EUGENE L. MORICE, ; J 2 . v : :*:.. THE STUDENT'S LYELL THE PRINCIPLES AND METHODS OF GEOLOGY, AS APPLIED TO THE INVESTIGATION OF THE PAST HISTORY OF THE EARTH AND ITS INHABITANTS L \ V EDITED BY JOHN W. JUDD C.B., LL.D., F.R.S. Formerly Professor of Geology and Dean of the Royal College of Science, London WITH HISTORICAL INTRODUCTION SECOND EDITION, REHSED AND ENLARGED WITH A PORTRAIT AND 736 ILLUSTRATIONS IN THE TEXT LONDON JOHN MURRAY, ALBEMARLE STREET 1911 [6] PREFACE most cherished recollections are those days of constant and friendly intercourse with Sir Charles Lyell during the period he was engaged in writing the ' Student's Elements ' and in preparing the second edition. The progress of geological science, during the last quarter of a century, has rendered necessary very considerable additions and corrections, and the re-writing of large por- tions of the book, but I have everywhere striven to preserve the author's plan and to follow the methods which charac- terise the original work. In spite of the expansion of the text and the intro- duction into it of more than one hundred new illustrations, it has nevertheless been found possible, by using smaller type for certain portions, to avoid increasing the bulk or the cost of the volume. It is hoped, moreover, that this employ- ment of different kinds of type will afford some assistance to the studenf. The beginner is advised, in his first perusal of the work, to devote his attention mainly to the portions printed in larger type ; and afterwards, in entering upon its more serious study, to read the whole through continuously. It may be well to mention, too, that although, for the very cogent reasons urged by Lyell (see p. 140), it is desirable in the first instance to study the newer and less altered strata before the older and greatly metamorphosed rocks, yet, in revising his studies at a later period, the reader may find it advantageous to take up the several systems in historical order. Some of the additional illustrations to the book have been taken from the writings of Lyell's lifelong friend and fellow-worker Poulett Scrope; for others my thanks are due to the Trustees of the British Museum and to Dr. Henry Woodward, to Professor 0. C. Marsh of Yale College, and to Dr. E. D. Roberts. The majority of the new illustrations have, however t been specially drawn for PREFACE [7] the work by Mr. Gilbert Cullis, and to that gentleman and to Dr. W. Eraser Hume I am also indebted for much care in reading portions of the proof-sheets, while Professor T. Eupert Jones has supplied some valuable notes and corrections. Nor should I omit to mention that the work owes not a little to the labours of the late Dr. P. M. Duncan, who revised the fourth edition of the * Student's Elements,' to the late Messrs. Searles V. Wood and David Forbes, as well as to Mr. Eobert Etheridge and Professor T. G. Bonney, who aided Lyell in the preparation of the work for a second edition. The size of the book of course precludes its being made an exhaustive treatise on geology, with full references to original memoirs ; nor dare I anticipate that all my fellow-teachers will coincide with me in judgment as to what should be included in and what it is best to omit from a work with the scope and limits of the present one ; yet I venture to hope that the modifications, rearrangements, and additions now introduced into the book have served to bring it up to date, and that teachers and students may alike find that it continues to be what Lyell made it a convenient and trustworthy introduction to geological science. JOHN W. JUDD. KEW : March 1896. As it is impossible to enable the Museum, and in many provincial reader to recognise minerals, rocks, museums. Specimens, rock sec- and fossils by the aid of figures tions, and microscopical prepara- and verbal descriptions only, he tions, &c., specially arranged to will do well to refer to properly illustrate the different portions of labelled specimens. Such may be this work, may be procured from seen in the British Museum of Mr. F. H. Butler, 158 Brompton Natural History, the Jermyn Street Boad, London, S.W. PBEFACE TO THE SECOND EDITION Now that the doctrine of Evolution as applying alike to the Inorganic and the Organic world is universally accepted, the writings of Lyell, who was truly "the Forerunner of Darwin/' acquire a new interest, and are invested with a permanent value. The Origin of Species has been justly asserted by Huxley to be " the logical sequence to the Principles of Geology " ; it has therefore seemed fitting to prefix to a new edition of the present volume a history of the events which led up to the production of Ly ell's epoch-making work, Besides correcting some errors that had crept into the first edition, an endeavour has been made to bring the work up to date, by adding a series of notes, embodying some of the chief additions to our know- ledge made during the fifteen years since the book appeared, that give additional support to the teachings of Lyell. To Mr. F. H. Butler, Mr. J. M. Connor, and other correspondents, who have kindly called my attention to errors or oversights in the first edition, I gladly take this opportunity of expressing my warmest thanks. J. W. J. March 1911. CONTENTS HISTORICAL INTRODUCTION PART I GENERAL PRINCIPLES CHAPTER I GEOLOGY DEFINED HISTORY OF THE DEVELOPMENT OF GEOLOGICAL SCIENCE PAGE Geology compared to History Its relation to other Physical and Natural Sciences Not to be confounded with Cosmogony Opinions of Classical and Mediaeval writers Causes which have retarded the Progress of Geology ... 1 CHAPTER II THE CRUST OF THE GLOBE What geologists mean by the earth's crust Physical characters of the crust of the globe Chemical composition of the solid crust and of its liquid and gaseous envelopes Distribution of tempera- ture in the earth's crust Distribution of pressure in the earth's crust .......... CHAPTER III ROCKS AND THEIR CLASSIFICATION Classification of rocks according to their characters, origin, and age Epigene rocks Aqueous rocks Volcanic rocks Hypogene rocks Plutonic rocks Metamorphic rocks [10] CONTENTS PART II AQUEOUS ROCKS SECTION I. GENERAL BELATIONS OF THE STRATIFIED BOCKS CHAPTEB IV COMPOSITION AND CLASSIFICATION OF AQUEOUS R0CKS PAQB Chemical, mechanical, and organic deposits Arenaceous rocks Argillaceous rocks Calcareous rocks Other varieties of aqueous rocks Phosphatic deposits Ironstones Gypsum Rock salt Carbonaceous deposits Peat Coal Anthracite 23 CHAPTEB V STRUCTURES PRODUCED IN AQUEOUS ROCK-MASSES DURING THEIR DEPOSITION Forms of stratification Original horizontality of strata False- bedding or oblique lamination Irregularities in the accumulation of strata Thinning-out and alteration in the characters of strata Ripple-marks, sun-cracks, footprints, tracks, trails, burrows, and worm-casts ......... , , 34 CHAPTEB VI ARRANGEMENT OF FOSSILS IN STRATA MARINE, FRESHWATER, AND TERRESTRIAL DEPOSITS Slow deposition of strata proved by- fossils Rocks formed of the remains of organisms Diatomaceous deposits Bog-iron ore and lake-ores of Sweden Importance of fossils as indicating the conditions under which strata were deposited Deep-sea deposits Radiolarian deposits, chalk and other limestones Distinction of freshwater from marine formations Genera of freshwater and land shells Rules for recognising marine shells Alternation of marine and freshwater deposits Terrestrial deposits and their fossils Origin of coal and other carbonaceous rocks ... 43 CHAPTEB VII CONSOLIDATION AND SUBSEQUENT ALTERATIONS OF STRATA AND PETRIFICATION OF ORGANIC REMAINS Consolidation of strata Concretionary structures: Jointed structure Mineralisation of organic remains Formation of casts Won.- CONTENTS [11] I'AOB derful preservation of the internal structures of fossil organisms Petrifactions and incrustations Pseudo-fossils .... 64 CHAPTEE VIII ELEVATION OF STRATA ABOVE THE SEA HORIZONTAL AND INCLINED STRATIFICATION FAULTING Why the position of marine strata, above the level of the sea, should be referred to the rising up of the land, not to the going down of the sea Strata of deep-sea and shallow-water origin alternate Also marine and freshwater beds and old land surfaces Vertical, inclined, and folded strata Anticlinal and synclinal curves Dip and strike Structure of the Jura Various forms of outcrop Synclinal strata forming ridges Connection of fracture and flexure of rocks Inverted strata Faults described Superficial signs of the same obliterated by denudation Great faults the result of repeated movements Arrangement and direction of parallel folds of strata Unconformability Overlap Dip-slopes and escarpments Outliers and inliers 75 CHAPTER IX DENUDATION AND ITS EFFECTS Denudation denned Its amount more than equal to the entire mass of stratified deposits in the earth's crust Subaerial denudation Action of the wind Action of running water Alluvium defined Different ages of alluvium Denuding power of rivers affected by rise or fall of land Littoral denudation Inland sea-cliffs Escarpments Submarine denudation Dogger-bank New- foundland bank Denuding power of the ocean during emergence of land . . .102 CHAPTEE X JOINT ACTION OF DENUDATION, UPHEAVAL, AND SUBSIDENCE IN REMODELLING THE EARTH'S CRUST How we obtain an insight, at the surface, of the arrangement of rocks at great depths Why the height of the successive strata in a given region is so disproportionate to their thickness Compu- tation of the average annual amount of subaerial denudation Antagonism of subterranean force to the levelling power of running water How far the transfer of sediment from the land to a neighbouring sea-bottom may affect subterranean movements Supposed permanence of continental and oceanic areas . .116 [12] CONTENTS SECTION II. CHRONOLOGICAL CLASSIFICATION OF AQUEOUS BOCKS CHAPTER XI PRINCIPLES ON WHICH THE CLASSIFICATION OF SEDIMENTARY ROCKS IS BASED PAGE Aqueous, Volcanic, Plutonic, and Metamorphic rocks considered chronologically Terms Primary, Secondary, and Tertiary; Palaeozoic, Mesozoic, and Cainozoic explained On the different ages of aqueous rocks Principal tests of relative age ; superposi- tion, mineral characters, fossils, and included fragments Faunas and floras determined by conditions, geographical ^position, and geological age William Smith's classification of British deposits by their organic remains Danger of extending the paleeontological \AJ method over wide areas Homotaxy ^Combination of physical and palseontological methods-i-Classification of tertiary strata /Tabular view of fossiliferous strata. ) . . . . 126 THE CAINOZOIC (TERTIARY) ERA CHAPTER XII THE PLEISTOCENE PERIOD WITH THE GLACIAL EPISODE Use of the terms ' pleistocene,' ' recent,' and ' human ' periods Mollusca of the Pleistocene period Mammalia of the Pleistocene period Shorter duration of mammalian as compared with mollus- can species Geographical distribution of mammalia in Pleisto- cene times similar to that at present day Remains of ma,n Flint implements Shell-mounds Cavern-deposits Valley gravels High- and low-level gravels Brick-earth Loess Lacustrine deposits Estuarine deposits Marine deposits Sub- divisions of the Pleistocene period Pre-glacial The Glacial period Origin of Boulder clay Glacial lakes and other pheno- mena of glaciated districts Post-glacial, Pluvial and Champlain periods Palaeolithic and Neolithic Copper, Bronze, and Iron 147 CHAPTER XIII THE NEWER TERTIARY STRATA (NEOGENE OR NEOCENE) Use of the terms Miocene and Pliocene, Neogene and Neocene Mollusca of the Newer-Tertiary Strata Mammalia of the Newer CONTENTS [13] PAGE Tertiaries The Newer Tertiary Flora British Newer and Older Pliocene Strata Forest-bed of Cromer Chillesford and Aldeby b e( j s Ked Crag "White or ' Coralline ' Crag Older Pliocene deposits of the North Downs and of St. Erth Relation of the Fauna of the Crag to that of the present day Proofs of denuda- tion between the periods of deposition of the British Older and Newer Tertiaries 171 CHAPTEE XIV THE OLDER TERTIARY (EOGENE OR EOCENE) Geographical Distribution of the Older-Tertiary Strata The London and Hampshire basins Foraminifera, corals, echinodermata, and crustaceans of the Older Tertiariei The Older Tertiary Mollusca The fish, reptiles, birds, and mammals of the period The Older Tertiary flora. The British Older- Tertiary Strata. Hempstead Beds The Bembridge Series The Headon Series- The Brocken- hurst Marine Group The Barton Sands and Clay The Brack- lesham Series The Bournemouth Beds The Plant-beds of Bovey Tracey and Mull The London Clay The Oldhaven beds and Woolwich arid Reading Series The Thanet Sands . . 191 CHAPTEE XV FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE CAINOZOIC OF THE BRITISH ISLES Tertiaries of France and Belgium Montian Argile plastique Calcaire grossier Gypsum of Montmartre Mammals of Oligocene of Northern and Central France Faluns of Touraine and Bor- deaux Pliocene of Northern France and Belgium Tertiaries of Central Europe Lower Brown Coal and Amber deposits* - Mayence Basin Pliocene of Eppelsheim Tertiaries of Alps and Switzerland Flysch and Nummulitic formations Lower, Middle, and Upper Molasse Plants and insects of Oeningen Tertiaries of Italy Oligocene and Miocene Subapennine strata Newer Pliocene of Sicily and the Val d'Arno Tertiaries of Eastern Europe Oligocene of Croatia Miocene (Leithakalk and Sarma- tian) of Vienna basin Pliocene (Congeria) strata Tertiaries of India Sind and Sivalik strata Post-pliocene deposits of Northern Europe and the Alps Scandinavia and Jlussia Central Europe Alps and Jura Older and Newer Palaeolithic periods Lake-dwellings Post-pliocene of India, New Zealand, and Australia Tertiaries of North America Eocene and Neo- cene of Eastern States Mammals and Plants of Tertiaries of the Western Territories American Post-pliocene deposits Glacial and Champlain periods Tertiary Zones in Europe . . . 219 [14] CONTENTS THE MESOZOIC (SECONDARY) ERA CHAPTER XVI THE CRETACEOUS SYSTEM PAGE Lapse of time between Eocene and Cretaceous Periods Classification of Cretaceous strata Foraminif era, Sponges, Corals, Bryozoa, and Mollusca of the Cretaceous Period Terrestrial Floras of the Cretaceous Keptiles, Birds, and Mammals of the Cretaceous- Chalk and Flint Zones of the Chalk with their fossils Chalk Marl Upper Greensand Gault Upper Neocomian Atherfield Clay Middle Neocomian Tealby Series Lower Neocomian Speeton Clay Spilsby Sands The Wealden and Hastings Sands Punfieldbeds ...... ... 248 CHAPTER XVII THE JURASSIC SYSTEM Classification of Jurassic strata Foraminifera, Sponges, Corals, Echinodermata, Brachiopoda Lamellibranchiata, Gastropoda, and Cephalopoda of Jurassic rocks Fishes, reptiles, birds, and mammals of the Jurassic rocks Terrestrial Flora of the Jurassic period Purbeck strata Purbeck mammals Dirt-beds Port- landian Kimeridge Clay Coralline Oolite Oxford Clay Cornbrash Forest Marble Great Oolite Stonesfield Slate with its Mammalia Inferior Oolite Upper Lias sand and clay Marlstoue and Middle Lias Lower Lias Ehsetic beds . 27S CHAPTER XVIII THE TRIASSIC SYSTEM Subdivisions of the Trias in England Corals, Echinodermata, Brachiopoda, Lamellibranchiata, Gastropoda, and Cephalopoda of the Trias Fish, Amphibians, and Keptiles Terrestrial Flora of the Trias Triassic Mammalia The Keuper and its Reptilia The Dolomitic Conglomerate Elgin Sandstones The Bunter Formation of Eed Sandstones and Clays Rock-salt, Gypsum, &c . 310 CONTENTS [15] CHAPTEE XIX FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE MESOZOIC STRATA OF THE BRITISH ISLES PAGE Secondary Strata of Central Europe Keuper, Muschelkalk, and Bunter The Black, Brown, and White Jura Planer and Quader Beds Chalk of Maestricht and Faxoe Freshwater Strata "Wealden of Hanover Strata of Aix-la-Chapelle Secondary Strata of the Alpine Eegions Hallstadt and St. Cassian Beds Alpine Jurassic, Tithonian, and Neocomian Hippurite Lime- stones Secondary Strata of Russia, India, and South Africa- Secondary Strata of North America Newark Formation Strata of the Eastern States and of the Western Territories . . . 823 THE NEWER PALEOZOIC ERA CHAPTER XX THE PERMIAN SYSTEM Subdivisions of the Permian Permian Marine Fauna of India, Texas, Russia, &c. Foraminifera Corals Brachiopoda Am- monites and other Cephalopoda Arthropods Fish, Amphibians, and Reptiles Terrestrial Flora Relations with Carboniferous and Trias respectively Upper Permian Magnesian Limestone and Marlslate, with their Fossils Lower Permian, with its breccias . 838 CHAPTER XXI THE CARBONIFEROUS SYSTEM Succession of Strata in the Carboniferous System Carboniferous Foraminifera and Corals, Echinodermata, Brachiopoda, Lamelli- branchiata, Gastropoda and Cephalopoda of the Period Carboni- ferous Fishes and Amphibians The Carboniferous Flora Peculiarity in Mode of Growth of the Cryptogams of the Period Ferns, Calamites, Lepidodendra, &c. Land-shells and Insects of the Carboniferous Period Carboniferous Strata of Britain- Coal-measures Millstone Grit Carboniferous Limestone and Yoredale Series Tuedian Series Scottish Carboniferous Calciferous Sandstone Series Mode of Formation of the Carbo- niferous Strata Coal-seams, Ironstones, &c. Marine and Fresh- water Strata of Carboniferous ....... 848 [16] CONTENTS CHAPTER XXII THE DEVONIAN SYSTEM PAGE Relations of the Devonian Devonian Corals, Brachiopoda, Cepha- lopoda, and Trilobites The Devonian Fish and their Relation- ships to Living Forms The Devonian Flora and its Relation to that of the Carboniferous Devonian Strata of Devon and Corn- wall Upper, Middle, and Lower Devonian Old Red Sandstone Relations to Devonian Proof of Freshwater Origin Old Red Sandstone of Scotland, Lower, Middle, and Upper Old Red Sandstone of England and Wales Old Red Sandstone of Ireland 374 CHAPTER XXIII FOREIGN DEPOSITS. WHICH ARE HOMOTAXIAL WITH THE NEWER PALAEOZOIC STRATA OF THE BRITISH ISLES The Devonian rocks of the Eifel of the Ardennes, and Brittany of the Carinthian Alps, the Iberian peninsula and Russia Carbo- niferous strata of France, Germany, and Russia- -Permian strata of Central Germany, the Alps, and the Ural Mountains Devonian strata of the United States, Canada, and the Arctic Regions Carboniferous strata of the United States Permian strata of Texas and Nebraska Devonian, Carboniferous, and Permian of India and Australia 891 THE OLDER PALEOZOIC ERA CHAPTER XXIV THE SILURIAN SYSTEM Classification of Silurian rocks Characteristics of the Marine Flora and Fauna of the Silurian Graptolites Corals Echinodermata Brachiopoda Gastropoda Cephalopoda Fish British representatives Shropshire North Wales Lake District Scotland Details of strata in the typical area Upper Ludlow Lower Ludlow Aymestry Limestone Oldest known fossil fish Wenlock Limestone Wenlock Shale Woolhope Limestone Tarannon Shales and Denbighshire Grits Upper and Lower Llandovery rocks May-Hill beds 897 CONTENTS [17] CHAPTER XXV THE ORDOVICIAN SYSTEM PAGE Classification of Ordovician strata Characteristics of the marine Fauna Foraminifera Grraptolites Echinodermata Brachio- poda Gastropoda Cephalopoda "Worms Trilobites and their organisation Bala or Caradoc strata Llandeilo beds Arenig beds or Stiper-stones Ordovician strata of the Lake District Ordovician strata of Scotland 410 CHAPTER XXVI THE CAMBRIAN SYSTEM Divisions of the Cambrian System Cambrian Flora and Fauna* Sponges Graptolites Echinodermata Brachiopoda Mollusca Annelida Trilobita The oldest known fossils of the Lower Cambrian Period Upper Cambrian, Tremadoc slates and Lingula Flags Middle Cambrian, Menevian beds, Harlech grits, and Llanberis slates Lower Cambrian, Comley Sandstone, Cambrian of Scotland, Durness Limestone, Girvan Limestone 'Fucoid* and Olenellus beds 419 CHAPTER XXVII FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE OLDER PALEOZOIC STRATA OF THE BRITISH ISLES Older Palaeozoic strata less altered than the British Basin of Bohemia ' Primordial' strata of Barrande Stages D, E, F Older Palaeozoic strata of Scandinavia Olenellus beds Alum Shales Limestones and Schists Russia and other parts of Europe North America Geographical distribution of life forms in Cambrian times Table showing equivalence of strata in different aieas 429 CHAPTER XXVIII SEDIMENTARY ROCKS OF PRE- CAMBRIAN AGE Existence of stratified and other Rock-masses underlying the Older Palaeozoic Deposits Rocks of both Igneous and Aqueous Origin Obscure Traces of Fossils Thickness and Extent of Pre- Cambrian Rocks Pre-Cambrian strata of the British Isles Pebidian Arvonian Dimetian Fundamental Gneiss, or Lewi- a [181 CONTENTS sian Caledonian, or Dalradian Malvernian Mqnian Urico- nian Longmyndian The Torridon Sandstone, or Torridenian Pre-Cambrian of Europe and North America Huronian Laurentian, Upper and Lower Algonkian and Archaean Pre- Cambrian of India Traces of Fossils in pre-Cambrian Rocks . 432 CHAPTEK XXIX GENERAL REVIEW OF THE SUCCESSION AND CHARACTERS OF THE SEDIMENTARY ROCKS Fossils not found uniformly distributed in Sedimentary Formations Imperfection of our Knowledge of Freshwater and Terrestrial Conditions during past Geological Times Existence of Organisms before Cambrian Times Illustrations of the great Imperfection of the Geological Record ' Time-ratios ' of the Geological Eras Date of Appearance of different Forms of Life as modified by new Discoveries of Fossils General Order in which Life-forms have appeared upon the Earth Groups of Animals and Plants which have predominated in successive Periods Synthetic Types Specialised Types Persistent Types Summary of Palaeonto- logical History Table of Fossiliferous Sedimentary Formations 439 PART III VOLCANIC ROCKS CHAPTER XXX VOLCANIC ROCKS, THEIR NATURE AND COMPOSITION Relation of volcanic Rocks to the sedimentary and hypogene Rocks Nature of Action taking place at Volcanic vents Lavas and thsir Varieties Fragmental materials ejected from Volcanoes Scoriae, lapilli, dust, pumice, bombs Formation of volcanic Tuffs Alteration of volcanic Rocks by solfataric and atmospheric agencies Chemical composition of lavas Acid, intermediate and basic lavas Rhyolites and Soda-rhyolites Andesites, Trachytes, Phonolites, and Tephrites Alteration of Andesites Propylites and Porphyrites Basalt and Melaphyres Tachylytes and Vario- lites Basaltic and Palagonite Tuffs 455 CONTENTS [19] CHAPTER XXXI ORIGIN AND STRUCTURE OF VOLCANIC ROCK-MASSES PAGE Explosive and effusive action of Volcanoes Origin of Volcanic Cones Internal structure of Volcanic Cones Origin of Volcanic Craters Formation of Volcanic Dykes Varieties of Volcanic Dykes Alteration of Eocks on the sides of Volcanic Dykes Contact Metamorphism Alteration of Sandstone, Shale, Lime- stone, and Coal Interbedded and contemporaneous Volcanic Rocks contrasted with intrusive or subsequent masses Columnar and globular structures in Lavas ....... 466 CHAPTER XXXII PRINCIPLES ON WHICH THE CHRONOLOGICAL CLASSIFICATION OF VOLCANIC ROCKS IS BASED Variations of mineral character in the volcanic rocks of different periods Not essential, but due to alteration Age of lava flows and intrusions Tests to be applied to determine relative age of \/ t volcanic masses Sources of error in drawing inference sWGreat value of fossils when found Test by included fragments-j-Order in which volcanic rocks have been erupted Views of Bunsen, Durocher, Richthofen, and later authors . . . . . 483 CHAPTER XXXIII VOLCANIC ROCKS OF CAINOZOIC AGE The latest exhibitions of volcanic energy in the British Islands The thermal springs of Bath, &c. Tertiary volcanoes of the west of Scotland and the north of Ireland First period of eruption Second period of eruption Third period of eruption Tertiai'y volcanic rocks of other parts of Europe Vesuvius Auvergne Newer Pliocene volcanoes of Italy Older Pliocene volcanoes of Italy and the Eifel Oligocene and Miocene volcanoes of Auvergne and the Eifel Eocene volcanic rocks of Monte Bolca Tertiary volcanic rocks of the Atlantic Islands of India of the United States and Australia .,..,.,.. 488 CHAPTER XXXIV VOLCANIC ROCKS OF THE MESOZOIC, PALAEOZOIC, AND ARCH^AN ERAS Absence of evidence of volcanic action in Cretaceous and Jurassic times in the British Isles and Western Europe Triassic Volcanoes of Devonshire Permian Volcanoes of Scotland Volcanoes of the Carboniferous Period Buried trees of Arran Volcanoes of the Old Red Sandstone and Devonian Period Volcanoes of the [20] CONTENTS PAGE Silurian, Ordovician, and Cambrian Periods Pre-Cambrian Volcanoes Pre-Tertiary Volcanoes of other parts of the globe Cretaceous and Jurassic volcanic rocks of Greece Newer and Older Palaeozoic Volcanoes of Central Europe Pre-Cambrian volcanic rocks of Canada .... ... 502 PART IV PLUTONIC ROCKS CHAPTEE XXXV PLUTONIC ROCKS, THEIR NATURE AND COMPOSITION Analogy of the Plutonic Rocks to those of volcanic origin Proofs of the deep-seated origin of Plutonic Rocks Chemical composition of the different classes of Plutonic Rocks Changes which they undergo Liquid cavities in the crystals of Plutonic Rocks Order in which the seveial minerals crystallise in Plutonic Rocks Granite and its varieties Syenites, &c. Diorites, &c. Nepheline syenites and Theralites Gabbro and its varieties Ultra-acid Rocks Ultra-basic Rocks Peridotites Pyroxenites Amphibo- lites Relations of the Ultra-basic Rocks to Meteorites . . 509 CHAPTER XXXVI STRUCTURE AND ORIGIN OF PLUTONIC ROCK-MASSES : THEIR RELATIONS TO ROCKS OF VOLCANIC AND SEDIMENTARY ORIGIN Plutonic Rocks can only be exposed at the earth's surface by denudation Latest formed Rocks of this class never seen at *he surface Relations of Plutonic masses to Volcanic extrusions Examples from the Western Isles of Scotland and Antrim Examples in other areas -Features exhibited by Plutonic Rock- masses Forms produced by weathering Voiiis and Dykes- Segregation Veins Result of segregative action in Plutonic Rock-masses Inclusions and Veins Differentiation in Igneous Magmas and its results ......... 520 CHAPTEE XXXVII PLUTONIC ROCKS BELONGING TO DIFFERENT GEOLOGICAL PERIODS Plutonic Rocks were formed during the whole of the geological periods Those of the most recent period seldom exposed at the CONTENTS [21] PAGE surface by denudation Test of the geological age of Plutonic Hock-masses Kelative position Intrusion and Alteration- Mineral composition Included fragments Tertiary Plutonic Kocks of Western Scotland, North-East Ireland, Elba, &c. Difficulty of determining the age of Plutonic Rock-masses in Mountain chains Plutonic Rocks of the Cretaceous the Jurassic the Carboniferous the Ordovician and Pre-Cambrian Periods 528 PART V METAMORPHIC ROCKS CHAPTER XXXVIII METAMORPHIC ROCKS, THEIR NATURE AND ORIGIN Contact Metamorphism and Regional Metamorphism Thermo- metamorphism and Hydrothermal action Dynamo-metamor- phism Different results of Metamorphic action Researches of Daubree and others on Thermo-metamorphic and Hydrothermal action Dynamo-metamorphic action and its results Slaty cleavage Its nature and origin Investigations of Phillips, Sharpe, Sorby, &c. Experimental proofs of origin of slaty cleavage Foliation, its nature and origin Relations between Cleavage and Foliation Experimental researches of Daubree, Spring, and others upon the action of pressure in producing Metamorphism .......... 537 CHAPTER XXXIX CONTACT METAMORPHISM AND REGIONAL METAMORPHISM I THE VARIETIES OF ROCKS RESULTING FROM THESE TWO KINDS OF ACTION Qlustrations of the action of Contact Metamorphism Distance to which Contact Metamorphism can be traced from the intrusive mass Minerals produced by Contact Metamorphism Chief types of Rocks produced by Contact Metamorphism Andalusite, Kyanite, Sillimanite, Staurolite Rocks, J V logical evidence Attempt to determine Time Ratios for the several geological periods Attempts to measure geological periods in years The doctrines of Catastrophism, Evolution, and Uniformitarianism Evidences in favour of Uniformitarianism during the periods covered by the geological history The Science of Geology limited to the study of the crust of the Globe Bearing of speculations concerning the earth's interior on the Science of Geology The Geological Record may only cover a small portion of the history of the Globe during past times . , 589 NOTES 601 APPENDIX A. THE COMMON ROCK-FORMING MINERALS . . 611 APPENDIX B. CLASSIFICATION OF PLANTS, LIVING AND FOSSIL 617 APPENDIX C. CLASSIFICATION OF ANIMALS, LIVING AND FOSSIL 618 INDEX . 623 HISTORICAL INTRODUCTION ' THE SCIENCE OF GEOLOGY is ENOEMOUSLY INDEBTED TO LYELL MORE so, AS I BELIEVE, THAN TO ANY OTHER MAN WHO EVER LIVED.' ' SIR CHARLES LYELL'S GRAND WORK ON THE PRINCIPLES OF GEOLOGY . . . THE FUTURE HISTORIAN WILL RECOGNISE AS HAVING PRODUCED A REVOLUTION IN NATURAL SCIENCE.' IN the first of these two passages, Charles Darwin writing only a few months before he passed away summed up his oft-expressed opinions concerning the value and importance of Lyell's geological labours ; while in the second passage he has placed on record, in his * Origin of Species,' his no less often repeated conviction that the great work of his own life the establishment of the doctrine of Evolution in the Organic world was prepared for by the teacher and friend, who had demonstrated the truth of the same principle in its application to the Inorganic world. All the other great pioneers of Evolution Wallace, Hooker, Huxley, and Haeckel have, with equal warmth, acknowledged their indebtedness to Lyell and his great work, the ' Principles of Geology ' as Huxley has well expressed it, ' Lyell was for others, as for me, the chief agent in smoothing the road for Darwin.' Dr. Francis Darwin has rendered an inestimable service to the student of the history of science by publishing extracts from his father's early notebooks and diaries, and, with the assistance of Professor Seward, of his corre- T26T HISTORICAL INTRODUCTION fc l ^, ~ J - 4. " V SponcTenCe *vtfith scientific friends ; most of all are we in- debted to him for rescuing from oblivion the two early drafts of his father's great work and publishing them in a notable volume, ' The Foundations of the Origin of Species.' It is now possible, by the aid of materials sup- plied from these different sources, not only to trace the development of Darwin's ideas on the Evolution question, but to realise his various fluctuations of opinion on the subject, at different periods of his life. It would be equally interesting to follow the mental processes by which Lyell was led to the important con- clusions enunciated by him in the ' Principles of Geology.' If this be not possible, we have nevertheless in the two volumes of letters and journals, published by Mrs. Katherine Lyell in 1881, many valuable materials furnished to us, which enable us to arrive at definite conclusions as to the influences which were at work on LyelPs mind influences that in the end brought about that bold revolt from the doctrines so stoutly maintained by his teachers and con- temporaries. These materials I have been able to sup- plement from other sources, and especially from frequent conversations with Lyell himself, during the later years of his life. It is the more necessary that this historical retrospect should be undertaken, seeing that much misconception has prevailed upon the subject. Again and again has it been assumed that Lyell's task was little more thar that of making known and illustrating the teachings of the illus- trious James Hutton, as contained in the justly celebrated ' Theory of the Earth ' ; and that to the younger author, who was born in the year (1797) in which his distinguished predecessor died, but little credit is due for originality. The truth is that Lyell, as we shall show in the sequel, owed very little to Hutton : his conclusions were arrived at quite independently from those of the great Scottish philosopher; they were based on different premises, and reached by very different trains of reasoning. The prevalent erroneous views on this subject are, no HISTORICAL INTRODUCTION [27] doubt, to a great extent due to the influence of the valuable and important contributions to the history of geological science which were made by Dr. Fitton. That able writer, unfortunately, seems never to have realised the truth of the fact that scientific discoveries are often made by different workers, independently of one another, and sometimes almost simultaneously. The truth of this has been very strikingly illustrated, in more recent years, in the case of the recognition of the principle of 'Natural Selection/ which was arrived at quite independently by Darwin and Wallace, after being adumbrated by earlier writers like Wells and Matthew. Fitton has placed side by side passages from the writings of the learned and eloquent Carmelite friar, Generelli, written in 1749, and from the great work of Hutton, which appeared in 1795, and has argued that the latter author must be regarded as having been indebted to the former for his splendid generalisations. In the same way, he has insisted that Hutton must be credited with having inspired the views so admirably expounded by Lyell in the ' Prin- ciples of Geology.' But it is extremely improbable that Hutton had read, or even heard of, the writings of Generelli ; while, as I shall show, Lyell when he first arrived at his anti-catastrophic opinions had certainly never read the ' Theory of the Earth.' In studying the history of science, we are constantly impressed by the conviction that all great discoveries are gradually ' led up to ' by many small advances in know- ledge ; and that even the final step which, at a distance, appears to have been a very great one has really been taken independently by a number of thinkers, often ap- proaching the subject from somewhat different points of view. The most important service rendered to science, however, is not in accomplishing this last step, but in realising the new views so thoroughly and expressing them so forcefully that fellow-workers become impressed and convinced ; and thus the stage reached becomes a starting-point for fresh advances. This was the great [28] HISTORICAL INTRODUCTION service which was rendered to natural science by the labours of Lyell. Charles Lyell, the son of a Scottish father and an English mother, was born in Forfarshire, but, having been taken by his parents to reside in the New Forest of Hampshire, when only a few months old, he was educated amid South- country surroundings, in English schools and at the Uni- versity of Oxford ; and it was in England that nearly the whole of Lyell's active life, not spent in travel, was passed. His father was a man of cultivated literary tastes a trans- lator and commentator of Dante and had also some scientific distinction as a botanist, doing useful work in the study of the cryptogams ; and this led to his becoming the friend and correspondent of the Hookers and other natura- lists. Lyell's mother appears to have been a woman of great force of character : it was to her determination to remove her husband and family from the influences sur- rounding the Forfarshire home, where all the 'lairds' of that time were addicted to the gross intemperance so painfully depicted by Dean Ramsay, that the abandonment of the ancestral home of the family and the flight to Eng- land events which had such a marked influence on her distinguished son's career must be ascribed. While a schoolboy at Salisbury, Lyell, at the age of ten, first had his attention attracted to minerals by seeing quartz-crystals and chalcedony exposed in flints, which broke up on being rolled, by himself and his playfellows, down the walls of Old Sarum. About the same time, he found in his father's library a number of well-illustrated books on Entomology, and, like Darwin and Wallace, he soon became an ardent and enthusiastic collector of insects. Of their structure he knew little, but from their external characters he framed a rough classification for his own use, and he developed considerable skill in detecting the minute differences which distinguish British from continental speci- mens. It was, however, the habits, and especially the dis- tribution of the different species and varieties, that chiefly excited hig interest, and in this way the characteristics HISTORICAL INTRODUCTION [29] and bent of his mind were betrayed. In spite of the claims of geology in after-years, Lyell always maintained his love of Entomology ; and I well remember how in the last year of his life he showed me with delight the collections which his devoted sisters had so carefully preserved, dwelling lovingly on specimens which he regarded with especial interest. To realise the nature and magnitude of the task upon which Lyell entered in writing the ' Principles of Geology,' it is necessary to bear in mind the condition of geological science in the early years of last century. At that time almost all students of the Science had ranged themselves in one or other of two opposing camps that of the Neptunians (Wernerians) or that of the Vulcanists or Plutonians (Huttonians). It is hard to say which of these two' rival sects for such they must be regarded was most extravagant in maintaining its special tenets : those who advocated the exclusive claims of water, declaring basalt to be an aqueous precipitation and volcanic eruptions only the result of the spontaneous ignition of coal-seams, or those who saw everywhere nothing but the effect of heat, insisting that chalk-flints, and even beds of rock-salt, were formed by igneous fusion, and that organic remains had become mineralised by the same agency. Although Hutton had combined with his wild theories of igneous action a very profound and beautiful system of cosmology, while Werner's teachings on the subject were of the most crude and unsatisfactory character, yet the speculations of the Saxon professor almost everywhere found favour among geologists, while the splendid theory of the Scottish philo- sopher was treated with ridicule or neglect. Dr. Fitton, writing of what was taking place in the geological world at that date, very justly states the con- dition of affairs as follows : * Without going so far as to say that in this con- troversy (Huttonian v. Wernerian) all the practical know- ledge was on one side and all the sound philosophy on the [30] HISTORICAL INTRODUCTION other, it is no exaggeration to assert that the Wernerians had very much the advantage over most of their opponents in their acquaintance with the characters and relations of rocks.' While Werner's teachings were proclaimed all over Europe, with missionary zeal, by his enthusiastic students, Hutton's doctrines were accepted only by a few faithful friends and disciples, while to the general public they were known only through the misrepresentations of his bitter adversaries. Even in his own land of Scotland, a * Wernerian ' Society was formed to propagate the system of the great Saxon mineralogist and to oppose the Huttonian ' heresies/ On the Continent, largely owing to the interrup- tion of international intercourse through the war, Hutton's great work remained almost unknown, and this was true, though to a less extent, of the eloquent ' Illustrations ' of Playfair, Hutton's friend and commentator, who so stoutly defended his master's views. Fitton, writing in 1839, says that * the original work of Hutton (in two volumes) is in fact so scarce that no very great number of our readers can have seen it. No copy exists at present in the libraries of the Royal Society ; the Linnaean, or even the Geological Society of London ! ' But the greatest bar to the dissemination of Hutton's philosophical views was undoubtedly the theological opposition, which their author had, innocently but most unfortunately, excited. All previous cosmologica 1 hypo- theses published in this country ' Sacred Theories of the Earth ' they used to be called were either directly based on the Mosaic accounts of the Creation and the Flood, or at least professed to be in perfect harmony with the early chapters of Genesis. Hutton not only abstained from referring to early traditions, but, in his enthusiasm for the beautifully compensated system of nature, which he so clearly perceived and expounded, declared that he failed to recognise in it either ' trace of a beginning or any sign of an end.' This gave great umbrage to his detractors, who, most unjustly, asserted that the phrases used by him were HISTORICAL INTRODtJCtlON [3l] tantamount to the denial of the fact of Creation, and even of the existence of a Creator. It was in this way that Neptunism or Wernerianism came to be connected with that system of theoretical geology which demanded the repeated destruction of the face of the globe, with all its inhabitants, to be followed by a re-creation, after every cataclysm, of all plants and animals. For this geological system Whewell suggested the name of ' Catastro- phism ' ; it became completely identified with the teachings of Werner (many of which were of great value), and in the end was regarded as the only theory conformable to the teaching of Scripture, its adherents receiving the crown of Orthodoxy. On the other hand, the much distorted and vilified theory of the gentle Scottish philosopher was branded as infidel and even atheistical. LyelPs early life was passed at the period when, in this country at least, all geological study was suspect, while Hutton's teachings were roundly anathematised ; it was a fortunate circumstance, therefore, that the work by which he was first introduced to the science to which he was to devote his future life, was one singularly free from the violent partizanship and prejudice which characterised most of the writings on the subject at that period. Kobert Bakewell was, like his great contemporary, William Smith, an agricultural and mining agent. In the year 1813 he issued his ' Introduction to Geology,' a work which passed through five editions in this country, three in America, and was translated into German a work, we may add, the great merits of which have not been so fully recognised by later authors as they certainly deserve to be. Dr. Karl von Zittel has very justly stated concerning this book that, ' while following Werner in the general treatment of the subject, Bakewell took up a neutral attitude on most contested points and showed a just appreciation of Hutton's views.' It was this book which, shortly after its appearance, Lyell, then seventeen years of age, found in his father's library, and he became greatly excited by its contents. [32] HISTORICAL INTRODUCTION If Ly ell's first introduction to geological studies was thus singularly free from the distorting prejudices which at the time enveloped the whole subject, the same can certainly not be said of the influences under which in following years his mind successively came. Immediately after reading Bakewell, he went to Oxford, and his zeal for geology was at once found fresh fuel in the lectures of the versatile and eccentric Dr. Buckland, whose devoted pupil he became, accompanying him on many geological excursions in all parts of the country. Buckland was the most bitter and determined opponent of the Huttonian doctrines. As his ' Vindicise Geologicse ' shows, he sought to make all his geological teachings fit in with the Mosaic account of creation ; while in his ' Reliquiae Diluvianse ' he endeavoured to show that the whole of the enormous superficial deposits of the globe are to be ac- counted for by Noah's flood. Whewell has given an amusing picture of the great Oxford professor riding forth ' with a cavalcade of forty horsemen ' and eloquently holding forth concerning ' the impotence of modern causes.' Nor is there wanting evidence that for a time Lyell was carried away by this teaching, for we find the young under- graduate recording in his journal that a deep narrow glen, at the time without a river in it, seemed to him to be a fact ' unfortunate for the Huttonians ! ' While still a student at Oxford, Lyell accompanied his father to the Continent, and from that time forward made frequent and often prolonged visits abroad. At Paris he was introduced to Cuvier, who showed him much kindness, and he was made free of the brilliant scientific society of the French capital at that day the Humboldts, Brong- niarts and others. Lyell was greatly interested in the results of Cuvier's splendid researches ; but in Paris, as at Oxford, he found geological history regarded as a succession of great cataclysms, for none could more stoutly maintain than did Cuvier and his disciples that each great geological era had been brought to a close by a tremendous convulsion of which the Noachian deluge was the last and that HISTORICAL INTRODUCTION [33] after the complete destruction of all plants and animals, which occurred at each of these paroxysms, a fresh creation of entirely new organisms took place. Upon leaving Oxford and beginning to prepare for his proposed career as a barrister, Lyell, at the age of twenty, joined the Geological Society, then in its infancy. There he found the cataclysmal doctrines of Buckland and Cuvier all prevalent. Besides Buckland himself, a number of other clergymen Sedgwick, Conybeare, Henslow, and others were leaders of the science, and all of them honestly, however mistakenly, were convinced that the Huttonian views were not only absurd but impious. The founder of the Society, Greenough, was a fanatical Wernerian ; the Society, as Buckland said, had ' a very landed manner, the professors of Oxford and Cambridge being admitted only on sufferance ' ; while men of wider views, like Bakewell, Farey, and William Smith, were never welcomed to its select circle. It is only right to remember, however, that there was at this time one active worker in the Geological Society who was to a great extent free from the prejudices of his contemporaries. John Macculloch, who may justly be regarded as { the Father of British Petrography,' although he did not proclaim himself a follower of Hutton, really did more than any geologist of that time to illustrate and supplement the conclusions of the author of the ' Theory of the Earth.' He not only produced that meritorious work, the first geological map of Scotland, but in a number of books and memoirs added very largely to our knowledge of the structure of that country. Although Macculloch's work was to some extent marred through his unfortunate peculiarities and defects of character, Lyell was able to assert of him, ' I may acknowledge with gratitude that I have received more instruction from his labours in geology than from those of any living writer.' Such being the influences operating on the mind of Lyell at Oxford, Paris, and in London, at the Geological Society to what, we may ask, must we ascribe that revolu- b [34] HISTORICAL INTRODUCTION tion in thought which enabled him to emancipate himself from the teachings of such men as Buckland and Olivier, and, in the end, to proclaim a revolt against the cherished ideas of nearly all his fellow- workers at the Geological Society, in the affairs of which he soon began to take a very prominent part ? While still an undergraduate at Oxford, at the age of twenty, Lyell paid a visit to his father's friend, Dawson Turner, at Yarmouth. There he met the young botanist Dr. Joseph Arnold, a man of great originality of mind, whose scientific career of great promise was, only a year later, to be cut off by a fatal attack of fever, but not before he had made himself famous by the discovery of that wonderful plant the Rafflesia Arnoldi. Arnold had made a collection of East Anglian fossils, which proved an irre- sistible attraction to Lyell ; and soon the two young men found occupation for their similar tastes in speculating on the origin and the changes that had taken place in the delta of the River Yare, extending their discussions to the alterations going on along the whole East- Anglian coast- line, and even to the question of the mode of separation of the British Isles from the Continent of Europe. During another of his geological excursions, while still an Oxford student, Lyell had the good fortune to make the acquaintance of Dr. Mantell, whose researches have added so much to our knowledge of the geology of the south-east of England. Mantell, who was at that time practising as a medical man at Lewes, was able to in- corporate in his ' Geology of the Isle of Wight ' many observations made by young Lyell, in this and following years, and the two friends devoted much attention to the Wealden and overlying formations. But while studying these older formations, LyelFs attention was largely devoted to the study of changes taking place in the country during recent times, of which he found very striking evidence at Romney Marsh and in the pebble beaches of Dungeness. The fact of Lyell's early studies being so largely devoted to the study of marine action, may account for what was HISTORICAL INTRODUCTION [35] Certainly the greatest defect in his views as first published. Unlike Hutton, who found, in the sub-aerial waste of the precipices and crags of his own mountain-land^ the evidence of constant destruction and change, I/yell's first proofs of the potency of existing forces were derived from the study of the coast-lines of the southern half of our island. It Was long before he fully realised how much must be attributed also to the work of rain and rivers, acting over the whole surface of the country. But at a very early date there was another, and much more powerful, influence at work on I/yell's mind which gradually weaned him more and more from the teachings of Buckland and Cuvier. On the occasional visits which he paid to his father's estate the ancestral home of Kin- nordy, in Forfarshire he met with a set of phenomena which produced a deep impression on his mind. The Vale of Strathmore he found to be covered with great masses of boulder clay, on the irregular surface of which were a number of small lakes, many of which had been drained. In most of these lakes, deposits consisting of peat and 4 marl ' had accumulated ; the calcareous beds, which sometimes attained a thickness of sixteen to twenty feet, being at that time extensively dug for the purpose of * marling ' the land. Lyell appears to have been greatly struck by the magnitude, and especially by the crystalline character of these calcareous beds, which sometimes contained the remains of insects and other organisms. Now, Buckland had always impressed on his pupils the fact that the whole of these boulder-clay and superin- cumbent deposits had been suddenly formed by Noah's deluge about four thousand years ago. And Cuvier had as strongly maintained that all calcareous deposits of recent age differ in many essential respects from the older limestones of the earth's crust. But Lyell, taking ad- vantage of the numerous excavations going on around his birthplace, carried on a series of careful studies during seven years, which in the end led him to recognise the [36] HISTORICAL INTRODUCTION fact that the views of Buckland and Cuvier were altogether untenable. In the first place, with the aid of Robert Brown the botanist, and his lifelong friend Michael Faraday, he was able to convince himself that the formation of these thick calcareous deposits must be ascribed to the action of plants (Chara), which possess the power of separating the calcium salts from their state of solution in water and of incorporating them in their stems and seed-vessels. And, in the second place, he was able to show, by chemical analyses made for him by Dr. Daubeny, that the actual amount of calcium salts present in the waters of these lakes was almost infimtesimally small, and therefore that the deposit of the thick beds of limestone must have taken an enormous time 'to accomplish. On the other hand, he proved that these ' marls ' had undoubtedly acquired in some cases the solid and crystalline character which Cuvier had maintained to be characteristic of and found only in the older limestones of the globe. This important work of Lyell was communicated to the Geological Society on two evenings at the end of 1824 and the beginning of 1825, when his paper was read and discussed. Thus, while Hutton and Lyell arrived at very similar conclusions concerning the reality and great effects of changes going on upon the earth's surface, the lines of reasoning on which they founded those conclusions were very different. Not only was Lyell more impressed, as we have seen, by the destructive agencies of the sea than by those of the atmospheric waste due to rain and rivers, which his great predecessor so clearly recognised, but he was even more influenced by the study of the slow and almost imperceptible, yet very real, result of organic agencies in separating material from a state of solution in the waters of the globe. It may be right to mention, however, that while studying this wonderful process upon his father's land, Lyell also sought to estimate the rate of the erosive action going on in the same district, by having his initials, with the date, carved on rocks in HISTORICAL INTRODUCTION [37] the beds of some of the streams as a basis for estimation in later years. In France, Lyell seems never to have met Cuvier's great rival, Jean Lamarck, then old and blind, or even to have heard of his works. Had he done so, he would have found in the ' Hydrogeologie,' published in 1801, views propounded strikingly similar to those of Hutton, though arrived at quite independently from those of the Scotch philosopher. In Constant Prevost, however, Lyell found a friend who had the courage to oppose some of Cuvier's catastrophic views, and Lyell and Prevost visited and studied together many districts in France and Great Britain. It was from Germany, however, that Lyell was destined to receive the most important aid of all. in investigating the changes now taking place on the globe. In 1818 the celebrated naturalist Blumenbach suggested to the Royal Society of Gottingen that a prize should be offered for an 1 investigation of the changes that have taken place in the earth's surface-conformation since historic times, and the application which can be made of such knowledge in investigating earth revolutions beyond the domain of history.' Karl von Hofi, a learned young German, took up the question with zeal and ability and produced a work of the greatest value, entitled ' The History of those Natural Changes in the Earth's Surface which are proved by Tradi- tion.' The first volume of this great work, dealing with changes between the relations of sea and land which have taken place in historical times, was published in 1822, and the second, treating of the results of earthquakes and volcanoes of which we have records, appeared in 1824. Von Hoff, in his scholarly work, derived his data almost wholly from literature, ' his modest circumstances,' as von Zittel tells us, ' not permitting him to visit the localities of which he wrote.' To Lyell, however, these books of von Hofi proved invaluable, stirring him up to visit the various localities and satisfy himself concerning the facts by per- sonal inspection. Lyell always fully acknowledged his indebtedness to von Hoff declaring in one of his letters. [38] HISTORICAL INTRODUCTION ' Von Hoff helped me most.' The cumbrous mass of historical evidence collected by von Hoff became the basis of arguments in the hands of Lyell, in which, by incorporating his personal observations, he was able to carry conviction to the minds of his readers. It is pleasant to record that Lyell's indebtedness to his German con- temporary was not only gratefully acknowledged, but fully repaid, as will be evident to any one who turns to von HofFs third volume, which did not appear till 1834, after the publication of the * Principles of Geology.' But there can be little doubt that another very powerful influence in confirming Lyell in his feelings of revolt against the current geological ideas of the day was his close associa- tion with George Poulett Scrope, who joined the Geological Society in 1824 ; while in the following year he and Lyell were elected joint secretaries. The two young men, of just the same age, had similar tastes and convictions ; both had travelled extensively and both, without having read the works of Hutton, had arrived at similar conclusions to those of that great man. The warmest friendship, which endured through their lives, sprang up between the two young men, though, unfortunately, Scrope soon abandoned science for a political career. In the preface to a some- what speculative work, Scrope had very clearly outlined a system of geology identical with that of Hutton ; but his greatest service to science consisted in the description, and illustration by panoramic views, drawn by himself, of the proofs of atmospheric and river erosion in the volcanic district of Auvergne. Not only must the continued inter- course with so congenial a mind as that of Scrope have greatly influenced Lyell in the great task before him, but, as we shall see, it led in the end to his paying a visit to Auvergne and completely remodelling the first draft of his great masterpiece. 