, 349 Melanatria, 530 Melania, 530 Melanopsis, 530 Mel bourn Rock, 505, 507 Melocrinus, 195 Melongena, 576 Melonites, 247 Mendip Hills, 397, 416 Mendola dolomite, 374 Menevian Beds, 72, 81, 82 Merevale Shales, 93 Merioneth, 80, 82, 85, 116, 117, 121 Meristella, 155, 157, 160 Meristina, 156 Mesacanthus, 226 Mesarchaean rocks, 35 Mesieres, 406 Mesodiadema, 350 Mesolithic age, 638 Mesophyllum, 195 Mesozoic time, 5, 347 Messinian stage, 587, 605 Metoptoma, 249 Meudon, 551 Meyer, C. J. A., 540 Meyeria, 466, 470 Michelinia, 195, 247 Micmaccia, 91 Micrabacia, 466, 496 Micraster, 466, 496 Microchcerus, 529 Microdiscus, 75 Microdon, 410 Microlestes, 384 Microtus, 602 Midlothian, 269, 302 Milanovitz, 432 Miliolina Limestone, 568, 570 Millburn Beds, 131 Millepore Beds, 420, 422 Millstone Grit, 283, 294, 295, 300, 302, 304, 305, 323 Miocene Series, 11, 526, 575 Mithraeia, 528 Mithracites, 466 Mixosaurus. 352 Modiola, 246 Modiolopsis, 114, 115, 158 Moel Ferna Beds, 172 Moel Tryfaen, 631 Moelwyn Mountains, 119 Moffat area, 136 Mohicana, 91 Moine schists, 41, 44, 47 thrust, 86, 39 Molasse, 565 Monckton, H. W., 542, 543 Monobolina, 112 Monodacna, 606 Monodonta, 386 Monograptus, 155, 157 Monopleura, 466 Monotis, 350, 351 Mons, limestone of, 350 Montagne Noir, 61, 281 Montian, 527, 549, 551 Monticulipora, 115 Montigny, greywacke of, 216 Montlivaltia, 385 Montmartre, 505 Moravia, 312, 582 Moray, 339 Morecambe Bay, 630 Morigny, sands of, 566 Mortefontaine, 554 Morte Slates, '200, 204 Morton, G., 255 Morvan, 376 Morvern, 368, 369, 513 Mosasaurus, 465 Moscovian Series, 245, 313 Moscow area, 281, 312, 455, 521 Mottled Sandstone, 355, 357 Mousterian, 638 Mull, Island of, 369, 403, 404, 513, 547 Mundesley, 627 Murchison, Sir R., 70, 109, 154, 324 Murchisonia, 97, 115, 158, 162, 197, 249, 349 Murex, 532, 576 Muschelkalk, 348, 354, 370, 378 Mwelrea Grits, 140 Mya Beds, 648 Myalina, 195, 249 Mydrim Limestone, 124 Mylor Series, 144 Myoconcfia, 385 Myophoria, 195, 350, 351, 371, 388 Myrica, 464 Mytilus, 350 Naiadites, 291 Namur, 145, 215, 217, 279 Nant-glyn Flags, 172 Narberth, 124 Nare Head, 202 Natica, 371, 385, 437, 492 Naticella, 350, 351 Naticopsis, 196, 249, 327, 349 Nautilus, 349, 390, 411, 494, 531 Nebulipora, 115 Necrocarcinus, 466 Nemacanthus, 388 Nemagraptus, 110 Nemdtinion, 466 Nematophycus, 155 Neocomian, 463 Neogene System, 12, 526, 572 Neolimulus, 156 Neolithic, 638 Neozoic era, 5, 348 Neptunea, 590 Nerincea, 385, 410, 412 Nerita, 386 Neritopsis, 386 Neumayr, 460 Neuropteridium, 350 Neuropteris, 226, 287, 290 Newark, 335, 336 Newbourn Crag, 592 Newcastle Beds, 297 New Red Sandstone, 324 Newton, E. T., 364 Nicholson, H. A., 174 Nikitin, S., 343, 344, 455 Nilssonia, 414 Niobe, 76, 78 Nipa, 528, 531 Nithsdale, 303, 338 Noisy, Calcaire de, 554 Nomenclature of systems, 4 Nomismoceras, 249 Norfolk, 484,' 486, 500, 504, 511, 597 Norian stage, 374 Normandy, Cambrian of, 100 Carboniferous of, 317 Cretaceous of, 514, 515 Devonian of, 214 Jurassic of, 406, 430, 457 Ordovician of, 142 Permian of, 341 Trias of, 377 North Sea basin, 608, 634 Northampton, 362, 400, 419 Northampton Sand, 419 Northumberland, 263, 3CO, 334, 622 Norway, 64, 103, 146, 238 Norwich Crag, 597 Notgrave Freestone, 417 Nothe Beds, 442 Notosamrus, 352 Nottingham, 334, 335, 359 Nucleolites, 410, 413 Nucleospira, 156 Nucula, 492 Nummulitic Sea, 568 Nuneaton, 91, 93 Nyssa, 578 Obolella, 73, 74, 75, 76 Ochil Hills, 232 Oculina, 52S Odessa, 606 Odontopteris, 287, 290, 325 Odontopteryx, 528 (Eglina, 100, 112 CElhert, D. P., 99,213 (Eninghen Beds, 574, 580 0/aster, 497, 508 Ogwell, 208 Ogygia, 76, 111, 113 Oil Shale Group, 267, 268 Oland, Island of, 146, 181 Old bury Shales, 93 Oldhaven Beds, 537 Old Red Sandstone, 6, 8, 193, 223 Olenellns, 73, 74 INDEX 663 Olenian, 84 Olenus, 75, 77 Oligocene Series, 526, 555 Oliva, 528, 57(3 OliveUa, 533 Olonetz, 65, 67 Omosaurus, 434 Omphalotrochus, 158, 315 Omphyma, 156, 159, 160 Onchotrochiis, 466 Onchus, 158 Onychiopsis, 467 Opalinum zone, 409, 415 Operculina, 570 Ophidioceras, 158 Ophioceras, 386 Ophioderma, 385, 391 Ophioclerpeton, 292, 315 Ophiolepis, 385 Opis, 385 Oppelia, 410 Opponitz Limestone, 375 Orbiculoidea, 73, 75, 162, 388, 436 Orbitoides, 570 Orbitolina, 517 Orcadian area, 236, 242 Ordovician System, 8, 12. 