1 At what date Lyell became so convinced of the un- 1 For further details concerning the relations of Lyell with his predecessors and contemporaries, see ' The Coming of Evolution,' Cambridge, 1910. HISTORICAL INTRODUCTION [39] tenability of the cosmological theories taught by Buckland alid Cuvier, as to be induced to prepare a work to oppose the views almost universally accepted by his contem- poraries and friends in the Geological Society, we have no means of determining. By the spring of 1827, however, when not quite thirty years of age, Lyell had made the discovery that he had neither the taste nor the qualifications for a barrister, the profession which he had adopted by his father's desire on leaving Oxford ; and when returning for the last time from circuit, he wrote to his friend Mantell as follows : ' I devoured Lamarck en voyage. His theories delighted me more than any novel I ever read, and much in the same way, for they address themselves to the imagination, at least of geologists who know the mighty inferences which would be deducible were they established by observations. But though I admire even his nights, and feel none of the odium theologicum which some modern writers in this country have visited him with, I confess I read him rather as I hear an advocate on the wrong side, to know what can can be made of the case in good hands. I am glad he has been courageous enough and logical enough to admit that his argument, if pushed as far as it must go, if worth any- thing, would prove that men may have come from the Ourang-Outang. But after all, what changes species may really undergo ! How impossible will it be to dis- tinguish and lay down a line, beyond which some of the so-called extinct species have never passed into recent ones. That the earth is quite a* old as he supposes, has long been my creed, and I will try before six months are over to convert the readers of the "Quarterly" to that heterodox opinion.'' In this last sentence, which we have italicised, Lyell evidently refers to his review of Scrope's book on Auvergne, which was then in preparation. From a letter written a few months after the above, it is evident that Lyell had not up to that time read Hutton's works, for he speaks of that author as having run ' unnecessarily counter to the feelings and prejudices [40] HISTORICAL INTRODUCTION of the age.' This is manifestly only a reflection of current opinion, and certainly not a judgment which Lyell would have delivered had he at that time himself perused the 'Theory of the Earth.' Indeed, as late as 1839 he informed Fitton in a letter : ' I found it difficult to read and remember Hutton, and though I tried, I doubt whether I even fairly read more than hah his writings, and skimmed the rest.' Even Hutton's devoted friend and disciple, John Play- fair, insisted on the necessity of ' explaining Hutton's " Theory of the Earth " in a manner more popular and perspicuous than is done in his own writings. The obscurity of these has been often complained of ; and thence, no doubt, it has arisen that so little attention has been paid to the ingenious and original speculations which they contain.' In writing the ' Principles of Geology ' Lyell was deeply impressed with the necessity of adopting a method and style which would make his work readable ; and he was no less determined that, while not hiding his convictions, he would avoid everything that tended to rouse theo- logical prejudice. He wrote to Scrope : ' I conceived the idea five or six years ago ' (that is in 1824 or 5) ' that if ever the Mosaic geology could be set down without giving offence, it would be in an historical sketch.' In order to carry out this plan he prefaced his wcrk with an account of the various absurd theories which Moham- medan and Christian writers had constructed, in order not to appear to do violence to their sacred writings; but he tells Scrope : * I was afraid to point the moral about Moses. Per- haps I should have been tenderer about the Koran.' It was while preparing this historical introduction that Lyell first became aware of the splendid prescience of Hutton as displayed, not so much in his own work as in Playfair's more lucid ' Illustrations ' ; and he did his utmost to rescue the doctrines of the Scottish philosopher HISTORICAL INTRODUCTION [41] from the slough of misrepresentation and oblivion into which they had fallen. Nothing can be more emphatic than his acknowledgment of priority and his testimony to the value of Hutton's theory ; while he adopted as mottoes for his own volumes passages from the eloquent writings of Playfair. At one time Lyell appears to have entertained the project of embodying his views in a work to be entitled 4 Conversations in Geology ' ; he probably contemplated adopting the same artifice for defending geological ideas, generally regarded as heterodox, as Copernicus and Galileo had employed in the case of astronomy. But, in the end, he abandoned this plan and decided to divide his work into two portions ; devoting the first to a discussion of the changes which are taking place on the earth's surface at the present time ; and, in the second, demonstrating that the events of the geological record are capable of being explained by the action of similar causes to those now at work, operating during boundless periods in the past. By the close of the year 1827, Lyell had completed what he afterwards termed ' a first sketch only ' of the ' Principles of Geology,' and the manuscript was delivered to the publisher under an arrangement by which it was to appear in two octavo volumes. But various causes were at work which led ere long to the suspension of the publication of the book. In the first place, he was so greatly impressed by the evidence adduced by his friend Scrope concerning the excavation of valleys by rivers in Auvergne, and of the work of organisms in building up limestones in that country on a far grander scale, indeed, than in his native Angus that he determined to visit and study for himself this most interesting country before proceeding further with his task. In the second place, he saw clearly that the fullest and most convincing evidence of the continuity of the processes by which the deposition of strata in past and present times was accomplished would be found in the youngest (Tertiary) rocks of the globe, [42] HISTORICAL INTRODUCTION of which no complete classification had at that time been attempted. Thus, early in the year 1828, Lyell was in- duced to stop the printing of his book, and after completing a very lucid sketch of his views, in a review of Scrope's work on Auvergne, to betake himself to travel and further investigation. Setting out with his friend Murchison in May 1828, he spent the summer in studying the strata of the Limagne of Auvergne and the volcanic rocks associated with them ) taking note, as Desmarest and Scrope had done before him, of the great results produced by the work of rain and rivers in this now classic geological district. In the autumn he put into practice a plan which he had conceived of correlat- ing all the scattered patches of Tertiary strata by estimating the proportion of recent to extinct forms of Mollusca which they severally contained. For this purpose he examined all the Subapennine deposits from the extreme north to the farthest point south in the Italian peninsula, and thence into Sicily ; he studied the collections of the museums and private workers in all the Italian cities ; and he devoted himself diligently to conchological study under all the naturalists he met with. In returning through Switzerland and France, he had opportunities, diligently used by him, for studying other large collections of Tertiary fossils, and for making comparisons with those of Italy. When, in March 1829, Lyell resumed his work upon the book, it is evident, from his correspondence, that the whole had to be rewritten in order to embody his new observa- tions. Thus it was that the publication of the first volume was delayed till June 1830. 1 In the original plan of the ' Principles ' it was proposed, as we have seen, to complete the discussions concerning 1 In the preface to the third volume of the ' Principles,' written in April 1833, Lyell, by a slip of the pen, writes January instead of June as the date of the publication of the first volume : and the statement is repeated in the ' Life and Letters.' But the correspondence proves without doubt that June was the actual date of publication, and Mr. Murray has obligingly con- firmed this by a reference to the books of the firm. HISTORICAL INTRODUCTION [43] the changes now taking place on the globe in the first volume. At that time Lyell thought that the bulk of the volume might be devoted to processes going on in the Inorganic world, and that a chapter or two at the end would suffice for dealing with those operating in the Organic world. But it is evident that as he warmed to his task he became more and more deeply engrossed in the latter subject, and, in the end, felt constrained to so far modify his first plan as to allow the first volume to appear without the discussions on the actions and mutations of organisms, reserving these great questions concerning plants and animals for treatment in a second volume. 1 This did not make its appearance till January 1832. Although this volume commences with a very trenchant criticism of v Lamarck's hypothesis, every thoughtful student of it must read between the lines that Lyell had become a convinced - evolutionist, though he could not accept any of the hypo- theses that had up to that time been proposed to explain the appearance of new species. In his private corre- spondence with Sedgwick, Whewell, Herschel, and other friends, Lyell never hesitated to affirm his strong con- viction ' that the creation of new species is going on at the present day,' though he adds, * I have studiously avoided laying the doctrine down dogmatically as capable of proof.' In another place, alluding to ' the probable origination of new species through the intervention of natural causes,' he says, ' I left this rather to be inferred, not thinking it worth while to offend a certain class of persons by embodying in words what would only be a speculation ' ; and again, ' I should have raised a host of prejudices against me, which are unfortunately opposed at every step to any philosopher who attempts to address the public on these mysterious subjects.' Although Lyell, with the example of Hutton's fate before him, showed a justifiable reticence in dealing with matters at that time not ripe for solution, yet the discussions in the second volume of the ' Principles ' of such questions 1 See the Mter quoted on page [56]. [44] HISTORICAL INTRODUCTION as the variation of plants and animals, the struggle for existence among them, the methods and results of their distribution, with the problems of hybridity and of single or multiple ' centres or foci of creation,' undoubtedly served to bring before the minds of Darwin, Wallace, and other naturalists the real nature of the evidence which was re- quired, before the ' mystery of mysteries,' as Herschel called the evolution of species, could be unveiled. Some may be disposed to blame Lyell for a caution which they may consider as extreme ; but it is certain that he would never have succeeded in his great task of inclining public and scientific opinion in the direction of evolution, had he been less circumspect in avoiding what would arouse the odium theologicum. Aided by the judicious reviews of Scrope in the Quarterly Reriew, the contents of the first and second volumes of the ' Principles,' while they excited much opposition in certain quarters, were generally felt to be animated by a sincere desire to avoid giving offence, while at the same time they clearly and eloquently ex- pounded views that had been so long regarded with suspicion or treated with ridicule. Having disposed of the first portion of his great task the investigation of the causes of change operating both in the organic and inorganic world Lyell was now free to concen- trate his attention on the second portion. Before, however, the third volume of the ' Principles ' appeared, in May 1833, Lyell had to engage in the study of the extensive collections of mollusca which he had accumulated from all the European Tertiary formations, and he was fortunate in being able to obtain instruction and aid in his con- chological studies from Deshayes in Paris and Beck in Copenhagen. Lyell wisely devoted the greater part of this third volume to the study of the most recent era of the world's history the Tertiary periods. The geological record resembles that of human history in the circumstance that, the nearer we are to our own times, the more com- plete and satisfactory is the evidence that is available : as we recede into the past that evidence, both in human HISTORICAL INTRODUCTION [45] and geological records, becomes more scanty, fragmentary, and unsatisfactory. By an appeal to the great collections of Tertiary mollusca, Lyell was able to supply the most complete refutation of the catastrophic doctrines of Buckland and Cuvier. Not only did the study of these mollusca reveal no trace of the supposed revolutionary changes, in which all species were destroyed and complete assemblages of new ones created, but all the evidence collected was shown by Lyell to be consistent only with the conclusion that species had died out singly and gradually, and that new ones had appeared as sporadically and imperceptibly. Based on this im- portant conclusion was the classification which Lyell proposed for the Tertiary deposits a classification which, in its main principles, is still followed. With regard to formations older than the Tertiary, Lyell was content to indicate that, making due allowance for the ever-growing imperfection of the record as we trace it backward, the conclusions are not inconsistent with those derived from the study of the Tertiary formations. In the last chapters of his book Lyell showed how the final disappearance of all available evidence during the oldest periods of the earth's history can be accounted for by his theory of ' Metamorphism.' The writing of the c Principles of Geology ' was the culminating effort in LyelPs very active career. Just as Darwin, after the publication of the ' Origin of Species,' devoted the remaining twenty years of his life to issuing a series of illustrations of the truth of the great principle he had enunciated, so Lyell spent the last forty years of his busy life in gathering additional facts and supplying new reasonings that bore on the theme of the ' Principles of Geology.' In addition to this, he had the satisfaction of seeing his teaching, in the hands of his friend Darwin, made the basis of new advances and of aiding, by counsel and criticism, in the full establishment of the doctrine of evolu- tion in the Organic as well as in the Inorganic world. Five years after the completion of the ' Principles of [46] mSTOEICAL INTRODUCTION Geology,' Lyell found that in the Mesozoic and Palaeozoic strata, which had been so summarily dealt with in his original work, fresh support of his doctrines of uniformity could be obtained, and he issued the second portion of his argument as an independent work, which in different editions were styled ' Elements ' and ' Manual ' of Geology. In a great number of separate memoirs, lectures, and ad- dresses, the outcome of his indefatigable travels and investi- gations, he furnished fresh illustrations or amplifications of his views ; and the two books in which he recorded his observations made during visits to the United States are directed in great part to the same end. Lastly, his ' Anti- quity of Man ' was devoted, not only to the discussion of the important new facts which had been discovered con- cerning the appearance of the human race on the earth, and closely related questions concerning the Glacial Period, but to his final avowal of agreement with the theory of Evolution announced by Darwin, the growth of which he had so eagerly watched in his long years of intimate intercourse with its author. Of the first portion of the ' Principles ' no less than twelve editions were issued ; of the second portion, or 1 Elements,' after its separation, six editions ; and of the ' Antiquity of Man,' four. In almost all cases, numerous additions and corrections were made to keep the works au courant with the rapidly growing science many por- tions being rewritten. It must be confessed that in these modifications, necessary as they were, much of the freshness and charm of the old ' Principles ' was lost, and he who would appreciate the real excellence of Ly ell's masterpiece and understand the influence it undoubtedly exercised, must read the first edition, bearing in mind the date and the state of knowledge when it was written. Second only to his interest in geological studies was Charles Ly ell's devotion to the promotion of education, and especially to all projects for introducing science into the ordinary courses of study in schools and universities. His earliest publication was an essay contributed to the HISTORICAL INTRODUCTION [47] Quarterly Review, ' On Various Scientific Institutions in England,' which, it is evident from his correspondence, he intended to be only the first of a, series of articles dealing with the problems of scientific education. Although his engrossing labours in geology prevented the carrying out of this project, it is evident from his letters and from his two works of travel in the United States that he never lost sight of educational questions. He took the greatest interest in the foundation of the University of London, of which his friend and father-in-law, Leonard Horner, became the first Warden, and was only prevented by the pressure of his geological work from falling in with the project of initiating the teaching of geology in that institution by becoming professor of the subject in University College. But in March 1831 he found himself sufficiently free from his literary tasks to accept the professorship of Mineralogy and Geology in King's College, London. It is evident from his correspondence that he felt very strongly that the discipline of his own mind, in preparing discourses that would arouse the attention, keep the interest, and convey instruction to audiences destitute of previous scientific training, would be invaluable to him as a preparation for addressing far wider audiences of the same kind in his published works. His lectures were continued for two years, and attracted much attention outside the class of University students. During this time he was writing the third volume of the ' Principles,' and all the questions connected with the Tertiary strata, metamorphism, &c., were discussed in his lectures, before his views were embodied in the chapters of his book. In 1870, when he had reached the age of seventy-three, Lyell felt that he might render a last service, alike to his favourite science and to education, by using his wide ex- perience to concentrate the teachings of fifty years into a volume of moderate size, adapted to the needs of students of the subject. Putting on one side the various discussions constituting side issues many of which were of only fugitive interest he determined to gather the fundamental 48] HISTORICAL INTRODUCTION portions of his works into a volume embodying a succinct statement of the methods and results of geological investi- gation, carried out upon those lines which he had so patiently and perseveringly advocated in his various writ- ings. That he was very sensible of the difficulties of such a task is shown by what he wrote in his preface to the first edition : * In executing this task I have found it very difficult to meet the requirements of those who are entirely ignorant of the science. It is only the adept who has already over- come the first steps as an observer, and is familiar with many of the technical terms, who can profit by a brief and concise manual. Beginners wish for a short and cheap book in which they may find a full explanation of the leading facts and principles of Geology. Their wants, I fear, some- what resemble those of the old woman in New England, who asked a bookseller to supply her with "the cheapest Bible in the largest possible print." ' But notwithstanding the difficulty of reconciling brevity with the copiousness of illustration demanded by those who have not yet mastered the rudiments of the science, I have endeavoured to abridge the work in the manner above hinted, so as to place it within the reach of many to whom it was before inaccessible.' The forty years which have elapsed since these words were written have certainly not lessened the difficulty of presenting to a beginner an accurate survey of tlie whole field of geological science. Vast tracts of the land-surface of the globe, unvisited in Lyell's day, have since yielded up to enterprising travellers many of the secrets of their rock-masses ; while the depths and floors of the oceans have been so far explored as to have afforded much new evidence concerning the conditions under which new rocks are being formed at the present day. In addition to all this, fresh methods of chemical, physical, and microscopical investigation have added largely to our knowledge of the molecular processes by which the structure of rocks has been determined, while the application of the principle of HISTORICAL INTRODUCTION [49] evolution has revolutionised the study of the organic remains which they contain. It is only necessary to examine any of the extended treatises on Geology which have appeared in recent years to realise how impossible it is to furnish, in a single work, a full outline of the state of our knowledge on the subject at the present day. The utmost that can be attempted by the ablest exponents of the science is the giving of a descrip- tion of the succession of events, in past time, in one country, accompanying this with more or less satisfactory compari- sons with what has been discovered in other areas. For the beginner, however, the teacher may be wise to avoid the confusion arising from a multiplicity of details, by contenting himself, at the outset, with an exposition of the principles of the science and the methods of investi- gation available for its cultivator, making such a selection of the results of study, of many workers in widely separated lands, as may seem to him best suited to illustrate these principles and methods. This I may safely assert, from frequent conversations with Lyell during the writing and revisions of the work, was the great object he steadily kept in view during the preparation of his ' Student's Elements of Geology,' on which the present work is mainly based. For all those who come to the study of the science with a desire, in the end, to extend the bounds of our know- ledge, this plan may still be recommended as more profit- able, at the outset, than any attempt to grapple with the whole of the vast and multifarious masses of fact which have now been accumulated in our ever-growing science. The necessity for a work like the present is shown by the fact that many of the misconceptions concerning LyelPs teaching, which were so rife in his day, may still be detected in the writings of recent authors on geological subjects. Again and again do we find it stated that Lyell's position was, that the forces operating on the globe in past times were the same as those we see in action around us at the present day, differing neither in nature nor in degree : and [50] HISTOKICAL INTKODUCTION that the past course of nature has been for all time absolutely uniform with her present course. Fortunately, Lyell' has left behind him not only in his books but in his correspondence abundant evidence that this was not the position which he adopted. The line which he took and it was one in which Darwin was always in the most hearty accord with him was that Geology is a historical science ; as Huxley has declared, ' as much a historical science as Archaeology.' Lyell maintained that, just as the historian is not justified in admitting the inter- vention of giants or demigods in order to account for past events, so long as an explanation of them can be found, based on what we know of the powers, passions, and modes of procedure of mortals like ourselves, so it is not logical to invent ' catastrophies ' to account for geological pheno- mena which can be explained by the action of causes of 'the same order of magnitude' (to use a phrase suggested by Wallace) as those now operating even though it be necessary to admit that they must have acted during incalculably vast periods of time. Although the facts and reasonings by which Hutton and Lyell reached their conviction concerning the uniformity of action in past and present times were as we have seen so very different, yet the positions finally taken up by the two philosophers were identical. Hutton summed up his famous work in the words : ' The result of this physical inquiry is that we find no vestige of a beginning no prospect of an end.' I venture to italicise the words that have so often been, unintentionally or maliciously, overlooked. Lyell writing, just before the issue of the ' Principles of Geology,' to his friend Scrope, who was fond of indulging in cosmological speculations, said, ' All I ask is, that at any given period of the past, don't stop inquiry when puzzled, by refuge to a " beginning," which is all one with " another state of nature," as it appears to me. But there is no harm in your attacking me, provided you point out that it is the proof I deny, not the probability of a beginning. HISTORICAL INTRODUCTION [51] Mark, too, my argument, that we are called upon to say in each case " Which is now most probable, my ignorance of all possible effects of existing causes," or that " the begin- ing" is the cause of this puzzling phenomenon? It is not the beginning I look for, but proofs of a progressive state of existence in the globe, the probability of which is proved by the analogy of changes in organic life.' After this periphrasis for ' evolution ' Lyell adds : ' 'Tis an easy come-off to refer gneiss to " the beginning, chaos," &c., and put back the finding an encrinite for half a century.' Writing to Whewell in 1837, Lyell said in reference to his doctrine of ' Uniformity,' as enunciated in the ' Principles of Geology ' : ' I did not lay it down as an axiom that there cannot have been a succession of paroxysms and crises, on which "a priori reasoning" I was accused of proceeding, but I argued that other geologists have usually proceeded on an arbitrary hypothesis of paroxysms and the intensity of geological forces, without feeling that by this assumption they pledged themselves to the opinion that ordinary forces and time could never explain geological phenomena. The reiteration of minor convulsions and changes is, I contend, a vera causa, a force and mode of operation which we know to be true. The former ' (greater ?) ' intensity of the same or other terrestrial forces may be true ; I never denied its possibility ; but it is conjectural. I complained that in attempting to explain geological phenomena, the bias has always been on the wrong side ; there has always been a disposition to reason d priori on the extraordinary violence and suddenness of changes, both in the inorganic crust of the earth, and in organic types, instead of attempting strenuously to frame theories in accordance with the ordinary operations of nature.' Lyell in his conversations always declared himself ready to accept any valid evidence which could be brought for- ward to show that, in the past, existing forces acted with greater intensity than at present. He was always ready [52] HISTORICAL INTRODUCTION to admit that the portion of t/lie earth occupied by com- petent observers being so small, and the time covered by our historical records so limited, it was quite possible, nay, even probable, that convulsions greater than the floods cf Tivoli or Bagnes, volcanic outbursts more tremendous than those of ' Skaptar Jokull ' or Krakatoa, and earth- quakes more devastating than those of Lisbon or Messina, may have occurred in comparatively recent times. And he was never tired of insisting, as in the passages re- produced in the last chapter of this work, on the probability that agencies existed in nature of the potency of which we are unaware. When Charles Darwin, as a schoolboy, had his attention arrested by a big boulder in his native town, a worthy local geologist solemnly assured him ' that the world would come to an end before any one would be able to explain how this stone carne where it now lay.' Yet within a very few years a solution was found, and in later life Darwin could write : ' This produced a deep impression on me, and I meditated over this wonderful stone. So that I felt the keenest delight when I first read of the action of ice- bergs in transporting boulders, and I gloried in the progress of Geology.' And now we know that icebergs afford not the only means of adequately accounting for the transport of this great stone ! Our knowledge of the actions going on upon the surface of the land and in the depths of the ocean has increased, and is still increasing, every day since Hutton and Lyell wrote; and in 1865 the latter author signalised his complete acquiescence in the conclusions of his great predecessor in words he addressed to the British Association at Bath : ' I will not venture on speculations respecting " the signs of a beginning " or the prospects of an end of our terrestrial system that wide ocean of scientific conjecture on which so many theorists before my time have suffered shipwreck.' Apart from theological prejudices, the greatest obstacle to the full acceptance of LyelPs doctrine of 'Uniformity' HISTORICAL INTRODUCTION [53] or ' Continuity ' has been the reluctance to grant the neces- sarily vast periods of time for the production of such grand results by the seemingly insignificant existing agencies. By an amusing anecdote Lyell endeavoured to illustrate the causes of this reluctance to concede the enormous ex- tension of past time which the advocates of the doctrine of uniformity admit to be necessary. He said : ' It is related of a great Irish orator of our day, that when he was about to contribute somewhat parsimoniously to- wards a public charity, he was persuaded by a friend to make a more liberal donation. In doing so he apologised for his first apparent want of generosity, by saying that his early life had been a constant struggle with scanty means and that " They who are born to affluence cannot easily imagine how long a time it takes to get the chill of poverty out of one's bones." In like manner we of the living generation, when called upon to make grants of thousands of centuries in order to explain the events of what is called the modern period, shrink naturally at first from making what seems so lavish an expenditure of past time. Throughout our early education we have been accustomed to such strict economy in all that relates to the chronology of the earth and its inhabitants in remote ages, so fettered have we been by old traditional beliefs, that even when our reason is convinced, and we are persuaded to make more liberal grants of time to the geologist, we feel how hard it is to get the chill of poverty out of our bones.' The prejudices of our early education were powerfully reinforced when mathematicians and physicists announced that geological time must be restricted within certain limits somewhat narrow limits in the eyes of evolutionists on the ground that all the known sources of heat, both in the sun and the earth, would be exhausted within the periods of time assigned by them. (See Note AA, p. 610.) The discovery of new and unsuspected sources of energy in the universe has entirely met this difficulty, and, with old Lamarck, geologists and biologists may now ex- claim, ' For Nature, time is nothing. It is never a diffi- [54] HISTORICAL INTRODUCTION culty, she always has it at her disposal ; and it is for her the means by which she has accomplished the greatest as well as the smallest results. For all the evolution of the earth and of living beings, Nature needs but three elements space, time, and matter.' And in another place he says : * Nature to perfect and to diversify animals requires merely matter, space, and time.' The historian wisely pursues his researches into the past without troubling himself greatly concerning the problem of the origin of the human race, leaving such riddles to the anthropologist. And the geologist will act with equal prudence if he resigns questions concerning the origin of our planet, and its condition before life began on its surface, to the cosmologist and geophysicist. Doubtless, the great revolution in our knowledge of physics and chemistry in recent years will, in the end, profoundly modify cosmo- logical theories whether those theories be based on the planetismal, the meteoritic, the nebular, or any other hypothesis.' Such being the case, the geologist of the present day may be well advised to leave problems concerning the ' beginning ' to the physicist, the chemist, and the astronomer ; contenting himself with the historian's task of unravelling the story of our globe and tracing it back till it is lost in 'the mists of antiquity.' And in regard to this task surely a sufficiently great and important one he may still be guided by the considerations laid down by Lyell in 1833, in the concluding words of his monu- mental work, the ' Principles ' words which close the last chapter of the present volume (see pp. 599-600). While writing the first volume of the * Principles/ Lyell r / clearly recognised, as we have shown, that organic evolution is the necessary corollary to that principle of continuity which he was establishing for the inorganic world. This tact has been strongly insisted upon both by Huxley and Haeckel, to the latter of whom Lyell, in 1868, wrote as follows : HISTORICAL INTRODUCTION [55] * Most of the zoologists forget that anything was written between the time of Lamarck and the publication of our friend's " Origin of Species." * I am therefore obliged to you for pointing out how clearly I advocated a law of continuity even in the organic world, so far as possible without adopting Lamarck's theory of transmutation. I believe that mine was the first work (published in January 1832) in which any attempt had been made to prove that while the causes now in action continue to produce unceasing variations in the climate and physical geography of the globe, and endless migration of species, there must be a perpetual dying out of animals and plants, not suddenly and by whole groups at once, but one after another. I contended that this succession of species was now going on, and always had been ; that there was a constant struggle for existence, as De Candolle had pointed out, and that in the battle for life some were always increasing their numbers at the expense of others, some advancing, others becoming exterminated. * But while I taught that as often as certain forms of animals and plants disappeared, for reasons quite intel- ligible to us, others took their place by virtue of a causation which was beyond our comprehension; it remained for Darwin to accumulate proof that there is no break between the incoming and the outgoing species, that they are the work of evolution, and not of special creation.' He then points out how the great authority of Cuvier, both on the Continent and in this country, was influential with all naturalists as it certainly was with Lyell himself in preventing them from seriously entertaining the idea of the formation of new species by the modification of pre- existing forms. The following, hitherto unpublished, letter of Lyell to Scrope, written on June 22, 1830, when the first volume of the ' Principles ' was on the eve of publication, and his friend was preparing a review of the book for the Quarterly, is of especial interest, as it shows the keenness with which Lyell was entering upon the task of applying his line of [56] HISTOKIOAL INTKODUCTION reasoning to the phenomena of Organic nature in the second volume : 'MY DEAR SCROPE, ' I am off to-morrow and will try to get to Olot. Some will say I have shirked the most difficult subject, the change in organic life. It is, however, a favourite one of mine. Bear in mind that it is all to come. The present distribution of species the permanency of specific char- acter partial extinction of certain species since historical era total of others modern introduction of man and perhaps of others their exterminating effects migratory powers, why given and how limited whether any reason to believe that no change now in progress in organic world why to be presumed imbedding of animals and plants in new strata, in deltas, lakes, deep seas, fissures, caves, &c. what classes most imbedded process of petrifaction, &c., &c. On all these I promise you interesting and novel matter, with plenty of original theory and speculation, without touching fossil animals, on which plenty of other things will be broached. ' Point out in conclusion that the inorganic world alone is discussed. . . . The volume is large enough, and more than some at least will digest before another appears. ' Believe me very truly yours, 'CHA. LYELL.' [The signature to this letter has been reproduced below the portrait forming the frontispiece of the present volume. The portrait itself is copied from the oldest-known photo- graph of Sir Charles Lyell, one taken by D. 0. Hill in 1846.] THE STUDENT'S LYELL PAET I GENERAL PRINCIPLES CHAPTER I GEOLOGY DEFINED HISTORY OF THE DEVELOPMENT OF GEOLOGICAL SCIENCE Geology compared to History Its relation to other Physical and Natural Sciences Not to be confounded with Cosmogony Opinions of Classical and Mediaeval writers Causes which have retarded the Pro- gress of Geology. GEOLOGY is the science which investigates the successive changes that have taken place in the inorganic and organic kingdoms of nature ; it inquires into the causes of these changes, and the influence which they have exerted in modify- ing the surface and external structure of our planet. By these researches into the state of the earth and its inhabitants at former periods we acquire a more perfect know- ledge of its present condition, and more comprehensive vieg^s concerning the laws now governing its animate and inanimate productions. When we study history, we obtain a more pro- found insight into human nature, by instituting a comparison between the present and former states of society. We trace the long series of events which have gradually led to the actual posture of affairs, and, by connecting effects with their causes, we are enabled to classify and retain in the memory a multitude of complicated relations the various peculiarities of national character, the different degrees of moral and intellectual refine- ment, and numerous other circumstances which, without historical associations, would be uninteresting or imperfectly understood. As the present condition of nations is the result of B 2 INTRODUCTOKY . [CH. i. many antecedent changes some extremely remote and others recent, some gradual, others sudden and violent so the state of the natural world is the result of a long succession of events, and, if we would enlarge our experience of the present economy of nature, we must investigate the effects of her operations in former epochs. We often discover with surprise, on looking back into the chronicles of nations, how the fortune of some battle has in- fluenced the fate of millions of our contemporaries, when it has long been forgotten by the mass of the population. With this remote event we may find inseparably connected the geo- graphical boundaries of a great State, the language now spoken by the inhabitants, their peculiar manners, laws, and religious opinions. But far more astonishing and unexpected are the connections brought to light when we carry back our researches into the history of nature. The form of a coast, the configura- tion of the interior of a country, the existence and extent of lakes, valleys, and mountains can often be traced to the former prevalence of earthquakes and volcanoes in regions which have long been undisturbed. To these remote convulsions the pre- sent fertility of some districts, the sterile character of others, the elevation of land above the sea, the climate, and various peculiarities may be distinctly referred. On the other hand, many distinguishing features of the surface may often be ascribed to the operation at a remote era of slow and tranquil causes to the gradual deposition of sediment in a lake or in the ocean, or to the prolific growth in the same of shells or corals. To select another example, we find in certain localities subterranean deposits of coal consisting of the remains of plants which have grown upon the spot, or have been drifted into seas and lakes. These seas and lakes have since been filled up, the rivers and currents which floated the vegetable matter can no longer be traced, the lands whereon the forests grew nave dis- appeared or changed their form, and the species of plants that supplied the materials have for ages passed away from the surface of our planet. Yet the commercial prosperity and numerical strength of a nation may now be mainly dependent on the local distribution of fuel determined by that ancient state of things. "M Geology is intimately related to almost all the physical sciences, as is history to the moral. An historian should, if possible, be at once profoundly acquainted with ethics, politics, jurisprudence, the military art, theology ; in a word, with all branches of knowledge whereby any insight into human affairs, or into the moral and intellectual nature of man can be ob- CH. i.] INTRODUCTORY 3 tained. It would be no less desirable that a geologist should be well versed in physics, chemistry, mineralogy, zoology, com- parative anatomy, botany ; in short, in every science relating to inorganic and organic nature. With these accomplishments the historian and geologist would rarely fail to draw correct and philosophical conclusions from the study of the various monu- ments of former occurrences. They would know to what com- bination of causes analogous effects were referable, and they would often be enabled to supply, by inference, information concerning many events unrecorded in the defective archives of former ages. But the brief duration of human life and our limited powers are so far from permitting us to aspire to such extensive ac- quirements that excellence, even in one department, is within the reach of few ; and those individuals most effectually promote the general progress who, after obtaining a general knowledge of the whole field of inquiry, concentrate their efforts upon some limited portion of it. As it is necessary that the historian and the cultivators of moral or political science should reciprocally aid each other, so the geologist and those who study physics or the biological sciences stand in equal need of mutual assistance. A comparative anatomist may derive some accession of knowledge from the bare inspection of the remains of an extinct quadruped, but the relic throws much greater light upon his own science when he is informed to what relative era it belonged, what plants and animals were its contemporaries, in what degree of latitude it once existed, and other historical details. A fossil shell may interest a conchologist, though he be ignorant of the locality whence it came ; but it will be of more value when he learns with what other species it was associated, whether they were marine or freshwater, whether the strata containing them were at a certain elevation above the sea, and what relative position they held in regard to other groups of strata, with many other particulars determinable by an experienced geologist alone. On the other hand, the skill of the comparative anatomist and conchologist is often indispensable to those engaged in geological research, although it may rarely happen that the geologist will himself combine these different qualifications in his own person. The analogy between the objects of research of the geologist and the historian extends no farther, however, than to one class of historical monuments those which may be said to be un- designedly commemorative of former events. The buried coin fixes the date of the reign of some Roman emperor; the ancient encampment indicates the district once occupied by an 4 INTRODUCTORY [CH. i. invading army, and the former method of constructing military defences ; the Egyptian mummies throw light on the art of embalming, the rites of sepulture, or the average stature of the human race in ancient Egypt. The canoes and the hatchets, called celts, found in peat bogs and estuary deposits, afford an insight into the rude arts and manners of a prehistoric race, to whom the use of metals was unknown, while flint implements of a much ruder type point to a still earlier period, when man coexisted in Europe with many quadrupeds long since extinct. This class of memorials yields to no others in authenticity, but it constitutes a small part only of the resources on which the historian relies, whereas in geology it forms the only kind of evidence which is at our command. For this reason we must not expect to obtain a fall and connected account of any series of events beyond the reach of history. But the testimony of geological monuments, if frequently imperfect, possesses at least the advantage of being free from either unconscious bias or intentional misrepresentation. We may make mistakes in our observations and the inferences which we draw from them, in the same way as we often misunderstand the nature and import of phenomena noticed in the daily course of nature ; but our liability to go astray in geological inquiries is confined to errors of observation and to faults of interpretation ; if obser- vations and reasonings be alike correct, our conclusions are certain. It was long before the distinct nature and legitimate objects of geology were fully recognised, and geology was at first con- founded with many other branches of inquiry, just as the limits of history, poetry, and mythology were ill defined in the infancy of civilisation. Even in Werner's time, or at the close of the eighteenth century, Geology * appears to have been regarded as little other than a subordinate department of Mineralogy; and Desmarest included it under the head of Physical Geography. But the most common and serious source of confusion arose from the notion that it was the business of geology to discover the mode in which the earth originated, or, as some imagined, to study the effects of those cosmological causes through the action of which our planet passed from a nascent and chaotic state into a more perfect and habitable 1 The word ' geology ' did not with the general relations of rock- come into general use till the corn- masses, without respect to their mencement of the present century. sequence in time, the old term Before that time the term ' geo- geognosy may still be conveniently gnosy' was employed for the branch employed. The term 'geology' of knowledge which we now desig- was first used, about 1779, by De nate as geology. For the sub- Luc and De Saussure. division of the science which deals CH. i.] INTKODUCTOKY 5 condition. The first who endeavoured to draw a clear line of demarcation between Cosmogony and Geology was Dr. James Hutton, 2 who declared that geology was in no way concerned * with questions as to the origin of things.' But his doctrine on this head was vehemently opposed at first, and, though it has been continually gaining ground, it cannot even yet be said to be universally accepted. History of the development of geological science. In 1830 Lyell gave a sketch of the early history of geological thought and speculation, to show how mischievous had been the effects of confounding the objects of geology with those of cosmogony. He showed that the Indian, Egyptian, Grecian, and Eoman philosophers with some conspicuous exceptions had altogether neglected the study of the monuments left to us of former changes in the earth's crust, to indulge in the more attractive but fruitless discussions con- cerning the origin of the world and the great catastrophes to which it was supposed to have been subjected. Certain Arabian writers of the tenth century, and afterwards the philosophers of Italy in the sixteenth, seventeenth, and eighteenth centuries Leonardo da Vinci, Fracastoro, Steno, Scilla, Lazzaro Moro, Donati, &c. with Hooke, Boyle, and Michell in England, Palissy and Buffon in France, and Kaspi, Fuchsel, and others in Germany, laid the true foundation of a geological science, based on the observation of the existing order of nature. The growth of just geological ideas was hindered, however, by many prejudices. Fossils were long maintained to be, not the remains of organised beings, but strange ' freaks of nature,' the productions of a fancied ' plastic force.' When their organic nature was at last accepted, many gravely argued that fossils must be regarded, not as the remains of beings that had existed in the past, but as the prototypes of creatures that were to receive the en- dowment of life in the future ! Even when all doubts had been removed concerning the true nature of fossils, it was only after con- siderable progress had been made in the study of the forms and structure of living beings that naturalists were able to realise the fact that many fossils represent forms of life which no longer exist upon the earth. Still more detrimental to the progress of geology were the fixed prepossessions in the minds of nearly all men that the earth's exis- tence had been limited to a few thousand years, and similar prejudices with regard to great catastrophes that were supposed to have happened in comparatively recent times. Though these prepossessions are now to some extent removed, yet others still exist, from which it is not easy to free ourselves. Most of the relics of animal and vegetable life, preserved to us in a fossil state, have lived in the watars of the ocean, while our obser- vations of existing nature are nearly all confined to the land surfaces on which we dwell. Many of the most important changes, occurring in the earth's crust, take place deep beneath the surface, 2 Hutton's Theory of the Earth Huttonian Theory of the Earth appeared in 1788, and in a more was published in 1802, and Lyell's complete form in 1795. Dr. John Principles of Geology in 1830-33. Playfair's Illustrations of the 6 INTRODUCTORY [CH. i. and concerning the nature and action of the operations going on there our experience on the earth's surface leads us to form only very inadequate conceptions. Last, and most important of all, perhaps, among the prejudices which have retarded the study of geological problems, we must regard those which are due to the effects produced on our minds by the magnificence of the phenomena themselves. It is very difficult at first sight to believe that the making of lofty mountains and deep valleys, the piling together of many thousands of feet of materials, and the passing away of whole generations of living creatures, have not been brought about by great and convulsive throes of nature rather than by simple causes operating through vast periods of time. The student of geology, however, must be prepared, upon due cause being shown, to lay aside these prepossessions, and to guard his mind during all his inquiries against the influence of these pre- judices. Two dangers, and two dangers alone, beset him the chance of erroneous observation, and the risk of incorrect inference from observation. He who is not prepared to give up prepossession and prejudice, when just reasoning on careful observations demands the sacrifice, is unworthy to enter upon the study of science. Since Lyell wrote the lines of this opening chapter, the advance of our physical and astronomical knowledge has enabled cosmogony to pass from the region of wild speculation, and to enter the circle of exact science. It has been maintained by some that the time has arrived when a judicious cosmology, which is the outcome of the application of correct physical principles to the explanation of cosmic phenomena, may be safely employed as the foundation of geology. But all experience shows that the dangers pointed out by Lyell as inherent to such a method still exist ; and that just as it is wise of the historian to pursue his investigations into the events of the past, without any reference to the fascinating question of the origin of the human race, so the geologist is justified, when tracing back the story of the earth and its inhabitants, to avoid allowing any theoretical views concerning the beginnings of matter or life to influence his conclusions. Especially is it desirable for the student and beginner that this distinction between the objects of geology and cosmology should be kept in view. There is one other point of resemblance between history and geology which it may be instructive to consider. The historian, when engaged in writing the annals of the more modern periods, is apt to be embarrassed by the abundance of the materials at his disposal. But as he passes backwards to the study of earlier times, this wealth of contemporary records manuscripts, monuments, and inscrip- tions gradually diminishes, till at last only a few inscribed stones, papyri, parchments, and coins are all he can rely upon in at- tempting to reconstruct the story of the earlier races of mankind. In the same way the geologist finds the-most recent periods of the earth's history richly illustrated in deposits formed on the land as well as in the sea materials accumulated by rivers, lakes, and glaciers retaining all their characteristic features and structures, and replete with the relics of almost every class of animals and plants. But as he pursues his investigations farther back into the past, the evidence on which he has to rely becomes smaller and more fragmentary. Instead of the varied materials of later periods, CH. i.] INTRODUCTORY 7 he finds only deposits that have been formed in the ocean, con- taining scarcely any remains of organisms besides those of marine animals, and these often very imperfectly preserved. Of still earlier periods the only records preserved consist of masses of rock, which have evidently undergone such an amount of chemical change that all traces of life, if they existed, must inevitably have been destroyed. The historical and the geological records alike commence in dimness and obscurity ; and, interesting as the study of these beginnings of the two records must always be, it would be manifestly unwise to allow the imperfect ideas we are able to form of the events of these early times to unduly influence us in our conclusions concerning the later periods, of which we have such abundant and unmistakable evidence (Note A, p. (>( 1). For further details concerning and Ramsay. Obituary notices the History of Geology the reader with the accounts of the work of is referred to the ' Principles of other geologists will be found in Geology,' Chapters II. to V. ; to the volumes of the ' Quarterly Fitton's ' Notes on the Progress of Journal ' of the Geological Society Geology in England ' in ' Phil. in the Anniversary Addresses of Mag.,' 1832-33, and to the biogra- the Presidents. See also Zittel's phies of William Smith, Lyell, ' History of Geology and Palaeon- Darwin, Murchison, Sedgwick, tology ' and Sir A. Geikie's ' The Buckland, Owen, Edward Forbes, Founders of Geology.' CHAPTER II THE CRUST OF THE GLOBE What geologists mean by the earth's crust Physical characters of the crust of the globe Chemical composition of the solid crust and of its liquid and gaseous envelopes Distribution of temperature in the earth's crust Distribution of pressure in the earth's crust. OF what materials is the earth composed, and in what manner are these materials arranged ? These are the first inquiries with which geology is occupied. We might have imagined at first sight that investigations of this kind would relate exclusively to the mineral kingdom, and to the various soils, rocks, and minerals, which occur upon the surface of the earth, or at various depths beneath it. But, in pursuing such researches, we soon find ourselves led on to consider the successive changes which have taken place in the former state of the earth's surface and interior, and the causes which have given rise to these changes ; and, what is still more singular and unexpected, we eventually become engaged in researches into the history of the organic world, or of the various tribes of animals and plants which have, at different periods of the past, inhabited the globe. All are aware that the solid parts of the earth consist of dis- ijinct substances, such as clay, chalk, sand, limestone, coal, slate, 8 THE EARTH'S CRUST [CH. n. granite, and the like ; but it is commonly imagined that all these have remained from the first in the state in which we now see them that they were created in their present form and in their present position. The geologist soon comes to a different conclusion, discovering proofs that the external parts of the earth were not all produced, in the beginning of things, in the state in which we now behold them, nor in an instant of time. On the contrary, he can show that they have acquired their actual configuration and condition gradually under a great variety of circumstances, and at successive periods, during each of which distinct races of living beings have flourished on the land and in the waters, the remains of these creatures still lying buried in the crust of the earth. By the ' earth's crust ' is meant that small portion of the exterior of our planet which is accessible to human observation. It comprises not merely all the parts of the earth which are laid open in precipices, or in cliffs overhanging a river or the sea, or which the miner may reveal in artificial excavations ; but the whole of that outer covering of the planet on which we are enabled to reason by observations made at or near the sur- face. These reasonings may extend to a depth of perhaps ten or fifteen miles ; and this is a very small fractional part of the distance from the surface to the centre of the globe. But al- though the dimensions of such a crust are, in truth, insignificant in comparison with those of the entire globe, yet they are vast and of magnificent extent in relation to man and to the organic beings which people our globe. Keferring to this standard of magnitude, the geologist may admire the ample limits of his domain, and admit at the same time that not only tho exterior of the planet, but the entire earth, is only an atom in the midst of the countless worlds surveyed by the astronomer. The materials of this crust are not thrown together con- fusedly ; but distinct mineral aggregates, called rocks, are found to occupy definite spaces, and to exhibit a certain order of arrangement. The term rock is applied indifferently by geolo- gists to all these substances, whether they be soft or stony, for clay, sand, and peat are included under this denomination. Our old writers endeavoured to avoid offering such violence to our language, by speaking of the component materials of the earth as consisting of rocks and soils. But there is often so insensible a passage from a soft and incoherent state to that of stone, that geologists of all countries have found it indispensable to have one technical term to include both, and in this sense we find roche applied in French, rocca in Italian, a,ndfclsart in German. The beginner, however, must constantly bear in mind that the CH. ii.] ITS PHYSICAL CHARACTERS 9 term ' rock ' by no means invariably implies that a mineral mass is in an indurated or stony condition. 