71 Organic remains, destruction of, 18 ,, preservation of, 16, 17 Orgon, 489 Origin of species, 14 Orleanais, 579 Orinoy, 556, 566 Ornithocheirus, 465 Ornithopus, 466 Ornithosuchus, 352, 369 Orophocrinus, 247 Ortowr, 75, 76, 77, 78, 113, 114, 115, 156, 157, 161, 162 Orthisina, 110 Orthoceras, 76, 78, 158, 160, 161, 162, 249, 349 Orthoceras Limestone, 47 Orthonota, 160 Orthotheca, 73 Osborne Beds, 561, 562 Osmeroides, 464 Osmington Oolite, 442 Osmunda, 528, 578 Ostende, 550 Osteohpis, 226 Ostian Beds, 312 Ostrcea, 385, 413, 438, 528, 576 Otopteris, 391 Otnzamit.es, 350, 385, 414, 491 Ottweiler Beds, 317, 318, 341 Ouen, Calcaire de St., 554 Oural Mountains, 223 Ovibos, 602 Ox Mountains, 49 Oxford, 400 Oxford Clay, 432, 440, 441 Oxford Limestone, 264, 265 Oxfordian stage, 383, 451, 452, 459 Oxynoticeras, 386 Pabba Shales, 403 Pachydiscus, 466 Pachynolophus, 529 Pachypora, 195 Pachyrisma, 410 Pachystella, 511 Pachytheca, 155 Pad stow, 208 Palcearca, 76, 78, 112 Palceaster, 162 Palceasterina, 76 Palcechinus, 156, 160, 247 PalfKga, 466 Palceocardita, 350 Palceochcerus, 577 PalcKocoma, 162 Palceocorystes, 466 Pal&ocycius, 156 Palceodiscus, 156 Palaeogene System, 12, 526 Pala?ogeography, 32 Palceohatteria, 327 Palaeolithic implements, 638 Palceomastodon, 557 Palceomutela, 238 Palceonictis, 528 PaLceoniscus, 317, 327, 354, 364 Palseontographical Society, 32 Palaeontology, Books on, 31 Pcdceophis, 528 PalcKophonus, 178 Palwopteris, 212, 226 Palceosaunis, 352, 366 Palceotherium, 557 Palaeozoic era, 5 Pcdmacites, 578 Paloplotherium, 529 Pattopleurocenis, 386 Paludina Beds, 587, 606 Panopea, 531 Parabolina, 75, 77 ParacyatJms, 528 Paradoxides, 75, 81 Paradoxidian Series, 81 Parcdlelodon, 195, 248 Parasmilia, 466 Pareiasaurus, 344 Parezus, 196, 226 Paris basin, 552, 565 Par7ra, 226 Parkinsonia, 410, 411 Parona, Signor, 63 Partnach Beds, 373, 375 PateZZa, 386 Pavlow, A., 455, 490 Pays de Bray, 520 Pea-grit, 415 Peach and Home, Messrs., 176, 301 Pebble Beds, the, 356, 359, 361, 377 Pebidian Series, 53 Peckforton Hills, 357 Pecopteris, 290, 315, 325 Pecten, 350, 371, 388, 437, 464, 494 Pecten Bed, 400, 401 Pectunculus, 466, 529 Pedina, 410 Pelagosaurus, 386 Pellengare Series, 307 Peloneustes, 434 Pelosaums, 342 Pcltastes, 436 Peltoceras, 436 Peltura, 75, 76 Pembrokeshire, 53, 79, 80, 84, 109, 121, 228, 230, 255, 632 Penceus, 385 Pendleside Beds, 253, 255, 259 Pengelly, W., 644 Penkill Beds, 176 Pennant Grits, 304, 307 Pennine area, 294, 314, 569, 629 Penrhyn, 81, 86, 121 Penrith Sandstone, 337, 338 Pentacrinus, 385, 391 Pentactinella, 350 Pentainerus, 156, 157, 160, 161, 162 Pentland Hills, 177, 179 Pen-y-glog Beds, 172 Percostoma, 528 Pericydus, 249 Periechocrinus, 156, 162 Perisphinctes, 436, 464 Permian System, 12, 324 Perna, 385, 464, 469 Perna Bed, 477 Perodinella, 410 Perrier, 580, 604 Perthshire, 233 Peterborough, 441 Petherwin Beds, 208, 209 PJiacops, 110, 114, 160, 162, 195, 197 Phcenicites, 528, 574 Phaneropleuron, 225, 226 Pharetrospongia, 466 Phascolotherium, 384, 415 Phillipsastrea, 195, 247 Phillipsia, 246, 248 Philoxene, 196 Pholadomya, 350, 364, 385, 411, 413, 435, 529 Pholidophorus, 386, 435 Phragmoceras, 158, 162 Phylloccenia, 410 Phylloceras, 349, 386 Phyllodus, 528 Phyllograptus, 110, 111 Phyllopora, 115, 343 Physa, 438 Pickwell Down Beds, 209 Pikermi Beds, 606 Pileolus, 410 Piloceras, 96 Pilsen, 317 Pilton Beds, 209, 276 Pinacoceras, 349, 351 Pinacopora, 159 Pinites, 471 Pinna, 385 Pinus, 527 Pirbright, 542 Pitharella, 530 Pitys, 245, 290 Placenticeras, 466 664 STRATIGRAPHICAL