1 Concerning the ' crust of the globe,' or that outer portion of the earth which is accessible to the geologist in his studies, it is desirable that we should bear in mind its general form, its density, chemical composition, and the distribution within it of temperature and pressure." Physical characters of the earth's crust. The crust of the globe may be regarded as being made up of three portions, solid, liquid, and gaseous. The solid crust forms a complete envelope to the unknown interior, which may be solid, liquid, or even gaseous in parts. This part of the crust of the globe, which is built up of the solid materials we call rocks, is sometimes spoken of as the ' lithosphere.' The liquid materials of the earth's crust consist of water, with various salts held in solution in it; these masses of water occupy most of the depressions in the solid crust, but do not form a continuous envelope. They are sometimes spoken of as the ' hydrosphere.' The gaseous materials of the earth's crust consist of air, with some gases diffused through it, forming the ' atmosphere,' a continuous layer which envelops the solid and liquid portions of the crust, and has a depth of over 200 miles. It must be borne in mind, however, that the limits between the solid, liquid, and gaseous portions of the earth's crust are by no means absolute and unchangeable. The waters not only flow over the surface of the solid crust, but penetrate it to great depths, and are returned to the surface in springs or sometimes, in volcanic districts, as vapour ; portions of the water are also dissolved in the atmosphere, or are held in suspension by it as clouds. In the same way, the gases of the atmosphere are found dissolved in the waters of oceans, lakes, and rivers, and imprisoned in the rocks of the earth's solid crust. Lastly, the materials of the solid crust itself are found dissolved or suspended in a finely divided state in the waters, and even held up in suspension by the atmosphere. In judging of the relations of the sea and land, the only standard we can employ is the level of the ocean. The highest mountains of the globe rise 29,000 feet above that level, and the deepest known oceanic depressions lie about the same distance below it. But the average height of the land is estimated as being only 2,200 feet, while the average depth of the ocean is no less than 12,600 feet. As, according to the most recent estimates, the land occupies only 28 1 If all the materials of the materials of the earth's crust, earth's crust are designated as though this restriction is sometimes ' rocks,' it seems impossible to avoid attended with considerable incon- calling by that name liquids, like venience. mineral oils and the waters of 2 The study of the globe in its springs, rivers, and the ocean. present condition, which is known Even gases, like those emitted from as ' Erdkunde ' by the Germans, is volcanoes and locked up in the now generally designated in this cavities of many solid rocks, must country as ' physiography.' The also be included under the same study of physiography, or ' physical term. It is usual, however, to geography,' as some people prefer avoid departing so far from the to call it, should, of course, precede popular use of the word, and to that of geology, while the study of confine the term rock to the solid cosmogony should follow it. 10 THE EARTH'S CRUST [CH. n. per cent, of the whole surface of the globe, the volume of the ocean masses is about fifteen times as great as that of the land masses rising above the ocean level. This is a very important consideration which the geologist must take into account when he is studying the phenomena which result from changes in the relative positions of land and water on the globe. While the earth, taken as a whole, has been shown to have a density or specific gravity of 5-5, the density of the solid crust is certainly much less than this. Most rocks have densities ranging from 2-0 to a little over 3. In a few rocks, like coal, peat, &c., the specific gravity is lower, and in some rocks, rich in compounds of the heavy metals, it is higher. Taking into account the relative abundance of rocks of different density, we may estimate the average density of all the rocks of the earth's solid crust at 2-75. This is exactly one half of the density of the whole globe. The density of the hydrosphere is less than two-fifths of that of the solid crust, and of the atmosphere at the earth's surface is y^ part of it, while at great elevations above the surface it is much less than this. Chemical composition of the earth's crust. The atmo- sphere is composed of the two gases, oxygen and nitrogen, mixed, but not combined with one another; the oxygen exists in some cases in its allotropic form of ozone, while the nitrogen is mixed to the extent of T T of its weight with still more inert gases, argon, &c. ; small but variable amounts of water vapour, carbon dioxide, and other gases are diffused through the atmosphere, while a few solid and liquid substances are held in suspension in it. The hydrosphere consists of water (a compound of hydrogen and oxygen) in which various soluble salts, gases, and suspended solids are present. The solid crust is much more complex in its composition, but is principally made up of various oxides, among which that of silicon plays the most prominent part, while the oxides of aluminium, calcium, magnesium, iron, potassium, sodium, and hydrogen, make up together by far the largest part of the remainder. From the general composition of rocks it has long been manifest that one half of the earth's solid crust consists of oxygen, one quarter of silicon, one-fourteenth of aluminium, while calcium, magnesium, and iron together make up one-tenth, sodium, potassium, and hydrogen together one-twentieth part of the whole. Mr. F. W. Clarke, the chemist to the Geological Smvey of the United States, has recently published a careful discussion of no less than 880 analyses of rocks with a view to determining the average composition of the earth's crust. 3 Assuming a thickness of solid crust of ten miles, he has arrived at the results given in the table on the opposite page. The only other elements, besides those mentioned in the table, which are at all widely diffused in the earth's crust are fluorine, boron, nickel, zirconium, beryllium, the metals of the cerium group, and some of the heavy metals which are found to be present in very minute quantities in most rocks. As all the elements form soluble compounds, they must be present in the waters of the ocean, though in very minute propor- tions. Thus gold and silver have both been detected in sea-water ; 3 ' The Relative Abundance of GeoL Survey, No. 78 (1891), pp. 34- the Chemical Elements,' Bull. U.S. 42, II.] ITS CHEMICAL COMPOSITION 11 and it has even been estimated that the quantity of the former metal now distributed through the ocean must be many million times greater than that which has been extracted by man from the solid crust. - Solid crust 93 per cent. Ocean 7 per cent. Mean including air Oxygen 47-29 85-79 49-98 Silicon 27-21 25-30 Aluminium .... 7-81 7-26 Iron ..... 5-46 5-08 Calcium .... 3-77 05 3-51 Magnesium .... 2-68 14 2-50 Sodium 2-36 1-14 2-28 Potassium .... 2-40 04 2-23 Hydrogen .... 21 10-67 94 Titanium .... 33 30 Carbon 22 002 -21 Chlorine .... 01 2-07 Bromine .... 008 15 Phosphorus .... 10 09 Manganese .... 08 07 Sulphur 03 + 09 04 + Barium 03 _ 03 Nitrogen .... 02 Chromium .... 01 01 Distribution of temperature in the earth's crust. In rising from the earth's surface into the highest points in the atmosphere which have been reached, we experience a constant and very rapid fall of temperature, but how far this progressive diminu- tion in temperature continues is not known. In the same way, in descending through the waters of the ocean, a remarkable fall in temperature is found to take place, even in the warmest seas, till at a moderate depth water at only a little above 0C. is found ; and it is ice-cold water which occupies all the deeper parts of the oceanic de- pressions. When we penetrate downwards into the solid crust of the globe, however, we everywhere experience a rise of temperature. At very moderate depths in all latitudes a stratum of invariable temperature is found, beneath which no changes due to daily or seasonal fluctua- tions are experienced. Beneath this stratum of invariable tempera- ture we find (alike in tropical, temperate, and polar regions) a progressive rise in temperature in going downwards, and this is con- tinued to the lowest points that have been reached. The instruments employed in determining the temperature of the earth's crust are either slow-action thermometers, or some form of maximum thermometer. Slow-action thermometers are instruments surrounded with some badly conducting material, which prevents any appreciable change taking place in the indications of the instru- ment while it is being drawn to the surface. The forms of maximum thermometers which have been used for the determination of earth- DISTRIBUTION OF TEMPERATURE [CH. II. temperatures in this country are the inverted Negretti, and Phillips's thermometer improved by Sir William Thomson (Lord Kelvin). On the Continent, various forms of overflow thermometers are usually employed for this class of observation the Magnus thermometer in Germany and the Walferdin thermometer in France (see fig. 1). Fig. 1. I I. n. in. iv. v. Thermometers employed in determining Earth -temperatures. I. Slow-action thermometer, with its bulb surrounded with stearine. II. The Inverted Negretti thermometer, with a constriction where the bulb joins the tube. III. The Phillips thermometer with very narrow bore. (The above instruments are employed in this country.) IV. The Magnus overflow thermometer used in Germany. V. The Walferdin overflow thermometer used in France. (In the overflow thermometers it is necessary to raise the temperature after the instrument is brought to the surface, and when the mercury reaches the open end of the tube a reading is taken on a standard instrument placed beside it.) All these instruments are protected from pressure by being enclosed in strong glass tubes, and from mechanical injury by being placed in metal cases. The figures are about half the size of the instruments themselves. CH. ii.] IN THE EARTH'S CRUST IB The observations by which earth -temperatures are determined fall into two classes : those made in mines, tunnels, &c., where access can be obtained by the observer to the spot where the temperature is to be taken, and those made in deep wells or bore- holes (usually filled with water), into which the thermometer has to be let down by a cord or wire. In the first class of observations, a slow- action or maximum thermometer is allowed to remain for a considerable time in a hole bored in the solid rock, the mouth of the hole being closed with a tamping of fragments of the same rock. The chief sources of error in observations in mines and tunnels consist in the changes of temperature due to ventilation and other processes going on within them. In the case of deep wells and bore- holes, the thermometer is let down and allowed to remain for a con- siderable time at the depth where it is desired to make a temperature observation. In such observations the principal risk of error is due to convection currents which must exist in all holes filled with water, but are least felt in boreholes of very small diameter. The action of such convection currents may, however, be neutralised by introdu- cing water-tight plugs above and below the thermometer, and thus isolating the water in the part of the well or borehole where the observation is being made. Although we everywhere find a more or less rapid rise in tem- perature in going downwards in the earth's crust, the rate of increase varies greatly in different localities. Many of the discrepancies can doubtless be accounted for by errors of observation, but when the fullest allowance is made for these the variations in different areas and even within the same area are often of a very startling character. Professor Everett and the Committee on Underground Tempera- tures appointed by the British Association for the Advancement of Science have calculated that the average rate of the rise of tempera- ture in going down into the earth's crust is 1F. for every sixty feet of descent. Professor Prestwich has been led to regard many of the published observations as altogether untrustworthy, and his discussion of all the best observations leads him to the conclusion that the average rise of temperature is as rapid as 1 for 47'5 feet of descent, or even 1 for 45 feet of descent. Another respect in which the results of underground temperature determinations show very marked discrepancies is in the uniformity or variation in the rate of increase in going downwards. In the case of the Sperenberg boring which was carried to a depth of 4,172 feet, there was exhibited a decided tendency to a diminution in the rate of rise in temperature in going downwards. In the Schladebach boring which reached a depth of 5,628 feet, the rate of increase (1F. for 67 feet) was very uniform. In the case of the deep well at Wheeling (West Virginia), with a depth of 4,500 feet, the temperature increased from 1F. for 82 feet in the upper part to 1F. for 58 feet in the lowest portion (Note B, p. ^601). The effects of variations in the specific heat and conductivity of rocks, and the influence of increased temperature and susceptibility to percolation of water, have all to be taken into account in consider- ing the varied results given by observations on earth-temperatures. Lines drawn through points in the earth's crust having the same temperature are called ' Isogeothermal lines ' or 'Isogeotherms.' At present, the data for drawing such lines are very imperfect, but their distribution within the earth's crust is a matter of great 14 PRESSURE IN THE EARTH'S CRUST [CH. n. interest to geologists. There is reason to believe that beneath the ocean floors covered with ice-cold water, and in those parts of the continents enveloped in snow and ice, the isogeotherms are depressed and crowded together ; while under the areas exposed to the atmo- sphere, and especially in mountain chains, the isogeotherms rise and become separated from one another. Distribution of pressure In tne earth's crust. The effects of pressure on the density of different parts of the atmosphere are of the most marked character. When we rise to a height of 3.V miles, we have passed through one half of the atmosphere, and the tenuity of the higher portions of the gaseous envelope must be extreme. Water, though so comparatively incompressible, does yield to the weight of the column above in the deep oceans. Every thousand fathoms of descent is equivalent to an increase of pressure of one ton to the square inch. From the experiments which he made upon this subject, Professor Tait has concluded that the com- pression of the oceanic waters by the superincumbent mass leads, in the case of the deepest oceans, to a depression of the surface of the ocean of one furlong (660 feet), and that the average height of the ocean is 116 feet less than it would be if water were an absolutely incompressible substance. Rocks, having a density 2'75 times greater than that of water, must produce a pressure of nearly three tons per square inch for every mile of descent. The effects of such pressures, not only in increasing the density of rocks, but in causing the penetration of ordinary liquids and gases (the latter often in a liquefied condition) through them, must be enormous. The minerals of nearly all the deeper-seated rocks, as we shall hereafter see, contain cavities filled with liquids, among which carbon-dioxide plays a very important part. Fuller information concerning length in the 'Bull. U. S. Geol. the physical characters of the earth's Survey,' No. 78, and for a complete crust will be found by students in discussion of earth-temperatures various treatises on physiography, the student is referred to Professor such as that of Professor Huxley, Everett's reports on Underground and in Dr. H. Mill's ' The Realm of Temperatures in ' Rep. Brit. Assoc.' Nature.' Much valuable informa- vols. from 1868 ownwards (see sum- tion on these questions is contained mary in vol. for 1886) ; also to in the ' Reports of the Voyage of Professor Prestwich's memoir on H.M.S. " Challenger." ' The ques- the subject (' Proc. Roy. Soc.' 1886), tion of the chemical composition of reprinted in ' Essays on Contro- the earth's crust is treated of at verted Questions of Geology,' 1895. CHAPTER III ROCKS AND THEIR CLASSIFICATION Classification of rocks according to their characters, origin, and age Epigene rocks Aqueous rocks Volcanic rocks Hypogene rocks Plutonic rocks Metamorphic rocks. ROCKS, or the solid materials of the earth's crust, may be classi- fied according to several different principles. All rocks are built CH in.] CLASSIFICATION OF KOCKS 15 UD of minerals ; but while the individual minerals making up a rock can sometimes be distinguished with the naked eye, or by the aid of a lens, it is, in most cases, necessary to prepare thin sections of rocks and to examine them with a microscope (often with very high powers) in order to discover the nature and peculiarities of the minerals which compose them. Kocks which are built up of distinct crystals are said to be crystalline ; l those made up of broken fragments are called clastic. A. much more useful classification of rocks, however, is based on a consideration of their origin or of the conditions under which they have been formed. Nearly all the crystalline rocks, like granite and gneiss, are found to have originally underlain the rocks which have been formed at the surface, like sandstones, clays, and the different kinds of lava ; while these latter do not as a rule occur underneath the highly crystalline rocks. The reason of this is, not that the crystalline rocks are necessarily older than the surface-formed rocks, but that they originated at considerable depths within the earth's crust. Hence we may distinguish all rocks as either liypogene or nether-formed rocks or epigene or surface -formed rocks. The epigene or surface-formed rocks include the aqueous rocks, formed by the action of water (of which class aerial or dEolian rocks, that is, materials accumulated by the action of wind, may be regarded as a subordinate group), and volcanic rocks pro- duced by igneous agencies operating at or near the surface. The hypogene or nether-formed rocks include materials, like granite and diorite, which must have crystallised from a molten condition at great depths below the surface, and are called plutonic rocks, and rocks, which though originally aqueous or igneous in origin, have undergone great changes, and often complete recrystallisation of their materials, and are therefore called metamorpliic, like gneiss and schist. We thus arrive at the following general tabulation of the rock masses of the globe : 2 Epigene (or surface-formed rocks) j Aqueous (and 2Eolian). Volcanic. Hypogene (or nether-formed rocks) Epigene or surface -formed rocks. The rocks of this class comprise materials the origin of which is obvious to us, living as we do upon the earth's surface. 1 Vitreous, hyaline, or glassy ' metamorpliic ' were proposed by rocks form a small class subordinate Lyell in 1833, in the first edition of to the crystalline rocks. the Principles of Geology. 2 The terms ' hypogene ' and 16 AQUEOUS ROCKS [CH. m. Aqueous rocks. The aqueous rocks, sometimes called also sedimentary or fossiliferous, cover a large part of the earth's surface, and they have evidently been formed under water. Some consist of mechanical deposits (pebbles, sand, and mud), and others are of organic origin, especially the limestones and coals. A few are of chemical origin, like rock-salt and gypsum. These rocks are usually stratified, or divided into distinct layers, or strata. The term stratum means simply a bed, or anything spread out or strewn over a given surface ; and we infer that these strata have been generally spread out by the action of water, from what we daily see taking place near the mouths of rivers, or on the land during temporary inundations. For, whenever a running stream charged with mud or sand has its velocity checked as when it enters a lake or sea, or overflows a plain the sediment, previously held in suspension by the motion of the water, sinks by its own gravity to the bottom. In this manner layers of mud and sand are thrown down one upon another. If we drain a lake which has been fed by a small stream, we frequently find a series of deposits at the bottom, disposed with considerable regularity, one above the other; the uppermost, perhaps, may be a stratum of peat, next below is a more dense and solid variety of the same material; still lower a bed of shell- marl, alternating with peat or sand, and then other beds of marl, divided by layers of clay. Now, if a second pit be sunk through the same continuous lacustrine formation at some distance from the first, nearly the same set of beds is met with, yet with slight variations; some, for example, of the layers of sand, clay, or marl, may be wanting, one or more of them having thinned out and given place to others, or sometimes one of the layers first examined is observed to increase in thickness to the exclusion of other beds. The term ' formation? which has been used in the above ex- planation, expresses in geology any assemblage of rocks which have some character in common, whether of origin, age, or composition. Thus we speak of stratified and un stratified, freshwater and marine, aqueous and volcanic, ancient and modern formations. In the estuaries of large rivers, such as the Ganges and the Mississippi, we may observe, at low water, phenomena analogous to those of the drained lakes above mentioned, but on a grander scale, and extending over areas several hundred miles in length and breadth. When the periodical inundations subside, the river hollows out a channel to the depth of many yards through horizontal beds of clay and sand, the edges of which are seen CH. in.] THEIR MODE OF FORMATION 17 exposed in perpendicular cliffs. These beds vary in their mineral composition, colour, and in the fineness or coarseness of their particles, and some of them are occasionally characterised by containing drift wood. At the junction of the river and the sea especially in lagoons, nearly separated by sand-bars from the ocean deposits are often formed in which brackish and salt-water shells are included. In Egypt, where the Nile has added to its delta by filling up part of the Mediterranean with mud, the newly deposited sedi- ment is stratified, the thin layer thrown down in one season differing slightly in colour from that of a previous year, and being separable from it, as has been observed in excavations at Cairo and other places. When beds of sand, clay, and marl, containing shells and vegetable matter, are found arranged in a similar manner in the interior of the earth, we ascribe to them a similar origin ; and the more we examine their characters in minute detail, the more exact do we find the resemblance. Thus, for example, at various heights and depths in the earth, and often far from seas, lakes, and rivers, we meet with layers of rounded pebbles, composed of flint, limestone, granite, or other rocks, resembling the shingles of a sea-beach, or the gravel in a torrent's bed. Such layers of pebbles frequently alternate with others formed of sand or fine sediment, just as we may see in the channel of a river descend- ing from hills bordering a coast, where the current sweeps down at one season coarse sand and gravel, while at another, when the waters are low and less rapid, fine mud and sand alone are carried seaward. If a stratified arrangement, and the rounded form of pebbles, are alone sufficient to lead us to the conclusion that certain rocks originated under water, this opinion is confirmed by the distinct and independent evidence of fossils, often very abundantly in- cluded in the earth's crust. By a fossil is meant any body, or the traces of the existence of an organic body whether animal or vegetable which has been buried in the earth by natural causes. Every stratum was the burial-ground of its time. Now the remains of animals, especially of aquatic species, are found almost everywhere embedded in stratified rocks, and sometimes, as in the case of many limestones, they are in such abundance as tovconstitute the entire mass of the rock itself. Shells and corals are the most frequent, and with them are often associated the bones and teeth of fish, fragments of wood, impressions of leaves, and other organic substances. Fossil shells, of forms such as now abound in the sea, are met with, far inland, both near the surface, and at great depths below it. They occur at c 18 JEOLIAN ROCKS [CH. in. all heights above the level of the ocean, having been observed at elevations of more than 8,000 feet in the Pyrenees, 10,000 in the Alps, 13,000 in the Andes, and above 18,000 feet in the Himalaya. 3 These shells belong mostly to marine forms of life, but in some places exclusively to forms characteristic of lakes and rivers. Hence it is concluded that some ancient strata were deposited at the bottom of the sea, and others in lakes and estuaries. Aerial or jEolian rocks did not attract much attention in the early days of geology, but it is evident that they are forming at the present time over large surfaces of the earth, and that this was also the case in former ages. Changes take place on the surface of the earth which cannot be attributed to move- ments by water, and deposits accumulate which are also not referable to that agent. The vast deposits of loess in Eastern Asia have been attributed to t blown dust; the desert sands of rainless regions, the sand dunes of many coasts and inland areas, and of the sides of lakes, are due to removal, by air in move- ment, of substances which have often been entirely eroded by atmospheric action and sometimes by water. Soils and thick deposits, like the laterite of Hindostan, are the result of chemical and other changes in the rocks exposed to the atmosphere. The accumulation of organic remains, both vegetable and animal in masses, often takes place without the intervention of an aqueous agency, and coal, peat, and some collections of bones, are examples of this action. Volcanic ash is wafted far arid wide by wind, and gives rise to important deposits, many of which were formed on dry land. Frost breaks up rocks, and the debris may accumulate without being subjected to the action of moving water. Moraine matter, the product of land glaciers, and the blocks carried by ice, or simply remaining as the relics of sub- aerial denudation, must be regarded as belonging to this group of aerial rocks. Many of these rocks, however, assume the stratified form, and contain organic remains. That aerial or JEolian rocks are not more commonly found among the stratified masses of the earth's crust is due to the circumstance that, before they can be covered up by marine deposits, such accumulations on the land must be subjected to the action of the waves and currents, and thus have their materials distributed to form ordinary aqueous masses. Volcanic rooks. The rocks which we may next consider are the volcanic, or those which have been produced at or near the surface, whether in ancient or modern times, by the action 3 Gen. Sir R. Strachey found Jurassic fossils at an altitude of 18,400 feet in the Himalaya. en. in.] VOLCANIC ROCKS 19 of heat ; such rocks are for the most part unstratified, and are devoid of fossils. Many volcanic rocks exhibit a parallel or banded structure, however ; and we find lava streams alternating with beds of scoriae and ash. These latter may sometimes be sorted while falling through air or water, and thus become stratified ; when accumulated in seas and lakes, deposits formed in this way may occasionally include fossils. The volcanic masses are more partially distributed than aqueous formations, at least in respect to horizontal extension. Among those parts of Europe where they exhibit characters not to be mistaken, may be mentioned not only Sicily and the country round Naples, but Auvergne, Velay, and Vivarais, now the departments of Puy-de-D6me, Haute-Loire, and Ardeche, towards the centre and south of France, in which are several hundred conical hills having the forms of modern volcanoes, with craters more or less perfect on many of their summits. Besides the parts of France above alluded to there are other countries, as the north of Spain, the south of Sicily, the Tuscan territory of Italy, the lower Rhenish provinces, Hungary, and many parts of Western America and Australia, where extinct volcanoes may be seen, still preserving, in many cases, a conical form, and having craters and often lava-streams connected with them. These cones are composed, moreover, of lava, scoriae, and ashes, similar to those of active volcanoes. Streams of lava may some- times be traced from the cones into the adjoining valleys, where they have choked up the ancient channels of rivers with solid rock, in the same manner as some modern flows of lava in Ice- land have been known to do the rivers either flowing beneath or cutting out a narrow passage on one side of the lava. Although none of the volcanoes of Central France have been in activity within the period of human history, their forms are often very perfect. There are some volcanoes, however, which have been compared to skeletons, in which rain, streams, and torrents have washed their sides, and removed all the loose sand and scoriae, leaving only the harder and more solid materials. By this erosion, their internal structure has occasionally been laid open to view, in fissures and ravines; and we then behold not only many successive sheets and masses of lava, sand, and porous scoriae, but also perpendicular walls or dikes, as they are called, of volcanic rock, which have burst through and filled up the cracks in the other materials. Such dikes may also be observed in the structure of Vesuvius, Etna, and other active volcanoes. There are also other rocks in almost every country in Europe, which we infer to be of igneous origin, although they do not c2 20 SYPOGENE EOCKS [CH. m. form hills with cones and craters. Thus, for example, we feel assured that the rock of Staffa, and that of the Giant's Cause- way, called basalt, is volcanic, because it agrees in its structure and mineral composition with streams of lava which we know to have flowed from the craters of recent volcanoes. We find also similar basaltic and other igneous rocks associated with beds of tuff in various parts of the British Isles and also forming dikes, such as have been spoken of; and some of the strata through which they cut are occasionally altered at the point of Contact, as if there had been an exposure to the intense heat of melted matter. The older writers were in the habit of calling the volcanic rocks of earlier geological periods by the name of ' trap rocks,' from the circumstance that the hills formed when such rock masses are denuded are apt to assume a terraced or step-like contour, from the Swedish trappa or stair. This term is now, however, seldom employed by geologists. The absence of cones and craters, and of long narrow streams of superficial lava, in England and many other countries is to be attributed to the circumstance that, owing to the long period which has elapsed since their eruption, all the loose accumulations have been swept away by the action of rain and rivers. But this question must be enlarged upon more fully in the chapters on igneous rocks, in which it will also be shown that, as different sedimentary formations, containing each their characteristic fossils, have been deposited at successive periods, so also volcanic dust and scoriae have been thrown out, and lavas have flowed over the land or bed of the sea, or have been injected into fissures, at many different epochs ; so that the igneous as well as the aqueous and aerial rocks may be classed as a chrono- logical series of monuments, throwing light on a succession of events in the history of the earth. Hypogene or nether-formed rocks. If we examine a large portion of a continent, especially if it contain within it a lofty mountain range, we rarely fail to discover two other classes of rocks, very distinct from either of those above alluded to, and which we can neither assimilate to deposits such as are now accumulated in lakes and seas nor to those generated by ordinary volcanic action. The members of both these classes of rocks agree in being highly crystalline and destitute of organic remains. The rocks of one class have been called plutonic, comprehending all the granites, diorites, gabbros, and certain ' porphyries,' which are allied in some of their characters to volcanic rocks. The members of the other class are more or less perfectly foliated or schistose in structure. They are the gneisses and crystalline schists, or metarnorphic rocks in which CH. in.] PLUTONIC AND METAMORPHIC 21 group are included gneiss, mica-schist, hornblende -schist, sta- tuary marble, and other rocks afterwards to be described. Plutonic rocks. As it is admitted that nothing strictly analogous to these crystalline rocks can now be seen in the progress of formation on the earth's surface, it will naturally be asked on what data we can find a place for them in a system of classification founded on the origin of rocks. It may be stated, as the result of careful study, that the various kinds of rocks, such as granite, diorite, and gabbro, which constitute the plutonic family, are of igneous or aqueo-igneous origin, and have been formed under great pressure, at a considerable depth in the earth. The Germans speak of these rocks as Tiefengesteine, while the French geologists apply to them the name of * roches de profondeur.' Like the lava of volcanoes, they have been melted, and have afterwards cooled and crystallised but with extreme slowness, and under conditions very different from those producing the volcanic rocks. Hence they differ from the volcanic rocks, not only by their more crystalline texture, but also by the absence of tuffs and breccias, which are the products of eruptions at the earth's surface, or beneath seas of inconsiderable depth. They differ also by the absence of those pores or cavities, to which the expansion of the entangled gases and steam gives rise in ordinary lava. Metamorphic rocks. The last great division of rocks in- cludes the foliated crystalline rocks and schists, called gneiss, mica-schist, chlorite-schist, talc-schist, quartzite, marble, and the like, the origin of which is more doubtful than that of the other classes. They rarely contain either pebbles, or sand, or scoriae, or angular pieces of embedded stone, or traces of organic bodies, and they are often as crystalline in their structure as granite itself ; they sometimes form bed-like masses, somewhat similar in form and arrangement to those of sedimentary formations. The bands or 'folia' of which they are made up consist of alternations of minerals varying in colour, composition, and thickness. According to the Huttonian theory, which is here adopted as the most probable, and which will be afterwards more fully explained, the materials of these rocks were originally deposited from water in the form of sediment, or thrown out from volcanoes as lava or dust, or consolidated beneath volcanoes as plutonic masses; but they have been subsequently so altered by heat, chemical action, and pressure, as to assume a new texture, and acquire a new mineral composition. It is demon- strable, in some cases at least, that such a complete conversion has actually taken place, fossiliferous strata having exchanged a.n earthy for a. highly crystalline texture for the distance of a 22 ROCK-FORMING MINERALS [CH. in. quarter of a mile from their contact with granite. In some cases dark limestones, replete with shells and corals, have been turned into white statuary marble, and hard clays, con- taining vegetable or other remains, into rocks approaching in character to mica- schist, every vestige of the organic bodies having been obliterated. In accordance with the hypothesis above alluded to, it was proposed in the first edition of the ' Principles of Geology ' (1833) to employ the term ' Metamorphic ' for the altered strata, the word being derived from /zero, meta, trans, and p.op(f>r), morphe, forma. From what has now been said, the reader will understand that each of the great classes of rocks may be studied from two distinct points of view. First, they may be regarded simply as mineral masses owing their origin to particular causes, and having a certain chemical composition, form, and position in the earth's crust, or exhibiting other characters, such as the presence or absence of organic remains. In the second place, the rocks of each class may be viewed as constituting a grand chronological series of monuments attesting a long succession of events in the former history of the globe and of its living inhabitants. We shall accordingly proceed to treat of each class of rocks ; first, in reference to those characters which are not chronolo- gical, and then in particular relation to the several periods when they were formed. If we desire to make a more special classification of rocks, it is necessary to determine the species of minerals of which they are built up, and the relations of these minerals to one another. Except in the case of some coarse-grained rocks, this can only be done by preparing thin transparent sections of the rock. Such transparent sections of rocks are produced by grinding down one side of a rock-fragment to a smooth and polished surface, cementing it upon a piece of glass, and then grinding away the exposed portion of the rock till nothing but a thin film remains. By the use of a lapidary's wheel and other apparatus, specially devised for the purpose, the work of making such rock-sections may be greatly facilitated. The characters of the chief rock-forming minerals are described in Appendix A. For works in which rocks are and the Treatises on Petrography, systematically described, the stu- published by Von Lasaulx, Zirkel, dent is referred to Mr. Barker's 'Pe- and Rosenbusch in Germany, and trology for Students,' Mr. Rutley's by Fpuc[ue, Michel Levy, and 'Granites and Greenstones,' Dr. Lacroix in France, and by Iddings, Hatch's ' Text Book of Petrology,' Pirsson, and others in the United Mr. Teall's 'British Petrography,' States. PART II AQUEOUS BOCKS SECTION I. GENERAL RELATIONS OF THE STRATIFIED BOCKS CHAPTER IV COMPOSITION AND CLASSIFICATION OF AQUEOUS ROCKS Chemical, mechanical, and organic deposits Arenaceous rocks Argilla- ceous rocks Calcareous rocks Other varieties of aqueous rocks Phosphatic deposits Ironstones Gypsum Rock salt Carbonaceous deposits Peat Coal Anthracite. IN pursuance of the arrangement explained in the last chapter, we shall begin by examining the aqueous (and aerial) or sedi- mentary rocks, which are for the most part distinctly stratified and often contain fossils. We may first study them with reference to their mineral composition, external appearance, position, mode of origin, organic contents, and the other characters which belong to them as sedimentary formations independently of their age ; and we may afterwards consider them chronologically or with reference to the successive geological periods in which they originated. We have already given an outline of the data which led to the belief that the stratified and fossiliferous rocks were originally, with rare exceptions, deposited under water ; but, before entering into more detailed investigations, it will be desirable to say something of the ordinary materials of which such strata are composed. They may be said to belong prin- cipally to three divisions the arenaceous, the argillaceous, and the calcareous. Of these the arenaceous are chiefly made up of sand or siliceous grains ; and the argillaceous of clays or compounds of silica, alumina, and water ; while the calcareous rocks consist of calcium carbonate, with sometimes magnesium carbonate also. Chemical, mechanical, and organic deposits. A dis- tinction has been made by geologists between deposits of a 24 MECHANICAL, CHEMICAL, [CH. iv. mechanical and those of a chemical origin. Under the term mechanical deposits are designated beds of mud, sand, or pebbles, produced by the action of running water, as well as accumula- tions of lava, fragments, scoriae, and dust thrown out of a volcano. These materials have been held in suspension in water or air, and have acquired their present disposition through the action of gravity. But the matter which forms a chemical deposit has not been mechanically suspended in water but held in solution in the water till separated from it by chemical action. In this way calcium carbonate is sometimes precipitated in a solid form around springs, as may be well seen in many parts of Italy. In these springs the calcium carbonate is usually held in solution by an excess of carbon dioxide dissolved in the water ; and, on the water escaping from the earth, the excess of gas passes off into the air, causing the dissolved calcareous matter to separate and be deposited on shells, fragments of wood, leaves, &c., encrusting and binding them together. The rock thus formed is called 'Travertine' (Tiber stone). Caves often have 'stalactites,' or pendent icicle- like masses, projecting from their roofs with layers on their floors (' stalagmite '), and these calcareous sub- stances are in process of formation at the present time. Bain- water which has taken up carbon dioxide from the air, percola- ting through limestone rocks (in which caves are so often formed), takes up a certain amount of the calcium carbonate, and forms a soluble bicarbonate. The water thus charged drops from the roof, and gives off some carbon dioxide to the air, a corresponding amount of calcium carbonate being set free to form the pendent stalactites. The excess of water which drops on the floor of the cave, in some instances, gives off more carbon dioxide, and a further precipitation of calcium carbonate takes place to form the layers of stalagmite. There is, how- ever, reason for believing, as shown by Professor Cohn, that in nearly all cases in which travertine is formed an important part is played by vegetable organisms; these extract carbon dioxide from the water and thus facilitate the precipitation of the calcium carbonate. No similar travertine ever appears to be formed upon the bed of the ocean, for, as a general rule, the quantity of calcium carbonate diffused through sea-water is so minute that direct chemical precipitation cannot take place. The separation of calcium carbonate from sea-water and the fresh water of many lakes and rivers appears to be due entirely to vital agency. Many plants and animals have the power of taking up from water the minute proportions of calcium carbonate, calcium phosphate, silica, &c., which it contains, and of building CH. iv.] AND ORGANIC DEPOSITS 25 these materials into their tissues. On the death of the organ- isms, the solid skeletons remain behind to form great rock- masses. In this way chalk and other forms of foraminiferal rock, various coral and shell-rocks, as well as bone-beds and certain siliceous deposits, are formed. Rocks thus produced by the action of vital agencies are known as organic deposits. Arenaceous rocks (psammites of some authors). These consist of masses of loose sand or of coarser materials which may become cemented together so as to form rocks of great hardness. We find many varieties dependent on the form and size of the constituent grains, the nature of the minerals forming the grains, and the substances by which the grains are bound together. Most sands are composed of grains of quartz. These are sometimes perfectly angular, at other times subangular, and not unfrequently completely rounded into microscopical pebbles (' millet-seed ' sand}. There is reason to believe that all per- fectly rounded sand-grains have at some period of their history been subjected to the action of the wind. The grains of sand found in deserts which have been acted upon by the wind are usually rounded and polished ; and both Daubree and J. A. Phillips have shown that but little rounding takes place in fine particles of quartz when suspended in water. We occasionally find sands made up of grains which have the external form of quartz crys- tals (crystalline sands and sandstones). It has been shown by Sorby and others that the original form of these sand-grains was irregular, and that their beautiful crystalline faces have been acquired by the deposition upon them of silica held in solution. In this way the fragments of old quartz crystals become enlarged and have their crystalline forms restored to them. By the aid of the microscope we can, indeed, see the old sand-grain lying in the midst of the crystal of quartz which has enveloped it. The sandstone of Penrith is a beautiful example of a crystalline sandstone. The red, brown, yellow, and other tints exhibited by sands are usually due to thin films of iron oxide more or less hydrated which have enveloped them. By the action of acids these surface films may be re- moved and a white or colourless sand left behind. Sands mingled with water are known as running or quick sands ; when moved about by the air they are called blown sands and form rounded hills (dunes). Dry sands sometimes give out a distinct note when struck or walked upon (musical sands). The sound produced by these musical sands appears to be due to a great number of particles of uniform size rubbing or striking against one another. Desert sands are 26 ARENACEOUS KOCKS [CH. iv. largely made up of well rounded and polished particles of quartz. Sands are usually composed of particles of the mineral quartz or crystallised silica. Quartz is a very abundant and a very hard mineral, which has no tendency to split up into thin flakes, or, as the mineralogist says, it has no cleavage, and it is for these reasons that the great bulk of most sands is made up of quartz grains. Other minerals, however, often enter, sometimes very largely, into the composition of sands. Thus the fragments of quartz, felspar, and mica formed by the dis- integration of a granite may accumulate to form a granitic sand. Such granitic sand when reconsolidated forms the rock known as arkose. . Many sandstones contain a considerable proportion of particles of felspar, and these are known as felspathic sand- stones, or greywacke. Other sandstones contain much mica, generally disposed along the planes of bedding, and are called micaceous sandstone and flagstone. Barer minerals are found in sands and sandstones, by sifting out the minuter and heavier particles and separating these, according to their density, by dropping them into heavy liquids. Zircons, garnets, tourma- lines, and many other minerals are thus shown to be often present in these rocks. By the study of sand and sandstones under the microscope, it is often possible to determine the nature of the rocks from which the loose materials have been derived by the action of denudation. Sands differ much in the size of the grains of which they are made up. When the grains are very coarse many authors speak of the rock as a grit ; but this name is given by other geologists to sandstones made up of angular grains. Hocks made up of loose fragments of all sizes, usually siliceous, are called gravels, and these are distinguished as angular, sub- angular, or pebbly, according to the degree of rounding of the fragments. Pebbly gravels, when consolidated into hard rocks, are known as conglomerates or pudding stones ; angu- lar fragments, when consolidated, form breccias. The sili- ceous particles of arenaceous rocks are sometimes bound together by calcareous matter (calcareous sandstones and calcareous grits] ; at other times iron oxide forms the cementing material, giving rise to what are known as car- stones. Most usually, however, the cementing material in arenaceous rocks is silica, either partially or wholly crys- tallised. In such rocks the original boundaries of the constitu- ent grains may sometimes be made out under the microscope, but are not unfrequently wholly lost. In this way the sand- stone is found insensibly passing into the rock known as quartz- CH. iv.] ARGILLACEOUS ROCKS 27 rock or quartzite. The rocks known as grey-wethers or sar sen- stones are composed of sand cemented into a hard rock by silica deposited between the grains. Various foreign admixtures may be found in sands and sandstones. When sand is largely mingled with argillaceous matter it is called a loam but this term is more employed by agriculturists than by geologists. Sandstones may contain particles of silicates (glauconite &c.), usually of a green colour, which have been deposited in the interiors of organisms. These ' greensands ' have sometimes been called ' chloritic sands ' and * glauconitic sands,' but neither name is very appro- priate. The presence of other kinds of foreign materials gives rise to carbonaceous, ferruginous, or argillaceous sands and sandstones. Argillaceous rocks (pelites of some authors) include all the varieties of mud, clay, and their hardened representatives, such as shale and clay -slate. These rocks are composed essen- tially of silicate of alumina, with varying quantities of water. The purest clay is kaolin, or porcelain clay, which contains 46 per cent, of silica, 40 per cent, of alumina, and 14 per cent, of water. In Fuller's earth the proportion of silica is higher and of alumina less, but the material contains 80 per cent, of water and considerable quantities of other substances (iron oxide, lime, and magnesia), which may be regarded as im- purities. Most clays probably consist of these and other hydrated silicates of alumina mingled with minute fragments of many other minerals. On account of the minuteness of the mineral particles which compose them, it is often difficult, even with the highest powers of the microscope, to make out the mineralogical constitution of clays. Many of the hydrated silicates of alumina form crystalline scales like mica, and these can be detected by the microscope. By carefully washing clays in water, fine needles of rutile (oxide of titanium) and fragments of other minerals may be isolated. Most clays exhibit the important property of plasticity, which renders them so valuable for making bricks, tiles, and various kinds of pottery. One general character dis- tinguishing the argillaceous rocks is that of giving out a peculiar earthy odour when they are breathed upon. Pipe clays are white clays nearly free from the hydrated oxides of iron which communicate red, yellow, and brown colours to most argillaceous rocks. Many varieties of clay when dug at some depth from the surface have a dark-blue colour, which is due, as was shown by Ebelmen and Church, to the presence of finely divided iron disulphide (iron pyrites). Fire-clays or refrac- tory clays contain a considerable amount of uncombined silica, 28 ARGILLACEOUS BOOKS [cir. iv. which makes them difficult to fuse ; such clays are used for making crucibles and lining furnaces. Clays frequently contain large quantities of foreign matter, and are known as car- bonaceous, micaceous, sandy, or ferruginous clays. Clays con- taining much calcareous matter are properly called marls ; but this name is often incorrectly applied to true clays containing little or no calcium carbonate. Hardened clays which are not fissile are often called mud* stones. When induration is accompanied with the development of a laminated structure along the planes of bedding, the rock is called a shale. Some carbonaceous shales yield hydrocarbons when subjected to distillation, and these are known as oil-shales. Torbanite is a valuable oil-shale found in the carboniferous rocks in the south of Scotland. Near great igneous masses argillaceous rocks pass into a material of great hardness, den- sity, and fineness of grain, which is called flinty-slate, Lydian stone (Lydite), and porcellanite or argillite. Some of the argillaceous rocks which have been altered by the contact of great igneous masses are found to be filled with microscopic crystals of garnets and other hard minerals. In consequence of the presence of these the rocks are employed for grinding and polishing purposes (ivhetstones, novaculites). In other cases, larger but ill-defined crystalline particles separate in such rocks, giving rise to what are known as spotted slates. When distinct minerals like chiastolite, ottrelite, &c., can be made out with the naked eye, the rocks are called chiastolite slate, ottrelite slate, &c. Some argillaceous rocks split up along planes distinct from the planes of bedding. These rocks constitute slates or clay- slates. When minerals like mica, talc, chlorite, &c., are developed along the planes of separation or cleavage, the clay-slates pass into what are called phyllites, or, as they are often railed by English writers, mica-slate, talc-slate, chlorite-slate, &c. These rocks constitute a transition between the classes of aqueous and metamorphic rocks. Calcareous rocks (or limestones) consist of calcium car- bonate often combined with more or less magnesium carbonate. They are usually of organic, but occasionally of chemical origin. When composed of calcium carbonate they effervesce freely when a drop of dilute acid is placed upon them. If the geologist finds it inconvenient to carry a bottle of liquid acid in the field, he may use solid substances like phosphoric, oxalic, and citric acids, adding a drop of water. When the quantity of magnesium carbonate in a rock is large, the effervescence with a.cid ip decidedly less brisk ; such rocks are called magnesian. CH. iv.] CALCAREOUS KOCKS 29 limestones. When we have the definite compound of the mag- nesium and calcium carbonates known as dolomite, we get no effervescence at all with cold dilute acid. Even dolomites, however, effervesce and dissolve when the acid is warmed. When limestones are heated they give off the carbon dioxide, and anhydrous calcium oxide (quick-lime) is left behind. If water be added to the quick-liine a hydrated calcium oxide (slaked lime) is formed with great evolution of heat. Travertine, and its varieties stalactites and stalagmite, have already been mentioned as examples of chemically formed lime- stones. Pisolite and oolite (roestone) are made up of rounded grains composed of concentric coats of calcium carbonate enve- loping a fragment of shell or other foreign substance. Recent studies point to the conclusion that the formation of all these substances is not due to chemical action alone, but that various lowly vegetable organisms play an important part in removing the excess of carbon dioxide in the water, and causing the de- position of the calcium carbonate. Most of the limestone rocks found in the earth's crust are undoubtedly of organic origin, and are built up of the remains of various plants (calcareous algae), or of the skeletons of animals, such as foraminifera, corals, bryozoa, mollusca, &c. Some organisms have their skeletons composed of calcium carbonate in the form of the mineral calcite, others in the form of the mineral aragonite, while some skeletons are made up of both these minerals. Aragonite is an unstable mineral, and cal- cite a stable one, but the former may be converted into the latter. Organic structures composed of aragonite are either dissolved away (leaving empty casts) or are converted into ' pseudornorphs ' of calcite. Chalk is a soft foraminiferal limestone. Other limestones made up of foraminifera are the nummulitic limestones, the orbitoidal limestones, the fusulina limestones, &c. Entro- chial limestones are made up of the stems of crinoids; and various kinds of shell limestones consist of the remains of different species of mollusca ; limestones made up of bryozoa (like the so-called ' coralline crag ') have also received dis- tinctive names. Oolite limestones are made up of small rounded grains, like the roe of a fish. When the grains are of larger size approaching that of a pea the rock is called a pisolite (' pea-grit '). Eecent investigations tend to show that oolites and pisolites probably owe their formation to the action of minute aquatic plants (algae). In thin sections, oolitic and pisolitic grains are seen to exhibit a remarkable concentric and radiated structure. Rocks made up 30 OOLITIC LIMESTONES [OH. iv. of oolitic grains are found of all ages, and similar rocks are being formed at the present day in the coral reefs of the Bahamas and in the Great Salt Lake of Utah. Fig. 2. Fig. 3. Fig. 2. Section of oolitic granules, the formation of which can be seen ut the present day, x 70. The four figures on the left show both radiated and con- centric structure. These specimens were obtained from the Great Salt Lake of Utah. Their origin is ascribed by Dr. Rothpletz to the action of minute alga?. The figures on the right are from oolitic grains found on the coral- reefs of the Bahamas. They exhibit a concentric structure and are developed around nuclei, which may be grains of sand, foraminiferal shells, or other minute objects. The grain partially seen at the top on the right shows a number of branching tubes formed by burrowing algas which are found perforating all calcareous organisms. Fig. 3. Section of an oolitic limestone from near Bath, also x 70. In general characters the oolitic granules agree with those formed at the present day, but they are bound together by crystalline calcite. Many limestones contain foreign substances ; and thus we get argillaceous, ferruginous, siliceous, and sandy limestones, car- bonaceous, glauconitic, and pyritous limestones, &c. The cal- cium carbonate is often more or less crystallised, and when sufficiently hard to bear polishing the rock is called ' marble.' When completely crystallised we get either the pure white statuary or saccharoid limestone, or a similar material coloured by various foreign minerals which are present as impurities. Other varieties of aqueous rocks. In addition to the three principal classes of aqueous rocks which pass into one another by insensible gradations, we find several other materials present in much smaller quantities as stratified masses. Beds of calcium phosphate, often made up of bones and teeth (bone-beds), occur, but are of limited thickness and extent. Beds of iron carbonate or of iron oxide, with or without water, are by no means rare ; the ferruginous materials being variously combined with calcareous, argillaceous, and arenaceous substances. In most cases it can be shown that the ferrous carbonate has replaced calcium carbonate in the rock, even the en. iv.J IRONSTONES, GYPSUM, SALT, &c. 31 remains of shells and other calcareous organisms being con- verted into iron carbonate. Some of these rocks, as at Cleve- land in Yorkshire, and Scunthorpe in Lincolnshire, form very valuable iron ores. Rocks which once consisted of ferrous car- bonate are often found converted into the brown hydrated ferric oxide. Gypsum, or hydrated calcium sulphate, forms beds of con- siderable extent. When crystalline or nearly compact it forms the ornamental stone known as alabaster, which is distinguished from marble by its much greater softness. In clays exposed to the action of the weather, crystals of gypsum (selenite) are often formed by sulphuric acid, produced by the oxidation of pyrites, coming into contact with the calcium carbonate of fossil shells. Beds of anhydrite, which is gypsum without the water, also occur in some places. Bock-salt and some allied substances are found in extensive beds in certain places. Lastly, deposits of peat, lignite, coal, anthracite, and graphite, with others of cannel coal, and solid and liquid hydrocarbons, are found in layers sometimes of considerable thickness ; while the whole substance of porous rock-masses may be impregnated with various liquid and gaseous hydrocarbons. Varieties of coal. Ordinary coal is more or less amorphous ; it only occasionally shows something of a fibrous structure, and it has a tendency to cleave in cubical or prismatic blocks. The divisional planes often contain small films of calcite, gypsum, and iron pyrites. The coals spoken of as ' bituminous ' are those which soften or fuse when heated at a less temperature than that required for combustion ; it must be remembered, however, there is nothing like bitumen in coal, and the proportion of carbon in such coals is from 80 to 90 per cent., of hydrogen 4-5 to 6 per cent., and oxygen 8 to 14 per cent. It appears, from the researches of Liebig and other eminent chemists, that when wood and vegetable matter are buried in the earth exposed to moisture, and partially or entirely excluded from the air, they decompose slowly and evolve carbon dioxide gas, thus parting with a portion of their original oxygen. By this means they become gradually converted into lignite or wood- coal, which contains a smaller proportion of hydrogen and oxygen than wood. A continuance of decomposition changes this lignite into common or bituminous coal, chiefly by the escape of carburetted hydrogen, or the gas by which we illuminate our streets and houses. According to Bischoff, the inflammable gases which escape from coal, and are so often the cause of fatal 32 COMPOSITION OF COAL [CH. IV. accidents in mines, always contain carbon dioxide, carburetted hydrogen, and nitrogen. The disengagement of all these gra- dually transforms ordinary or bituminous coal into anthracite. The chemical composition of the several varieties of coal, with their relations to one another and to the vegetable tissues out of which they are formed, are illustrated in the following tables, which are based on data collected by Prof. Thorpe. MEAN COMPOSITION OF CARBONACEOUS DEPOSITS, THE ASH BEING DEDUCTED Wood Humus Peat Lignite Brown coal Caking coal Steam coal Anthra- cite Carbon 50-2 54-8 60-8 67-4 72-8 80-5 86-5 95-2 Hydrogen . 6-2 4-8 5-9 5-6 5-4 5-3 5-2 2-5 Oxygen and Nitrogen . 43'6 40-4 33-3 27-0 21-8 14-2 8-3 2-3 That the conversion of vegetable tissues into peat and coal and thence into anthracite is brought about by a diminution in the quantity of hydrogen, oxygen, and nitrogen, and an increase of the residual carbon, is shown by the following table, in which the proportion of the gaseous constituents to the carbon is cal- culated, the ash being omitted : Specific gravity Carbon Hydrogen Oxygen and Nitrogen Wood (average) . 0-50 100 12-3 86-8 Peat (average) . * 0-85 100 9-7 54-7 Lignite (average) 1-04 100 8-3 40-0 Brown coal (average) . 1-15 100 7.4 29-7 Common coal (average) 1-30 100 6-4 13-4 Anthracite (average) . " * 1-50 100 2-6 2-3 Graphite (average) . . 2-20 100 It must be remembered, however, that while the oxygen, hydrogen, and nitrogen are passing off, a portion of the carbon goes too, not only water and ammonia being formed but carbon dioxide and various hydrocarbons. The gaseous elements, how- ever, pass off at a greater rate than the carbon, and thus the proportion of the latter element is being continually augmented in the residual mass. The existence of occasional seams of coal almost wholly made up of the macrospores and microspores of the great cryptogams of the period will be noticed in the sequel ; such beds occur in the Yorkshire and Leicestershire Coal-fields, and in many other districts. (See p. 61, fig. 66.) The composition of the nearest modern representatives of CH. IV.] ORIGIN OF COALS, &c. 33 the coal-measure plants, and of their spores, is compared with that of the spore-coals in the following table : Lycopods Lycopod spores ' Better bed ' spore coal Carbon Hydrogen . Oxygen and Nitrogen . Ash . 46-8 6-2 42-1 4-9 61-5 8-4 27-7 2-4 85-1 3-4 5-2 6-3 There is an intimate connection between the extent to which the coal has in different regions parted with its gaseous contents, and the amount of disturbance which the strata have undergone. In the eastern part of the South Wales Coal-field we find beds of ordinary or ' bituminous ' coal, which further west are replaced by the harder coals containing a higher proportion of carbon and a smaller percentage of oxygen and hydrogen, and constituting the well-known ' steam-coals ' of the district. Further west, in Pembrokeshire, where the disturbance of the strata has been very great, we find the coals replaced by beds of anthracite, in which almost all traces of oxygen and hydrogen have disappeared. In Pennsylvania, the strata of coal are horizontal to the westward of the Appalachian Mountains, where Professor H. D. Rogers pointed out that they were most bituminous ; but as we travel south-eastward, where they no longer remain level and unbroken, the same seams become progressively 'debituminised' in proportion as the rocks become more bent and distorted. At first on the Ohio River the proportion of hydrogen, oxygen, and other volatile matters, ranges from forty to fifty per cent. Eastward of this line, on the Monongahela, it still approaches forty per cent., where the strata begin to experience some gentle flexures. On entering the Appalachian Mountains, where the distinct anticlinal axes begin to show themselves, but before the dislocations are considerable, the volatile matter is generally in the proportion of eighteen or twenty per cent. At length, when we arrive at some isolated coal-fields associated with the boldest flexures of the Appalachian chain, where the strata have been actually turned over, as near Pottsville, we find the coal to contain only from six per cent, of volatile matter, thus becoming a genuine anthracite. Besides the general descriptions of the different varieties of aqueous rocks in the several petrographical works already referred to, the stu- dent will find much valuable in- formation in the addresses of Mr. Sorby to the Geological Society in 1879-80. He will also do well to consult the memoir on ' Oceanic Deposits,' forming one of the volumes of the 'Reports of the Challenger Expedition.' P 34 STKATIFICATION [CH. v. CHAPTER V STRUCTURES PRODUCED IN AQUEOUS ROCK-MASSES DURING THEIR DEPOSITION Forms of stratification Original horizontally of strata False bedding or oblique lamination Irregularities in the accumulation of strata* Thinning-out and alteration in the characters of strata Ripple marks, sun-cracks, footprints, tracks, trails, burrows, and worm-casts. WHEN we study a rock-mass of aqueous origin, we find that it presents certain characters which must be the result of causes acting while its materials were being accumulated, and other features which are as certainly the consequence of changes that have taken place in the rock long subsequently to its deposition. It is the first -mentioned class of characters which we propose to consider in the present chapter. Forms of stratification. A series of strata sometimes con- sists of one of the varieties of rocks mentioned in the preceding chapter, sometimes of two or more kinds in alternating beds. Thus, for example, in the coal districts of England, we often pass through various beds of sandstone, some of finer, others of coarser grain, some white, others of a dark colour, and below these, alternating layers of shale and sandstone or beds of shale, divisible into leaf- like laminae, and containing beautiful impres- sions of plants. Then again we meet with beds of pure and impure coal, also alternating with shales and sandstones, and underneath the whole, perhaps, are beds of limestone, filled with corals and marine shells, each bed distinguishable from the others by certain fossils, or by the abundance of particular species of shells or zoophytes. This alternation of different kinds of rock produces the most distinct stratification ; and we often find beds of limestone and marl, conglomerate and sandstone, sand and clay, recurring again and again, in nearly regular sequence, throughout a series of many hundred strata. The causes which produce these phenomena are various, and may be either changes in the nature and degree of fineness of the material deposited, or interruptions in the regular course of deposition, when the layer first formed may have had time to consolidate before the next layer was spread over it, thus causing an imperfect adhesion between successive strata of the same composition. Rivers flowing into lakes and seas are found to be charged with sediment, varying in quantity, composition, colour, and grain according CH. v.] LAMINATION 35 to the seasons ; the waters are sometimes flooded and rapid, at other periods low and feeble. Different tributaries, also, draining peculiar countries and soils and therefore charged with peculiar sediment are swollen at distinct periods ; but all these different kinds of sediment will be deposited successively over the same area. The waves of the sea and currents also undermine the cliffs, during wintry storms, and sweep away the materials into the deep, after which a season of tranquillity succeeds, when nothing but the finest mud is spread by the movements of the ocean over the same submarine area. It is not the object of the present work to give a description of these operations, repeated as they are year after year and century after century ; but we may explain by way of illustration the manner in which some micaceous sandstones have origi- nated, namely, those in which we see thin layers of mica dividing layers of fine quartzose sand. This arrangement of materials may be observed in recent mud deposited upon the shore near La Roche St. Bernard in Brittany, at the mouth of the Loire. The surrounding rocks are of gneiss, which, by its waste, supplies the mud ; when this dries, it is found, at low water, to consist of brown laminated clay, divided by thin seams of mica. The separation of the mica in this case, or in that of micaceous sandstones, may be illustrated in the following manner. If we take a handful of quartzose sand, mixed with mica, and throw it into a clear running stream, we see the materials immediately sorted by the moving water, the grains of quartz falling almost directly to the bottom, while the plates of mica take a much longer time to sink through the water, and are carried farther down the stream. At the first instant the water is turbid, but almost im- mediately the flat surfaces of the plates of mica are seen all alone, reflecting a silvery light as they descend slowly, to form a distinct micaceous lamina. Although the mica is the heavier mineral of the two, it remains a longer time suspended in the fluid, owing to its greater extent of surface. It is easy, therefore, to perceive that where such mud is acted upon by a river or tidal current, the thin plates of mica will be carried farther, and not deposited in the same places as the grains of quartz ; and since the force and velocity of the stream varies from time to time, layers of mica or of sand will be thrown down successively on the same area. Original horizontallty. It is said generally that the upper and under surfaces of strata, or the ' planes of stratification,' are parallel. Although this is not strictly true, they make an approach to parallelism, for the same reason that sediment is usually deposited at first in nearly horizontal layers, whatever 86 IRREGULARITIES IN STRATIFICATION [CH. V. Fig. 4. may be the state of the floor on which the deposit rests. Yet if the sea should go down, as when there is very low tide, near the mouth of a large river where a delta has been forming, we see extensive plains of mud and sand laid dry, which, to the eye, appear perfectly level, although, in reality, they slope gently from the land towards the sea. This tendency in newly formed strata to assume a horizontal position arises principally from the motion of the water, which forces particles of sand or mud over the bottom, and causes them to settle in hollows or depressions where they are less exposed to the force of a current than when they are resting on elevated points. The velocity of the current and the motion of the superficial waves diminish from the surface downwards, and are least in those depressions where the water is deepest. A good illustration of the principle here alluded to may be sometimes seen in the neighbourhood of a volcano, when a sec- tion, whether natural or artificial, has laid open to view a succession of various- coloured layers of sand and ashes, which have fallen in showers upon uneven ground. Thus let A B (fig. 4) be two ridges with an intervening valley. These original inequalities of the surface have been gradually effaced by beds of sand and ashes, e, d, e, the surface at e being quite level. Now, water in motion can exert this levelling power on similar materials more easily than air, for almost all stones lose in water more than a third of the weight which they have in air, the specific gravity of rocks being in general as 2 when compared with that of water, which is taken as 1. But the buoyancy of sand or mud would be even greater in the sea, as the density of salt water exceeds that of fresh. Yet, however uniform and horizontal may be the surface of new deposits in general, there are still many disturbing causes, Fig. 5. Section of strafa of sandstone, grit, and conglomerate. such as eddies in the water, and currents moving first in one and then in another direction, which frequently cause irregu- larities. We may sometimes follow a bed of limestone, shale, CUBBENT-BEDDING 87 or sandstone for a distance of many hundred yards continuously, but we generally find that, sooner or later, each individual stratum thins out, and allows the beds which were previously above and below it to meet. If the materials are coarse, as in grits and conglomerates, the same beds can rarely be traced many yards without varying in size, and often rapidly thinning out and coming to an end (see fig. 5). False bedding or oblique lamination. There is also another phenomenon of frequent occurrence in stratified masses. We find a series of larger strata, each of which is composed of a number of minor layers placed obliquely to the general planes of stratification. To this diagonal arrangement the name of ' false or cross bedding ' or ' oblique lamination ' has been given. Thus in the annexed section (fig. 6) we see many beds of loose Fig. 6. False bedding in Great Oolite. After Jukes- Brown. sand, yellow and brown, and some of the principal planes of stratification are nearly horizontal. But the greater part of the subordinate laminae do not conform to these planes, but have often a steep slope, the inclination being sometimes to- wards opposite points of the compass. When the sand is loose and incoherent, as in the case here represented, the deviation from parallelism of the slanting laminae cannot possibly be accounted for by any rearrangement of the particles acquired during the consolidation of the rock. In what manner, then, can such irregularities be due to original deposition ? We must suppose that at the bottom of shallow seas, as well as in the beds of rivers, the motions of waves, currents, and eddies often cause mud, sand, and gravel to be thrown down in heaps on par- ticular spots instead of being spread out uniformly over a wide 38 CONTEMPORANEOUS EROSION OH. v. area. Sometimes, when banks are thus formed, currents may cut passages through them, just as a river forms its bed. Suppose the bank A (fig. 7) to be thus formed with a steep sloping side, and, the water being in a tranquil state, the layer of sediment Fig. 7. No. 1 is thrown down upon it, conforming nearly to its surface. Afterwards the other layers, 2, 3, 4, may be deposited in succes- sion, so that the bank B C D is formed. If the current then increases in velocity, it may cut away the upper portion of this mass down to the dotted line e, and deposit the materials thus Fig. removed farther on, so as to form the layers 5, 6, 7, 8. We have now the bank B C D E (fig. 8), of which the surface is almost level, and on which the nearly horizontal layers, 9, 10, 11, may then accumulate. It was shown in fig. 6 that the diagonal layers of successive strata may sometimes have an opposite slope. This is well seen in some cliffs lg ' of loose sand on the Suf- ^"~^ j ^ r _HI folk coast. A portion of one of these is represented in fig. 9, where the layers, of which there are about six in the thickness of an inch, are composed of quartzose grains. This ar- rangement may have been due to the altered direction of the tides and currents in the same place. Irregularities in the accumulation of strata. The description above given of the slanting position of the minor layers constituting a single stratum is in certain cases appli- Cliff between Mismer and Dunwich. CH. v.] LENTICULAR FORM OF STRATA 39 cable on a much grander scale to masses several hundred feet thick, and many miles in extent. A fine example may be seen at the base of the Maritime Alps near Nice. The moun- tains here terminate abruptly in the sea, so that a depth of one hundred fathoms is often found within a stone's throw of the beach, and sometimes a depth of 3,000 feet within half a mile. But at certain points strata of sand, marl, or conglomerate in- tervene between the shore and the mountains, as in the section (fig. 10), where a vast succession of slanting beds of gravel and sand may be traced from the sea to Monte Calvo, a distance of no less than 9 miles in a straight line. The dip of these beds is remarkably uniform, being always southwards or towards the Mediterranean, at an angle of about 25. They are exposed to view in nearly vertical precipices, varying from 200 to 600 feet in height, which bound the valley through which the river Monte Calvo. Fig. 10. Sea Section from Monte Calvo to the sea by the valley of Magnan, near Nice. A. Dolomite and sandstone of Masozoic age. , &, y attending to the nature of these remains, we are often enabled to determine whether the deposition was slow or rapid, whether it took place in a deep or shallow sea, near the shore or far from land, and whether the water was salt, brackish, or fresh. Some limestones consist almost exclusively of corals, and in many cases it is evi- dent that the present position of each fossil zoophyte has been determined by the manner in which it grew originally. The axis of the coral, for example, if its natural growth is erect, still remains at right angles to the plane of stratification. If the stratum be now horizontal, the round spherical heads of certain species continue uppermost, and their points of attachment are directed downwards. This arrangement is sometimes repeated throughout a great succession of strata. From what we know of the growth of similar zoophytes in modern reefs, we infer that the rate of increase was extremely slow, and some of the fossils must have flourished for years, like forest trees, before they attained so large a size. During these ages, the water must have been clear and transparent, for such corals cannot live in turbid water. In like manner, when we see thousands of full-grown shells dispersed everywhere throughout a long series of strata, we cannot doubt that time was required for the multiplication of CH. VI.] INDICATED BY FOSSILS 45 Fig. 13. successive generations ; and the evidence of slow accumulation is rendered more striking from the proofs, so often discovered, of fossil bodies having lain for a time on the floor of the ocean after death, before they were embedded in sediment. Nothing, for example, is more common than to see fossil oysters in clay, with serpulffi, or barnacles (acorn-shells), or corals, and other creatures attached to the inside of the valves, so that the mollusk was certainly not buried in argillaceous mud the moment it died. There must have been an interval during which it was still surrounded with clear water, when the crea- tures whose remains now adhere to it grew from an embryonic to a mature state. Attached shells which are merely external, like some of the serpulaB (a) in fig. 13, may often have grown upon an oyster or other shell while the animal within was still living ; but if they are found 011 the inside, it could only happen after the death of the inhabitant of the shell which affords the support. Thus, in fig. 13 it will be seen that two serpulse have grown on the interior, one of them ex- actly on the place where the adductor muscle of the Gryplice.a (a kind of oyster) was fixed. Some fossil shells, even if simply attached to the outside of others, bear full testimony to the conclusion above alluded to, namely, that an interval elapsed between the death of the creature to whose shell they adhere and the burial of the same in mud or sand. The sea-urchins, or Echini, so abundant in white chalk, afford a good illustration of this remark. It is well known that these animals, when living, are invariably covered with spines supported by rows of tubercles. These last are only seen after the death of the sea-urchin, when the spines have dropped off. In fig. 15 a living specimen of Spatangus, common on our coast, is represented with one half of its shell stripped of the spines, Fossil Gnjphcea (nat. size), covered both on the outside and inside with fossil serpulee. 46 SLOW DEPOSITION OF STBATA [CH, VI. In fig. 14 a fossil of the genus Micraster found in the white chalk of England shows the naked surface which the individuals of this species exhibited when denuded of their spines. The full-grown Serpula, therefore, which now adheres externally, Fig. 14. Pig. 15. Serpula attached to a fossil Micraster, J nat., from the chalk. Becent Spatangm, \ nat., with the spines removed from one side. b. Spine and tubercles, nat. size. a. The same magnified. Fig. 16. could not have begun to grow till the Micraster had died and the spines became detached. Now the series of events here attested by a single fossil may be carried a step farther. Thus, for example, we often meet with a sea-urchin (Ananchytes or Echinocorys) in the chalk, (see fig. 16), which has the lower valve of a Crania, a genus of Brachiopoda, fixed to it. The upper valve (b, fig. 16) is almost invariably wanting, though occasionally found in a perfect state of preservation in the chalk at some dis- tance. In this case, we see clearly that the sea-urchin first lived from youth to age, then died and lost its spines, which were carried away. Then the young Crania adhered to the bared shell, grew and perished in its turn ; after which the upper valve was sepa- rated from the lower before the Ananchi/tes a. Echmoconus (Ga~ , , n . . .... , leritis), from the became enveloped in chalky mud. The rate vai' of ith cJS of accumulation of the chalk must, therefore, attached, J nat. have been excessively slow. 6> U rLtach e ed. f It may be well to mention one more illus- tration of the manner in which single fossils may sometimes throw light on a former state of things, both in the bed of the ocean and on some adjoining land. We meet with many fragments of wood bored by ship-worms, at various depths in the clay on which London is built. Entire branches and stems of trees, several feet in length, are some- times found drilled all over by the holes of these borers, the tubes and shells of the moljusk still remaining in the cylindri- CH. VI,] INDICATED BY FOSSILS 47 cal hollows. In fig. 18, e, a representation is given of a piece of recent wood pierced by the Teredo navalis, L., or common ship-worm, which destroys wooden piles and ships. When the cylindrical tube d has been extracted from the wood, the valves are seen at the larger or anterior extremity, as shown at c. In like manner, a piece of fossil wood (a, fig. 17) has been perforated by a kindred but distinct genus, the Teredina of Lamarck. The calcareous tube of this mollusk was united and as it were soldered on to the valves of the shell (6), which therefore cannot be detached from the tube, like the valves of the recent Teredo. The wood in this fossil specimen is now Fossil and recent wood drilled by perforating Mollusca. Fig. 17. a. Fossil wood from London clay, bored by Teredina, \ nat. size. b. Shell and tube of Teredina personata, Lam. sp., the right-hand figure the ventral, the left the dorsal view. Fig. 18. e. Recent wood bored by Teredo, \ nat. size. d. Shell and tube of Teredo navalis, L., from the same. c. Anterior and posterior view of the valves of same detached from the tube, nat. size. converted into a stony mass ; but it must once have been buoyant and floating in the sea, when the Teredince lived upon and perforated it. Again, before the infant colony settled upon the drift wood, part of a tree must have been floated down to the sea by a river, uprooted, perhaps, by a flood, or torn off and cast into the waves by the wind; and thus our thoughts are carried back to a prior period, when the tree grew for years on dry land, enjoying a fit soil and climate. The present rate of accumulation of deep-sea sediment is exceedingly slow, as is proved by the growths of coral that occur on electric cables. The corals grow at great depths very much 48 ORGANIC DEPOSITS [ CH . vi. more quickly than the accumulation of the foraminiferal ooze. But rapid accumulation of some sediments must have taken place formerly, for tree stems standing erect are found in strata of coal, sand, and grit which gathered around them. Minuteness of some of tbe organisms 'which build up great rock-masses. It has been already remarked that there are rocks in the interior of continents, at various depths in the earth, and at great heights above the sea, almost entirely made up of the remains of zoophytes and mollusca. Such masses may be compared to modern coral-reefs and oyster-beds ; and, as in their case, the rate of increase must have been extremely gradual. But there are certain deposits in the earth's crust, now proved to have been derived from plants and animals of which the organic origin was not at one time suspected, even by naturalists. Great surprise was created half a century ago by the discovery of Professor Ehrenberg, of Berlin, that a kind of siliceous material, called tripoli, was entirely composed of millions of the remains of organic beings, which were formerly referred to microscopic Infusoria, but which are now known to be plants. They abound in rivulets, lakes, and ponds in England and other countries, and are termed Diatomaceae. The substance alluded to has long been well known in the arts, under the name of Infusorial Earth or Mountain Meal, and is used in the form of powder for polishing stone and metal. It has been procured, among other places, from Bilin, in Bohemia, in which place a single stratum, extending oyer a wide area, is no less than 14 feet thick. This stone, when examined under high powers of the micro- scope, is found to consist of the siliceous tests of the Diatomacese figured on the next page, united together without any visible cement It is difficult to convey an idea of their extreme minute- ness ; but Ehrenberg estimates that in the Bilin tripoli there are 41,000 millions of individuals of the Gallionella distans, Ehb., (see fig. 20) in every cubic inch (which weighs about 220 grains), or about 187 millions in a single grain. At every stroke, there- fore, that we make with this polishing powder, several millions, perhaps tens of millions, of perfect fossils are crushed to atoms. A well-known substance, called bog-iron ore often met with in peat mosses, has been shown by Ehrenberg to consist of in- numerable articulated threads, of a yellow-ochre colour, com- posed of silica, argillaceous matter, and peroxide of iron. These threads are the remains of a minute microscopic plant, called Didymohelix fcrruginea, Ehb. sp. (fig. 19), associated with the siliceous remains of other freshwater algre. Layers of this iron ore occurring in Scotch peat bogs are often called ' the pan ' ; similar beds of iron ore which have been formed of vegetable. CH. vi.] STKATA MADE UP OF FOSSILS 49 organisms are found at the bottom of certain lakes in Sweden, and occur between the basalts of Antrim, and these are of con- siderable economical value. It is clear that much time must have been required for the accumulation of strata to which countless generations of Diato- maceae and similar microscopic algae have contributed their remains ; and these discoveries lead us naturally to suspect that other deposits, of which the materials have been supposed to be inorganic, may in reality be composed chiefly of micro- scopic organic bodies. That this is the case with the white chalk has often been imagined, and is now proved to be the fact. It has, moreover, been lately discovered that the chambers into which these Foraminifera are divided are actually often filled with thousands of well-preserved organic bodies (see figs. 27, 28), which abound in every minute grain of chalk, Fig. 19. Pig. 20. Fig. 21. Didymohelix Oallionella Badllaria paradoxa, Gmel. erruginea, Ebb. sp. distans, Ehb. a. Front view. 5. Side view. and are especially apparent in the white coating of flints, often accompanied by innumerable needle-shaped spiculse of sponges. The dust we tread upon was once alive ! BYBON. How faint an idea does this exclamation of the poet convey of the real wonders of nature ! for here we discover proofs that the calcareous and siliceous dust of which whole hills are com- posed has not only been once alive, but almost every particle albeit invisible to the naked eye still retains the organic struc- ture which, at periods of time incalculably remote, was impressed upon it by the powers of life. Importance of fossils as indicating the conditions under which strata were deposited. It is a well-known fact that peculiar forms of mollusca, corals, crustaceans, &c., are confined to certain depths in the ocean ; some charac- terise shallow water between tide-marks, others are found in moderately deep water, and others, again, only in the very deepest parts of the ocean. If, then, we find in a par- ticular stratum an assemblage of organisms which we recog- nise as always occurring at a given depth of water in our existing seas, we may fairly conclude that the stratum containing 50 GLOBIGERINA OOZE AND CHALK [CH. VI. the assemblage of fossils must have been deposited in a similar depth of water. Caution, of course, is required in applying this reasoning seeing that in most cases the organisms found as fossils in rock masses are not identical with but only closely related to those found in the sea at the present time. Deep-sea deposits. Besides the shells, corals, fish, &c., that have longbeen known as characterising thelittoral, shallow water, Fig. 22. Fig. 23. Fig. 22. Globigerina ooze from the North Atlantic Ocean, dredged from a depth of 1,990 fathoms ( x 70). The mass is seen to be made up of the cal- careous shells of Globigerina and other Foraminifera. entire or broken, with fragments of larger organisms, and numerous minute calcareous particles, which are shown much more highly magnified in figs. 27 and 28. Pig. 23. Washings from the white chalk of Kent (also x 70), showing simi- lar organisms to those found in the Globigerina ooze. Pig. 24. Fig. 25. Fig. 26. Fig. 27. Fig. 28. Pig. 24. Coccosphere found floating on the ocean surface a very minute organism (Calcareous alga ?), x 1,000. Figs. 25, 26. Two forms of Rhabdospheres. Similar minute organism from the ocean surface, x 1,000. (These three figures are taken from the ' Chal- lenger ' Reports.) Fig. 27. Isolated Coccoliths, which build up Coccospheres, and are found both in the Globigerina ooze and the chalk (see figs. 22 and 23), x 1,000. Fig. 28. Isolated Khabdoliths, which build up Rhdbdospheres, and are found both in the Globigerina ooze and the chalk (see figs. 22 and 23), x 1,000. and deep-water parte of the ocean, we have become acquainted through the explorations carried on by the 'Challenger' and other surveying vessels with organisms that live in the CH, vi.] RADIOLARIAN AND DIATOMACEOUS OOZE 5l abysmal recesses of the ocean, at depths down to nearly 5,000 fathoms. It is interesting to note that among the stratified rocks we find many examples of materials made up of the remains of organisms like those found in the deeper parts of the ocean. In the chalk we have an example of a rock almost entirely made up of the calcareous shells of Forarninifera (Grlobigerina, &c.) with numerous small bodies, Coccoliths and Rhabdoliths, supposed to be the remains of calcareous algae (see figs. 22-28). In the Barbadoes Earth, we have an ooze made up of the siliceous skeletons of Radiolarians precisely similar to the material dredged up from great depths in the Pacific and Indian Oceans (see figs. 29-80). Fig. 29. Fig. 30. Fig. 29. Radiolarian ooze from the Pacific Ocean, depth 2,425 fathoms, almost wholly made up of the remains of the siliceous skeletons of Radiolarians, x 70. Fig. 30. The Barbadoes Earth, a siliceous rock of Tertiary age, almost entirely made up of similar organisms (Radiolarians), x 70. In the white earth of Richmond, Virginia, we have a material almost identical with the white diatomaceous ooze of the Ant- arctic Ocean, which is made up of the siliceous frustules of microscopical algse (see fig. 31). Similar freshwater algae are found building up white siliceous deposits at the bottom of lakes in this and other countries (see fig. 82). At the greatest depths, a reddish or chocolate-coloured clay of great fineness often containing the teeth and bones of marine animals, and curious nodules composed of iron and manganese oxides is found covering the ocean floors. Freshwater deposits and their fossils. Strata, whether deposited in salt or fresh water, have the same forms ; but the embedded fossils are very different in the two cases, because the aquatic animals which frequent lakes and rivers are, as a rule, E2 DISTINCTION OF FRESHWATER [CH. VI. distinct from those inhabiting the sea. In the northern part of the Isle of Wight formations of marl and limestone, more than 50 feet thick, occur, in which the shells are of extinct species. Yet we recognise their freshwater origin, because they are of the same genera as those now abounding in ponds, lakes, and rivers, either in our own country or in warmer latitudes. In many parts of France, as in Auvergne, there occur strata of limestone, marl, and sandstone hundreds of feet thick, which contain exclusively freshwater and land shells, together with the remains of terrestrial quadrupeds. The number of land shells scattered through some of these freshwater deposits is exceedingly great ; and there are districts in Germany where the rocks contain scarcely any other fossils than snail-shells, Fig. 31. Fig. 32. Fig. 31. Marine forms of Diatomacece (unicellular algae with siliceous skele- tons) found at the bottom of the Antarctic Ocean (Diatomuceous ooae), and in certain siliceous rocks like the Tertiary White Earth of Richmond, Vir- ginia ( x 250). Fig. 32. Freshwater forms of Diatomacece, found in rivers and lakes, and also making up siliceous rocks known as ' mountain-meal,' ' Kieselguhr ' ' tripoli ' &c. See also figs. 20 and 21. (Helix) ; as, for instance, the limestone on the left bank of the Khine, between Mayence and Worms, at Oppenheim, Findheim, Budenheim, and other places. In order to account for this phenomenon, the geologist has only to examine the small deltas of torrents which enter the Swiss lakes when the waters are low, such as the newly formed plain where the Kander enters the Lake of Thun. He there sees sand and mud strewn over with innumerable dead land-shells, which have been brought down from the valleys in the Alps in the preceding spring, during the melting of the snows. Again, if we search the sands on the borders of the Khine, in the lower part of its course, we find countless land-shells mixed with others of species belonging to lakes, stagnant pools, and marshes. These organisms have CH. Vl.J FROM MARINE FORMATIONS 53 been washed away from the alluvial plains of the great river and its tributaries, some from mountainous regions, others from the low country. Although freshwater lormations are often of great thickness, yet they are usually very limited in area when compared with marine deposits, just as lakes and estuaries are of small dimen- sions in comparison with seas. Fig. 34. Cyclas (Sphcerium) corneus, Sow. ; living and fossil, nat. size. Cyrena (Corbicula) fuminaUs, Mull. ; fossil, Grays,' Essex, and living in the Nile, nat. size. The absence of many fossil forms usually met with in marine strata affords a useful negative indication of the freshwater origin of a formation. For example, there are no sea-urchins, no corals, no chambered shells, such as the nautilus, nor micro- scopic foraminifera in lacustrine or fluviatile deposits. In dis- tinguishing the latter from formations accumulated in the sea, Pig. 35. Fig. 36. Fig. 37. Anodonta Cordieri, D'Orb. Paris, $. Anodonta latimarginata, Lea. ; recent. Bahia, \. Unio littoralis, Lam. recent. Auvergne, we are chiefly guided by the forms of the mollusca. In a fresh- water deposit, the number of individual shells is often as great as in a marine stratum, if not greater ; but there is a smaller variety of species and genera. This might be anticipated from the fact that the genera and species of recent freshwater and land shells are few when contrasted with the marine. Only a very small number of genera of bivalve shells inhabit 54 FRESHWATER UNIVALVES [CH. VI. fresh water. Among these last, the four most common forms, both recent and fossil, are Cyclas (Sphteriuni), Cyrena, Unio, and Anodonta (see figures 33-37). Fig. 38. Fig. 39. Gryphcea incurva, Sow. (G. arcuata, Lam.) : smaller valve. Lias, nat. size. Marine. Planorbis euomphalits, Sow. ; fossil. Isle of Wight, . Fig. 40. Fig. 41. Limnoea longiscata, Brong. ; fossil. Isle of Wight, A. Paludina viviparn. "Brand ; living and fossil, nat. size. Lamarck divided the bivalve mollusca into the Dimyaria, or those having two large muscular impressions in each valve, as Fig. 42. Fig. 43. Sucdnea amphibia, Ancylus velletia (A. Drap.(S. putris, L.) ; elegans). Sow. ; fossil, fossil. Loess, Rhine, Isle of Wight, nat. size. Fig. 44. Fig. 45. Valcata piscina- Phusti hypno- lis, Mull. ; fos- rum, L. ; re- sil. Grays, cent. Isle of Wight, nat. size. a b in the Cyclas, fig. 33, and Unio, fig. 87, and the Monomyaria such as the oyster and scallop, in which there is only one of these impressions, as seen in fig. 38. Now, as none of these CH. VI.] FEESHWATER UNIVALVES 55 last, or the unimuscular bivalves, are freshwater, 1 we may at once presume a deposit containing any of them to be marine. 3 Fig. 46. Pig. 47. Fig. 48. Fig. 49. A uriatla, recent. Cerithiv.m funatum, Physa columnaris, Melanopsisbuccinoidea^ Ava, i Mant.;fossil. Wool- Desh. ; fossil. Ferr. ; recent. Asia, vtich beds, nat. size. Paris basin, $. nat. size. The univalve shells most characteristic of freshwater deposits are Planorbis, Limncea, and Paludina (Vivipara). But to these are occasionally added Physa, Succinea, Ancylus, Valvata, Fig. 50. Fig. 51. Fig. 52. Neritina globulus, Def. Paris basin, nat. size. Nerita granulosa, Desh. Paris basin, . Melanopsis, Melania, Potamides, and Neritina (see figs. 39-49), the last four being usually found in estuaries. Some naturalists include Neritina (fig. 50) and the marine Nerita (fig. 51) in the same genus, it being scarcely possible to distinguish the two by good generic characters. But, as a general rule, the fluviatile species are smaller, smoother, and more globular than the marine ; and they have never, like the Neritcz, the inner margin of the outer lip toothed or crenulated. (Compare figs. 50 and 51.) 1 The freshwater Mulleria, which when young has two muscular im- pressions, has only one in the adult state, thus forming a single excep- tion to the rule. 2 It must be remembered, how- ever, that marine shells are oc- casionally found living in brackish water, and sometimes in water that is nearly fresh. Thus oysters and cockles are sometimes found in water with but little salt ; in these cases, however, the dwarfed and imperfectly developed character of the shells indicates the abnormal conditions under which they lived. 