GEOLOGY Placodus, 353 Placoparia, 110, 145 Placoparia Beds, 146 Placocystis, 156, 160 Plagiaulax, 434 Plagiozamites, 315, 325 Plaisantian stage, 587, 605, 606 Planera, 574 Planorbis, 4S9, 559 Plant-assemblages, 322 Plasmopora, 195 Platanus, 464, 527 Plateau gravels, 640 Plattenkalk, 432 Platyceras, 158, 196 Plaiychisma, 158 Platycrinus, 247, 248 Platysomus, 250, 292 Pleistocene Series, 526, 611, 637 Plesiosaurus, 386, 392 Plethodus, 466 Pleuracanthus, 292 Pleurodictyum, 195 Pleurograptus, 111, 115 Pleuromya, 413, 464 Pleuronautilus, 293 Pleurophorus, 195, 249, 327 Pleurosternon, 438 Pleurotoma, 531, 576 Pleurotomaria, 158, 196, 249, 327, 390, 436, 464 Plicatula, 350, 385, 492 Plinthosella, 466 Pliocene Series, 11, 526, 586 Pliopithecus, 577 Pliosaurus, 386, 434 Plocoscyphia, 466 Plougastel Grits, 213 Pluckley, 450 Plutonia, 75 Plymouth, 204, 207, 208 Plynlimmon, 632 Polacanthus, 464 Poland, 183, 222, 241, 312, 431, 521 Polesworth, 332 Poly pty chiles, 436 Polyptychodon, 464 Pomeroy, 139, 178, 237 Pontefract, 297, 335 Pontesford Hill, 56 Pontian stage, 587, 605, 607 Pontypool, 307 Popanoceras, 327, 344 Populus, 464, 527, 574 Porcellia, 196 Porlock, 367 Porsguen Slates, 213 Portheus, 464 Portlandian, 383, 432, 445, 455, 460 Portland Beds, 432, 445 Bill, 645 Portscatho Beds, 144 Portsmouth, 546 Portugal, 100, 144, 187 Portunites, 528 Posidoniella, 248 Posidonomya, 246, 350 Potamides, 530, 559, 576 Potamomya, 528 Poterioceras, 196, 249 Poteriocrinus, 247 Potton, 483 Prague, 222 Preservation of organic re- mains, 16, 17 Presteign, 165 Preston, 359 Prestwich, Sir J., 593 Prestwi'cMa, 292 Primofilices, 289 Prionocyclus, 466 Prionotropis, 466 Pristis, 528 Probus Grits, 202 Procervulus, 577 Productus, 195, 246, 248, 327 Proetus, 110, 156, 195, 248 Prolecanites, 196, 249 Pronorites, 249 Protapirus, 557 Protarchsean Rocks, 35 Proterosaurus, 327 Prothyris, 195 Protocardia, 350, 385, 388 Protnplasma, 315 Protosphyrcena, 464 Protospondylus, 347 Protospongia, 73, 74 Provence, 431, 487, 489, 517, 555, 600 Proviverra, 528 Przibram Shales, 64 Psammodus, 251 Psammosteus, 226 Psaronius, 289, 325 Pseudocrinus, 156 Pseudodiadema, 385 Pseudomelania, 350, 386 Pseudomonotis, 327, 351, 390 Psiloceras, 386, 389 Psilophyto-n, 226 Pteranodon, 465 Pteraspis, 196, 204, 226, 228 Pteria, 350, 388 Pteridosperrns, 289, 325 Pterlchthys, 224 Pterincea, 158, 161, 162, 195 Pterinopecten, 248 Pteris, 528 Pterocerian stage, 452 Pterodactyles, 384, 433, 434 Pterodon, 528 Pteroperna, 410 Pterophyllum, 315, 325, 350, 371, 385, 412 Pterosauria, 384 Pterygotus, 156, 162, 224, 226 Ptychites, 349, 351 Ptychodus, 466 Ptychoparid;, 91 Ptychopteria, 195, 198 Ptychostoma, 350 Pugnax, 248 Pugnelhis, 466 Punfield Beds, 477, 481 Purbeck Beds, 432, 448, 449 Purbeckian, 452, 460 Purley Shales, 93 Purpura, 591 Purpuri'na, 386 Pygcister, 410 Py gurus, 410, 464 Pyrenees, the, 187, 281 Pyrula, 528 Quaderstein, 518 Quantoek Hills, 207, 367 Queen Camel, 395 Quercus, 574, 578 Raasey, 368, 369, 403, 404, 428 Radiolarian cherts, 135, 130, 150, 186, 276 Radiolites, 466 Radstock Beds, 304 Raibl Beds, 373, 375 Raised Beaches, 645 Ramipora, 115 Ramsay, Sir A. C., SO, 87, 88, 331 Ramsdell Clay, 541 Ramsey Island, 116 Rastall, R. H., 402, 443 Rastrites, 155, 157 Rastrites Shales, 182 Raugraben Shales, 375 Reade, T. M., 629 Reading Beds, 527, 537, 539 Reculvers, 537 Red Chalk, 500 Red Crag, 584, 592, 596 ,, Head Group, 232 Redon, 141, 142 Eedonia, 112 Reed, E. Cowper, 140 Reef Limestone, 191 Reid, C., 276, 562, 593, 596, 599, 600, 601, 603, 609, 649 Reiffling Limestone, 373, 374 Reigate, 478, 479 Reinach, von, 333 Relative age of strata, 2 Remopleurides, 110 Rennes, 99 Rensellceria, 195 Reptilia, 384 Retzia, 162 Revinian Series, 101 Reynolds, S. H., 139, 180 Rhabdocidaris, 385 Rhacopteris, 245 RhacophyllUes, 349 Rhaetic Beds, 375, 383, 395, 399, 405, 408, 456 Rhamphorhynchus, 434 Rhine, the, 600, 60S Rhine