56 TERRESTRIAL UNIVALVES [CH. vi. The Potamides inhabit the mouths of rivers in warm lati- tades, and are distinguishable from the marine Cerithia by their orbicular and multispiral opercula. The genus Auricula (fig. 46) is both marine and freshwater, frequenting swamps and marshes within the influence of the tide. The terrestrial shells are all univalves. The most important genera among these, both in a recent and fossil state, are Helix (fig. 53), Cyclostoma (fig. 54), Pupa (fig. 55), Clausilia (fig. 56), Bulimus (fig. 57), Glandina, and Achatina. Fig. 53. Fig. 54. Fig. 55. Fig. 56. Fig. 57 Helix turonensis, Desh. Faluns, Touraine, nat. size. Cyclostoma eleyans, MU11. Loess, nat. size. Pupa tridens, Drap. nat. size. Claustlia bid ens, Drap. Loess, nat. size. Bulimus lubricus, Miill. Loess, Rhine. Fig. 58. Ampullaria (fig. 58; is another genus of shells, inhabiting rivers and ponds in hot countries. Many fossil species for- merly referred to this genus, which have been met with chiefly in marine formations, are now considered by conchologists to belong to Natica, and other marine genera. All univalve shells of land and freshwater species, with the exception of Melanopsis (fig. 49), and Achatina, which show a slight indentation, have entire mouths ; and this circumstance may often serve as a con- venient rule for distinguishing freshwater from marine strata ; since if any univalves occur of which the mouths are not entire, we may presume that the formation is marine. The aperture is said to be entire in such shells as the freshwater Ampullaria (fig. 58) and the land shells (figs. 53-57), when its outline is not in- terrupted by an indentation or notch, such as that seen at b in Ancilla (fig. 60) ; or is not prolonged into a canal, as that seen at a in Pleurotoma (fig. 59). The mouths of a large proportion of the marine univalves have these notches or canals, and almost all the species are carnivorous ; whereas nearly all gastropoda having entire mouths are plant-eaters, whether the species be marine, freshwater, or terr strial. Ampullaria (jlaiica, from the Jumna, J. CH. vi.] FRESHWATER PLANTS 57 There is, however, a genus which affords an occasional exception to one of the above rules. The Potamides (fig. 52), a subgenus of Cerithium, although provided with a short canal, comprises some species which inhabit salt, others brackish, and others fresh water, and they are said to be all plant-eaters. Among the fossils very common in freshwater deposits are the shells of Cypris, a minute bivalve crustaceous animal. Man;y minute living species of this genus swarm in lakes and stagnant pools in Great Britain ; but their shells are not, if considered separately, conclusive as to the freshwater origin of a deposit, because the majority of species in another kindred genus of the same order, the Cytlierina of Lamarck, inhabit salt water ; and, although the animal differs slightly, the shell is scarcely dis- tinguishable from that of the Cypris. Fig. 59. Fig. 60. Plenrotoma exorta, Brand. Upper Ancilla buccinoidet, Lam. and Middle Eocene, Barton and Barton clay. Eocene, Bracklesham, nat. size. nat. size. Freshwater fossil plants. The seed-vessels and stems of Chara, a genus of calcareous plants, are very frequent in fresh- water strata. The seed-vessels were called, before their true nature was known, gyrogonites, and, like many similar fragments of calcareous algae, were supposed to be foraminiferous shells. (See fig. 61, a.} The Chara inhabit the bottom of lakes and ponds, and flourish mostly where the water is charged with calcium car- bonate. Their seed-vessels are covered with a very tough integument, containing calcium carbonate, to which circumstance we may attribute their abundance in a fossil state. The annexed figure (fig. 62) represents a branch of one of many new species found by Professor Amici in the lakes of Northern Italy. The stems, as well as the seed-vessels, of these plants occur both in modern shell marl and in ancient freshwater formations. 58 FRESHWATER FISH [CH. VI. They are generally composed of a large central tube surrounded by smaller ones, the whole stem being divided at certain inter- vals by transverse partitions or joints. (See 6, fig. 61.) It is not uncommon to meet with layers of vegetable matter, impressions of leaves, and branches of trees in strata contain- ing freshwater shells ; and we also find occasionally the teeth and bones of land quadrupeds, of species now unknown. The manner in which such remains are sometimes carried by rivers into lakes, especially during floods, has been fully treated of in the ' Principles of Geology.' Freshwater and marine fish. The remains of fish are oc- casionally useful in determining the freshwater origin of strata. Certain genera, such as Cyprinus (carp), Perca (perch), Esox (pike), Cobitis (loach), and Lebias, are peculiar to fresh water. Pig. 61. Fig. 62. Chara medicaginula, Brong. ; fossil. Chara elastica, Amici ; recent. Italy. Upper Eocene, Isle of Wight. a. Sessile seed-vessel between the divisions a. Seed-vessel magnified of the leaves of the female plant. 20 diameters. b. Magnified transverse section of a branch 6. Stem magnified. with five seed-vessels seen from below upwards. Other genera contain some freshwater and some marine species, as Coitus, Mugil, and Anguilla (eel). The rest are either common to rivers and the sea, as the salmon ; or are ex- clusively characteristic of salt water. The above observations respecting fossil fishes are applicable only to the modern or tertiary deposits ; for in the more ancient rocks the forms depart so widely from those of existing fishes that it is very difficult, at least in the present state of science, to derive any positive information from ichthyolites respecting the nature of the water in which strata were deposited. The alternation of marine and freshwater formations, both on a small and large scale, are facts well ascertained in geology. When it occurs on a small scale, it may have arisen from the successive occupation of certain spaces by river water and the CH. vi.] MARINE AND FRESHWATEK BEDS 59 sea ; for in the flood season the river forces back the ocean, and freshens it over a large area, depositing at the same time its sediment ; after which the salt water again returns, and, on resuming its former place, brings with it sand, mud, and marine shells. There are also lagoons at the mouths of many rivers, as the Nile and Mississippi, which are divided by bars of sand from the sea, and which are filled with salt and fresh water by turns. They often communicate exclusively with the river for months, years, or even centuries ; and then, a breach being made in the bar of sand, they are for long periods filled with salt water. The Lym-Fjord in Jutland offers an excellent illustration of analogous changes ; for, in the course of the last thousand years, the western extremity of this long frith, which is 120 miles in length, including its windings, has been four times fresh and four times salt, a bar of sand between it and the ocean having been as often formed and removed. The last Fig. 63. N. Coal with upright trees. Sandstone and shale. S. d e f Section of the cliffs of the South Joggins, near Minudie, Nova Scotia. c. Sandstone used for grindstones, rf, g. Alternations of sandstone, shale, and coal containing upright trees. e,f. Portion of cliff, given on a larger scale in fig. 64. /. 4-foot coal, main seain. h, i. Shale with freshwater shells. irruption of salt water happened in 1824, when the North Sea entered, killing all the freshwater shells, fish, and plants ; and from that time to the present, the seaweed Fucus vesiculosus, together with oysters and other marine mollusca, has succeeded the Cyclas, Limncea, Paludina> and Chara. But changes like these in the Lym-Fjord, and those before mentioned as occurring at the mouths of great rivers will only account for some cases of marine deposits of partial extent and thickness resting on freshwater strata. When we find, as in the south-east of England, a great series of freshwater beds, 1,000 feet thick, resting upon marine formations and again covered by other rocks, such as the cretaceous, also more than 1,000 feet thick, and of deep-sea origin, we shall find it necessary to seek for a different explanation of ihe phenomena. Terrestrial deposits and their fossils. Although deposits formed on the land are rare, they are not quite unknown to geologists. Beds of peat, lignite, and coal consist of the remains of land plants which in many cases can be shown to have grown 60 TERRESTRIAL DEPOSITS [CH. VI. in the spot where they are now found. In some cases the trunks of trees are still found attached to their roots and rising through the strata above them, as shown in the diagrams of the South Joggins Coal-field of Nova Scotia (see figs. 63, 64). Fig. 64. Erect fossil trees, a, c, rf,/, g. Coal-measures, Nova Scotia (after Sir J. W. Dawson). In some cases, great numbers of trunks of trees have thus been found in connection with the masses of vegetable matter forming a bed of coal. Thus in South Staffordshire a seam of Fig. 65. Ground plan of a fossil forest, Parkfield Colliery, near Wolverhampton, showing the position of 73 trees in a quarter of an acre. coal was laid bare in the year 1844, in what is called an open work at Parkfield Colliery, near Wolverhampton. In the space of about a quarter of an acre the stumps of no less than seventy- three trees with their roots attached appeared, as shown in the CH. VI.] FOKMATION OF COAL 61 annexed plan (fig. 65), some of them more than 8 feet in cir- cumference. The trunks, broken off close to the root, were lying prostrate in every direction, often crossing each other. One of them measured 15, another 30 feet in length, and others less. They were invariably flattened to the thickness of one or two inches, and converted into coal. Their roots formed part of a stratum of coal 10 inches thick, which rested on a layer of clay 2 inches thick, below which was a second forest, resting on a 2-foot seam of coal. Again, five feet below this, was a third forest with large stumps of Lepidodendra, Catamites, and other trees. In one instance it was found possible to determine the direction of the prevailing wind, at the time when the trees were growing, from the bending of all the trunks in one direction. Fig. 66, A C A. ' Better-Bed ' Coal, from a portion unusually full of Macrospores, which are here shown in transverse section. B. Same coal, section parallel with bedding ; showing Macrospores, e, and Microspores, / ; the latter (which are here represented somewhat too large) appear as bright rings enclosing a dark spot. C. Australian ' White Coal,' showing Macrospores in transverse section. D. External view of Macrospores separated from the ' White Coal.' All these figures are enlarged about 16 diameters. The dirt-beds of the Isle of Portland (see figs. 345, 348, p. 291) offer examples of similar terrestrial deposits. While many beds of coal consist of the compressed stems, leaves, and other parts of plants, some particular bands are almost entirely made up of their spores or organs of fructifica- tion. Professor Huxley has ascertained that in the Better-Bed coal of Lowmoor, near Bradford (see A, B, fig. 66), the spores (ma- crospores and microspores) of the great plants of the Carboni- ferous age constitute a very large portion of the rock, and this is also the case with the recent * White Coal ' of Australia (see C, D, fig. 66). 62 PUEITY OF COAL [CH. vi. It must be remembered, however, that these ' spore-coals ' are somewhat exceptional, though all coals probably contain spores as well as other portions of the plants of which they are made up. Some beds of coal, moreover (like the cannel coals), were certainly not formed of plants growing insitu,l>nt must have consolidated from masses of black carbonaceous mud like those produced by the bursting of peat bogs. Coal seams also alter- nate with strata containing marine, freshwater, or brackish- water fossils, showing that, even when accumulated on land, this land was but little above the sea-level like the islands forming portions of deltas. Strata accumulated under such conditions as these are spoken of as ' estuarine deposits.' The student must bear in mind that while some seams of coal were certainly formed from plants undergoing decay upon the spot where they grew as is indicated by the existence of * underclays ' or old soils beneath them, containing the roots of plants other beds of coal would seem to have been produced from masses of drifted vegetable matter that have accumulated in hollows, have been covered up by other strata, and have then slowly undergone chemical change. The purity of the coal itself, or the absence from it of earthy particles and sand, throughout areas of vast extent, is a fact which appears very difficult to explain when we attribute each coal-seam to a vegetation growing in swamps. It has been asked how, during river inundations capable of sweeping away the leaves of ferns and the stems and roots of SigiUarice and other trees, could the waters fail to transport some fine mud into the swamps ? One generation after another of tall trees grew with their roots in mud, and their leaves and prostrate trunks formed layers of vegetable matter, each of which was afterwards covered with mud since turned to shale. Yet the coal itself, or altered vegetable matter, remained all the while uncontaminated by earthy particles. The difficulty will be removed if we consider what is now taking place in deltas. The dense growth of reeds and herbage which encompasses the margins of forest-covered swamps in the valley and delta of the Mississippi is such that the fluviatile waters, in passing through them, are filtered and made to clear themselves entirely before they reach the areas in which vegetable matter accumulates ; and this accumulation may go on for centuries, forming coal if the climate be favourable. There is little chance of the intermixture of earthy matter in such cases. Thus in the large submerged tract called the ' Sunk Country,' near New Madrid, forming part of the western side of the valley of the Mississippi, erect trees have been standing ever since the year CH. vi.] SLOW FORMATION OF COAL 63 1811-12, killed by the great earthquake of that date ; lacustrine and swamp plants have been growing there in the shallows, and several rivers have annually inundated the whole space, and yet have been unable to carry in any sediment within the outer boundaries of the morass, so dense is the marginal belt of reeds and brushwood. It may be affirmed that, generally, in the ' cypress swamps ' of the Mississippi no sediment mingles with the vegetable matter accumulated there from the decay of trees and semi-aquatic plants. As a singular proof of this fact, it may be mentioned that whenever any part of a swamp in Louisiana is dried up, during an unusually hot season, and the wood is set on fire, pits are burnt into the ground many feet deep, or as far down as the fire can descend, without meeting with water, and it is then found that scarcely any residuum of earthy matter is left. At the bottom of all these ' cypress swamps ' a bed of clay is found, with roots of the tall cypress (Taxodium di- stichum, Rich.), just as the underclays of the coal are filled with Stigma/rid-. The separation from the carbonaceous strata in the earth's crust of water, ammonia, carbon dioxide, and the various hydro- carbons must be attended with a great diminution in the bulk of the mass. LTnger calculated that.it would require a thickness of 8-76 feet of vegetable matter to make a bed of coal one foot in thickness. With regard to the time taken for the growth of the materials forming carbonaceous deposits, Heer has estimated that the growth of one foot of peat requires about a century. There is at Petrosene in Transylvania a bed of coal of tertiary age ninety feet in thickness. If the above estimates be correct, the ninety feet of coal would represent 788 feet of vege- table matter, the accumulation of which would require 78,800 years ! The facts of the distribution of tains a useful summary of all the the forms of vegetable and animal facts which were ascertained by life in the ocean have been much the ' Challenger ' and the deep-sea more fully made known by the exploring expedition concerning the various deep-sea exploring expedi- distribution of life forms at various tians. The ' " Challenger " Report depths in the ocean. On the sub- on Oceanic Deposits,' by Dr. John ject of the formation of coal, the Murray and Prof. A. Renard, con- student will find much valuable tains much valuable information information in ' Coal : its History illustrating the origin of limestones and Uses,' by Professors Green, and other marine deposits. The Miall, Thorpe, Ru'cker, and Mar- final volume of the Reports con- shall, 1878. 64 [OH. vii. CHAPTER VII CONSOLIDATION AND SUBSEQUENT ALTERATIONS OF STRATA AND PETRIFACTION OF ORGANIC REMAINS Consolidation of strata Concretionary structures Jointed structure Mineralisation of organic remains Formation of casts Wonderful preservation of the internal structures of fossil organisms Petrifac- tions and incrustations Pseudo-fossils. HAVING spoken in the preceding chapters of the characters of sedimentary formations, both as dependent on the deposition of inorganic matter and the distribution of fossils, we may next treat of the consolidation of stratified rocks, and the petrifac- tion of embedded organic remains. Consolidation of strata. In the case of some calcareous rocks, solidification takes place at the time of deposition. But there are many deposits in which a cementing process comes into operation long afterwards. We may sometimes observe, where the water of ferruginous or calcareous springs has flowed through a bed of sand or gravel, that iron or lime compounds have been deposited in the interstices between the grains or pebbles, so that in certain places the whole has been bound together into a stone, the same set of strata remaining in other parts loose and incoherent. Proofs of a similar cementing action are seen in a rock at Kellaways in Wiltshire. A peculiar sandy stratum, belonging to the Jurassic formation of geologists, may be traced through several counties, the sand being for the most part loose and unconsolidated, but becoming stony near Kellaways. In this district there are numerous fossil shells which have decomposed, having for the most part left only their casts. The calcareous matter hence derived has evidently served, at some former period, as a cement to the siliceous grains of sand, and thus a solid sandstone has been produced. If we take fragments of many sandy or argillaceous rocks, retaining the casts of shells, and plunge them into dilute acid, we see the mass immediately breaks up into sand or mud, the cement of calcium carbonate, derived from the shells, having been dissolved by the acid. Traces of impressions and casts are often extremely faint. In some loose sands of recent date we meet with shells in so advanced a stage of decomposition as to crumble into powder when touched. It is clear that water percolating such strata may soon remove the calcareous matter of the shell; and, unless CH. vii.] CONSOLIDATION OF STRATA 65 circumstances cause the calcium carbonate to be again deposited, the grains of sand will not be cemented together ; in which case no memorial of the fossil will remain. It is evident that silica and calcium carbonate are widely diffused in small quantities through the waters which permeate the earth's crust, and thus a stony cement is often supplied to sand, pebbles, or any fragmentary mixture. In some con- glomerates, like the pudding-stone of Hertfordshire (a Lower Eocene deposit), pebbles of flint and grains of sand are united by a siliceous cement so firmly that, if a block be broken, the fracture passes as readily through the pebbles as through the cement. While it is true that many strata become solid owing to the pressure of the superincumbent rocks under which they have been buried, yet their consolidation is often largely due to the chemical action of the water and gases which penetrate their minutest pores, and the consequent deposition of calcium carbonate, iron carbonate or oxide, silica, and other minerals previously held in solution. Most stones on being freshly quarried are found to be soft and easily cut, but harden by exposure. The marl recently deposited at the bottom of Lake Superior, in North America, is soft and often filled with freshwater shells ; but if a piece be taken up and dried, it becomes so hard that it can be broken only by a smart blow of the hammer. If the lake, therefore, were drained, such a deposit would be found to consist of strata of marlstone, like that observed in many ancient European formations, and, like them, containing freshwater shells. Concretionary structure. It is probable that some of the heterogeneous materials which rivers transport to the sea may at once set under water, like the artificial mixture called poxzo- lana, which consists of fine volcanic sand, to which a small quantity of lime has been added. This substance hardens, and becomes a solid stone in water, and was used by the Romans in constructing the foundations of buildings in the sea. Conso- lidation, in such cases, is brought about by the chemical reaction which takes place between the silica and lime. After deposi- tion particles of similar chemical composition seem often to exert a mutual attraction for each other, and segregate in particular spots, forming lumps, nodules, and concretions. Thus, in many argillaceous deposits there are calcareous balls, or spherical con- cretions, ranged in layers parallel to the general stratification ; an arrangement which took place after the shale or marl had been thrown down in successive laminae ; but these laminae are often traceable through the concretions, remaining parallel to those of F 66 FORMATION OF CONCRETIONS [CH. VII. the surrounding unconsolidated rock (see fig. 67). Sue nodules of argillaceous limestone have often a shell or other foreign body in the centre. In some cases these nodules exhibit a series of ramifying cracks which are usually filled with calcite. Nodules of this kind are called ' Septaria.' The calcareo-argilla- ceous nodules often contain much ferrous carbonate, and some- times constitute valuable iron-ores. Among the most remarkable examples of concretionary struc- ture are those described by Professor Sedgwick as abounding in the magnesian limestone of the north of England. The spherical balls are of various sizes, from that of a pea to a dia- meter of several feet, and they have both a concentric and radiated structure, while at the same time the laminae of origi- nal deposition pass uninterruptedly through them. In some cliffs this limestone resembles a great irregular pile of cannon- balls. Some of the globular masses have their centre in one stratum, while a portion of their exterior passes through to the stratum above or below. Thus the larger spheroid in the an- Fig. 67. fig, 68. Calcareous nodules in Lias seen in section. Spheroidal concretions in inagnesian limestone. nexed section (fig. 68) passes from the stratum b upwards into a. In this instance we must suppose the deposition of a series of minor layers, first forming the stratum 6, and afterwards the incumbent stratum a ; then a movement of the particles took place, and the calcium and magnesium carbonates separated from the more impure and mixed matter forming the still un- consolidated parts of the stratum. Crystallisation, beginning at the centre, must have gono on forming concentric coats around the original nucleus without interfering with the lami- nated structure of the rock (Not3 C, p. 601). By similar processes of segregation and crystallisation in masses of mixed materials, the structures known as ' cone-in- cone,' ' beef,' * stylolites,' &c., have evidently been formed. The crystallising material in these cases is usually either calcium car- bonate or ferrous carbonate, and the clay or other foreign materials are caught up and included in the crystals. In the Fontainebleau sandstone we have a mixture of sand and calcium carbonate, and in the midst of the rock, groups of large and perfect crystals CH. vii.] AND JOINTS 67 of calcite are formed, which are crowded with sand grains caught up by the growing crystals. The curious structures known to geologists as ' botryoidal,' ' rnammillated,' &c., are due to similar selective and crystallising agencies operating in the midst of great rock masses. The rocks of the earth's crust have often been subiected to great pressure for long periods of time from the accumulation of thousands of feet of rock above them. Such pressures are capable of bringing about the coherence of finely divided materials like clay ; this is illustrated by the process of making lead pencils by subjecting powdered graphite to pressure, and by the very suggestive experiments of M. Spring, who has shown how many powdered materials may be converted into hard and coherent masses by the action of pressure alone. Analogous effects of consolidation and condensation have arisen when the solid parts of the earth's crust have been forced in various directions by those mechanical movements hereafter to be described, by which strata have been bent, broken, and raised above the level of the sea. Rocks of more yielding materials must often have been forced against others previously consolidated, and may thus, by compression, have acquired a new structure. Finely laminated and cleaved structures in rocks are produced and intensified by the action of pressure. Jointed structure. Joints are natural fissures which often traverse rocks in straight and well-determined planes, more or less at right angles to the planes of bedding or stratification. If a sufficient number cross each other, the whole mass of rock is split into symmetrical blocks, and they afford to the quarryman the greatest aid in the extraction of blocks of stone. The faces of the joints are for the most part smoother and more regular than the surfaces of true strata. The joints are straight-cut fissures, sometimes slightly open, and often pass not only through layers of successive deposition, but also through balls of limestone or other matter, which have been formed by con- cretionary action since the original accumulation of the strata, and in the case of conglomerates even through quartz pebbles. Such joints, therefore, must often have resulted from one of the last changes superinduced upon sedimentary deposits. In the diagram on page 68 (fig. 69), the flat surfaces of rock, A, B, c, represent exposed faces of joints, to which the walls of other joints, J, J, are parallel, s s are the lines of stratification ; D D are lines of slaty cleavage, which intersect the rock at a considerable angle to the planes of stratification. In the Swiss and Savoy Alps, as Mr. Bakewell has remarked, enormous masses of limestone are cut through so regularly by 68 CHARACTERISTICS OF JOINTS [CH. vn. nearly vertical partings, and these joints are often so much more conspicuous than the planes of stratification, that an inexpe- rienced observer will almost inevitably confound them, and suppose the strata to be perpendicular in places where in fact they are almost horizontal. Jukes observed joints in recently formed coral rock in the Australian and other reefs. Joints are due to contraction of strata during consolidation, and also to great movements which take place in the earth's crust. Joints in aqueous rock-masses are supposed to be analogous to the partings which separate volcanic and plutonic rocks into cuboidal and prismatic masses. On a small scale we see clay and starch when dried split into similar shapes ; this is often caused by simple contraction, whether the shrinking be due to the evaporation of water, or to a change of temperature. It is well known that sandstones and other rocks expand by the Fig. 69. Stratification, joints, and cleavage. (From Murchison's ' Silurian System.' p. 245.) application of heat, and then contract again on cooling ; and there can be no doubt that large portions of the earth's crust have, in the course of past ages, been subjected again and again to very different degrees of heat and cold. These alternations of temperature have probably contributed largely to the pro- duction of joints in rocks. In many countries where masses of basalt rest on sandstone or shale, the aqueous rock has for the distance of several feet from the point of junction assumed a columnar structure similar to that of the igneous mass. In like manner some hearthstones, after exposure to the heat of a furnace without being melted, have separated into prismatic blocks. Mineralisation of organic remains. The changes which fossil organic bodies have undergone since they were first em- bedded in rocks throw much light on the consolidation of strata. In some modern deposits, shells and other organic CH. VII.] FOSSILISATION 69 remains have been scarcely altered in the course of centuries, having simply lost a part of their animal matter. Such slightly altered organisms are spoken of as sub-fossil forms. But in other cases the shell may have disappeared, and left an impression only of its exterior, and perhaps also a mould of its interior ; in other cases again we find a reproduction of the shell itself in some new material taking the place of the original matter which has been removed. These different forms of fossilisation may easily be understood if we examine the mud recently thrown out from a pond or canal in which there are shells. If the mud be argillaceous, it acquires consistency on drying, and on breaking open a portion of it we find that each shell has left impressions of its external form. If we then remove the shell itself, we find within a solid nucleus of clay, having the form of the interior of the shell. This form is often very different from that of the Fig. 70. Fier. 71. Chemnitzia Heddingionensis, Sow. sp., f nat. , and cast of the same. Coral Rag. Pleurotomaria anglica, Sow. sp., and cast, J nat. size. Lias. outer shell. Thus a cast such as a-, fig. 70, commonly called a fossil screw, would never be suspected by any one but a concholo- gist to be the internal shape of the fossil univalve, 6, fig. 70. Nor should we have imagined at first sight that the shell a and the cast fe, fig. 71, belong to one and the same fossil. The reader will observe in the last-mentioned figure (&, fig. 71), that an empty space shaded dark, which the shell itself once occupied, now intervenes between the enveloping stone and the cast of the smooth interior of the whorls. In such cases the shell has been dissolved and the component particles removed by water per- colating the rock. If the nucleus were taken out, a hollow mould would remain, on which the external form of the shell with its tubercles and striae, as seen in a, fig. 71, would be found embossed. Now, if the space alluded to between the nucleus and the impression, instead of being left empty, has been filled up with calcareous spar, flint, pyrites, or other mineral, we then. 70 REPLACEMENT OF ORGANIC TISSUES [CH. vn. obtain from the mould an exact reproduction both of the external and internal form of the original shell. In this manner silicified casts of shells have been formed ; and if the material of the nucleus happen to be incoherent, or soluble in acid, we can then procure in flint an empty shell, which in shape is the exact counterpart of the original. This cast may be compared to a bronze statue, representing merely the superficial form, and not the internal organisation; but there is another description of petrifaction by no means uncommon, and of a much more wonderful kind, which may be compared to certain anatomical models in wax, where not only the outward forms and features, but the nerves, blood-vessels, and other internal organs, are also shown. Thus we find corals, originally calcareous, in which not only the general shape, but also the minute and complicated internal organisation, is retained in flint. Such a process of fossilisation is still more remarkably ex- hibited in fossil wood, in which we often perceive not only the rings of annual growth, but all the minute vessels and medullary rays. Many of the minute cells and fibres of plants, and even those spiral vessels which in the living vegetable can only be discovered by the microscope, are preserved. Among many instances of the kind may be mentioned a fossil tree, seventy-two feet in length, found at Transverse section of a tree >-, /. ,-, XT ., . , . from the coal measures, Goslorth, near Newcastle, in sandstone turf of fi ^ood'' Sh Wing teX " Strata associated with coal - % cutting a transverse slice so thin as to transmit light, and magnifying it about fifty-five times, the structure as seen in fig. 72 is exhibited. A texture equally minute and complicated has been observed in the wood of large trunks of fossil trees found in the Craigleith quarry, near Edinburgh, where the stone was not siliceous, but consisted chiefly of calcium carbonate. The parallel rows of vessels here seen are the rings of annual growth, but in one part they are imperfectly preserved, the wood having probably decayed before the mineralising matter had penetrated to that portion of the tree. In attempting to explain the process of fossilisation in such cases, we may first assume that strata are very generally per- meated by water charged with minute portions of calcareous, siliceous, and other earths in solution. In what manner they 1 Witham, ' Fossil Vegetables,' 1831. Plate IV. fig 1. CH. vii.] BY MINERAL MATTER 71 become so impregnated will be afterwards considered. If an organic substance is exposed in the open air to the action of the sun and rain, it will in time decay, or be resolved into its component elements, consisting usually of oxygen, hydrogen, nitrogen, and carbon, which pass into the atmosphere as water, carbon dioxide, and ammonia. But if the same substance be submerged in water, it will decompose more gradually; and if buried in earth, still more slowly ; as in the familiar example of wooden piles or other buried timber. Now, if as fast as each particle is set free by decomposition, a particle of calcium carbo- nate, silica, or other mineral is at hand ready to be precipitated, we may imagine this inorganic matter to take the place just before left unoccupied by the organic molecule. In this manner a cast of the interior of certain vessels may first be taken, and afterwards the more solid walls of the same may decay and suffer a like transmutation. Yet when the whole is petrified, it may not form one homogeneous mass of stone. Some of the original ligneous, osseous, or other organic elements may remain mingled in certain parts, or the fossilising substance itself may be differently coloured at different times, or so crystallised as to reflect light differently, and thus the texture of the original body may be faithfully exhibited. The student may perhaps ask whether, on chemical princi- ples, we have any ground to expect that mineral matter will be thrown down precisely in those spots where organic decompo- sition is in progress. The following interesting experiments may serve to illustrate this point. Professor Goppert, of Breslau, with a view of imitating the natural process of fossilisation, steeped a number of animal and vegetable substances in waters, some holding siliceous, others calcareous, others metallic matter in solution. He found that in the period of a few weeks, or sometimes even days, the organic bodies thus immersed were mineralised to a certain extent. Thus, for example, thin ver- tical slices of deal, taken from the Scotch fir (Pinus sylvestris, L.), were immersed in a moderately strong solution of sulphate of iron. When they had been thoroughly soaked in the liquid for several days, they were dried and exposed to a red heat until the vegetable matter was burnt up and nothing remained but iron oxide, which was found to have taken the form of the deal so exactly that casts even of the dotted vessels peculiar to this family of plants were distinctly visible under the micro- scope. The exact reproduction of the minute internal structures of organisms is doubtless facilitated by the extreme slowness with which the changes take place. The molecules of the animal 72 PETRIFACTIONS AND INCRUSTATIONS [CH. vn. and vegetable tissues are one by one broken up a-nd removed in a gaseous state, and a particle of mineral matter takes its place. The highest powers of our microscopes are far from reaching the ultimate chemical molecules, and hence we are not surprised to find the individual cells and vessels of plants, and even the markings in these cells and vessels, exquisitely repro- duced in mineral matter; and all the minute structures in corals, shells, and bones preserved with the same delicacy. Even flowers and the wings of insects may sometimes be found exquisitely preserved in a fossil state. The chief substances which replace vegetable and animal remains, and thus form fossils or true petrifactions, are calcium carbonate (in the forms of the minerals calcite and aragonite), ferrous carbonate (which is often converted into various iron oxides more or less hydrated), iron disulphide (in the forms known as pyrites and marcasite), and silica (in the forms of opal, chalcedony, and quartz). But besides these four very common fossilising materials almost any mineral may be found replac- ing the substance of an animal or vegetable organism, or occu- pying the empty space left by its removal. These true ' fossils ' must not be confounded with the in- crustations of organic bodies by calcium carbonate which are found in districts where calcareous springs abound, and which are often erroneously spoken of as ' petrifactions.' In Derbyshire, Auvergne, and other similar districts, leaves, twigs, birds' nests, and even larger structures may be found coated over and pre- served by a thin stalagmite-like layer of calcium carbonate, which has been deposited from a so-called ' petrifying spring.' Indeed, such springs are sometimes employed for coating artificial sub- stances, and producing casts or reproductions of objects of art. We must also be on our guard against treating as fossils those accidental representations of natural objects which some- times occur in rocks. Thus in irregular flints and other con- cretions a more or less fanciful resemblance to twigs or nuts, fingers or toes, and various other parts of plants and animals may sometimes be traced. "When there is no ground for believing that such resemblances are more than accidental, we speak of the object as a ' pseudo-fossil.' There are some pseudo-fossils, however, which it is extremely difficult to distinguish from real fossils. On exposed rock-sur- faces we often find a curious ' mimic vegetation ' (like the frost on a window-pane), known as dendrites (figs. 73-75), and these have often been taken for the remains of plants. Similar structures are found enclosed within the so-called ' moss-agates.' On the surface of slates at Wicklow, Ireland, peculiar mark- CH. VII.] PSEUDO-FOSSILS 73 ings supposed to be the remains of plants, bryozoa or some other organisms, have been found, and have received the name of Oldhamia, two distinct species being recognised. Professor Sollas has shown grounds, however, for believing that these Fig. 73. Fig. 7L Dendrites on surfaces of flint hatchets in the drift of St. Acheul, near Amiens. a. Natural size. 6. Natural size. c. Magnified. d. Natural size. e. Magnified. markings are not really of organic origin, but may be ' pseudo-fossils,' formed by a peculiar wrinkling of the surfaces of a fine mud in drying. There are many tracks and burrows formed by worms and other creatures living on muddy and sandy shores that have been mistaken for fossil plants as has been shown by Nathorst. The most remarkable example of a pseudo-fossil, however, is the famous ' Eozoon canadense,' Daw., which was long regarded Fig. 76. OldJtamia radiata, Forbes. Wicklow, Ireland. Oldhamia antiqua, Forbes. Wicklow, Ireland. as the oldest known fossil. This consists of layers of white calcite and of a green silicate so remarkably intergrown as to simulate in a very striking manner the internal structure of certain organisms (foraminifera). Many very able naturalists have 74 EOZOON CANADENSE [CH. VII. been deceived by the curious resemblances of ' eozoon ' to real organisms. In the accompanying drawing the following struc- tures may be noted : (1) the arrangement of cells like those of foraminiferal shells ; (2) the rod-like connecting processes re- sembling stolons ; ' (3) the curious series of branching tubuli, like what is known in the * intermediate skeleton ' of those organisms ; and (4) the apparently finely perforated shell sub- stance (' nummuline layer '). These can all be made out under the microscope. In spite of these striking resemblances, however, the most recent investigations of microscopists and Fig. 78. Fig. 79. Eozoon canadense,' Daw. (after Carpenter). A pseudo-fossil. The structures sup- posed to be organic are indicated in these two figures by Dr. Carpenter, which give a diagrammatic representation of what was thought to be foraminiferal structure. Fig. 78. . Chambers of lower tier communicating at +, and separated from ad- joining chambers at by an intervening septum, traversed by passages. b. Chambers of an upper tier. c. Walls of the chambers, traversed by fine tubules (' Nummuline layer '). (These tubules pass with uniform parallelism from the inner to the outer surface, opening at regular distances from each other.) d. Intermediate skeleton, composed of homogeneous shell substance, traversed by stoloniferous passages (/), connecting the chambers of the two tiers, e. Canal system in intermediate skeleton, showing the arborescent sarcodic prolonga- tions. (Fig. 79 shows these bodies in a decalcified state.) /. Stoloniferous Fig. 79. Decalcified portion of natural rock, showing the supposed canal system and the several layers. Natural size. mineralogists leave little room for doubt that the structures are all of inorganic origin and of an imitative character. These and similar cases illustrate the necessity of great caution on the part of the geologist in discriminating between true fossils showing real organic structures and the curiously imitative structures which sometimes result from the action of segregative and crystallising forces acting within rock-masses. The chemical processes involved in the consolidation of strata, in the formation of nodules, casts, &c., and in the petrifaction of organic remains, are described in Professor Haughton's ' Manual of Geology,' Lecture IIL.and inDajia's Manual of Geology,' 4th edition, 1895. The mineralisation of plant remains has been very ably discussed by Professor Solms-Laubach in the introductory chapter to his ' Fossil Botany ' (English edition, 1891). CH. VIII.] 75 CHAPTER VIII ELEVATION OF STRATA ABOVE THE SEA HORIZONTAL AND INCLINED STRATIFICATION FAULTING Why the position of marine strata, above the level of the sea, should be referred to the rising up of the land, not to the going down of the sea Strata of deep-sea and shallow- water origin alternate Also marine and freshwater beds and old land surfaces Vertical, inclined, and folded strata Anticlinal and synclinal curves Dip and strike Struc- ture of the Jura Various forms of outcrop Synclinal strata forming ridges Connection of fracture and flexure of rocks Inverted strata Faults described Superficial signs of the same obliterated by denuda- tion Great faults the result of repeated movements Arrangement and direction of parallel folds of strata Unconformability Overlap Dip-slopes and escarpments Outliers and inliers. L.ind has been raised, not the sea lowered. It has been already stated that the aqueous rocks containing marine fossils extend over wide continental tracts, and are seen in mountain chains rising to great heights above the level of the sea. Hence it follows, that what is now dry land was 'once under w r ater. But if we admit this conclusion, we must imagine, either that there has been a general lowering of the waters of the ocean, or that the solid rocks, once covered by water, have been raised up bodily out of the sea, and have thus become dry land. The earlier geologists, finding themselves reduced to this alternative, embraced the former opinion, assuming that the ocean was originally universal, and had gradually sunk down to its actual level, so that the present islands and continents were left dry. It seemed to them far easier to conceive that the water had gone down than that solid land had risen upwards into its present position. It was, however, impossible to invent any satisfactory hypothesis to explain the disappearance of so enormous a body of water throughout the globe, it being neces- sary to infer that the ocean had once stood at whatever height marine shells might be detected. It moreover appeared clear, as the science of Geology advanced, that certain areas on the globe had been successively sea, land, estuary, and then sea again, to finally become once more habitable land; and that they re- mained in each of these states for considerable periods. In order to account for such phenomena, without admitting any movement of the land itself, we are required to imagine several retreats and returns of the ocean ; and even then our theory applies merely to cases where the marine strata composing the dry- land are horizontal, leaving unexplained those more common 76 LAND HAS BEEN KAISED, [CH. vm. instances where strata are inclined, curved, or placed on their edges, and evidently not in the position in which they were first deposited. Geologists, therefore, were at last compelled to have recourse to the doctrine that the solid land has been repeatedly moved upwards or downwards, so as permanently to change its position relatively to the sea. There are several distinct grounds for pre- ferring this conclusion. First, it will account equally for the position of those elevated masses of marine origin in which the stratification remains horizontal, and for those in which the strata are disturbed, broken, inclined, or vertical. Secondly, it is consistent with human experience that land should rise gradually in some places and be depressed in others. Such changes have actually occurred in our own days, and are now in progress, having been accompanied in some cases by violent convulsions, while in others they have proceeded so insensibly as to have been ascertainable only by the most careful scientific observations, made at considerable intervals of time. On the other hand, there is no evidence from human experience of a rising or lowering of the sea's level in any region, and the ocean cannot be raised or depressed in one place without its level being changed all over the globe. The vast bulk of the oceans as compared with that of the land rising above the sea-level renders it improbable that great changes in the relative position of land and water can be due to changes in the sea-level. At the same time, it must be remem- bered that minor alterations in the relations of land and sea may be due to local variations in the sea-level ; for it has been shown that the attraction of the land-masses and other causes prevent the ocean level being that of a true sphere. These preliminary remarks will prepare the reader to under- stand the great theoretical interest attached to all facts con- nected with the position of strata, whether horizontal or in- clined, curved or vertical. Now the first and most simple appearance is where strata of marine origin occur above the level of the sea in horizontal position. Such are the strata which we meet with in the south of Sicily, filled with shells for the most part of the same species as those now living in the Mediterranean. Some of these rocks rise to the height of more than 2,000 feet above the sea. Other mountain masses might be mentioned, composed of horizontal strata of high antiquity, which contain fossil remains of animals wholly dissimilar to any now known to exist. Tn the south of Sweden, for example, near Lake Wener, the beds of some of the oldest fossiliferous deposits, called Silurian and Cambrian CH. viii.] NOT THE SEA LOWERED 77 by geologists, occur in as level a position as if they had recently formed part of the delta of a great river, and been left dry on the retiring of the annual floods. Instead of imagining that such fossiliferous rocks were always at their present level, and that the sea was once high enough to cover them, we suppose them to have constituted the ancient bed of the ocean, and to have been afterwards uplifted to their present height. This idea, however startling it may at first appear, is quite in accordance, as before stated, with the analogy of changes now going on in certain regions of the globe. Thus in parts of Sweden, and the shores and islands of the Gulf of Bothnia, proofs have been obtained that the land is experiencing, and has experienced for centuries, a slow up- heaving movement. It appears, from the observations of Mr. Darwin and others, that very extensive regions of the continent of South America have been undergoing slow and gradual upheaval, by which the level plains of Patagonia, covered with recent marine shells, and the Pampas of Buenos Ayres, have been raised above the level of the sea. On the other hand, the gradual sinking of the west coast of Greenland, for the space of more than 600 miles from north to south, during the last four centuries, has been established by the observations of a Danish naturalist, Dr. Pingeh And while these proofs of continental elevation and subsidence, by slow and insensible movements, have been brought to light, the evidence has been daily strengthened of continued changes of level effected by violent convulsions in countries where earthquakes are frequent. Mr. Darwin has also inferred that, in those parts of the Pacific and Indian Oceans where circular coral islands (atolls) and barrier reefs abound, there is a slow and continued sinking of the submarine mountains on which the masses of coral are based, while there are other areas where the land is on the rise, and where coral has been upheaved far above the sea-level. l The long submerged river valleys known as fiords in Scandi- navia and as firths in this country bear striking testimony to the fact that on many coasts subsidence has taken place. These long winding valleys are quite different from any features that are produced by marine denudation ; they have evidently been formed by the erosion of the streams which still occupy their higher and unsubmerged portions. Along our coasts we find numerous submerged forests, only visible at low water, having the trunks of the trees erect and 1 See Note D, p. 601. 78 ALTERNATIONS OF STRATA [CH. vin. their roots attached to them and still spreading through the ancient soil as when they were living. They occur in too many places, and sometimes at too great a depth, to be explained by a mere change in the level of the tides, although, as the coasts waste away and alter in shape, the height to which the tides rise and fall is always varying, and the level of high tide at any given point may, in the course of many ages, differ by several feet or even fathoms. It is this fluctuation in the height of the tides, and the erosion and destruction of the sea-coast by the waves, that makes it exceedingly difficult for us in a few cen- turies, or even perhaps in a few thousand years, to determine whether there is a change by subterranean movement in the relative level of sea and land. We often behold, as on the coasts of Devonshire and Pem- brokeshire, facts which appear to lead to opposite conclusions. In one place is a raised beach with marine littoral shells, and in another immediately adjoining may be a submerged forest. 1 Alternations of marine and freshwater strata. It has been shown in the sixth chapter that there is such a difference between land, freshwater, and marine fossils as to enable the geologist to determine whether particular groups of strata were formed at the bottom of the ocean or in estuaries, rivers, or lakes. If surprise was at first created by the discovery of marine corals and shells at the height of several miles above the sea-level, the imagination was afterwards not less startled by observing that in some successive strata composing the earth's crust, with a thickness amounting to thousands of feet, were comprised formations of littoral as well as of deep-sea origin, of beds of brackish or even of purely freshwater forma- tion, and of others containing vegetable matter or coal which accumulated on ancient land. In these cases we as frequently find freshwater beds below a marine series, or shallow water under those of deep-sea origin, as the reverse. Thus in boring an Artesian well in London, we pass through a marine clay (the London clay), and then reach, at the depth of several hundred feet, a shallow-water and fluviatile sand (the Woolwich and Beading beds) beneath which comes the white chalk originally formed in a deep sea. Or, if we bore vertically through the marine Lower Greensand of Surrey, we come upon a freshwater formation many hundreds of feet thick, called the Wealden, such as is seen in Kent and Surrey, and this is known in its turn to rest on other purely marine beds. In like manner, in various parts of Great Britain we sink vertical shafts through lacustrine deposits of great thickness, and come upon coal, which was formed by the growth of plants on an ancient land-surface. 1 See Note E, p. G02. CH. viii.] CHANGES IN THE POSITION OF STRATA 79 Fig. 80. Vertical, inclined, and curved strata. It has been stated that marine strata of different ages are sometimes found at a considerable height above the sea, yet retaining their original horizontality ; but this state of things is quite exceptional. As a general rule, strata are inclined or bent in such a manner as to imply that their original position has been altered. The most unequivocal evidence of such a change is afforded by their standing up vertically showing their edges, which is by no means a rare phenomenon, especially in mountainous coun- tries. Thus we find in Scotland, on the southern skirts of the Grampians, beds of conglomerate alternating with thin layers of fine sand, all placed vertically to the horizon. When De Saussure first observed certain conglomerates in a similar position in the Swiss Alps, he remarked that the pebbles, being for the most part of an oval shape, had their longer axes parallel to the planes of stratification (see fig. 80). From this he in- ferred that such strata must, at first, have been horizontal, each oval pebble having settled at the bottom of the water, with its flatter side parallel to the horizon. Some few, indeed, of the rounded stones in a con- glomerate occasionally afford an exception to the above rule, for the same reason that in a river's bed, or in a shingle beach, some pebbles rest on their ends or edges; these having been shoved stones by a wave or current so as to assume this position. Anticlinal and synclinal curves. Vertical strata, when they can be traced continuously upwards or downwards for some depth, are almost invariably seen to be parts of great curves, which may have a diameter of a few yards or of several miles. We will first describe two curves of considerable regularity, which occur in Forfarshire, extending over a country twenty miles in breadth, from the foot of the Grampians to the sea near Arbroath (fig. 81). The mass of strata exhibited may be 2,000 feet in thick- ness, consisting of red and white sandstone and various coloured shales, the beds being distinguishable into four principal groups namely : No. 1, red marl or shale ; No. 2, red sandstone, used for building ; No. 3, conglomerate; and No. 4, grey paving-stone, and tile-stone with green and reddish shale, containing peculiar organic remains. A glance at the section will show that each of the formations 2, 3, 4 is repeated thrice at the surface, twice Vertical conglomerate and sandstone, against or between other 80 ANTICLINALS AND SYNCLINALS [CH. vin. with a southerly, and once with a northerly inclination or dip ; and the beds in No. 1, which are nearly horizontal, are still brought up twice. by a slight curva- ture to the surface, once on each side of A. Beginning at the north- west extremity, the tile-stones and conglomerates, No. 4. and No. 3, are vertical, and they generally form a ridge parallel to the southern skirts of the Grampians. The superior strata, Nos. 2 and 1, become less and less inclined on descending to the valley of Strathmore, where the yi in I ~s strata, having a concave bend, are I/ Hli I & said by geologists to lie in a ' trough ' or ' basin.' Through the centre of this valley runs an ima- ginary line A, called technically a ' synclinal axis,' where the beds, which are tilted in opposite direc- tions, may be supposed to meet. It is most important for the observer to mark the position of such axes, for he will perceive by the diagram that, in travelling from the north to the centre of the basin, he is always passing from older to newer beds ; whereas, after crossing the line A, and pursuing his course in the same southerly direction, he is continually leaving the newer, and advancing upon older strata. All the deposits which he had before examined begin then to recur in reversed order, until he arrives at the central axis of the Sidlaw hills, where the strata are seen to form an arch or saddle, having an anti- clinal axis B in the centre. On passing this axis, and continuing towards the S.E., the formations 4, 3, and 2 are again repeated, in the same relative order of superposition, but with a southerly dip. At Whiteness (see diagram) it will be seen that the inclined strata are covered by a newer deposit, a, CH. VIII.] TROUGHS AND SADDLES 81 in horizontal beds. These are composed of red conglomerate and sand, and are newer than any of the groups, 1, 2, 3, 4, before described, ' and rest unconformably upon strata of the sandstone group No. 2. Strata which are bent into a vertical, or nearly vertical position, and afterwards resume their original horizontality, are said to exhibit a ' unicliiial ' or ' monoclinal ' fold. A -good example of such a monoclinal fold is exhibited by the beds of the Isle of Wight ; and the same phenomenon is often presented on a much grander scale in the Western territories of the United States. An example of curved strata, in which the bends or plica- tions of the rock are sharper and far more numerous within an equal space, has been well described by Sir James Hall. It Fig. 82. Curved strata of slate near St. Abb's Head, Berwickshire. (Sir J. Hall.) occurs near St. Abb's Head, on the east coast of Scotland, where the rocks consist principally of a bluish slate, having frequently a ripple -marked surface. The undulations of the beds reach from the top to the bottom of cliffs from 200 to 300 feet in height, and there are sixteen distinct bendings in the course of about six miles, the curvatures being alternately concave and convex upwards. All these strata were once horizontal. Folding- by lateral movement. An experiment was made by Sir James Hall, with the object of illustrating the manner in which such strata, assuming them to have been originally horizontal, may have been forced into their present position. 2 A set of layers of clay were placed under a weight, and their opposite ends pressed towards each other with such force as to cause them to approach more nearly together. On the removal 2 Similar experiments have Lord Avebury, and others* been described by Mr. Cadell, 82 DENUDATION OF CURVED STRATA [en. vnr. of the weight the layers of clay were found to be curved and folded, so as to bear a miniature resemblance to the strata in the cliffs of St. Abb's Head. We must, however, bear in mind that in the natural section or sea-cliff we only see the foldings imperfectly, one part being invisible beneath the sea, and the other, or upper portion, being supposed to have been carried away by denudation, or that action of water which will be ex- plained in the next chapter. The dark lines in the accom- panying plan (fig. 83) -represent what is actually seen of the strata in the line of cliff alluded to ; the fainter lines indicate that portion which is concealed beneath the sea-level, as also that which is supposed to have once existed above the present level. In some cases the flexures found in rocks form regular and sweeping curves; in other cases sharp angular foldings are produced, and in other cases the axis plane of the fold becomes inclined, and overfolding with inversion is the result. We may still more easily illustrate the effects which a lateral thrust must produce on flexible strata, by placing several pieces of differently coloured cloths upon a table, and when they are spread out horizontally, covering them with a book ; then applying other books to each end, and forcing them towards each other. The folding of the cloths (see fig. 84) will imitate those of the bent strata ; the incumbent book being slightly lifted up, and no longer touching the two volumes on which it rested before, because it is supported by the tops of the anticlinal ridges formed by the curved cloths. In like manner there can be no doubt that the squeezed strata, although late- rally condensed and more closely packed, are yet elongated and made to rise upwards in a direction perpendicular to the pressure. en. vin.] CUEVATUEE DUE TO LATEEAL PEESSUEE 83 Whether the analogous flexures in stratified rocks have really been due to similar lateral movements is a question which we cannot decide by reference to our own observation. Our inability to explain the nature of the process is, perhaps, not simply owing to the inaccessibility of the subterranean re- gions where the mechanical force is exerted, but to the extreme slowness of the movement. The changes may sometimes be due to variation in the temperature and chemical constitution of mountain masses of rock, causing them, while still solid, to expand or contract. If such be the case, we have scarcely more reason to expect to witness the operation of the process within the limited periods of our scientific observation than to see the swelling of the roots of a tree, by which, in the course of years, a wall of solid masonry may be lifted up, rent, or thrown down. In both instances the force may be irresistible, but though adequate, it need not be visible to us, provided the time re- Fig. 84. quired for its development be very great. The lateral pressure arising from the unequal expansion of rocks by heat may cause one mass lying in the same horizontal plane gradually to occupy a larger space so as to press upon another rock, which, if flexible, may be squeezed into a bent and folded form. It will also ap- pear, when the volcanic and plutonic rocks are described, that some of them, when melted in the interior of the earth's crust, have been injected forcibly into fissures, and after the solidifi- cation of such intruded matter, other sets of rents, crossing the first, have been formed and in their turn filled by melted rock. Such repeated injections imply a stretching, and often upheaval, of the whole mass. We also know, especially by the study of regions liable to earthquakes, that there are causes at work in the interior of the earth capable of producing a sinking in of the ground, some- times very local, but often extending over a wide area. The continuance of such a downward movement, especially if partial o2 84 STRIKE AND DIP [CH. vin. and confined to linear areas, may produce regular folds in the strata. But the cause of the great flexures and curvatures of strata that are such grand features in mountains is the same as that which produces elevation and subsidence on the greatest scale, and is that which produced the continents and sea-floors. The force was directed tangentially to the earth's surface, and lateral compression resulted ; the original horizontal strata were forced into anticlinal and synclinal curves, and the breadth of area was diminished. The force was the outcome of the energy of heat within the globe. As the internal heat was conducted to the surface through cooling rocks, to be radiated into space, contraction occurred. The contraction was unequal, because rocks contract at different rates in cooling. These irregular contractions produced dragging down of the superficies, and a resolved force was produced, the direction of which was tangential. The phenomena of slaty cleavage, and some metamorphism, hereafter to be considered, are the proofs of the direction of the force and of its effects. The positions assumed by strata in mountain -chains are also evidences of the same forces. It is in mountain-chains, indeed, that we find the most striking examples of the extreme results of lateral pressure upon stratified rock-masses. The complicated folds have their axes greatly inclined, and the middle limb is frequently dragged out or crushed, so that the fold is converted into a fault, as was shown by H. D. Rogers. Very exaggerated examples of such broken folds are called by some authors thrusts ; and examples of them have been described in the Appalachians, the Scottish Highlands, and the Alps. This subject will be more fully con- sidered in connection with the study of the metamorphic rocks. Dip and strike. In describing the manner in which strata depart from their original horizontality, the technical terms F 85 such as ' dip ' and ' strike ' are used by geologists. These we shall now pro- ceed to explain. If a stratum or bed of rock, instead of being quite level, be incHned to the horizon, it Is said to dip the point of the compass to which it is inclined is called the direction of dip, and the degree of deviation from a horizontal plane is called the amount of dip, or the angle of dip. Thus, in the annexed diagram (fig. 85), a series of strata are in- clined, and they dip to the north at an angle of forty-five CH. viii.] APPARENT AND TRUE DIP 85 degrees, The strike, or line of bearing, is the prolongation or extension of the strata in a direction at right angles to the dip. Thus, in the above instance of strata dipping to the north, their strike must necessarily be east and west. We have borrowed the word from the German geologists, streiclien signifying to extend, to have a certain direction. A stratum which is horizontal, or quite level in all directions, has neither dip nor strike. It is always important for the geologist, who is endeavouring to comprehend the structure of a country, to learn how the beds dip in every part of the district ; but it requires some practice to avoid being occasionally deceived, both as to the direction of dip and the amount of it. If the upper surface of a hard stony stratum be uncovered, whether artificially as in a quarry, or by the waves at the foot of Fig. 86. Apparent horizontality of inclined strata. a cliff, it is easy to determine towards what point of the compass the slope is steepest, or in what direction water would flow, if poured upon it. This is the true dip. But the edges of highly inclined strata may give rise to perfectly horizontal lines in the face of a vertical cliff, if the observer see the strata in the line of their strike, the dip being inwards from the face of the cliff. If, however, we come to a break in the cliff, which exhibits a section exactly at right angles to the line of the strike, we are then able to ascertain the true dip. In the drawing above (fig. 86), we may suppose a headland, one side of which faces to the north, where the beds would appear perfectly horizontal to a person in the boat; while on the other side, facing the west, the true dip would be seen by the person on shore to be at an angle of 40. If, therefore, our observations are confined to a vertical precipice facing in one direction, we endeavour \g fiji4 a. ledge or portion of the plane of one 86 MEASUREMENT OF DIP AND STKIKE [CH. vm. of the beds projecting beyond the others, in order to ascertain the true dip. The true dip is always at right angles to the strike ; any inclination of strata measured on a plane which is not at right angles to the strike we call apparent dip. Many of the in- clinations of strata seen in sea-cliffs, quarries, &c., are evidently apparent and not true dips. From one or more apparent dips, the relation of which to the strike is known, it is always possible to calculate the true dip of a bed. Dips (apparent and true) are measured by means of instruments called clinometers, in which the vertical is given by a plumb-line or the horizontal plane by a spirit level. If not provided with a clinometer a most useful instrument when it is of consequence to determine the inclination of the strata with precision the observer may measure the angle within 87 a few degrees, by standing exactly opposite to a cliff where the true dip is exhibited, holding the hands immediately before the eyes, and placing the fingers of one in a perpendicular and of the other in a horizontal position, as in fig. 87. It is thus easy to discover whether the lines of the inclined beds bisect the angle of 90, formed by the meeting of the hands, so as to give an angle of 45, or whether it would divide the space into two equal or unequal portions. You have only to change hands to get the dip indicated by the lower dotted line on the upper side of the horizontal hand. It has been already seen, p. 80, in describing the curved strata on the east coast of Scotland, in Forfarshire and Berwick- shire, that a series of concave and convex bendings are occa- sionally repeated several times. These usually form part of a series of parallel waves of strata, which are prolonged in the same direction, throughout a considerable extent of country. Thus, for example, in the Swiss Jura, that lofty chain of moun- tains has been proved to consist of many parallel ridges, with intervening longitudinal valleys, as in fig. 88, the ridges being formed by curved fossiliferous strata, the nature and dip of which are occasionally displayed in deep transverse gorges, called * cluses,' caused by fractures at right angles to the direction of the chain. Now let us suppose these ridges and parallel valleys to run north and south, we should then say that the CH. VIII.] OUTCROP OF STRATA 87 strike of the beds is north and south, and the dip east and west. Lines drawn along the summits of the ridges A, B, would be anticlinal axes, and one following the bottom of the adjoining valleys a synclinal axis. It frequently happens that while the inclination of a series of strata is, on the whole, in one particular direction, we find Section illustrating the structure of the Swiss Jura. many irregularities in the amount of dip at certain points, and that occasionally the dip may be reversed. The careful study of such strata shows that, while having a general slope in one direction, the beds really lie in a series of very flat folds. The prevailing inclination of the beds we speak of as the general dip ; minor exhibitions of slope in the beds we call local dip. Outcrop of strata. It will be observed that some of these ridges. A, B (fig. 88), are unbroken on the summit, whereas one of them, C, has had its . -, Fig. 89. Fig. 90. upper portion carried away by denudation, so that the ridges of the beds in the formations * Slope of valley 20, dip of strata 20, in opposite directions. inclination, the newer beds will appear the highest, as in the first and second cases. This is shown by the drawing (fig. 93), which exhibits strata rising at an angle of 20, and crossed by a valley, which declines in an opposite direction at 20. These rules may often be of great practical utility ; for the different degrees of dip occurring in the two cases represented in figs. 91 and 92 may occasionally be encountered in following the same line of flexure at points a few miles distant from each other. A miner unacquainted with the rule, who had first explored the valley (fig. 91), may have sunk a vertical shaft below the coal seam A, until he reached the inferior bed B. He might then pass to the valley (fig. 92), and discovering there also the outcrop of two coal seams, might begin his workings in the uppermost in the expectation of coming down to the other bed A, which woulcl be observed cropping out lower cjowr* th 90 RIDGES FORMED BY SYNCLINAL STRATA [CH. vm. valley. But a glance at the section will demonstrate the futility of such hopes. 3 Synclinal strata forming- ridges. Although in some cases an anticlinal axis forms a ridge, and a synclinal axis a valley, as in A, B, fig. 88, p. 87, yet this can by no means be laid down as a general rule, as the beds very often slope inwards from either side of a mountain, as at a, b, fig. 94, while in the intervening valley c they slope upwards, forming an arch. Synclinal. Grits and shales. Mountain limestone. Grits and shales. Section of carboniferous rocks of Lancashire. (E. Hull.) At the western extremity of the Pyrenees, great curvatures of the strata are seen in the sea-cliffs, where the rocks consist of marl, grit, and chert. At certain points, as at a, fig. 95, some of the bendings of the flinty chert are so sharp that specimens might be broken off, well fitted to serve as ridge -tiles on the roof of a house. Although this chert could not have been brittle as now, when first folded into this shape, it presents, nevertheless, here and there at the points of greatest flexure small cracks, which show that it was solid, and not wholly in- capable of breaking, at the period of its displacement. The numerous rents alluded to are not empty, but filled with chalce- dony and quartz. Fig. 95. Strata of chert, grit, and marl near St. Jean de Luz. It would be natural to expect the fracture of solid rocks to take place chiefly where the bending of the strata has been sharpest ; the entire absence, however, of such cracks at points where the 3 Sir C. Lyell was indebted to the late T. Sopwith, Esq., for the models which he had copied in the above diagrams; but the be- ginner may find it by no means easy- to understand such copies, although, if he were to examine and handle the originals, turning them about in different ways, he would at once comprehend their meaning as well as the import of others far more complicated, which the same engineer has constructed to illustrate faults. CH. VIII.] BENDING OF STEATA 91 Fig. 96. strain must have been greatest, as at a, fig. 95, is often very remarkable and not always easy of explanation. We must imagine that many strata of limestone, chert, and other rocks which are now brittle, were pliant when bent into their present position. It must be remembered that large masses of matter behave very differently from small fragments when force is applied to them. Ice, sealing-wax, and glass are brittle substances, but long rods of these substances are capable of being bent and twisted without breaking. In many cases, too, such substances behave very differently when a force is slowly applied and when it is suddenly brought into action, and when changes of tem- perature are taking place in a mass it yields much more easily than when maintained at a uniform temperature. The great rock-masses of the earth's crust are of enormous dimensions, they have been subjected to extraordinary variations in tempe- rature, and the forces which have operated on them have acted with extreme slowness. Between Santa Caterina and Castrogiovanni, in Sicily, bent and undulating gypseous marls occur, with here and there thin beds of solid gypsum interstratified. Some- times these solid layers have been broken into detached fragments, still preserving their sharp edges (g g, fig. 96), while the con- tinuity of the more pliable and ductile marls, in m, has not been interrupted. We sometimes find that pebbles, fossils, and other objects included in bent and folded strata exhibit in their crushed and dislocated appearances clear evidence of the great pressure to which the rocks have been subjected. In some cases pebbles of limestone, and even of quartzite, have been thrust against one another with such irresistible force as to cause mutual im- pressions to be produced upon them ; these are called impressed pebbles by geologists. Slickensides are grooved or polished surfaces of rock produced by the grinding of one part of the rock against another during the movements which have taken place. We have already explained (fig. 94) that stratified rocks have their strata usually bent into parallel folds forming anticlinal and synclinal curves, a group of several of these folds having often been subjected to a common movement, and having ac- quired a uniform strike or direction. In some disturbed regions g gypsum m marl. 92 FOLDING AND INVERSION [cm. vni these folds have been doubled back upon themselves in such a manner that it is often difficult for an experienced geologist to determine the relative age of the beds correctly by superposition. Thus, if we meet with the strata seen in the section, fig. 97, we should naturally suppose that there were twelve distinct beds, or sets of beds, No. 1, the uppermost, being the newest, and No. 12 the oldest of the series. But this section may perhaps exhibit merely six beds, which nave been folded in the manner seen in fi g- W, so that each of them is repeated, the^ position of one half being reversed, and part of No. 1, originally the uppermost, having now become the lowest of the series. The upper part of the curves seen in this diagram, fig. 98, and expressed in fainter lines, has been removed by denudation. The phenomena of folding, inversion, and reversal of strata are seen on a magnificent scale in certain regions in Switzer- land, in precipices often more than 2,000 feet in perpendicular height, and there are flexures not inferior in dimensions in the Pyrenees. Ordinary inversion of strata is well seen near Milford, and is explained in the diagram, fig. 99. On passing from N to S the Fig 98 topmost strata, 3, are lower than 2 and 1. The folding is on such a grand scale and has been so sharp in the Alps that old meta- morphic rocks, whose place is below the sedi- mentary strata, have become included in the folds and exposed by denudation. The old rocks then appear newer than some of the younger strata. In the Mont -Blanc range the lateral crush has been sufficient to cause the sedimentary strata to dip under the old crystalline schists, as will be explained when treating of metamorphic rocks. Fractures of the strata and faults. "When the force to which a rock-mass has been subjected has resulted in the fracture and displacement of its two portions, we have the phenomenon known to geologists as a fault. Numerous rents may often be seen in rocks which appear to have been simply , the fractured parts still remaining in contact j \>M we CH. VIII.] FRACTURE OF STRATA 93 often find a fissure, several inches or yards wide, intervening between the disunited portions. These fissures are sometimes filled with fine earth and sand, or with angular fragments of stone, evidently derived from the crushing of the contiguous rocks. The face of each wall of the fissure is often beautifully polished, as if glazed, striated, or scored with parallel furrows and ridges, such as would be produced by the continued rubbing together of surfaces of unequal hardness. These are the polished surfaces already referred to as ' slickensides.' It is supposed that the lines of the striae indicate the direction in which the rocks were moved. During one of the minor earthquakes in Chili, in 1840, the brick walls of a building were rent vertically in several places, and made to vibrate for several minutes during each shock, after which they remained uninjured, and without any opening, although the line of each crack was still visible. When Fig. 99. S 1 2 8 3 2 Inverted beds near Milford Haven. (After Green.) 3. Top. Carboniferous limestone. 2. Carboniferous shale. 1. Bottom. Old red sandstone. all movement had ceased, there were seen on the floor of the house, at the bottom of each rent, small heaps of fine brick- dust, evidently produced by trituration. It is not uncommon to find the mass of rock, on one side of a fissure, thrown up above or down below the mass with which it was once in contact on the other side. This mode of dis- placement is called a fault, shift, slip, or throw. Playfair, in describing a fault, remarks : ' The miner is often perplexed, in his subterraneous journey, by a derangement in the strata, which changes at once all those lines and bearings which had hitherto directed his course. When his mine reaches a certain plane, which is sometimes perpendicular, as in A B, fig. 100, sometimes oblique to the horizon (as in C D, ibid.), he finds the beds of rock broken asunder, those on the one side of the plane having changed their place, by sliding in a particular direction along the face of the others. In this motion they have sometimes 94 VARIETIES OF FAULTS [CH. VIII, preserved their parallelism, as in fig. 100, so that the strata on each side of the faults A B, C D, continue parallel to one another ; in other cases, the strata on each side are inclined, as in a, b, c, d (fig. 103), though their identity is still to be recog- nised by their possessing the same thickness and the same in- ternal characters. rig. 100. A ^ s^ V" "- \ v N 1 B D Faults. A B vertical. C D hading towards the downthrow. Faults are sometimes vertical, as at A B, fig. 100, but usually they are inclined (C D). The inclination of a fault from the vertical is called its hade. Ordinary faults are those in which the 'hade ' is towards the downthrow side of the fault (see fig. 101). Reversed faults are those in which the hade is in the opposite direction (see fig. 102). In the case of an ordinary fault a pit may Fig. 101. Fig. 102. Ordinary fault, a b is the throw or amount Reversed fault, of vertical displacement. In both cases the bending near the fracture indicates the direction in which the dislocated portion must be sought for. be sunk so as to avoid the faulted bed altogether ; while in the case of a reversed fault a boring or pit may pass through the same bed twice. As seen from these sketches, the strata on the upthrow side of a fault are often bent towards the downthrow, and the opposite is the case on the downthrow side. Lateral displacement of strata occurs in relation to the departure of the fault from the vertical. Usually, the ends of strata close to a OH. viii.] EFFECTS PRODUCED BY FAULTING 95 fault are more or less bent : those which have dropped down are bent up against the line of fault, and those which have been pushed up have their edges forced downwards (see figs. 101, 102). In Coalbrook Dale, deposits of sandstone, shale, and coal, several thousand feet thick and occupying an area of many miles, have been shivered into fragments, and the broken rem- nants have been placed in very discordant positions, often at levels differing several hundred feet from each other. The sides of the faults, when perpendicular, are commonly several yards apart, and are sometimes as much as 50 yards asunder, the in- terval being filled with broken debris of the strata (fault-rock). In following the course of the same fault it is sometimes found to produce in different places very unequal changes of level, the amount of shift being in one place 300 and in another 700 feet ; this may arise from the union of two or more faults. In other cases, the disjointed strata may in certain districts have been subjected to renewed movements, which they have not suffered elsewhere. Fig. 103. E F, fault or fissure filled with crushed material (fault-rock) on each side of which the shifted strata are not parallel. We may occasionally see exact counterparts of these slips, on a small scale, in pits of loose sand and gravel, many of which have doubtless been caused by the drying and shrinking of ar- gillaceous and other beds, slight subsidences having taken place from failure of support. Sometimes, however, even these small slips may have been produced by the subterranean movements which are occasionally accompanied by earthquakes ; for land has been moved, and its level, relatively to the sea, considerably altered, since much of the alluvial sand and gravel now covering the surface of continents was deposited. A remarkable instance of the occurrence of the changes just alluded to, in modern times, was observed in New Zealand during the earthquake of January 1855. In the course of the subterranean disturbances a fracture in the strata was produced, extending for a distance of 90 miles. On one side of this fissure, the land was elevated in places as much as 9 feet, so as to form an inland cliff of that height, but on the other side the strata 96 FAULTS AND EARTHQUAKES [CH. vin. were unaffected. At the same time, a large district in the North Island, in the neighbourhood of Wellington, was upraised, while on the opposite side of Cook's Strait a subsidence of 5 feet took place, so that ships were obliged to go three miles higher up the river Wairau to obtain a supply of fresh water. A still more remarkable example of movements of the nature of a fault being observed in connection with an earthquake was that which has been described by Dr. B. Koto as occurring in Japan in 1891. In this case a great rent was produced which could be traced for more than 50 miles, and the surface of the ground was up- heaved on one of the sides of the fracture, in some cases to the extent of 20 feet. In the roads which traversed the country, moreover, lateral shifting could be seen to have taken place on opposite sides of the fissure, exactly like that which is rendered manifest when faulted beds are accurately mapped. Pig. 104. Apparent alternations of strata caused by vertical faults. We have already stated (p. 92) that a geologist must be on his guard, in a region of disturbed strata, against inferring repeated alternations of rocks, when, in fact, the same strata, once con- tinuous, have been bent round so as to recur in the same section, and with the same dip. A similar mistake has often been occasioned by a series of faults. If, for example, the dark line A H (fig. 104) represents the surface of a country on which the strata a b c frequently crop out, an observer, who is proceeding from H to A, might at first imagine that at every step he was approaching new strata, whereas the repetition of the same beds has been caused by vertical faults, or downthrows. Thus, suppose the original mass, A, B, C, D, to have been a set of uniformly inclined strata, and that the different masses under E F, F G, and G D, sank down successively, so as to leave vacant the spaces marked in the diagram by dotted lines, and to occupy those marked by en. viii.] EXTENT OF DISPLACEMENT BY FAULTS 97 the continuous lines, then let denudation take place along the line A H, so that the protruding masses indicated by the fainter lines are swept away a miner who has not discovered the faults, finding the mass a, which we will suppose to be a bed of coal four times repeated, might hope to find four beds, workable to an indefinite depth, but first on arriving at the fault G he is stopped suddenly in his workings, for he comes partly upon the shale b, and partly on the sandstone c ; the same result awaits him at the fault F, and on reaching E he is again stopped by a wall composed of the rock d. The very different levels at which the separated parts of the same strata are found on the different sides of the fissure, in some faults, are truly astonishing. One of the most celebrated faults in England is called the ' ninety-fathom dike,' in the coal- field of Newcastle. This name has been given to it because the same beds are ninety fathoms (540 feet) lower on the northern than they are on the southern side. The fissure has been filled by a body of sand, now converted into sandstone, which is sometimes very narrow, but in other places more than twenty yards wide. The walls of the fissure are scored by grooves, such as would have been produced if the broken ends of the rock had been rubbed along the plane of the fault. In the Tynedale and Craven faults, in the north of England, the vertical displacement, or ' amount of throw,' as it is tech- nically called, is still greater, and the fracture has extended in a horizontal direction for a distance of thirty miles or more. In the district of Morvern, on the shores of the Sound of Mull, tertiary basalts are faulted against the gneiss of the district; the throw of the fault being about 2,000 feet. Sir Andrew Eamsay described a fault in North Wales as having a throw of 12,500 feet, or nearly 2.j miles ! Some faults run in the same direction as the dip of the strata; they produce a lateral shift of the beds. Others are along the strike (strike faults), and often blot out strata by not allowing them to reach the surface. Step faults carry down a stratum, which may be near the surface, by a series of parallel dislocations, so that it becomes deeper and deeper, as it were, along a set of steps ; while trough faults let down a portion of a stratum, which is brought back nearly to its normal position by dislocation in opposite directions (see fig. 261, p. 235). 4 Great faults the result of repeated movements. It must not, however, be supposed that faults generally consist of single linear rents ; there are usually a number of faults springing off 4 The results produced by fault- in models, like those of Sop with, ing, especially when seen on de- Similar models may be made in unded surfaces, are best studied folded cardboard. 98 OUTCROPS AFFECTED BY FAULTS [OH. vm. from the main one, and sometimes a long strip of country broken up into fragments by sets of parallel and connecting transverse faults. Oftentimes a great line of fault has been repeated or the movements have been continued through suc- cessive periods, so that, newer deposits having covered the old line of displacement, the strata both newer and older have given way along the old line of fracture. Protruding masses of rock forming precipices or ridges along the lines of great faults may occur ; but they have usually been removed by denudation. This is well exemplified in nearly every coal-field which has been extensively worked. It is in such districts that the former relation of the beds which have been shifted is determinable with great accuracy. Thus in the coal-field of Ashby-de-la-Zouch, in Leicestershire (see fig. 105), a fault occurs, on one side of which the coal-beds, abed, must have been raised to the height of 500 feet above the cor responding Fig. 105. Faults and denuded coal strata, Ashby-cle-la-Zouch. pfammatt.) beds on the other side. But the uplifted strata do not stand up 500 feet above the general surface ; on the contrary, the out- line of the country, as expressed by the line z z, is uniformly undulating without any break, and the mass indicatpd by the dotted outline must have been denuded off and carried away. In the Lancashire coal-field the vertical displacement has amounted to thousands of feet, and yet all the superficial in- equalities which must have resulted from such movements have been obliterated by subsequent denudation. It appears that there are proofs of there having been two periods of vertical movement in one of the faults one, for example, before, and another after the Triassic epoch. An hypothesis which attributes such a change of position to a succession of movements is far preferable to any theory which assumes each fault to have been accomplished by a single up- cast or downthrow of several thousand feet. For we know that there are operations now in progress, at great depths in the CH. vni.J UNCONFORMITY 99 interior of the earth, by which both large and small tracts of ground are made to rise above and sink below their former level, some slowly and insensibly, others suddenly and by starts, a few feet or yards at a time ; whereas there are no reasons for believing that, during the last 3,000 years at least, any regions have been either upheaved or depressed, at a single stroke, to the amount of several hundred, much less several thousand feet. Faulting on a very grand scale accompanied mountain for- mation, and appears to have occurred as the result of the actfon of the tangential thrust, or lateral force, which curved and upheaved the mass. The most remarkable of the folds and faults seen in mountain chains are found affecting rock -masses that have suffered metamorphism, and will be discussed in the division of this work which deals with the metamorphic rocks. Conformable and unconformable stratification. When strata rest one upon the other horizontally or with the same Fig. 106. Unconformable junction of old red sandstone and Silurian schist at the Sicarr Point, near St. Abb's Head, Berwickshire. dip, they are conformable. But strata are said to be unconform- able when one series is so placed over another that the planes of the superior repose on the edges of the inferior (see fig. 106). In this case it is evident that a period had elapsed between the production of the two sets of strata, and that, during this in- terval, the older series had been tilted and disturbed. After- wards the upper series accumulated, in horizontal strata, upon it. If these superior beds, d d, fig. 106, are also inclined, it is plain that the lower strata, a a, have been twice displaced first, before the deposition of the newer beds, d d, and a second time when the same strata were upraised out of the sea, and thrown slightly out of the horizontal position. It often happens that in the interval between the deposition of two sets of unconformable strata, the inferior rock has not only been denuded, but drilled by perforating shells. Thus, for example, at Autreppe and Gusigny, near Mons, beds of an ancient (palaeozoic) limestone, highly inclined, and often bent, H '2 100 OVERLAP [CH. Vltt. are covered with horizontal strata of greenish and whitish marls of the Cretaceous formation (fig. 107). The lowest, and therefore the oldest, bed of the horizontal series is usually the sand and conglomerate, a, in which are rounded fragments of stone, from an inch to two feet in diameter. These fragments have often adhering shells attached to them, and have been Fig. 107. Junction of unconformable strata near Moiis, in Belgium. bored by perforating mollusca. The solid surface of the inferior limestone has also been bored, so as to exhibit cylindrical and pear-shaped cavities, as at c, the work of saxicavous mollusca ; and many rents, as at &, which descend several feet or yards into the limestone, have been filled with sand and shells, simi- lar to those in the stratum a. Overlap of strata. Strata are said to overlap when an upper bed extends beyond the limits of a lower one. Sediment spread over a region of subsidence has the area of deposit gra- dually increased, and the newest formed strata will overlap the next below them if these be inclined and their edges denuded. Thus, as shore lines have subsided, shallow-water marine deposits have crept over the land, and as subsidence has pro- gressed, deep-water deposits have been laid down upon these Fig. 108. Overlap of strata. ab cde, Jurassic rocks. 1. Wealden. 2. Lower greensand. 3. Gault. 4. Upper greensand. 5. Chalk. (From Jukes-Brown, Phys. Geol. p. 388.) last. Unconformable overlap (' overstep ' of some authors) re- sults when one set of strata rest upon others with a different angle of dip. When unconformable overlap is noticed, lapse of time and alterations in the physical geography of the area are inferred to have taken place between the deposition of the last stratum of the lower formation and the first of the upper formation ; and this is more obvious when erosion of a lower Vlll.l ESCARPMENTS * S s I*!? !fP I III! I fill 1 :--*- * stratum is seen to have taken place before -the dopceition ,of the upper. J '- J S*ti&~ '- - * * ' It is usually found that when two series of strata are unconformable or overlap, and thus exhibit a physi- cal break, their fossils differ con- siderably. This change in fossils is termed a palaeontological break, and it may be slight or very nearly absolute, as between the Chalk and the overlying Tertiaries. Dip-slopes and Escarpments. The action of denudation, or the wearing away of the surface of the land, will be fully discussed in the next chapter ; but it will be neces- sary to consider some of the effects of that action in this place in order to explain certain appearances pre- sented by the outcrop of strata. When one stratum is harder than those above and below it, < sloping surfaces determined by the ; dip of the strata are produced by \ denudation, and these are called dip-slopes. The steep slopes formed where such beds are worn away are called escarpments. In fig. 109 the upper Cretaceous strata are repre- sented overlying all the older beds down to the Palaeozoic; but the parts indicated by dotted lines have been removed by denudation. The beds of chalk, o, exhibit a good example of a dip-slope on the right and a steep escarpment to the I left. Outliers and inliers. The same diagram illustrates the forma- tion by denudation of those isolated patches of a stratum known to geologists as outliers. Although the greater part of the beds indi- cated by the dotted lines have been swept away, portions still remain \Uhsi .2 -2 102 DENUDATION [en. vm. lying beyovid the escarpment formed by the mass of the Chalk ' atid -Upper 'Grcer; sand: Thus there have been formed a number of outliers of the Upper Greensand, and on the left of the section is seen an outlier composed of both Upper Greensand and Chalk. On a map these outliers are seen as isolated patches of a stratum surrounded by older beds. Occasionally, when beds have been bent into folds, denudation causes the exposure of portions of an older stratum in the midst of a newer one. Such exposures were called by the older geologists ' outliers by protrusion,' but the officers of the English Geological Survey have introduced the use of the more convenient term inlier for masses of strata showing these relations. It is most important that the ' Physical Geology,' third edition, student should try to master the 1882, chapter xi., and of Professor problems of solid geometry involved James Geikie's 'Outlines of Geo- in the bending and fracture of logy,' second edition, 1888, chapter great stratified masses, and the xv. The names applied in different superficial appearances produced countries to various kinds of flex- by the subsequent planing away of ures and faults are explained in their surfaces by denudation. Great Heim and Margerie's ' Les Disloca- assistance will be obtained by the tions de 1'ecorce terrestre,' 1888. careful study of Professor Green's CHAPTEE IX DENUDATION AND ITS EFFECTS Denudation defined Its amount more than equal to the entire mass of stratified deposits in the earth's crust Subaerial denudation Action of the wind Action of running water Alluvium defined Different ages of alluvium Denuding power of rivers affected by rise or fall of land Littoral denudation Inland sea-cliffs Escarpments Sub- marine denudation Doggerbank Newfoundland bank- -Denuding power of the ocean during emergence of land. DENUDATION, which has been occasionally referred to in the preceding chapters, consists in the disintegration or breaking up of the earth's surface and the removal of the products by water in motion whether of rivers or of the waves and currents of the sea and by wind, and the consequent laying bare. of some inferior rock. This operation has exerted an influence on the structure of the earth's crust as universal and important as sedimentary deposition itself; for denudation is the necessary antecedent of the production of all new strata of mechanical origin. The forma- tion of every new deposit by the transport of sediment and pebbles necessarily implies that there has been, somewhere else, CH. ix.] AND ITS FACTOKS 103 a grinding down of rock into rounded fragments, sand, or mud, equal in quantity to the new stratum. All deposition, therefore except in the case of a shower of volcanic ashes, the outflow of lava, and the growth of certain organic formations is the sign of former superficial waste, or of that going on contem- poraneously, and to an equal amount, elsewhere. The gain at one point is no more than sufficient to balance the loss at some other. Disintegration and transport. From the preceding re- marks it will be apparent that denudation results from the joint operation of two distinct agencies, which we may speak of as disintegration and transport. By the action of rain and frost the hardest and most solid rocks are broken up, and their surfaces covered by debris or ' rubble.' The accumulation of these masses of rubble would, in time, check the work of dis- integration by protecting the surfaces of the solid rocks below them from the further action of rain or frost ; but now the other agencies of transport come into play, and by the action of streams and sea-waves the loose masses of disintegrated material are swept away, fresh surfaces of the rock being thus exposed to atmospheric waste. The materials produced by disintegration, and carried to new localities by the various agents of transport, accumulate to form new rocks ; this constitutes deposition. Denudation resulting from disintegration and transport, and deposition acting on materials supplied by denudation, are the two great processes constantly going on upon the earth as the result of the circulation of air and water over and through its solid crust. When we see a stone building, we know that somewhere, far or near, a quarry has been opened. The courses of stone in the building may be compared to successive strata, the quarry to a ravine or valley which has suffered denudation. As the strata, like the courses of hewn stone, have been laid one upon another gradually, so the excavation both of the valley and quarry has been gradual. But we occasionally find in a conglomerate large rounded pebbles of an older conglomerate, which had previously been derived from a variety of different rocks. In such instances we are reminded that strata have been formed by the deposition of denuded materials worn from older strata, and have been curved and elevated into hills and mountains. These in their turn have been worn down by the agents of denudation. In such cases it is evident that the same materials have been in very different conditions and positions over and over again during the mutations which have affected the surface of the globe. De- 104 DISINTEGRATING ACTION [CH. ix. nudation and re-deposition have persisted ever since the earth's crust has been covered by an atmosphere and has had its rivers and seas. Denudation may be classed as subaerial and marine, accord- ing as it takes place above or below the level of the sea ; and the agents which produce it are the sun's heat, frost, the atmosphere, rain, rivers, and the movements of the sea. Subaerial denudation. The sun acts on rocks by heating them, and when the component minerals expand and contract unequally, disintegration is the result. In tropical countries the hardest rocks, like granite, are broken up by this unequal expansion and contraction of the minerals which compose them. In the daytime the rock surfaces become intensely heated, in the night they cool rapidly by radiation ; and in consequence of the great strains set up in the mass, flakes of rock are violently torn off, the whole surface of the rock-mass seeming ta exfoliate. Similar action may be seen taking place on mountain peaks ex- posed to equally great vicissitudes of heat and cold. The sun also dries clay at the surface, producing cracks in it which enable other agents, like rain and frost, to act. Prolonged cold, especially of frost acting on the water present in rocks, is a great destroyer of the surface down to some depth, and the principal cause is the expansion of the water during the as- sumption of the crystalline state of ice. The atmosphere acts both chemically and mechanically, and is assisted by the moisture it contains. Weathering of rocks by the carbon dioxide of the air is assisted by the removal of the bicar- bonates by rain. The rapidity with which inscriptions on monuments in churchyards become effaced, when compared with similar records placed within the church, has often been pointed out as a striking illustration of the process of dis- integration. Professor Milne and other authors have shown how the sand-blast erodes the Arabian Wadys, scouring and polishing the rocks, and removing the particles ground away from their exposed surfaces ; and there are numerous examples of wind- borne and wind-polished rocks on many sea coasts. ' Weathering ' is often very conspicuous in crystalline rocks, such as granite and most volcanic rocks, which are composed of several mineral elements. Through the decomposition of the felspar and other minerals most liable to be chemically affected by air and rain, hard rocks like basalt sometimes crumble to pieces, and may be dug with a spade. Some of the most fertile districts in Italy and France owe their riches to the scoriae and lava that once issued in a molten condition from the craters CH. ix.] OF THE SUN AND WIND . 105 of volcanoes, destroying all the vegetation around, but which since then have cooled and crumbled into dust. In desert regions, where no rain falls, or where, as in partg of the Sahara, the soil is so salt as to be without any covering of vegetation, clouds of dust and sand attest the power of the wind to cause the shifting of the unconsolidated or disintegrated rock. In examining volcanic countries, one is much struck with the great superficial changes brought about by this power in the course of centuries. The higher peak of Madeira is about 6,050 feet above the sea, and consists of the skeleton of a volcanic cone now 250 feet high, the beds of which once dipped from a centre in all directions at an angle of more than 30. The summit is formed of a dike of basalt with much olivine, fifteen feet wide, apparently the remains of a column of lava which once rose to the crater. Nearly all the scoriae of the upper part of the cone have been swept away, those portions only remaining which were hardened by the contact or proximity of the dike. The wind is seen to be continually removing the dust and finer particles from this exposed mass of volcanic materials. On the highest platform of the Grand Canary, at an eleva- tion of 6,000 feet, there is a cylindrical column of hard lava, from which the softer matter has been carried away ; and other similar remnants of the dikes of cones of eruption attest the de- nuding power of the wind at points where running water could never ha>ve exerted any influence. The waste effected by .wind, aided by frost and snow, may not be trifling, even in a single winter, and, when multiplied by centuries, may become inde- finitely great. Action of running water. There are different classes of phenomena which attest in a most striking manner the vast spaces left vacant by the erosive power of water. I may allude, first, to those valleys on both sides of which the same strata are seen following each other in the same order, and having the same mineral composition and fossil contents. We may observe, for example, several formations, as Nos. 1, 2, 3, 4, in the accom- panying diagram (fig. 110) ; No. 1, conglomerate, No. 2, clay, No. 3, grit, and No. 4, limestone, each repeated in a series of hills separated by valleys varying in depth. When we examine the subordinate parts of these four formations, we find, in like manner, distinct beds in each, corresponding, on the opposite sides of the valley, both in composition and order of position. No one can doubt that the strata were originally continuous, and that some cause has swept away the portions which once 106 RAIN AND RIVERS [ca. ix. connected the whole series. A torrent on the side of a moun- tain produces similar interruptions ; and when we make artificial cuts in lowering roads, we expose, in like manner, corresponding beds on either side. But in nature, these appearances occur in mountains several thousand feet high, and separated by intervals of many miles or leagues in extent. In general, it is only when rivers are swollen by heavy rain that any considerable quantity of solid matter is removed by their waters. At these times they frequently undermine their banks and precipitate vast masses of earth into the stream; these are rapidly washed away, while in the bed of the river fine gravel and larger fragments of loose stone are swept along, as the transporting power of the current is intensified with each addition to its volume. But the erosive power of rivers would be comparatively in- significant if it were not aided by other causes, by means of which the hard and compact masses of rock, composing so great a part of the earth's crust, are reduced to fragments capable of Fig. no. a. Older alluvium, or drift. b. Modern alluvium. t>eing easily removed. All the subaerial agents of denudation tend to excavate the ordinary river valley, but canons, which are deep gorges and ravines, with perpendicular sides, have been excavated by the unassisted power of rivers. It must be remembered that rivers are mostly very old channels, and that in many instances they have lasted during the epoch of mountain formation which determined their existence, and ever since. Lowering of the surface, the formation of all the features of the hills, and the production of deep river gorges have progressed slowly and variably, but the main drainage has lasted on. In considering the erosive power of rivers, it must be remem- bered that the oscillation or meandering of streams from side to side in their flood plains has been and is an important factor in sweeping down gravels and muds towards the sea. Denuding powers of rivers affected by rise or fall of land. It has long been a matter of common observation that ynost rivers are now cutting their channels through alluvial de- CH. ix.] RAIN AND RIVERS 107 posits of greater depth and extent than could ever have been farmed by the present streams. From this fact it has been in- ferred that rivers in general have grown smaller, or become less liable to be flooded than formerly. It may be true that, in the history of almost every country, the rivers have been both larger and smaller than they are at the present moment. For the rain- fall in particular regions varies according to climate and physical geography, and is especially governed by the elevation of the land above the sea, or its distance from it, and other conditions equally fluctuating in the course of time. But the phenomenon alluded to may sometimes be accounted for by oscillations in the level of the land, experienced since the existing valleys origi- nated, even where no marked diminution in the quantity of rain and in the size of the rivers has occurred. Suppose, for example, part of a continent, comprising within it a large hydrographical basin like that of the Mississippi, to subside several inches or feet in a century. It will rarely happen that the rate of subsidence will be everywhere equal, and in many cases the amount of depression in the interior will regularly exceed that of the region nearer the sea. When- ever this happens, the fall of the waters flowing from the upland country will be diminished, and each tributary stream will have less power to carry its sand and sediment into the main - river, and the main river less power to convey its annual burden of transported matter to the sea. All the rivers, therefore, will proceed to fill up their ancient channels partially, and, during frequent inundations, will raise their alluvial plains by new deposits. If, then, the same area of land be again upheaved to its former height, the fall, and consequently the velocity, of every river will begin to augment. Each river then will be less given to overflow its alluvial plain ; and its power of carry- ing earthy matter seaward, and of scouring out and deepening its channel, will be sustained, until, after a lapse of years, a new channel or valley will be found to have been eroded through a fluviatile formation of comparatively modern date. The surface of what was once the river-plain at the period of greatest depression will then remain fringing the valley sides in the form of a terrace apparently flat, but in reality sloping down with the general inclination of the river. Every- where this terrace will present cliffs of gravel and sand, facing the river. That such a series of movements has actually taken place in the main valley of the Mississippi and in its tributary valleys during oscillations of level has been proved by geological investigations ; and the freshwater shells of existing species and bones of land quadrupeds, partly of extinct races, preserved in 108 FORMATION OF ESCARPMENTS |CH. ix. the terraces of fluviatile origin, attest the exclusion of the sea, during the whole process of filling up and partial re -excavation. Escarpments are the abrupt faces of rocks of various kinds which sometimes resemble sea-cliffs, but are often found far inland. They may extend for many miles and bound many valleys, and have more or less precipitous faces. They are due to subaerial denudation, and must be carefully distinguished from cliffs due to marine action. It was at one time supposed that the steep line of cliff-like slopes seen along the outcrop of the chalk, when we follow the edge of the North or South Downs, was due to marine action ; but Sir A. Ramsay and other authors have shown that the physical geography of the district points to the idea of the escarpments having been due to gradual waste since the rocks were exposed to the atmosphere, and to the action of rain and rivers. Mr. Whitaker has given a good summary of the grounds for ascribing these apparent sea-cliffs to waste in the open air. 1. There is an absence of all signs of ancient sea-beaches or littoral deposits at the base of the escarpment. 