valley, 219 Rhinoceros, 588, 601, 63S Rhinog Grits, 81 Rhipidocrinus, 195 Rhiwlas Limestone, 125 Rhizodopsis, 292 Rhizodus, 250, 292 Rhizophyllum, 156 Rhodocrinus, 247 Rhyader Slates, 16S INDEX 665 Rhynchonella, 110, 197, 198, 350, 390, 391, 411, 436, 494, 495 Rhynchosaurus, 353 Rhynchotreta, 156 Riaden Slates, 141, 142 Ribband Series, 138 Rlccarton Beds, 177 Richardson, L., 395 Richmond, 393 Rimella, 528 Ripple-marks, 364 Rissoina, 410 River gravels, 641 Roberts, T., 443, 451 Robinia, 574 Rochester, boring at, 163 Rock Salt, 343, 365, 370 Rocroi, 101 Rogers, T., 309 Roman Fell, 259, 261 Roslin Sandstone, 302 Ross, Archaean of, 36 Rostellaria, 528 Rostellec Shales, 213 Rotherham, 298 Rouen, 405 Rowe, A. W., 506, 507, 508, 510, 511 Roydon, 500 Rozan, limestone of, 141 Ruddy, T., 125 Rupelian, 363 Russia, Cambrian of, 104 Carboniferous of, 281, 312 Cretaceous of, 490, 519 Devonian of, 222 Jurassic of, 409, 455 Ordovician of, 148 Permian of, 324, 343 Silurian of, 183 Rutland, 399 Rutotia, 199 Sabal, 528, 574 Sables Moyens, 554 Saccammina, 245, 263 Sacco, F., 555, 567, 584 Sagenaria, 226 Saiga, 638 Sailmhor Group, 96 Salenia, 466, 494 Salmian Series, 101 Salopian Series, 154, 165, 170, 177 Salrock Beds, 180 Salter, J. W., 276, 477 Salterella, 73, 91, 96 Saltholm, 516 Salzburg, 408 Samara, 343 Sancats, 578 Sandgate, 477, 478 Sandgate Beds, 475, 478, 479 Sandringham sands, 486 Sanguinolaria, 528 Sanguinolites, 249 Sannoisian, 556, 565 Sao, 104 Sarmatian stage, 573, 575, 582, 583 Sarrebruck area, 311, 317, 341, 376 Sarthe, the, 514 Saxonian stage, 324, 341, 342, 345 Saxony, 325 Scalaria, 350, 386, 492 Scaldisian, 584, 602, 603 Scalpa, 404 Scandinavia, 64, 103, 146, 181, 2S2, 407, 634, 647 Scania, 147, 181 Scaphaspis, 156, 162, 226 Scapheus, 385 Scaphites, 466, 495 Scaphognathus, 433, 434 Scar Limestones, 259, 263, 265 Scarborough Limestone, 420, 422 Scelidosaurus, 386 Schatzlar Beds, 312 Schellwein, 320 Schizaster, 528 Schizocystis, 156 Schizodus, 249, 319, 326, 327, 350 Schizolepis, 350 Schizopteris, 325 Schlern Dolomite, 375 Schlier, the, 582, 585 Schlcenbachia, 466, 495 Schlotheimia, 386 Schmidt, 104 Scifenurus, 528 Sciuroides, 529 Sderomocklus, 369 Scoliostoma, 196 Scolithus, 73 Scotland, Archaean of, 36 Cambrian of, 96 Carboniferous of, 265, 301 Cretaceous of, 487, 513 Eocene of, 547 Jurassic of, 402, 450 O.R.S. of, 132 Ordovician of, 132 Pleistocene of, 617, 619 Silurian of, 176 Trias of, 368 Scremerston Beds, 264 Scunthorpe, 400 Scutella, 574, 579 Scytalia, 511 Seamraig Beds, 45 Seaton, 367 Sederholm, J. J., 65 Sedgley, 167 Sedgwick, A., 70, 79, 109, 153, 194 Seemann, F., 222 Selborne, 499 Selbornian, 462, 463, 498 Selenacodon, 466 Seliscothon, 511 Selous, F. C., 16 Seminula zone, 253, 259 Semionotus, 351 Senni Beds, 228 Senonian, 463, 491, 515 Sequanian, 383, 451, 452 Sequoia, 527, 574 Serpulites, 73 Serpulite Grits, 96, 97 Servia, 409, 432 Sevenoaks, 477, 478 Sewerby gravels, 625 Shells, stability of, 19 Shelve, 127, 128 Sheppey, Isle of, 540 Sherborn, 416, 424 Shineton Shales, 92 Shipman, J., 364 Sholeshook Limestone, 124 Shotover, 483 Shrewley Sandstone, 364 Shropshire, Archaean of, 55 Cambrian of, 89, 92 Carboniferous of, 295, 298 O.R.S. of, 227 Permian of, 328 Silurian of, 164 Trias of, 355, 363 Shumardia, 92 Sibertzew, 344 Sibly, T. P., 257 Sicilian stage, 584, 605, 606, 609 Sidmouth, 367, 499, 500 Sigillaria, 286, 288, 319 Silesia, 280, 312, 343, 518 Silurian System, 12, 70, 153 Simbirsk, 455, 490 Sinemurian, 383, 405, 408 Sion Slates, 141 Siphonia, 466, 494, 511 Siphonotreta, 110 Sisteron, 487 Skelgill Beds, 174 Skiddaw Slates, 93, 113 Skima, 146 Skye, Island of, 45, 97, 368, 369, 428, 547 Slade Beds, 124 Sleddale Beds, 132 Sligo, 272 Slimonia, 156 Smerwick Beds, 180 Smith, W., 3 Snettisham Clay, 486 Snowdon, 123 Soissonais, 552 Solarium, 386 Solenastrea. 528 Solenhofen, 454 Solenopleura, 75, 97, 104 Solenopora, 263 Solenopsis, 195 Solva Beds, 72, 81, 82 Somerset, Carboniferous of, 305 Jurassic of, 395, 397, 399, 422, 424 Permian of, 332 Trias of, 367 Sonninia, 410 Sorby, H. G., 19 Sorrel, Mount, 58 South Wales, see Wales 666 STRATIGRAPHICAL GEOLOGY Southwold, 599 Stomechinus, 410 Sowerbya, 410 Stonesfield Beds, 425 Spain, 100, 144, 187, 214 Stonehaven, 97 Spalacotherium, 434 Stotfield, 369, 429 Sparagmite Sandstone, 65 Straiton Beds, 177 Sparnacian stage, 527, 549 Stralian, A., 110, 230, 255, Spatangus, 528, 574 481, 629 Species, origin of, 14 Stramberg Limestone, 454 Speeton, 505, 507 Straparollus, 196, 249 Speeton Clay, 484 Strata, relative age of, 2, 3 Sperenberg, 343, 381 Spermophilus, 638 Sphcerexochus, 110 Strathmore, 232 Stratigraphical Breaks, 7 Stratiotes, 578 Sphceriola, 350 .Street, 397 Sphserocodium Limestone, Strepsodus, 292 182 Streptospondylus, 410 Sphceroceras, 410 Stricklandia, 156,. 157, 160 Sphcerocrinus, 195 Stringocephalus, 195 Sphceronites, 114, 115 Stroboceras, 249 Sphcerophthalmus, 76, 77 Sphcerulites, 466 Sphenodon, 353 Stromatopora, 156, 195 Stromatoporella, 195 Stromatoporoids, 181, 189, Sphenolepidium, 471 190, 207, 217 Sphenopteris, 226, 245, 287, Strombodes, 160 290, 414 Strophalosia, 195, 326, 327 Sphenosaurus, 317 Strophodus, 410 Sphyrcenodus, 528 Strophomena, 110, 156, 159 Spilsby, 445, 484, 486 Strophonella, 159, 162 Spirifer, 156, 197, 248, 326, 327 Spiriferina, 248, 319, 350 Stur, D., 280, 312 Stylastrea, 385 Spirorbis Limestone, 297, 298, Stylina, 410 301 Styliola, 208 Spirulirostra, 576 Stylodon, 434 Spitzbergen, 459 Subulites, 158 Spondylus, 410, 496 Succession of strata, 3 Spongiostroma, 182, 183 ,, of faunas, 21 St. Alban's Head, 444, 445 Suessonian, 549, 555 St. Bees, 337, 359 Suess, E., 585 St. Cassian Beds, 373, 385 Suffolk, 593, 596 St. David's, 79, 116 Superga Beds, 583 St. Erth Beds, 595 Surrey, 477 St. Etienne, 315 Sussex, 508, 540 St. George's Land, 282 Sutherland, 36, 96, 403, 404 St. Lo, schists of, 60, 100 Sutton (Glamorgan), 397 St. Ouen, Calcaire de, 554 Sutton (Suffolk), 595, 596 Stachyodes, 195 Swabia, 407, 431, 453 Staddon Grits, 204 Swagerina, 319 Staffordshire, Carboniferous Swanage, 448, 476, 503 of, 255, 296, 297, 298, 299 Swansea, 307 Permian of, 328 Sweden, 64, 65, 103, 146, 181, Silurian of, 167 407, 516, 647 Trias of, 356, 363 Swindon, 447, 449 Stampian stage, 556, 566 Switzerland, rocks of, 62 Starfish Bed, 398 ,, glaciers of, 637 Start Point, 59 Sycum, 528 Stauria, 156 Synchronism, 23 Staurocephalus, 110, 115 Synocladia, 327 Stauronema, 466 Syringopora, 195, 247 Stavelot, 101 Syringothyris, 246 Steneosaurus, 434 Syringothyris zone, 253, 259 Stenometopon, 369 Systems, definition of, 10, 12 Stenotheca, 73, 104 Stephanian Series, 245, 313, Tceniopteris, 315, 350 320 Tancredia, 385 Stepheoceras, 411 Tanne and Tannenthal, 185 Stereognattius, 415 Taplow, 510 Stigmaria, 226, 288, 322 Tarannon Shale, 152, 155, 162, Stiper stones, 128 168 Stobbs, J. T., 255 Tate, G., 263, 512 Stockdale Shales, 174 Tatra district, 408 Taunus quartzite, 216, 220 Taxocrinus, 195 Taxodium, 574 Tealby Beds, 484, 486 Teall, J. J. H., 39 Teignmouth, 333 Teilia Beds, 255 Telacodon, 466 Teleostean fish, 347 Teleosaurus, 410, 434 Telerpeton, 353 Tellina, 576 Temnechinus, 589 Temnocheilus, 249, 327, 349, 371 Tempskya, 471, 475 Tentaculites, 115 Teratosaurus, 352 Terebratella, 385 Terebratula, 195,350, 371, 412, 413, 589 Terebratulina, 410, 451 Terebrirostra, 466 Tertiary time, 5, 526 Tetractinella, 350, 351 Tetragraptus, 110, 111 Beds, 117, 131 Teutoburger Wald, 407 Text-books on Geology, 28 Thalassiceras, 344 Thame, 447 Thames valley, 641 Thamnastrea, 350, 385 Thanet Beds, 527, 535 Thanetian stage, 527, 549, 551 Thecia, 156 Thecidium, 385, 516 Thecodontosaurus, 352, 366 Ttiecosmilia, 350, 385, 435 Thecospira, 350 Thelodus, 158, 162 Theridomys, 529 Thetironia(= Thetis), 466 Thonschiefer, 62, 64 Thornhill basin, 338 Thracia, 350, 364, 385, 436 Thurgarton, 298 Thuringia, 185, 280, 342, 370 Thuringian stage, 225, 334, 342 Thursius, 226 Thurso Beds, 236 Thyle valley, the, 101 Tiarcchinus, 350 Tilgate Forest, 475 Till, 619, 620 Tirolites, 349 Tithonian stage, 454 Toarcian stage, 383, 405, 406, 408, 457 Tobermory, 403 Todites, 414 Tomichia, 559 Tongrian, 556. 