2. Great in- equality is observed in the level of the base line. 3. The escarp- ments do not intersect a series of distinct rocks like sea-cliffs, but are always confined to the boundary line of the same forma- tion. 4. There are sometimes different contiguous and parallel escarpments those, for example, of the greensand and chalk which are so near each other, and occasionally so similar in altitude, that we cannot imagine any existing archipelago, if converted into dry land, to present a similar outline. The above theory is by no means inconsistent with the opinion that the limits of the outcrop of the chalk and greensand, which the escarpments now follow, were originally determined by marine denudation. When the south-east of Eng q and last emerged from beneath the level of the sea, it was acted upon, no doubt, by the tide, waves, and currents, and the chalk would form, from the first, a mass projecting above the more destruc- tible clay called gault. Still the present escarpments so much resembling sea-cliffs have, no doubt, for reasons above stated, derived their most characteristic features, subsequently to emer- gence, from subaerial waste by rain and rivers. The vast results of denudation in past time are exhibited in a most impressive manner in those districts where we see some of the older strata of the earth appearing at the surface, as, for example, in the middle of an anticlinal curve (fig. 83, p. 82), on either side of which rest a long series of succeeding and con- formable strata. The newer beds must once have arched over en. ix.] DELTAS 109 the whole area, and have been stripped off, before the older strata could have been laid bare. In the ' Memoirs of the Geological Survey of Great Britain ' (vol. i.), Sir A. Bamsay has shown that the missing beds, removed from the summit of the Mendips, must have been nearly a mile in thickness ; and he has pointed out considerable areas in South Wales and some of the adjacent counties of England, where a series of very ancient or palaeozoic strata, not less than 11,000 feet in thickness, has been stripped off. All these materials have of course been transported to new regions, and have entered into the composition of more modern forma- tions. It is clear that such old rocks, mostly formed of mud and sand, and consolidated, were the monuments of denuding operations, which must have taken place at some of the remotest periods of the earth's history yet known to us. For whatever has been given to one area must always have been borrowed from another ; a truth which, obvious as it may seem when thus stated, must be repeatedly impressed on the student's mind, because in many doubtful geological speculations, it has been wrongly stated that the crust of the earth has been always growing thicker in consequence of the accumulation, period after period, of sedimentary matter, as if the new strata were not always produced at the expense of pre-existing rocks, stratified or unstratified. It is well known that deltas are forming at the mouths of some large rivers, and the land is encroaching upon the sea ; these deltas are monuments of recent denudation and depo- sition ; and it is obvious that if the mud, sand, and gravel were taken from them and restored to the continents, they would fill up a large part of the ravines and valleys which are due to the excavating and transporting power of torrents and rivers. By duly reflecting on the fact, that alL- deposits of mechanical origin imply the transportation from some other region, whether contiguous or remote, of an equal amount of solid matter, we perceive that the stony exterior of the planet must always have grown thinner in one place, whenever, by accessions of new strata, it was acquiring thickness in another. Alluvium. Between the superficial covering of vegetable mould and the subjacent rock there often intervenes, in many districts, a deposit of loose gravel, sand, and mud, to which, when it occurs in valleys, the name of alluvium has been popularly applied. The term is derived from alluvio, an inundation, or alluo, to wash, because the pebbles and sand commonly resemble those of a river's bed, or the mud and gravel washed over low lands by a flood. 110 FORMATION OF RIVER GRAVELS [CH. ix. In the course of those changes in physical geography which may take place during the gradual emergence of the bottom of the sea and its conversion into dry land, any spot may have been either a sunken reef, or a bay, or an estuary, or sea- shore, or the bed of a river. The drainage, moreover, may have been deranged again and again by earthquakes, during which temporary lakes may have been caused by landslips, and partial deluges occa- sioned by the bursting of the barriers of such lakes. For this reason it would be unreasonable to hope that we should ever be able to account for all the alluvial phenomena of each particular country, seeing that the causes of their origin are so various. And, further, the last operations of water have a tendency to disturb and confound together all pre-existing alluvia. Hence we are always in danger of regarding as the work of a single era, and the effect of one cause, what has in reality been the result of a variety of distinct agents during a long succession of geological epochs. Much useful instruction may therefore be gained from the exploration of a country like Auvergne, where Fig. 111. Lavas of Auvergne resting on alluvia of different ages. the superficial gravel of very different eras happens to have been preserved and kept separate by sheets of lava, which were poured out, one after the other, at periods when the denudation, and probably the upheaval, of rocks were in progress. That region had already acquired in some degree its present con- figuration before any volcanoes were in activity, and before any igneous matter was superimposed upon the granitic and fossili- ferous formations. The pebbles, therefore, in the older gravels are exclusively constituted of granitic and gneissic rocks ; and afterwards, when volcanic vents burst forth into eruption, those earlier alluvia were covered by streams of lava, which protected them from intermixture with gravel of subsequent date. In the course of ages, a new system of valleys was excavated, so that the rivers ran at lower levels than those at which the first alluvia and sheets of lava were formed. When, therefore, fresh eruptions gave rise to new lava, the melted matter was poured out over lower grounds ; and the gravel of these plains differed from the first or upland alluvium, by containing in it rounded CH. IX.] AT DIFFERENT HEIGHTS 111 fragments of various volcanic rocks, and often fossil bones belonging to species of land animals different from those which had previously nourished in the same country. The annexed drawing (fig. Ill) will explain the different heights at which beds of lava and gravel, each distinct from the other in composition and age, are observed, some on the flat tops of hills 700 or 800 feet high, others on the slope of the same hills, and the newest of all in the channel of the existing river, where there is usually gravel alone, although in some cases a narrow strip of solid lava shares the bottom of the valley with the river. The proportion of extinct species of quadrupeds is more numerous in the fossil remains of the highest gravel than in that lower down ; and in the bed of the river they agree with those of the existing fauna. The usual absence or rarity of Fig. 112. 1. Peat. 3'. Loam of same age. 2. Gravel of modern river. 4. Higher-level valley ground. 2'. Loam of brick earth (loess) of same 4'. Loam of same age. age as 2, formed by inuiidations of the river. 3. Lower-level valley gravel. 5. Upland gravel of various kinds and periods. 7, 8, 9. Older rocks. organic remains in beds of loose gravel and sand is owing partly to the friction which originally ground down the rocks into small fragments, and partly to the porous nature of alluvium, which allows the free percolation through it of rain-water, and pro- motes the decomposition and removal of fossil remains. But even in cases where the alluvia produced by successive stages of denudation are not sealed up, as in Auvergne, under beds of lava, we may frequently recognise the evidence of a sequence of deposits in a series of terraces on the sides of river valleys. As shown by Professor Prestwich, the upland or plateau gravels (see fig. 112) must have been spread out before the excavation of the valley, arid the higher level and lower level gravels must each have formed the bottom of the valley before it was excavated to its present depth. As in the case of Auvergne, this succession of events is confirmed by the study of the fossils found in these successive alluvia. 112 ACTION OF ICE [CH. ix. Under the name of diluvium or drift, the older geologists used to distinguish those masses of loose material which often attain great thicknesses, and are formed of materials that indicate more violent action than that which has accumulated the alluvia of our river valleys. Such deposits were at one time thought to have been produced by the action of violent floods sweeping over the land and carrying blocks of stone of great size and vast quantities of sand and mud from one region to another. The more careful study of these diluvial deposits or drifts has shown that they must have been accumulated by the action of ice either as glaciers, icebergs or shore-ice. Ice, as we shall see in a subsequent chapter, is a most important agent of disintegration and transport. Rocks have their surfaces scored, smoothed, and polished by rock-fragments frozen into the bottoms of glaciers, and these fragments are at the same time ground to the finest dust. Glaciers and icebergs transport blocks of the largest size, as well as sand and mud, to great dis- tances ; and, by the action of ice, vast masses of material are accumulated both on the land and under the sea, forming what are known as ' glacial ' deposits. Most of the deposits formerly classed as ' diluvium ' can now be shown to have resulted directly or indirectly from the action of glaciers, icebergs, and shore-ice. Marine denudation. The waves of the sea when driven by storms are continually wearing away the coastline, in some cases undermining the cliffs and hollowing out deep caverns. Cliffs are worn back leaving low foreshores, which are planed more or less level by the waves and tides. Part of the action of the waves between high- and low-water mark must be included in subaerial denudation, more especially as the undermining of cliffs by the waves is facilitated by land-springs, and these often lead to the sliding down of great masses of land into the sea. But the destruction wrought by these means would soon come to an end if the force of the waves and the tides did not break up whatever is brought within their reach, and, by sweeping the fragments to deep water, prepare the way for renewed gains upon the land. Though the denuding power of the waves is confined within the narrow limits between tide-marks, the phenomena of our raised beaches and submerged forests indicate oscillations of level, and as such movements are very gradual, they must have given repeated opportunities to the breakers to denude the land which was again and again exposed to their fury, although it is evident that the submergence was sometimes effected in such a CH. ix.] MAEINE DENUDATION 113 manner as to allow the trees which border the coast to be quietly covered up by sediment instead of being carried away. Ground- swell waves are important . agents of denudation when they come into shallow water. Scott Russell showed that a single roller of a ground-swell, 20 feet high, falls with the pressure of a ton on every square foot, and Stevenson stated that the force of the breakers of the Atlantic on the sea-coasts of Britain was 611 Ibs. per square foot in summer, and 2,086 Ibs. in winter. It is stated that ground swell will influence the bottom at 200 fathoms. But Delesse has proved that engineer- ing operations are scarcely disturbed at a greater depth than 16 - 4 feet in the Mediterranean Sea. and 26*24 feet in the Atlantic. All modern research tends to show that the greater part of the eroding action of the sea is restricted to within a few fathoms of the shore (Note F, p. 602). The sea removes the products of its own erosion, and most of the results of subaerial denudation. The mud of rivers sinks sooner or later when in contact with the sea, and clays readily sink in salt water ; but it appears that deep-sea deposits remote from land are singularly exempt from materials derived from the land. The vast volumes of soluble matters brought down by the rivers into the sea supply the material of the calcareous and siliceous skeletons of a host of marine organisms. The littoral deposits, as they are termed, are shingle beds and similar accumulations, and they are rarely stationary. Derived from the fall of cliffs, and worn by the rolling of water and by impact with other stones, the fragments become pebbles, while the sand, resulting from this wearing action, is carried off by tide and currents. Finally, the pebbles collect in masses, which resemble many geological formations, and were they cemented would be true conglomerates. The fine materials formed by the wearing down of the fragments and pebbles are spread out in layers, which resemble the sandstones of old with rain-prints and ripple -markings. Submarine denudation. When we attempt to estimate the amount of submarine denudation, we become sensible of the disadvantage under which we labour from our habitual inca- pacity of observing the action of marine currents on the bed of the sea. We know that the agitation of the waves, even during storms, diminishes at a rapid rate, so as to become very insig- nificant at the depth of a few fathoms ; but when large bodies of water are transferred by a current, from one part of the ocean to another, they are known to maintain at some depth such a velocity as must enable them to remove the finer, and sometimes even the coarser, materials of the rocks over which they flow. I 114 SUBMARINE DENUDATION [CH. ix. As the Mississippi when more than 150 feet deep can keep open its channel and even carry down gravel and sand to its delta, the surface velocity being not more than two or three miles an hour, -so a gigantic current like the Gulf Stream, equal in volume to many hundred Mississippis, and having in parts a surface velocity of more than three miles, may in moderately deep water act as a propelling and abrading power. But the efficacy of the sea as a denuding agent, geologically considered, is not dependent on the power of currents to preserve at considerable depths a velocity sufficient to remove sand and mud, because, even where the deposition or removal of sediment is cot in pro- gress, the depth of water does not remain constant throughout geological time. Every page of the geological record proves to us that the relative levels of land and sea, and the position of the ocean and of continents and islands, have been always varying, and we may feel sure that some portions of the sub- marine area are now rising and others sinking. The force of tidal and other currents and of the waves during storms was sufficient to prevent the emergence of many lands, even though they were undergoing continual upheaval. This must always have been the case when the reduction of level by the action of marine currents went on faster than its elevation by sub- terranean forces. It is not an uncommon error to imagine that the waste of sea-cliffs affords the measure of the amount of marine denudation, of which it probably constitutes an insig- nificant portion. Dogger-bank. That great shoal called the Dogger-bank, about sixty miles east of the coast of Northumberland, and occupying an area about as large as Wales, has nowhere a depth of more than ninety feet, and in its shallower parts is less than forty feet under water. It might contribute towards the safety of the navigation of our seas to form an artificial island, and to erect a lighthouse cm this bank ; but no engineer would be rash enough to attempt it, as he would feel sure that the ocean in the first heavy gale would sweep it away as readily as it does every temporary shoal that accumulates from time to time around a sunken vessel on the same bank. 1 No observed geographical changes in historical times entitle us to assume that where upheaval may be in progress it proceeds at a rapid rate. Three or four feet rather than as many yards in a century may probably be as much as we can reckon upon in our speculations ; and if such be the case, the continuance of the upward movement might easily be counteracted by the de- nuding force of such currents aided by such waves as during a 1 ' Principles,' 10th ed. vol. i. p. 569. CH. ix.] BANKS, 'NEEDLES,' AND CLIFFS 115 gale are known to prevail in the German Ocean. What parts of the bed of the ocean are stationary at present, and what areas may be rising or sinking, is a matter of which we are very igno- rant, as the taking of accurate soundings is but of recent date. Neivfoundland-bank. The great bank of Newfoundland may be compared in size to the whole of England. This part of the bottom of the Atlantic is surrounded on three sides by a rapidly deepening ocean, the bank itself being from twenty to fifty fathoms (or from 120 to 300 feet) under water. We are unable to determine by the comparison of different charts, made at distant periods, whether it is undergoing any change of level, but if it be gradually rising we cannot anticipate on that account that it will become land, because the breakers in an open sea would exercise a prodigious force even on solid rock brought up to within a few yards of the surface. We know, for example, that when a new volcanic island rose in the Mediterranean in 1831, the waves were capable in a few years of reducing it to a sunken bank. In the same way currents which flow over the Newfoundland- bank a great part of the year at the rate of two miles an hour, and are known to retain a considerable velocity to near the bottom, may carry away all loose sand and mud and make the emergence of the shoal impossible, in spite of the accessions of mud, sand, and boulders derived occasionally from melting ice- bergs which, coming from the northern glaciers, are frequently stranded on various parts of the bank. They must often leave at the bottom large erratic blocks which the marine currents ma} 7 be incapable of moving. * Needles ' and ' No Man's Lands ' are portions of cliffs left behind when surrounding parts have been worn down by the sea ; they indicate the former extension of the land up to and beyond them seawards. They are, as it were, measures of the strata which have been worn away, and which are recog- nised in the main cliffs of the land. Inland sea-cliffs. In countries where hard limestone rocks abound, inland cliffs have often retained the characters which they acquired when they constituted the boundary of land and sea. Thus, in the Morea, no less than three or even four ranges of cliffs are well preserved, rising one above the other at dif- ferent distances from the actual shore, the summit of the highest and oldest occasionally attaining 1,000 feet in elevation. A consolidated beach with marine shells is usually found at the base of each cliff, and a line of old shore caverns. But the beginner should be warned not to expect to find evidence of the former sojourn of the sea on all those lands i 2 116 TRACES OF MARINE ACTION [OH. ix. which we are nevertheless sure have been submerged at periods comparatively modern; for notwithstanding the enduring nature of the marks left by littoral action on some rocks, especially limestones, we can by no means detect sea-beaches and inland cliffs everywhere. On the contrary, they are, upon the whole, extremely partial, and are often entirely wanting in districts composed of argillaceous and sandy formations, which must, nevertheless, have been upheaved at the same time, and by the same intermittent movements, as the adjoining harder rocks. Equally necessary is it for the student to avoid confounding ordinary escarpments, formed by subaerial denudation, with true sea cliffs, to which they sometimes exhibit a superficial resemblance (Note G, p. 602). The importance of subaerial de- ' Scenery of Scotland, viewed in nudation in sculpturing the earth's connection with its Physical Geo- surface was first shown in Mr. logy,' may all be studied with ad- Bcrope's classical work on the vantage as supplying valuable illus- 1 Volcanoes of Central France.' trations of the principles laid down Sir Andrew Eamsay's ' Physical in this chapter. For an admirable Geology and Geography of Great summary of the question see Pro- Britain,' Col. Greenwood's ' Rain fessor Green's ' Physical Geology,' and Rivers,' and Sir A. Geikie's third edition, ch. xiii. CHAPTER X JOINT ACTION OF DENUDATION, UPHEAVAL, AND SUBSIDENCE IN REMODELLING THE EARTH'S CRUST How we obtain an insight, at the surface, of the arrangement of rocks at great depths Why the height of the successive strata in a given region is so disproportionate to their thickness Computation of the average annual amount of subaerial denudation Antagonism of sub- terranean forces to the levelling power of running water How far the transfer of sediment from the land to a neighbouring sea-bottom may affect subterranean movements Supposed permanence of continental and oceanic areas. How we obtain an insight, at the surface, of the arrange- ment of rocks at great depths. The reader has been already informed that in the structure of the earth's crust we often find proofs of the direct superposition of marine to fresh- water strata, and also evidence of the alternation of deep-sea and shallow-water formations. Sedimentary deposits cannot become thick if exposed to concurrent denudation. Darwin has suggested that all deep sediments must have accumulated during subsidence of the area in which they were formed. In order to explain how such a series of rocks could be made to give rise to our present continents and islands, we have to as- CH. x.] LATERAL COMPEESSION OF STRATA 117 sume not only that there have been alternate upward and down- ward movements of great vertical extent, but that the upheaval in the areas which we at present inhabit has, in later geo- logical times, sufficiently predominated over subsidence to cause these portions of the earth's crust to be land instead of sea, The sinking down of a delta beneath the sea-level may cause strata of fluviatile or even terrestrial origin, such as peat, to be covered by deposits of deep-sea origin. There is also no limit to the thickness of mud and sand which may accumulate in shallow water, provided that fresh sediment is brought down from the wasting land at a rate corresponding to that of the sinking of the bed of the sea. The succession of strata here alluded to would be consistent with the occurrence of gradual downward and upward move- ments of the land and bed of the sea without any disturbance of the horizontality of the several formations. But the arrange- ment of rocks composing the earth's crust differs materially from that which would result from a mere series of radial vertical movements. Had the internal energies of the globe only pro- duced such movements, and had the stratified rocks been first formed beneath the sea and then raised above it, without any lateral compression, the geologist would never have obtained an insight into the monuments of various ages, some of extremely remote antiquity. What we have said in Chapter VIII. of dip and strike, of the folding and inversion of strata, of anticlinal and synclinal flexures, and in Chapter IX. of denudation at different periods, whether subaerial or submarine, must be understood before the student can comprehend what may at first seem to him an anomaly, but which it is his business particularly to understand. We allude to the small height above the level of the sea attained by strata, often many miles in thickness, and about the chrono- logical succession of which, in one and the same region, there is no doubt whatever. Had stratified rocks in general remained horizontal, the waves of the sea would have been enabled during oscillations of level to plane off entirely the uppermost beds as they rose or sank during the emergence or submergence of the land. But the occurrence of a series of formations of widely different ages, all remaining horizontal and in conform- able stratification, is exceptional, and for this reason the total annihilation of the uppermost strata has rarely taken place. We owe, indeed, to the lateral movements produced by tan- gential thrust those anticlinal and synclinal curves of the beds already described (fig. 81, p. 80), which, together with denu- dation, subaerial and submarine, enable us t-P investigate the 118 VAST THICKNESS OF STEATA [CH. x. structure of the earth's crust many miles below those points which the miner can reach under other circumstances. It has already been shown in fig. 83, p. 82, how, at St. Abb's Head, 'a series of strata of indefinite thickness may become vertical, and then denuded, so that the edges of the beds alone shall be exposed to view, the altitude of the upheaved ridges being reduced to a moderate height above the sea-level. The breadth of an exposed edge of a stratum is equivalent to its thickness when ine vertical position is assumed. It may be observed that, although the incumbent strata of Old Ked Sandstone are nearly horizontal, yet they will in other places be found so folded as to present vertical strata, the edges of which are abruptly cut off, as in 2, 3,4 on the right-hand side of the diagram, fig. 81, p. 80. Why the height above sea-level of the successive strata in a given region is so disproportionate to their thickness. We cannot too distinctly bear in mind how dependent we are, for our power of consulting the different pages of those stony records of which the crust of the globe is composed, on the joint action of the internal energies and agents of denudation, the one in disturbing the original position of rocks, and the other in destroying large portions of them. Why, it may be asked, if the ancient bed of the sea has been in many regions uplifted to the height of two or three miles, and sometimes twice that altitude, and if it can be proved that some single formations are of themselves two or three miles thick, do we so often find several important groups resting one upon the other yet attaining only the height of a few hundred feet above the level of the sea ? The American geologists, after carefully studying the Appa- lachian mountains, have ascertained that the older fossilife- rous rocks of that chain (from the Silurian to the Carboniferous inclusive) are not less than 42,000 feet thick, and if ihey were now superimposed on each other in the order in which they were deposited, they ought to equal in height the Himalayas with the Alps piled upon them. Yet they rarely reach an altitude of 5,000 feet, and their loftiest peaks are no more than 7,000 feet high. The Carboniferous strata forming the highest member of the series, and containing beds of coal, can be shpwn to be of shallow-water origin, or even sometimes to have originated in swamps in the open air. But what is more surprising, the lowest part of this great Palaeozoic series, in- stead of having been deposited at the bottom of an abyss more than 40,000 feet deep, consists of sediment (the Potsdam sand- stone), evidently spread out on the bottom o fa shallow sea on which ripple-marked sands were occasionally formed, This vast CH. x.] NOT FORMING HIGH MOUNTAINS 119 thickness of 42,000 feet is estimated by measuring the denuded edges of the vertical strata forming the parallel folds into which the originally horizontal Silurian and Carboniferous rocks had been forced, and which ' crop out ' at the surface. A like phenomenon is exhibited in every mountainous coun- try, as, for example, in the European Alps ; but we need not go farther than the north of England for its illustration. Thus in Lancashire and central England the thickness of the Carboni- ferous formation, including the Millstone Grit and Yoredale beds, is computed to be more than 18,000 feet ; to this we may add the Mountain Limestone, at least 2,000 feet in thickness, and the overlying Permian and Triassic formations, 3,000 or 4,000 feet thick. How then does it happen that the loftiest hills of Yorkshire and Lancashire, instead of being 24,000 feet high, never rise to 3,000 feet ? The denuded edges of the strata, which are in great curves, are measurable, but the bulk of the thickness is below sea-level. A study of figs. 97 and 98, p. 92, will explain the relation of the thickness of strata to their height above sea-level. It is evident that the denuded edges of very thick masses of strata, which are in great curves, can be measured, although the bulk of the deposit is hidden. Hence masses of stratified rocks may be several miles in thickness, although the elevation attained by them may not be more than a mile above sea-level. Computation of tbe average annual amount of subaerial denudation. Attempts were made by Manfredi in 1736, and afterwards by Playfair in 1802, to calculate the time which it would require to enable the rivers to deliver over the whole of the land into the basin of the ocean. The data were at first too imperfect and vague to allow them even to approximate to safe conclusions. But in our own time similar investigations have been renewed with more prospect of success, the amount brought down by many large rivers to the sea having been more accu- rately ascertained. Mr. Alfred Tylor, in 1850, inferred that the quantity of detritus now being distributed over the sea-bottom would, at the end of 10,000 years, cause an elevation of the sea- level to the extent of at least three inches. Subsequently Mr. Croll in 1867, and again, with more exactness, in 1868, deduced from the latest measurement of the sediment transported by European and American rivers, the rate of subaerial denudation to which the surface of large continents is exposed, taking espe- cially the hydrographical basin of the Mississippi as affording the best available measure of the average waste of the land. The conclusion arrived at in his able memoir was that the whole terrestrial surface is denuded at the rate of one foot in 6,000 120 KATE OF DENUDATION BY MISSISSIPPI [CH. x. years, and this opinion was enforced by Sir A. Geikie, who pub- lished a valuable essay on the subject in 1868. The student, by referring to the ' Principles of Geology,' may see that Messrs. Humphreys and Abbot, during their survey of the Mississippi, attempted to make accurate measurements of the proportion of sediment carried down annually to the sea by that river, including not only the mud held in suspension, but also the sand and gravel forced along the bottom. It is evident that when we know the dimensions of the area which is drained, and the annual quantity of earthy matter taken from it and borne into the sea, we can affirm how much on an average has been removed from the general surface in one year ; and there seems no danger of our overrating the mean rate of waste by selecting the Mississippi as our example, for that river drains a country equal to more than half the continent of Europe, extends through twenty degrees of latitude, and therefore through regions enjoying a great variety of climate, and some of its tributaries descend from mountains of great height. The Mississippi is also more likely to afford us a fair test of ordinary denudation, because, unlike the St. Lawrence and its tributaries, there are no great lakes in which the fluvia- tile sediment is thrown down and arrested on its way to the sea In striking a general average we have to remember that there are large deserts in which there is scarcely any rainfall, and tracts which are as rainless as parts of Peru, and these must not be neglected as counterbalancing others, in the tropics, where the quantity of rain is in excess. From the careful observations of Messrs. Humphreys and Abbot it is found that the quantity of materials carried down to the sea every year, in suspension and in solution, would, if spread out over the vast area drained by the Mississippi and consolidated into rock, raise that basin by g^tf part of a foot. In other words, the whole Mississippi basin is being lowered by the action of denudation at the rate of one foot in 6,000 years. Small as this rate may seem to be, a little consideration will show what stupendous effects may be produced in long periods of time. The average height of the North American continent is (according to the most recent researches) 2,030 feet. It follows then that if the other rivers of North America are carrying on the work of denudation at the same rate as the Mississippi, the whole North American continent would be swept away and its materials deposited in the ocean in a period of 12,000,000 years. The results of these calculations are only trustworthy if it is true that the rainfall has not greatly increased or diminished, fm^ that the climate Jia remained approximately the same, CH. x.] AND OTHEE EIVERS 121 There can be little doubt that many rivers perform the work of denudation at a much quicker rate than the Mississippi. It has been estimated, in the case of the Ganges, that the quantity of mud carried down to the Bay of Bengal, during four months of wet season of each year, is so great that it would require a fleet of eighty ' Indiamen,' each of 1,400 tons, to set sail every hour of every day during the whole of those four months in order to carry the same amount of material as is done by this river. The estimates made in the case of some other rivers are as follows : To reduce the height of the river-basin by one foot would require the following periods in the case of the several rivers : Danube, 6,846 years ; Nith, 4,723 years ; Yang-tse-kiang, 2,700 years; Ganges, 2,358 years; Elbe, 1,600 years; Khone, 1,528 years ; Hoang Ho, 1,464 years ; Po, 729 years. A rate of 3,000 years for the removal of one foot thickness from the surface in the case of the whole of the rivers of the globe would probably be a very fair average ; and, as the mean height of all the land-masses of the globe is about 2,300 feet, it would require about 7,000,000 years, at the present rate of subaerial denudation, to carry away their materials and deposit them beneath the ocean. Action of hypog-ene forces in compensating- tbose of subaerial denudation. In all these estimates it is assumed that the entire quantity of land above the sea-level remains on an average undiminished in spite of annual waste. Were it other- wise, the subaerial denudation would be continually lessened by the diminution of the height and dimensions of the land exposed to waste. It was stated in 1830, in the ' Principles of Geology,' that running water and volcanic action are two antagonistic forces ; the one labouring continually to reduce the whole of the land to the level of the sea, the other to restore and maintain the inequalities of the crust on which the very existence of islands and continents depends. We must always bear in mind that it is not simply by upheaval that subterranean movements can counteract the levelling force of running water. For, whereas the transportation of sediment from the land to the ocean or the upheaval of its bed would raise the general sea-level, the subsidence of the sea-bottom by increasing its capacity would check this rise and prevent the submergence of the land. The average height and area of the land-masses can only be preserved if the increase occasioned by elevation in one part ex- ceeds the loss by subsidence elsewhere ; the amount removed by denudation from the whole surface of the land is the measure of this excess of elevation over subsidence. It is only by con- sidering the joint action of all the causes that determine the level 122 COMPENSATION BY SUBTERRANEAN FORCES [OH. x. of the sea and the height of the land that we can form some idea of the relation of these destroying and renovating energies. Unless we assume that there is, in volcanic districts, more sub- sidence than upheaval, we must suppose the volume of the land- masses to be always increasing, by that quantity of volcanic matter which is annually poured out in the shape of lava or ashes, and accumulated on the land, and which is derived from the interior of the earth. The abstraction of this matter causes, no doubt, in some instances, subsidence. Moreover it is possible that the globe has become smaller from contrac- tion during secular cooling. Hypog-ene action. The action of energies within the earth in counterbalancing denudation by producing great curvings of the crust in past times is not a mere matter of conjecture. The student will see in a future chapter that we have proofs of Carboniferous forests hundreds of miles in extent which grew on the lowlands or deltas near the sea, and which subsided and gave place to other forests, until in some regions fluviatile and shallow- water strata with occasional seams of coal were piled one over the other, till they attained a thickness of many thousand feet. These have often been preserved owing to their being forced into synclinal curves and removed out of the range of denudation. It will be also seen in another chapter that we have evidence of a rich terrestrial flora, the Devonian, even more ancient than the Carboniferous ; while, on the other hand, the later Triassic, Oolitic, Cretaceous, and successive Tertiary periods have all supplied us with fossil plants, insects, or terrestrial mammalia ; showing that, in spite of great oscillations of level and continued changes in the position of land and sea, the internal energies have maintained a due proportion of dry land. We may appeal also to freshwater formations, such as the Purbeck and Wealden, to prove that in the Oolitic and Neocomian eras there were rivers draining ancient lands in Europe in times when we know that other spaces, now above water, were submerged. How far tbe transfer of sediment from the land to a neighbouring sea-bottom may affect subterranean move- ments. It has been suggested that the stripping off by denu- dation of dense masses from one part of a continent and the delivery of the same into the bed of the ocean must have a decided effect in causing changes of temperature in the earth's crust below, or, in other words, in causing the subterranean isothermals to shift their position. If this be so, one part of the crust may be made to rise, and another to sink, by the expan- sion and contraction of the rocks, of which the temperature is altered. CH. X .] CONTINENTAL MASSES, OCEANIC DEPKE8SIONS 123 Persistence and mutability of continental and oceanic areas. If the thickness of more than 40,000 feet of sedimen- tary strata, before alluded to, in the Appalachians, proves a pre- ponderance of downward movements of the sea-floor in Palaeozoic times in a district now forming the eastern border of North America, it also proves, as before hinted, the continued existence and waste of some neighbouring continent, probably formed of Laurentian rocks, and situated where the Atlantic now prevails. Such an hypothesis would be in perfect harmony with the con- clusions forced upon us by the study of the present configuration of our continents, the relation of their height to the depth of the oceanic basins, also to the considerable elevation and extent sometimes reached by drift containing shells of recent species ; and still more by the fact of sedimentary strata, several thousand feet thick, as those of central Sicily, or such as flank the Alps and Apennines, containing fossil mollusca sometimes almost wholly identical with species still living. Movements of 1,000 feet or more would turn much land into sea, and sea into land, in the continental areas and their borders ; whereas oscillations of equal magnitude would have no corre- sponding effect in the bed of the ocean generally, believed as it is to have a mean depth of nearly 13,000 feet. The greatest depths of the sea do not exceed the greatest heights of the land ; it may, therefore, seem strange that the mean depth of the sea should exceed the mean height of the land six times, even taking the lowest estimate of the ocean depths as given by the late deep-sea soundings. This apparent anomaly arises from the fact that the extreme heights of the land are exceptional and confined to a small part of its surface ; while the ocean maintains its great depth over enormous areas. It is evident that, during the recent periods of the earth's history, there have been great subsidences and elevations of the land ; many raised beaches are 1,000 to 1,200 feet above sea- level. Dana, following Darwin's theory of Atoll formation, terms the Atoll a memorial of a departed land, and considers that the great Pacific subsidence was contemporaneous with the post-glacial upheaval in the north. From all that we know of the extreme slowness of the up- ward and downward movements which bring about even slight geographical changes, we may infer that it would require a great lapse of time to cause the submarine and supramarine areas to change places, even if the ascending movements in the one region and the descending in the other were continuously in one direction. But we have only to appeal to the structure of the Alps, where there are so many shallow and deep-water 124 PERMANENCE OR MUTABILITY [CH. x. formations of various ages crowded into a limited area, to con- vince ourselves that mountain chains are the result of great oscillations of level. High land is not produced simply by uniform upheaval, but by a predominance of elevatory over subsiding movements. Where the ocean is extremely deep it is because the sinking of the bottom has been in excess, in spite of interruptions by upheaval. Yet, persistent as may be the leading features of land and sea on the globe, they are not immutable. Some of the finest mud is doubtless carried to indefinite distances from the coast by marine currents, and we are taught by deep-sea dredgings that in clear water, at depths equalling the height of the Alps, organic beings may flourish, and their spoils slowly accumulate on the bottom. We also occasionally obtain evidence that submarine volcanoes are pouring out ashes and streams of lava in mid-ocean as well as on land, and that wherever mountains like Etna, Vesuvius, and the Canary Islands are now the site of eruptions, there are signs of accompanying upheaval, by which beds of ashes full of recent marine shells have been uplifted many hun- dred feet. We need not be surprised, therefore, if we learn from geology that the continents and oceans were not always placed where they now are, although the imagination may well be overpowered when it endeavours to contemplate the amount of time required for such revolutions. The chalk formation consists of masses of foraminiferal ooze, one to two thousand feet in thickness, and was certainly formed in an ocean of considerable depth ; but it now constitutes the surface over many thousands of square miles in the whole district of central Europe from Ireland to Eussia and thence into Asia. In the same way masses of Globigerina- and Eadio- larian-ooze accumulated in a deep ocean are now found at the height of several thousand feet above the sea-level in th j islands of the West Indies. It was at one time supposed that among the great masses of stratified materials forming the earth's crust there were no rocks comparable to the deposits which are now accumulating upon the floors of the great oceans. But the discoveries of the last few years have proved that such is not the case. Among the formations of the older, as well as among those of the newer periods of the earth's history, we find great masses of calcareous and siliceous rocks sometimes thousands of feet in thickness entirely made up, as shown by the microscope, of the minute forms of life that cover the existing deep-ocean floors. The comparatively modern chalk has its counterpart in a number of plder calcareous rocks, almost wholly built up of the shells of CH. x.] OF OCEANS AND CONTINENTS 125 foraminifera with coccoliths and similar minute organisms. Siliceous rocks crowded with the remains of radiolarians have been found of great thickness and at a number of different horizons among the older as well as among the younger stratified deposits of the earth's crust ; and these vast masses of calcareous and siliceous rocks are now found elevated to form portions not only of the dry land, but of great mountain-chains, and may be seen exposed to our study at the height of several miles above the sea-level. In the face of these facts, it seems impossible to doubt that great interchanges have taken place between oceanic and continental areas of the globe ; and this conclusion is placed beyond doubt when we come to study the distribution of the forms of terrestrial and marine life. We have gained a great step in obtaining an approximate estimate of the number of millions of years in which the average aqueous denudation going on upon the land would convey seaward a quantity of matter equal to the volume of our continents ; and this may afford us a gauge to the minimum of subterranean force necessary to counteract such levelling power of running water ; but to discover a relation between the periods required for the operation of these great physical agencies and the rate at which species of organic beings vary, is at present wholly beyond the reach of our computation though perhaps it may not prove eventually to transcend the powers of Man. The rate of denudation in the concerning the process of earth Thames Valley, so far as the re- sculpture in the North American moval of matter in solution is con- continent, and especially of afford- cerned, has been calculated, on what ing illustrations of the joint action appear to be very trustworthy data, of the internal and external forces by Prof. Prestwich (Anniversary of the globe, in giving rise to the Address to Geological Society, 1872), existing forms of its surface. In and by Mr. T. Mellard Reade for connection with this subject, the the whole of England (' Soluble De- essays of Prof. W. H. Davis on the nudation,' Address Geological So- structure of Pennsylvania and New ciety of Liverpool, 1877). For Jersey may be studied with advan- materials carried in suspension the tage, and also the writings of admirable memoir on the Mississippi Dutton, Gilbert, Spencer, and other by Messrs. Humphreys and Abbot American geologists on the warping supplied the first data that could of the earth's crust, and on the in- be relied upon by geologists. The fluence of this action in the forma- subject has been discussed, in re- tion of canons, lakes, and other spect to other river basins, in the surface features. The zoological essays of J. Croll and Sir A. Geikie, evidence upon the question of the and the average rate of subaerial permanence or mutability of oceanic denudation may be regarded as and continental areas has been dis- now fairly well ascertained. The cussed by Mr. Blanford. (Anni- various publications of the United versary Address to Geological States Geological Survey should be Society, ' Quart. Journ. Geol. Soc.' consulted by the student as sup- vol xlvi., 1890) (Note H, p. 602). plying the most valuable details 126 [CH. XI. SECTION II. CHRONOLOGICAL CLASSIFICATION OF AQUEOUS EOCKS CHAPTER XI PRINCIPLES ON WHICH THE CLASSIFICATION OF SEDIMENTARY ROCKS IS BASED Aqueous, Volcanic, Plutonic, and Metamorphic rocks considered chrono- logically Terms Primary, Secondary, and Tertiary ; Palaeozoic, Meso- zoic, and Cainozoic explained On the different ages of aqueous rocks - Principal tests of relative age : superposition, mineral characters, fossils and included fragments Faunas and floras determined by conditions, geographical position, and geological age William Smith's classification of British deposits by their organic remains Danger of extending the palosontological method over wide areas Homotaxy Combination of physical and palaeontological methods Classification of Tertiary strata Tabular view of fossiliferous strata. Chronology of rocks. In the first chapter it was stated that the four great classes of rocks the aqueous, the volcanic, the plutonic, and the metamorphic would each be considered, not only in reference to their mineral characters and mode of origin, but also to their relative age. In regard to the aqueous rocks, we have already seen that they are stratified, that some are calcareous, others argillaceous or siliceous, some made up of sand, others of pebbles ; that some contain freshwater, others marine fossils, and so forth ; but the student has still to learn which rocks, exhibiting some or all of these characters, have originated at one period of the earth's history, and which at another. To determine this- point in reference to the sedimentary and fossiliferous formations is more easy than in any other class ; and it is therefore the most convenient and natural method to begin by establishing a chronology for these strata, and then to refer, as far as possible, to the same divisions the several groups of volcanic, plutonic, and metamorphic rocks. Such a system of classification is not only recommended by its greater clear- ness and facility of application, but is also best fitted to strike the imagination by bringing into one view the contemporaneous CH. xi.] CHRONOLOGY OF ROCKS 127 evolution of the inorganic and organic creations of former times. For the sedimentary formations are most readily dis- tinguished by the remains of different species of animals and plants which they enclose ; and of these animals and plants one set after another has flourished and then disappeared from the earth, each set leaving its relics behind as * fossils,' or, as they have been termed not inaptly ' medals of creation.' In the present work, therefore, the four great classes of rocks will form four parallel, or nearly parallel, columns in one chrono logical table. They will be considered as sets of monuments relating to contemporaneous, or nearly contemporaneous, series of events. Just as aqueous and fossiliferous strata are now formed in certain seas or lakes, while in other places volcanic rocks break out at the surface, so, at every era of the past, fossiliferous deposits and superficial igneous rocks were in pro- cess of formation contemporaneously ; and at the same time deep-seated chemical and mechanical actions led to the com- plete crystallisation and recrystallisation of materials both of igneous and aqueous origin, thus giving rise to the rocks which we call plutom'c and metamorphic. The early geologists gave to all the crystalline and non-fossili- ferous rocks the name of Primitive or Primary, under the idea that their formation was anterior to the appearance of life upon the earth ; while the aqueous or fossiliferons strata were termed Secondary ; and alluvia or other superficial deposits, Tertiary. 