563 Topley, W., 450 Tornebohm, A. E., 67, 68 Tornoceras, 196 Torquay, 204, 206, 207, 644 Torridonian Series, 36, 38, 44 INDEX 667 Tortonian stage, 573, 576, 582, 583 Tortworth, 168 Totternhoe stone, 504 Toula coal, 321 Touraine, 515, 578 Tournaisian stage, 251, 278 Toxaster, 466, 490 Trachyceras, 349 Transition Beds, 285, 294, 307 Traouliors Shales, 213 Trapa, 635 Trapezium, 385 Trappoid breccia, 330, 331 Traquair, R. H., 302 Tremadoc Beds, 72, 84, 87 Trematis, 110, 115 Trematosaurus, 352 Tretoceras, 158, 160 Triassic System, 12, 348 Trichites, 410 Triconodon, 384, 434 Trigonia, 385, 411, 412, 435, 437, 464, 469 Trigonia Grits, 415, 417, 425 Tngonocarpus, 290 . Trigonodus, 359 Trigonosemus, 466 Trimerocephalus, 195 Trinudeus, 110, 115 Triplesia, 110, 156 'J'ritonium, 528 Tritylodon, 354 Trockammina, 245 Trochoceras, 115, 158, 162, 196 Trochocyathus, 385, 492 Trochotoma, 386 Trochus, 350 Trogontherium, 601, 638 Trophon, 528 Tropites, 349, 351 Trumpington, 642, 643 Tschernyschew, T., 319 Tuedian stage, 263, 264 Tuffeau de Ciply, 516 Tunbridge Sands, 475 Turbinolia, 528 Turbo, 160 Turbonitella, 249 Turin, 567 Turnford, 241 Turonian, 463, 491, 515 Turrilites, 466, 495 Turritella, 350, 386, 576 Twt Hill, 53 Typhis, 528 Tyrol, the, 373, 408 Tyrone, 49, 139, 178, 237, 304 Ufa, 343 Ullmannia, 325 Ulodendron, 286 Uncites, 195 Unconformities, 7 Underclays, 322 Ungulite Limestone, 104 Unicardium, 350, 385 Unio, 438, 467 Upper Greensand, 498 Wardie Shales, 267 Upware, 443, 483 Wardour, vale of, 447, 449 Upway, 445 Ware, boring at, 163 Ural Mountains, 281, 313, Warwickshire, 363 315, 319, 343 Uralian Series, 245, 315, 320 Watchet, 367, 395 Watcombe Clays, 333 Urgonian, 489 Waterstones, 362, 363, 364 Uriconian rocks, 56, 57 Watney and Welch, Misses, Uronectes, 327 174 Ursus, 602 Urycordylus, 292, 317 Watson, D. M. S., 339 Watts, Professor W. W., 58 Usher, W. A. B., 200, 206, 308, Waulsortian, 279 333 Weald, the, 473 Weald Clay, 475 Valenginian, 463, 489, 490 Wealden Beds, 462, 463, 473 Valentian Series, 154, 158, Wealden Lake, 520 168, 176, 178, 184 Weeley, 102 Valvulina, 245 Welton, 501 Vaughan, Dr. A., 251, 252, Wemmelian, 527, 549, 551 259, 260, 274, 278 Wengen Beds, 373, 374 Vectian, 462, 463 Wenlock Beds, 154, 158, 165, Fectisaurus, 464 170, 175, 177, 184 Velay, 566, 567, 579 Wensleydale, 260 Velopecten, 350, 385 Werfen Beds, 373, 374 Ventoux, Mont, 487 Werfenian stage, 374 Ventrlculites, 466 Westbury, 442 Verrucocoilia, 410 Westmoreland, 93, 124, 138, Verruculina, 466, 511 174, 262 Verticordia, 528 Westphalia, 285, 309, 453 Veryan Beds, 144 Westphalian Series, 245, 285, Vestinautilus, 249 309 Vicentin, 555, 567 Westrogothia, 103 Vienna, 432, 554, 582 Vienna Sandstone, 554 Wexford, 95, 137, 633 Weybourn Crag, 592, 600 Vieux Jones, 563 Weymouth, 442 Vigra Beds, 86 Whid borne, G. F., 207 Villedieu, Craie de, 515 Whitaker, W., 473, 537, 599 Vireux Beds, 216 White Lias, 395, 399 Virgloria Limestone, 373 Whitecliff Bay, 540, 542, 560, Virgulian, 452 561, 562 Visean stage, 251, 255, 278 Whitehaven, 301 Viviparus, 569 Wicklow, 95, 137, 633 Vobster Beds, 304, 307 Wickwar, 253 Volcanic rock, Archaean, 53, Williamsonia, 385, 414 56, 57, 58 Willoughby, 501 Cambrian, 107 Wilsonia, 156, 162 ,, Devonian, 206, 231 Wiltshire, 442, 444, 499, 501 Ordovician, 116, 121 ,131 Woburn sands, 483 Tertiary, 544 Wood, S. V., 595, 628 . Volgian, 455 Wood, Miss E. M., 170, 171, Vologda, 344 172 Voltzia, 350, 363 Woodocrinus, 246, 277 Voluta, 466, 528, 531, 533, Woodward, H. B., 400, 424, 559, 589 447, 449, 457 Vosges district, 376 Woolacott, D., 334 Woolhope Shales, 164 Walchia, 315, 325, 326 ,, Limestone, 165, 166 Waldenbury Beds, 312 Woolwich Beds, 527, 535, 537 Waldheimia, 391 Wrekin, The, 56 Wales, Archaean of, 50 Cambrian of, 79 Xanthopsis, 528 Carboniferous of, 253, 255 Xenophora, 528, 533 Old Red Sandstone, 227 Xiphodon, 537 Pleistocene of, 618, 629, 631 Trias of, 367 Yarbridge, 506 Walpen sands, 479 Yeovil, 398 Walton Crag, 592, 597 Yoldia Clays, 647, 648 Warbarrow Bay, 476, 477 Yoredale Beds, 259, 260 Warberry Hill, 204, 205 Yorkshire, Carboniferous of, Ward, Clifton, 93 258, 295 668 STRATIGRAPHICAL GEOLOGY Yorkshire, Cretaceous of, 484, 505, 507, 511 Jurassic of, 401, 420, 443, 445, 447 Permian of, 335 Pleistocene of, 623 Ypresian, 549, 550, 552 Yuccites, 363 Zamites, 350, 385, 471 Zanclodon, 352, 371 Zaphrentis, 195, 247 Zaphrentis zone, 253 Zechstein, 342 Zeiller, R., 311 Zenacanthus, 317, 327 Zones, 22, 158 ,, of fossil plants. 322 THE END Printed by R. & R. CLARK, LIMITED, Edinburgh A SELECTION OF BOOKS AND MAPS ON GEOLOGICAL AND KINDRED SUBJECTS THE BUILDING OF THE BRITISH ISLES BEING A HISTORY OF THE CONSTRUCTION AND GEOGRAPHICAL EVOLUTION OF THE BRITISH REGION BY A. J. JUKES-BEOWNE, B.A., F.R.S, F.G.S. Third Edition, Rewritten and Enlarged. 488 pages, large post 8vo, with 10 plates, 16 coloured maps, and 54 text illustrations. Cloth. Price 12s. net (12s. 4.d.). " A singularly fascinating story, told with spirit and yet with restraint by one of the great masters of British geology." Mining Journal. "The work is one which we consider essential to the student of the geology of the British Isles." Science Progress. STANFORD'S GEOLOGICAL MAP OF CENTRAL EUROPE SHOWING ALSO THE SOUTHERN AND MIDLAND COUNTIES OF ENGLAND AND IRELAND deduced from the " Carte geologique Internationale.'" One sheet, 20 by 14 inches, 100 miles to an inch (1 : 6,336,000). Price: Coloured sheet, 5s. (5s. 4d.). READY IN JUNE SUB=ATLANTIC PHYSIOGRAPHY A new Monograph by Professor EDWARD HULL, LL.D., F.R.S., with, one chapter by Professor J. W. SPENCER. An important work, illustrated by Eleven Charts, dealing with the form and structure of the ocean floor, with especial reference to the submarine valleys of the chief rivers entering the ocean from the land. Folio size. Price, One Guinea net. LONDON : EDWARD STANFORD, 12, 13, & 14, LONG ACHE, W.C. Cartographer to His Majesty the King. 1 A SELECTION OF BOOKS AND MAPS ON GEOLOGICAL AND KINDRED SUBJECTS CONTRIBUTIONS TO THE PHYSICAL HISTORY OF THE BRITISH ISLES ; with a Dissertation on the Origin of Western Europe and of the Atlantic Ocean. With 27 Coloured Maps. By EDWARD HULL, M.A., LL.D., F.R.S., late Director of the Geological Survey of Ireland. Medium 8vo. Cloth. 6s. net. (6s. 4d.). " Professor Hull has rendered a signal service to students of geology by preparing and publishing a series of maps of the British Islands, in which he indicates, by dis- tinctive colouring, the distribution of land and water at successive periods of geological history. No such series of maps, so far as we know, has ever before been published, or even attempted." Academy. BY THE SAME AUTHOR. THE PHYSICAL GEOLOGY AND GEOGRAPHY OF IRELAND. Part I. Geological Formation of Ireland. Part II. Physical Geography of Ireland. Part III. The Glaciation of Ireland. Second Edition, revised. With Maps and Illustrations. Post 8vo. Cloth. 7s. (5s. 7d.). PHYSICAL GEOLOGY AND GEOGRAPHY OF GREAT BRITAIN. By Sir ANDREW C. RAMSAY, LL.D., F.R.S., etc., late Director-General of the Geological Survey. Sixth Edition. Edited by HORACE B. WOODWARD, F.G.S., of the Geological Survey. With numerous Illustrations and a Geological Map of Great Britain, printed in Colours. Post Svo. Cloth. 10s. 6d. (8s. 5d.). "No better introduction to the principles of geology could possibly be recom- mended to the English reader. It is a work of the highest value, and one worthy to take a foremost place among popular manuals of science." Nature. GEOLOGY IN THE FIELD. ' The Jubilee Volume of the Geologists' Association (1858-1908). Edited by H. W. MONKTON and R. HERRIES. A Series of Articles by competent authorities on the various districts in England and Wales visited by the Association on their excursions during the last fifty years. In Four Parts, about 900 pages, with upwards of 30 Plates and 130 Figures in the Text. Demy Svo. Wrappers. Each Part, 5s. net. (5s. 3d.) Or bound in Cloth, in One volume, 23s. net (23s. 6d.); or in Two volumes, 12s. 6d. net each (12s. lid.). OUTLINES OF GEOLOGY : an Introduction to the Science for Junior Students and General Readers. By JAMES GEIKIE, LL.D., F.R.S. With 400 Illustrations. Fourth Edition. Revised. Large post 8vo. Cloth. 12s. (9s. 5d.). "The style is clear, simple, and unpretending ; the author has evidently striven to put the subject fairly before students rather than to express his own views, and has thus produced a book which cannot fail to be of great service." Saturday Review. PRE-HISTORIC EUROPE. A Geological Sketch. By JAMES GEIKIE, LL.D..F.R.S. With Maps and Illustrations. Medium Svo. Cloth. 25s. (19s. 3d.). " Dr. Geikie's object in the present volume is to give an outline of what appear to have been the most considerable physical changes experienced in Europe since the beginning of the Pleistocene or Quaternary period, the period immediately preceding that which is now being wrought out, and which Dr. Geikie divides into Post-glacial and Recent. He attempts to describe, in a more systematic manner than has hitherto been attempted, that succession of changes, climatic and geographical, which, taken together, constitute the historical geology of Pleistocene, Post-glacial, and Recent times. This he does in a more thorough and complete manner than, we venture to think, has ever been done before." Times. LONDON: EDWARD STANFORD, 12, 13, & 14, LONG ACRE, W.C. 2 A SELECTION OF BOOKS AND MAPS ON GEOLOGICAL AND KINDRED SUBJECTS SECOND EDITION REVISED AND ENLARGED Crown 8vo, cloth. Price 12s. 6d. net (12s. 10<2.) STANFORD'S GEOLOGICAL ATLAS OF GREAT BRITAIN AND IRELAND [Based on Reynolds'" s Geological Atlas] WITH PLATES OF CHARACTERISTIC FOSSILS PRECEDED BY A DESCRIPTION OF THE GEOLOGICAL STRUCTURE OF GREAT BRITAIN AND IRELAND AND THEIR COUNTIES; AND OF THE FEATURES OBSERVABLE ALONG THE PRINCIPAL LINES OF RAILWAY BY HOEACE B. WOODWAED F.R.S., F.G.S. 36 Coloured Maps and 16 double-page Plates of Fossils ; with Text, 200 pages, containing 20 Sections and Views. " Crammed full of information of the best quality, and written by an experienced practical geologist who tells his readers what is to be seen of geological interest on every line of railway and in every separate county in the kingdom." The Geological Magazine. LONDON: EDWARD STANFORD, 12, 13, & 14, LONG ACRE, W.C. Cartographer to His Majesty the King. 3 STANFORD'S INDEXED ATLAS OF THE COUNTY OF LONDON WITH PARTS OF THE ADJACENT BOROUGHS AND URBAN DISTRICTS WITH A PREFACE BY SIR LAURENCE GOMME Size 6j by 10 inches. 166 pages of Text, and 87 Map-Sections. With an Index of upwards of 12,000 names. SCALE OF MAPS 4 inches to one mile. Strongly bound in attractive red cloth, 7s. 6d. net (7s. IQd.) Presentation binding, half morocco, gilt, 10s. 6d. net (10s. lOd. " It is admirably executed and most useful. London geography is a most complicated matter, and this should help to disentangle it." Spectator Detailed Prospectus, with Specimen Section, gratis. Complete Catalogue gratis on application. LONDON: EDWARD STANFORD, 12, 13, & 14, LONG ACRE, W.C. Cartographer to His Majesty the King. 4 14 DAY USE RETURN TO DESK FROM WHICH BORROWED EARTH SCIENCES LIBRARY This book is due on the last date stamped below, or on the date to which renewed. Renewed books are subject to immediate recall. LD 21-40m-5,'65 (F4308slO)476 General Library University of California Berkeley