1 The meaning of these terms has, however, been gradually modi- fied with advancing knowledge, and they are now used to designate great chronological divisions under which all geological formations can be classed, each of them being characterised by the presence of distinctive groups of organic remains rather than by any physical peculiarities of the strata themselves. The use of the term ' Primary ' is now almost entirely abandoned, but the terms ' Secondary ' and ' Tertiary ' are still used, though with very different significations attached to them. To avoid the risk of misapprehension, geologists have introduced the term ' Palaeozoic ' for the rocks containing the oldest known forms of life, from r?a\aidv, 'ancient,' and 5oi/, 'an organic being,' still retaining the terms ' secondary ' and ' tertiary ; ' Professor Phillips, however, for the sake of uniformity, proposed ' Mesozoic ' for secondary, from //eVoy, 'middle,' &c. ; and ' Cainozoic ' for tertiary, from KCIIVOS, ' recent,' &c. ; the terms ' mesozoic ' or 1 At a very early date it was appeared to form a link between noticed that certain hard rocks the ' Primary ' and ' Secondary ' like slates, flagstones, and gray- rocks. These intermediate rocks wackes, while containing fossils, were called by Werner and the older were partially crystallised, and geologists ' Transition rocks.' 128 AGE OF STKATA. SUPERPOSITION [CHAP. xi. secondary and ' cainozoic ' or tertiary may be employed as useful synonyms. The periods of time covered by the Palaeozoic were so great, as shown by the enormous thickness of the strata, that it is con- venient to group the Palaeozoic rocks in two great divisions. Some authors propose to call these divisions Proterozoic and Deuterozoic; but as these names have not come into general use, it will be convenient to speak of them as Older Palaeozoic and Newer Palaeozoic respectively. We thus find that the series of fossiliferous rocks fall naturally into the following four grand divisions or classes : Cainozoic, or Tertiary. Mesozoic, or Secondary. Hewer Palaeozoic (Deuterozoic). Older Palaeozoic (Proterozoic). We shall see in the sequel that each of these great classes of strata is divided into three systems. It will also be shown that great masses of sedimentary rocks, some of them greatly metamorphosed, and associated with volcanic and plutonic rocks, are found underlying the Palaeozoic or the strata containing the oldest known fossils. Age of strata. For reasons already stated, we proceed first to treat of the aqueous or fossiliferous formations, considered in chronological order or in relation to the different periods at which they have been deposited. There are three principal tests by which we determine the age of a given set of strata: first, superposition; secondly, mineral character ; and, thirdly, organic remains. Some aid can occasionally be derived from a fourth kind of proof, namely, the fact of one deposit including in it fragments of a pre-existing rock, by which the relative ages of the two may, even in the absence of all other evidence, be determined. Superposition. The first and principal test of the age of one aqueous deposit, as compared with another, is relative posi- tion. It has been already stated that, where strata are horizon- tal, the bed which lies uppermost is the newest of the whole, and that which lies at the bottom the most ancient. Thus a series of sedimentary formations are like volumes of history, in which each writer has recorded the annals of his own times, and then laid down the book, with the last written page uppermost, upon the volume in which the events of the era immediately preceding were commemorated. In this manner a lofty pile of chronicles is at length accumulated ; and they are so arranged as to indicate, by their position alone, the order in which the events recorded in them have occurred. CH. xr.] TESTS OF THE AGES OP STRATA 129 In regard to the crust of the earth, however, there are some regions where, as the student has already been informed, the beds have been disturbed, and sometimes extensively thrown over and turned upside down. But an experienced geologist can rarely be deceived by these exceptional cases. When he finds that the strata are fractured, curved, inclined, or vertical, he knows that the original order of superposition may be doubtful, and he then endeavours to find sections in some neighbouring district where the strata are horizontal, or only slightly inclined. Here, the true order of sequence of the entire series of deposits being ascertained, a key is furnished for settling the chronology of those strata where the displacement is extreme. It should be remembered, however, that while this orier of sequence is invariable, all the members of the series may not everywhere be present. Certain formations may never have been deposited in a particular area, or, if deposited, they may have been removed by denudation before later ones were thrown down. Thus one of the youngest members of the series may be found resting directly on one of the oldest. Mineral character. The same rocks may often be observed to retain for miles, or even hundreds of miles, the same mineral peculiarities, if we follow the planes of stratification, or trace the beds, if they be undisturbed, in a horizontal direction. But if we pursue them vertically, or in any direction transverse to the planes of stratification, this uniformity ceases almost immediately. In that case we can scarcely ever penetrate a stratified mass for a few hundred yards without beholding a succession of extremely dissimilar rocks, some of fine, others of coarse, grain, some of mechanical, others of chemical, origin ; some calcareous, others argillaceous, and others siliceous. These phenomena lead to the conclusion that rivers, wind, and marine currents have dispersed the same sediment over wide areas at one period, but at successive periods have caused the accumulation, in the same region, of very different kinds of materials. The first observers were so astonished at the vast spaces over which they were able to follow the same homogeneous rocks in a horizontal direction, that they came hastily to the opinion that the whole globe had been environed by a succession of distinct aqueous formations, disposed round the nucleus of the planet, like the concentric coats of an onion. But although, in fact, some formations, like the chalk, may be continuous over districts as large as the half of Europe, or even more, yet most of them either terminate within narrower limits, or soon change their lithological charac- ter. Sometimes they thin out gradually, as if the supply of K 130 STRATA IDENTIFIED BY FOSSILS [CH. XT. sediment had failed in that direction, or they come abruptly to an end, as if we had arrived at the borders of the ancient sea or lake which served as their receptacle. It no less frequently happens that they vary in mineral aspect and composition, as we pursue them horizontally. For example, we trace a lime- stone for a hundred miles, until it becomes more arenaceous, and finally passes into sand, or sandstone. We may then follow this sandstone, already proved by its continuity to be of the same age, throughout another district a hundred miles or more in length. Organic remains. This character must be used as a test of the age of a formation or of the contemporaneous origin of two deposits in distant places, under very much the same restrictions as the test of mineral composition. First, the same fossils may be traced over wide regions if we examine strata in the direction of their planes, although by no means for indefinite distances. Secondly, while the same fossils prevail in a particular set of strata for hundreds of miles in a horizontal direction, we seldom meet with the same remains for many fathoms, and very rarely for several hundred yards, in a vertical direction, or a direction transverse to the strata. This fact has now been verified in almost all parts of the globe, and has led to the conviction that, at successive periods of the past, the same area of land and water has been inhabited by distinct assemblages of species of animals and plants. It appears that from the remotest periods there has been ever a coming in of new organic forms, and a dying out or extinction of those which pre-existed on the earth : some species have endured for a longer, others for a shorter, time; while none have ever re- appeared after once dying out. The law which has governed the succession of species, whether we adopt or reject the theory of evolution seems to be expressed in the verse of the poet, Natura il fece, e poi ruppe la stampa. AEIOSTO. Nature made him, and then broke the die. And this circumstance it is which confers on fossils their highest value as chronological tests, giving to each of them, in the eyes of the geologist, that authority which belongs to contemporary medals in history. The same cannot be said of each peculiar variety of rock ; for some of these, as red marl and red sandstone for example, may occur at once at the top, bottom, and middle of the entire sedimentary series, exhibiting in each position so perfect an identity of mineral aspect as to be undistinguishable. Such exact repetitions, however, of the same mixtures of sediment CH. xi.] LIFE-PROVINCES 131 have not often been produced, at distant periods, in precisely the same parts of the globe ; and, even where this has happened, we are not in any danger of confounding together the monuments of remote eras, when we have studied their embedded fossils and their relative position. Zoological provinces. It was remarked that the same species of organic remains cannot be traced horizontally, or in the direction of the planes of stratificationf^for indefinite dis- tances. This might have been expected from analogy ; for when we inquire into the present distribution of living beings, we find that the habitable surface of the sea and land may be divided into a considerable number of distinct areas or provinces, each peopled by a peculiar assemblage of animals and plants. The extent of these separate divisions and the origin of their* in- habitants depend on many causes, of which climate, though certainly an important, is by no means the only one. As, therefore, different seas and lakes are inhabited, at the same period, in different zones and at various depths, by distinct assemblages of aquatic animals and plants, and as the lands adjoining these may be peopled by varied terrestrial species, it follows that distinct fossils will be embedded in contempo- raneous deposits. If it were otherwise if the same species abounded in every climate, or in every part of the globe where, so far as we can discover, a corresponding temperature and other conditions favourable to their existence are found the identifi- cation of mineral masses of the same age, by means of their included organic contents, would be a matter of even greater certainty than it really is. Nevertheless, the extent of some single zoological provinces, especially those of marine animals, is very great; and our geological researches have proved that the same laws prevailed at remote periods ; for the fossils are often identical throughout wide spaces, and in detached deposits, consisting of rocks, vary- ing widely in their mineral nature. The doctrine here laid down will be more readily understood if we reflect on what is now going on in the Mediterranean. That entire sea may be considered as one zoological province ; for, although certain species of mollusca and zoophytes may be very local, and each region (according to its depth, the tem- perature and saltness of the water, and other conditions) has pro- bably some species peculiar to it, still a considerable number are common to the whole Mediterranean. If, therefore, at some future period, the bed of this inland sea should be converted into land, the geologist might be enabled, by reference to organic remains, to prove the contemporaneous origin of various K 2 132 EFFECTS OF VARYING CONDITIONS [CH. xi. mineral masses scattered over a space equal in area to the half of Europe. Deposits, for example, are well known to be now in progress in this sea in the deltas of the Po, Rhone, Nile, and other rivers, which differ as greatly from each other in the nature of their sedi- ment as does the mineral composition of the mountains which they drain. There are also other quarters of the Mediterranean, as off the coast of Campania, or near the base of Etna, in Sicily, or in the Grecian Archipelago, where another class of rocks is now forming ; where showers of volcanic ashes occasionally fall into the sea, and streams of lava overflow its bottom ; and where, in the intervals between volcanic eruptions, beds of sand and clay are frequently derived from the waste of cliffs, or the turbid waters of rivers. Limestones, moreover, such as the Italian travertins, are here and there precipitated from the waters of mineral springs. In all these detached formations, so diversi- fied in their lithological characters, the remains of the same species of shells, corals, Crustacea, and fish are becoming en- closed ; or at least, a sufficient number must be common to the different localities to enable the zoologist to refer them all to one contemporaneous assemblage of species. There are, however, certain combinations of geographical circumstances which cause distinct provinces of animals and plants to be separated from each other by very narrow limits ; and hence it must happen that strata, on the same geological horizon, will be sometimes formed in contiguous regions, differ- ing widely both in mineral contents and organic remains. Thus, for example, the testacea, zoophytes, and fish of the Red Sea are, as a group, distinct from those inhabiting the adjoining parts of the Mediterranean, the narrow isthmus of Suez having acted as an efficient barrier. Calcareous formations have accumulated on a great scale in the Red Sea in modern times, and fossil shells of existing species are well preserved therein ; and we know that at the mouth of the Nile large deposits of mud are amassed, in- cluding the remains of Mediterranean species. It follows, there- fore, that if at some future period the bed of the Red Sea should be laid dry, the geologist might experience great difficul- ties in endeavouring to ascertain the relative age of these forma- tions, which, although dissimilar both in organic and mineral characters, were of synchronous origin. But there are some species of molhisca common to the Medi- terranean and the Red Sea, and their presence would suggest to the geologist of the remote future a more or less complete synchronism. In some parts of the globe the line of demarcation between CH. xi.] EVIDENCE FKOM INCLUDED FKAGMENTS 133 distinct provinces of animals and plants is not very strongly marked, especially where the change is determined by tem- perature, as it is in seas extending from the temperate to the tropical zone, or from the temperate to the Arctic regions. Here a gradual passage takes place from one set of species to another. In like manner, the geologist, in studying particular formations of remote periods, has sometimes been able to trace the gradation from one ancient province to another, by care- fully observing the fossils of all the intermediate places. His success in thus acquiring a knowledge of the zoological or botanical geography of very distant areas has been mainly owing to this circumstance, that the mineral character has no tendency to be affected by climate. A large river may convey yellow or red mud into some part of the ocean, where it may be dispersed by a current over an area several hundred leagues in length, so as to pass from the tropics into the temperate zone. If the bottom of the sea be afterwards upraised, the organic remains embedded in such yellow or red strata may indicate the different animals or plants which once inhabited at the same time the temperate and equatorial regions. It is a general rule that groups of the same species of animals and plants may extend over wider areas than deposits of homogeneous composition ; and thus palaeoritological charac- ters are of more importance in geological classification than the test of mineral composition. Test by included fragments of older rocks. It was stated that proof may sometimes be obtained of the relative date of two formations, by fragments of an older rock being included in a newer one. This evidence may sometimes be of great use, where a geologist is at a loss to determine the relative age of two formations from want of clear sections exhibiting their true order of position, or because the strata of each group are vertical. In such cases we sometimes discover that the more modern rock has been in part derived from the degradation of the older. Thus, for example, we may find chalk in one part of a country, and in another strata of clay, sand, and pebbles. If some of these pebbles consist of that peculiar flint, of which layers more or less continuous are characteristic of the Chalk, and which include fossil shells, sponges, an'd forarninifera of Cretaceous species, we may confidently in/er that the chalk was the older of the two formations. Chronological groups. The separate groups into which the fossiliferous strata may be divided are more or less nume- rous, according to the views of classification which different geologists may entertain ; but when we have adopted a certain 134 UNCONFORMITY AND OVERLAP [CH. XI. system of arrangement we immediately find that a few only of the entire series of groups occur one upon the other in any single section or district. The thinning out of individual strata was before described (p. 37). But let the annexed diagram represent seven fossili- Fig. 113. ferous groups, instead of as many strata. It will then be seen that in the middle all the superimposed formations are present but in consequence of some of them thinning out, No. 2 and No. 5 are absent at one extremity of the section, and No. 4 at the other. In another diagram (fig. 114) a true section of the geological formations in the neighbourhood of Bristol and the Mendip Hills is presented to the reader, as laid down on a natural scale by Sir A. Kamsay, where the newer groups 1, 2, 3, 4 rest Pig. 114. Dundry Hill. Section South of Bristol. (A. C. Ramsay.) Length of section 4 miles. a, b. Level of the sea. 1. Inferior Oolite. 2. Lias. 3. New Red Sandstone. 4. Dolomitic or magnesian conglomerate. 5. Upper coal measures (shales, &c.). 6. Pennant rock (sand- stone). 7. Lower coal measures (shales, &c.). 8. Carboniferous or mountain limestone, with lower limestone shale at its base. 9. Old Red Sandstone. unconformably on, and overlap, the formations 5, 6, 7, and 8. At the southern end of the lino of section we meet with the beds No. 3 (the New lied Sandstone) resting immediately on Nos. 7 and 8, while farther north, as at Dundry Hill in Somersetshire, we have eight groups superimposed one upon the other, compris- ing all the strata from the inferior oolite, No. 1, to the coal and CH. XT.] PAL^ONTOLOGICAL BREAKS 135 carboniferous limestone. The limited horizontal extension of the groups 1 and 2 is owing to subsequent, denudation, as these formations end abruptly, and have left outlying patches to attest the fact of their having originally covered a much wider area. In order, therefore, to establish a chronological succession of fossiliferous groups, a geologist must begin with a single section in which several sets of strata lie one upon the other. He must then trace these formations, by attention to their mineral character and fossils, continuously as far as possible, from the starting-point. As often as he meets with new groups, he must ascertain their age, relatively to those first examined, by super- position, and thus learn how to intercalate them in a tabular arrangement of the whole. By this means the German, French, and English geologists have determined the succession of strata throughout the greater part of Europe, and have adopted pretty generally the groups enumerated in the table at the end of this chapter, p. 145, almost all of which have their representatives in the British Islands. It must be understood, however, that, although in a given locality there may be a physical break unconformity and also a palasontological break change in fossils between two successive groups of strata, these evidences of lapse of time will not be discovered universally and wherever the twc groups are present. Somewhere or other, strata of intervening age will be found to exist ; or the groups will pass insensibly one into the other ; in this way our classificatory distinctions will be found to break down. All stratigraphical schemes are therefore more or less arti- ficial and arbitrary ; and they cannot be applied universally, for the ' breaks,' on which such schemes are based, did not occur contemporaneously over the whole globe. From what has been stated, it may be accepted as a general but not a perfectly strict truth, that strata of different countries which contain the same species of fossils are of similar geological age. Such strata are said to be ' equivalent,' or ' on the same geological horizon,' and these terms are used in a very wide sense. But the strata containing the same species of fossils may be widely separated, geographically, and this fact is opposed to the idea of exact contemporaneity, for it took time for the species to disperse themselves over wide areas. Chronological sequence of Britisb strata. The principle that strata, whatever their mineral characters, and however disturbed in their positions, may be identified by their organic remains, was first clearly enunciated at the end of the last 136 THE WORK OF WILLIAM SMITH [CH. xi. century by the famous William Smith, who has been justly called 'the Father of English Geology.' It so happens that in England we find, within a very small area, representa- tives of the whole series of sedimentary formations lying in their proper sequence, and crowded with exquisitely preserved fossils, but in slightly tilted positions, and with their edges exposed by denudation. We who live in this country have therefore exceptional facilities for studying stratigraphical geo- logy, and for making out the nature and succession of the faunas and floras which distinguish the several sedimentary formations. As a matter of fact, the order of succession of strata was first determined in the British Islands by William Smith and his followers, and the scheme of classification which he elaborated was gradually extended from this country to the continent of Europe, and thence to other parts of the world. The names still applied to the principal, and even to many of the subordinate, groups of strata are those which were given to them by William Smith or his followers. William Smith's table of strata published with the first edition of his geological map of England and Wales, in 1815-16, showed the true order of succession of the British formations from the carboniferous limestone to the chalk, inclusive. With respect to the rocks which underlie the carboniferous, however, this great pioneer in geological investigation found himself unable to apply the important principles he had discovered ; but a quarter of a century later the labours of Sedgwick, Murehison, and Lonsdale, carried on upon the lines laid down by Smith, resulted in the establishment of the older geological systems as understood at the present day. With respect to the strata over- lying the chalk, Smith fell into some serious errors, which were only finally got rid of by the extension of the palaeontological me- thod advocated in the first edition of the ' Principles of 3-eology.' Caution necessary in using 1 tbe Palaeontological Me- thod. But, from what we have said in preceding paragraphs, it will be obvious that the use of fossils for the identification of strata calls for a certain amount of caution. In the first place, it is obviously necessary that we should satisfy ourselves that the fossils we find in a bed are really the remains of organisms that were living when the beds containing them were deposited. William Smith made out his order of succession of strata in the first instance in the district between Bath and Bristol, while making a survey for the construction of a canal. He soon saw, however, that besides the regularly stratified formations, clay, sand, and limestone, each containing its peculiar (' characteris- on. XL] THE PAL^ONTOLOGICAL METHOD 137 tic ' ) fossils, there were masses of sand and gravel in which all these fossils were found mingled indiscriminately. Careful examination, however, soon convinced him that the fossils in the gravel were all derived ones, that is, had been washed out of older beds ; and this fact was inferred from their waterworn characters, the differences in their mode of mineralisation, and the circumstance that they often contained portions of a matrix quite different from that of the deposit in which they are now found. In appealing to fossils as indicating the geological age of a stratum, therefore, we must be perfectly satisfied that they belong to the formation, and are not derived. The importance of the principle will be seen when we come to study the strata known as the ' crags.' In the second place, we must remember that (inasmuch as distinct forms of shells, corals, sponges, &c., inhabit different depths of the ocean, and particular assemblages of animals and vegetables nourish on sandy or muddy bottoms respectively) strata formed in the same district, during a particular period cannot be expected to exhibit identical fossils. Geologists soon learn to recognise that each period has its shallow-water and its deep-water forms ; and that different assemblages of species occur in the clays, sands, and limestones of the same formation. In the third place, it must be remembered that, as we have a geographical distribution of life-forms at the present day, there are clear evidences of a similar geographical distribution of animals and vegetables during the earlier periods of the earth's history. Hence, while within a more or less limited area we may expect to find a particular assemblage of fossils in a geological formation, we must be prepared in distant areas to find these particular fossils more or less completely wanting, and their place taken by an equivalent or representative group of fossils, belonging to the same period, but to a different zoo- logical province. We thus see that the fossil flora or fauna l found in a forma- tion at a particular locality is a function (to use a mathematical expression) of three variables. The assemblage of life-forms depends first on the conditions that prevailed when the beds were deposited such as depth of water, climate, nature of sea- bottom, &c. ; secondly, on the particular zoological province in which the locality was situated ; and thirdly, on the geological period at which the beds were formed. We must always be on our guard to avoid assigning to differences of geological age 1 The assemblage of animal plants its ' flora.' Thus we speak of forms in a particular area or stra- ' the British flora,' ' the Mediter- tum is called by naturalists its ranean fauna,' 'the Cretaceous 'fauna,' and the assemblage of fauna,' 'the Carboniferous flora,' &c. 138 HOMOTAXY [CH. xi. changes of flora or fauna which may be due to differences of condition or of geographical position. Limits of the Palaeontologlcal Method, zxomotaxy. There is another important consideration to which the atten- tion of geologists was especially called by the late Mr. Godwin- Austen. Upon a gradually rising or sinking ocean-floor different areas may successively exhibit the same peculiarities of depth of water, temperature, &c. ; and the forms of life which affect those conditions may be naturally expected to migrate from the old areas where the conditions have become unfavourable, into those new areas where the favourable conditions appear. Thus, in the stratum known as the Upper Greensand, which was evidently deposited in shallow water near a sinking coast line, belts of similar sediment of different age, but containing the same fossils, would be successively formed, and these will be taken by the geologist to be of contemporaneous formation. It is nevertheless evident that long periods of time may have elapsed between the deposition of one part of the Upper Greensand and another part of the same stratum or formation. This brings us to the consideration of the meaning of the term ' synchronous ' or ' contemporaneous ' as employed by the geologist. Even the historian employs such a term with considerable latitude; but geologists, who are quite unable to assign terms of years for the great periods with which they have to deal, must necessarily use the word contemporary in a much more general sense even than the historian. Two beds are said to be contemporary by the geologist, when the time between the periods of their deposition does not appear to have been sufficient for any marked change in the forms of animal and vegetable life. But, as forms may migrate without change, the term ' geological contemporaneity ' can have only a very general application. Of two great systems of strata in distant parts of the globe it can frequently only be said that they have a like position in the great geological record. In these cases, as pointed out by the late Professor Huxley, it is safer to employ the term ' homotaxy,' signifying a similarity of arrangement, instead of ' contemporaneity ' or ' synchronism,' which conveys the idea of absolute identity in time. Frequent unconlormability of strata. Where the widest gaps appear in the sequence of the fossil forms, as between the Permian and Triassic rocks, or between the Cretaceous and Eocene, examples of stratigraphical unconformability are very frequent. But they are also met with in some part or other of the world at the junction of almost all the other principal formations, and sometimes the subordinate divisions of any one CH. xi.] THE GEOLOGICAL KECORD IMPERFECT 139 of the leading groups may be found lying unconforniably on another subordinate member of the same. Instances of such irregularities in the mode of succession of the strata are the more intelligible as we extend oar survey of the fossiliferous for- mations over wider areas, for we are continually bringing to light deposits of intermediate date, which have to be inter- calated between those previously known ; these deposits reveal to us a long series of events, which, antecedently to such dis- coveries, were quite unsuspected by us. But while unconformability invariably bears testimony to a lapse of unrepresented time, the conformability of two sets of strata in contact by no means implies that the newer formation immediately succeeded the older one. It simply indicates that the ancient rocks were subjected to no movements of such a nature as to tilt, bend, or break them before the more modern forma- tion was superimposed. It does not show that the earth's crust was motionless in the region in question, for there may have been a gradual sinking or rising, extending imiformly over a large area, and yet during such movement the stratified rocks may have retained their original horizontality of position. Strata possessing very different animal remains and different kinds of rock may still be conformable, yet great changes must have occurred. There may have been a conversion of a wide area from sea into land and from land into sea, and during these changes of level some strata may have been slowly removed by aqueous action, and after this new strata may be superimposed, differing perhaps in date by thousands of years or centuries, and yet resting conformably on the older set. There may even be a blending of the materials constituting the older deposit with those of the newer, so as to give rise to a passage in the mineral character of the one rock into the other as if there had been no break or interruption in the depositing process. Imperfection of the record. Although, by the frequent discovery of new sets of intermediate strata, the transition from one type of organic remains to another is becoming less and less abrupt, yet the entire series of records appears to the geologists now living far more fragmentary and defective than it seemed to their predecessors a century ago. The earlier inquirers, as often as they encountered a break in the regular sequence of formations, connected it, theoretically, with a sudden and violent catastrophe, which had put an end to the regular course of events that had been going on uninterruptedly for ages, annihilating at the same time all or nearly all the organic beings which had pre- viously flourished, after which, order being re-established, a new series of events was initiated. In proportion as our faith in 140 NEWER STRATA SHOULD BE STUDIED FIRST [CH. xi. these views grows weaker, and the phenomena of the organic or inorganic world presented to us by geology seem explicable on the hypothesis of gradual and insensible changes, varied only by occasional convulsions, on a scale comparable to, though it may be far greater than, any witnessed in historical times ; and in proportion as it is thought possible that former fluctuations in the organic world may be due to the indefinite variability of species without the necessity of assuming new and independent acts of creation, the number and magnitude of the gaps which still remain, or the extreme imperfection of the record, become more and more striking, and what we possess of the ancient annals of the earth's history appears insignificant when con- trasted with that which has been lost. It is observed that strata, in proportion as they are of newer date, bear the nearest resemblance in mineral character to those which are now in process of formation in seas or lakes, the newest of all consisting principally of soft mud or loose sand, in some places full of shells, corals, or other organic bodies animal or vegetable in others wholly devoid of such remains. The farther we recede from the present time, and the higher the antiquity of the formations which we examine, the greater, as a general rule, are the changes which the sedimentary deposits have undergone. Time, as has already been explained, has multiplied the effects of alteration by pressure and solution, and the modifications brought about by heat, pressure, contortion, upheaval, and denudation. The organic remains have sometimes been obliterated entirely, or the mineral matter of which they were composed has been removed and replaced by other sub- stances. Why newer groups should be studied first. We like- wise observe that the older the rocks the more widely do their organic remains depart from the types of the living creition. Thus we find in the newer Tertiary rocks a few species which no longer exist, mixed with many living ones, and then, as we go farther back, many genera and families at present unknown are met with, until we come to strata in which the fossil relics of existing species and genera are nowhere to be detected, while families and orders of animals and plants wholly unrepresented in the living world begin to be conspicuous. When we study, therefore, the geological records of the earth and its inhabitants, we find, as in human history, the defective- ness and obscurity of the monuments always increasing, the remoter the era to which we refer ; the rocks becoming more generally altered and crystalline the older they are, and the difficulty of determining their true chronological relations CH. xi.] CLASSIFICATION OF TERTIARY STRATA 141 becoming more and more enhanced, especially when we are comparing those which were formed in very distant regions of the globe. Hence we advance with securer steps when we begin with the study of the geological records of later times, proceeding from the newer to the older, or from the more to the less known. In thus inverting what might at first seem to be the more natural order of historical research, we must bear in mind that each of the periods above enumerated, even the shortest, such as the Post-tertiary, or the Pliocene, Miocene, or Eocene, embraces a succession of events of vast extent, so that to give a satis- factory account of what we already know of any one of them would require many volumes. When, therefore, we study one of the newer groups before endeavouring to decipher the monu- ments of an older one, it is like endeavouring to master the history of our own country and that of some contemporary nations, before we enter upon Roman History ; or like investiga- ting the annals of Ancient Italy and Greece before we approach those of Egypt and Assyria. The geological record is so much more complete in the case of the Tertiary or youngest strata, that geologists have been led to adopt principles of chronological classification with respect to them which are somewhat different from those that have been found suitable when dealing with the much more fragmentary records of the Mesozoic and Palaeozoic Eras. The Tertiary or Cainozoic strata were so called because they were all posterior in date to the Secondary series, of which last the chalk or Cretaceous constitutes the newest group. The whole of the Tertiaries were at first confounded with the super- ficial alluvia of Europe ; and it was long before their real extent and thickness, and the various ages to which they belong, were fully recognised. They were observed to occur in patches, some of freshwater, others of marine origin, their geographical extent being usually small as compared with that of the Secondary formations, and their position often suggesting the idea of their having been deposited in different bays, lakes, estuaries, or inland seas, after a large portion of the space now occupied by Europe had already been converted into dry land. The first deposits of this class of which the characters were accurately determined, were those occurring in the neighbour- hood of Paris, described in 1810 by Cuvier and Brongniart. They were ascertained to consist of successive sets of strata, some of marine, others of freshwater origin, lying one upon the other. The fossil shells and corals were found to be almost all of unknown species, but to have a general affinity with those now inhabiting warmer seas. The bones and skeletons of 142 THE GREAT DIVISIONS OF THE TERTIARY [CH. XT. land animals,, some of them of large size, and belonging to more than forty distinct species, were examined by Cuvier, and declared by him not to agree either specifically, or even gene- rically, with any hitherto observed in the living creation. Strata were soon afterwards brought to light in the vicinity of London, and in Hampshire, which, although dissimilar in mineral composition, were justly inferred by Webster to bo of the same age as those of Paris, because the greater number of the fossil shells were specifically identical. For the same reason, rocks found in the Gironde, in the South of France, and at certain points in the North of Italy, were suspected to be of contemporaneous origin. Another important discovery was soon afterwards made by Brocchi in Italy. He investigated the argillaceous and sandy deposits replete with shells, which form a low range of hills flanking the Apennines on both sides, from, the plains of the Po to Calabria. These lower hills were called by him the Sub- apennines, and were found to consist of strata chiefly marine, and newer than those of Paris and London. Another tertiary group occurring in the neighbourhood of Bordeaux and Dax, in the South of France, was examined by Basterot in 1825 ; and he described and figured several hun- dred species of shells, which differed for the most part both from the Parisian series and those of the Subapennine hills. It was soon, therefore, suspected that this fauna might belong to a period intermediate between that of the Parisian and Subapen- nine strata, and it was not long before the evidence of super- position was brought to bear in support of this opinion; for other strata contemporaneous with those of Bordeaux were observed in one district (the Valley of the Loire) to overlie the Parisian formation, and in another (in Piedmont) to underlie the Subapennine beds. The first example of these was pointed out in 1829 by Desnoyers, who ascertained that the sand and marl, full of sea-shells and corals, occurring near Tours, in the basin of the Loire, and called Faluns, rest upon a lacustrine formation, which constitutes the uppermost subdivi- sion of the Parisian group, extending continuously throughout a great table-land intervening between the basin of the Seine and that of the Loire. The other example occurs in Italy, where strata containing many fossils similar to those of Bordeaux, were observed by Bonelli and others in the environs of Turin, subjacent to strata belonging to the Subapennine group of Brocchi. Long afterwards, the superficial layers which cover many of these, and which have their stones scratched and polished, were found to contain Arctic shells. CH. si.] IMPOETANCE OF FOSSIL MOLLUSCA 14'd Value of fossil mollusca in classification. It will be observed that in the foregoing allusions to organic remains the shell-bearing mollusca are selected as the most useful and convenient class for the purposes of general classification. In the first place, they are more universally distributed through strata of every age than any other organic bodies. Those families of fossils which are of rare and casual occurrence are of little use in establishing a chronological arrangement. If we have plants alone in one group of strata and the bones of mammalia in another, we can draw no conclusion respecting the affinity or discordance of the organic beings of the two epochs compared ; and the same may be said if we have plants and vertebrated animals in one series and only shells in another. Although corals are more abundant, in a fossil state, than plants, reptiles, or fish, they are still rare in comparison with shells, because they are more dependent for their well-being on the constant clearness of the water, and are, therefore, less likely to be included in rocks which endure in consequence of their thickness and the copiousness of sediment which prevailed when they originated. The utility of the mollusca is, moreover, enhanced by the circumstance that some forms are proper to the sea, others to the land, and others to fresh water. Rivers scarcely ever fail to carry down into their deltas some land-shells, to- gether with species which are at once fluviatile and lacustrine. By this means we learn what terrestrial, freshwater, and marine species coexisted at particular eras of the past ; and having thus identified strata formed in seas with others which originated contemporaneously in inland lakes, we are then enabled to advance a step farther, and show that certain quadrupeds or aquatic plants, found fossil in lacustrine formations, inhabited the globe at the same period when certain fish, reptiles, and zoophytes lived in the ocean. Among other characters of the molluscous animals, which render them extremely valuable in settling chronological ques- tions in geology, may be mentioned, first, the wide geographical range of many species : and, secondly, what is probably a con- sequence of the former, the great duration in time of some species in this class, for they appear to have surpassed in longevity the greater number of the fish and mammalia. Had each species inhabited a very limited space, it could never, when embedded in strata, have enabled the geologist to identify deposits at distant points over large areas ; or had they each lasted but for a brief period, they could have thrown no light on the connection of rocks placed far from each other in the chronological, or, as it is sometimes termed, the vertical series. 144 NAMES GIVEN TO TERTIARY FORMATIONS [CH. XL Classification of Tertiary strata. In the first edition of the ' Principles of Geology ' the whole of the Tertiary formations were divided into four groups, characterised by the percentage of recent shells which they contained. The lower tertiary strata of London and Paris were thought by Deshayes to contain only 3^ per cent, of recent species, aud were termed Eocene. The middle tertiary of the Loire and Gironde had, according to the specific determinations of the same eminent ".onchologist, 17 per cent., and formed the Miocene divi- sion. The Subapennine beds contained 35 to 50 per cent., and were termed Older Pliocene, while still more recent beds in Sicily, which had from 90 to 95 per cent, of species identical with those now living, were called Newer Pliocene. The first of the above terms, Eocene, is derived from rjas, eos, dawn, and KCUI/O.C, cainos, recent, because the fossil shells of this period contain an ex- tremely small proportion of living species, which may be looked upon as indicating the dawn of the existing state of the mollus- can fauna, no recent species (with one or two exceptions) having been detected in the older or secondary rocks. The term Miocene (from /netoi', meion, less, and KUIVOS, cainos, recent) is intended to express a minor proportion of recent species (of mollusca), the term Pliocene (from TrXelov, pleion, more, and Kaivos, cainos, recent), a comparative plurality of the same. It may assist the memory of students to remind them, that the Miocene contain a minor proportion, and PZiocene a compara- tive ^Zurality of recent species ; and that the greater number of recent species always implies the more modern origin of the strata. Subsequently to this classification, Beyrich founded the ' Oligocene ' as a division intermediate between the Eoceni proper and the Miocene. This division includes the Lower Mio- cene formations of older writers, together with much jf their Upper Eocene Series. Nummulites, so abundant in the Eocene, became scarce and degenerated in the Oligocene series, which in Europe contains very important freshwater beds with mam- malian remains, as well as marine deposits. Since the year 1830 the number of known shells, both recent and fossil, has largely increased, and their identification has been more accurate. Hence some modifications have been required in the classifications founded on less perfect materials. The Eocene, Oligocene, Miocene, and Pliocene periods have been made to comprehend certain sets of strata, of which the fossils do not always conform strictly, in the numerical proportions of recent to extinct species, with the definitions first given to those divisions or which are indicated in the etymologies of the terms. CH. XT.] TABLE OF STRATA 145 There is such convenience in distinguishing between the earlier Tertiary strata in which only a small minority of the fossil shells are found living in the existing seas, and the later deposits in which a very considerable proportion of the shells are still living, that we shall follow the geologists of Eastern Europe and North America in adopting a twofold division for the great mass of the Cainozoic rocks. We shall speak of the earliest Tertiary strata as Older Tertiaries, as the term has long been in use in this country ; in the United States, and in Eastern Europe, this division is often called ' Eogene.' The Newer Ter- tiaries of English authors are called by the Austrian geologists ' Neogene,' and by those of the United States ' Neocene.' It will be convenient to give at this point a summary, in the ABRIDGED GENERAL TABLE OF FOSSILIFEROUS STRATA CLASSES of Strata representing ERAS of Time CAINOZOIC (Tertiary) MESOZOIC (Secondary) SYSTEMS of Strata representing PERIODS of Time PLEISTOCENE NEWER TERTI ARIES OLDER, TERTIARIES CRETACEOUS STAGES of Strata representing EPOCHS of Time Post-Glacial. Glacial. Pre-Glacial. Pliocene. Miocene (not known in Britain). Oligocene. Eocene. Paleocene (not known in Britain). Chalk. Upper Greensand and Gault. Neocomian. ( Oolites. JURASSIC \ Lias. ( Rluetic. ^ TRIASSIC ( Keuper j Muschelkalk (not known in Britain). ( Bunter. 1 PERMIAN f Upper Permian. \ Roth-todt-liegende. NEWER PALAEOZOIC CARBONI- FEBOUS C Coal measures. 4 Millstone grit. ( Lower limestones and shales. DEVONIAN (Upper Devonian. Middle Devonian. Lower Devonian. 1SILUBIAN f Ludlow. ] Wenlock. ( May Hill. OLDER PALAEOZOIC OBDOVICIAN ( Bala. 4 Llandeilo. ( Arenig. CAMBBIAN ( Upper Cambrian. 4 Middle Cambrian. ( Lower Cambrian. AGNOTOZOIC. Eparchian or Algoukian strata. 146 COMPARISON OF STRATA IN DISTANT AREAS [CH. xi. form of a table, of the general system of classification of strata, according to their geological age, which has now been generally adopted by geologists. It must be borne in mind that this classification of geological periods is in the mam the result of studies carried on in the British Islands and Western Europe, and that if the science of stratigraphical geology had originated in Eastern Europe, India, Australia, or the United States, the great divisions which would have been adopted and the limits between them would have been altogether different. Even among European geologists there is considerable diver- sity in opinion and practice as to the delimiting and naming of the great geological systems, and of their principal subdivisions. Thus, some authors make the lower portion of the Cretaceous a distinct system, calling it ' Neocomian,' while others divide the Jurassic into two, the Liassic and Oolitic. It may be some aid to the memory to adopt the four great classes of strata, each in- cluding three systems, as shown in the Table. Some English authors still follow Murchison in combining the Silurian and Ordovician, and naming the latter ' Lower Silurian.' The Table takes account only of marine formations ; but it must be remembered that in addition to these there are great systems of strata of freshwater origin, like the Wealden and the Old Red Sandstone. Eventually it may be necessary to have two distinct schemes of classification for stratified rocks, one to include strata of marine origin, the other for freshwater and terrestrial deposits. The limits of the systems and other sub- divisions in these two schemes, could not be expected to agree. In giving names to the groups of strata of different orders of magnitude, and the divisions of time which they represent, we have followed the scheme proposed by the International Geolo- gical Congress, with the modifications suggested by Mr. Blanford. The jjrinciples of geological cations and nomenclatures for the classification have been discussed strata for widely separated areas, by Professor Huxley in his address like South Africa, India, Austra- to the Geological Society in 1862, lia, and South America, rather on 'Geological Contemporaneity than to assign the terms estab- and Persistent Types of Life,' re- lished by the study of European printed in his collected essays. geology to formations at a great The student should also consult distance, which have so little in Mr. Blanford's addresses to the common with them either in their Geological Society in 1889-90. It mineral characters or their fossils. is wiser to adopt different classifi- 147 THE CAINOZOIC (TERTIAEY) EEA CHAPTER XII THE PLEISTOCENE PERIOD WITH THE GLACIAL EPISODE Use of the terms 'pleistocene,' 'recent,' and 'human' Mollusca of the Pleistocene period Mammalia of the Pleistocene period Shorter duration of mammalian as compared with molluscan species Geo- graphical distribution of mammalia in Pleistocene times similar to that at present day Remains of man Flint implements Shell mounds Cavern deposits Valley gravels High- and low-level gravels Brick earth Loess Lacustrine deposits Estuarine deposits Marine de- posits Subdivisions of the Pleistocene period Pre-glacial The Glacial period Origin of Boulder clay Glacial lakes and other phenomena of glaciated districts Post-glacial, Pluvial, and Champlain periods Palaeolithic and Neolithic Copper, Bronze, and Iron Ages, Nomenclature and classification of the Pleistocene deposits. The youngest of the divisions of the Newer-Tertiary system is known as the Pleistocene, or Post-pliocene. The terms ' Post-tertiary ' and ' Quaternary ' have also been applied to the period by some authors ; but these names may fairly be objected to on the ground that they imply the existence of differences between these youngest strata and the other Tertiary rocks, which are not borne out when a careful comparison is made of their organic remains. The term * Pleistocene,' pro- posed by Lyell in 1839 as a synonym for Newer Pliocene, was used by the late Edward Forbes as the equivalent of Post- pliocene, and has now passed into general use with that signifi- cation. The very latest deposits of this period are sometimes dis- tinguished by the terms ' recent ' and * human.' To the use of the former term it may be objected that cases constantly occur in which it is impossible to draw a boundary line between the recent and other Pleistocene deposits. The employment of the term 'human period' is equally inconvenient, seeing that geologists are by no means agreed as to the exact part of the Pleistocene period at which man made his appearance on the L 2 148 PLEISTOCENE MOLLUSCA [CH. XII. earth, while some observers have even maintained that there is evidence of his existence in pre-Pleistocene times. Characteristics of the fauna and flora of the Pleisto- cene deposits. The shells found in these Pleistocene deposits belong, almost without exception, to species still living on the earth. It is worthy of remark, however, that the geographical distribution of these rnollusca was in Pleistocene times very different from that of the present day. In not a few cases we find in the Pleistocene deposits of the British Islands and North America an assemblage of shells now only found in much higher latitudes, where the temperature of the sea is much colder than that both of the British Islands and of the Atlantic shore of the United States. The shells figured below are only a few out of a large assemblage of living species, which, taken as a whole, bear Fig. 115. Astarte borealis, Chem. sp. J. Fig. 116. Leda lanceolata, Sow. |. Fig. 117. Fig. 118. Fig. 119. Fig. 120. Sazicava rugosa, Pecten islandicus, Natica clausa, Trophon clathra- Lam. 4. Mull. . Brod. & Sow. \. turn, L. Mag. 2 diatns. Northern shells common in the drift of the Clyde, in Scotland. testimony to conditions far more arctic than those now pre- vailing in the Scottish seas. But a group of marine shells, indicating a still greater excess of cold, has been brought to light from glacial drift or clay on the borders of the estuaries of the Forth and Tay. This clay occurs at Elie in Fife, and at Errol in Perthshire ; and has already afforded about thirty-five shells, all of living species, and now inhabitants of Arctic regions, such as Leda truncata, Brown, Tellina calcarea, Chem. (see figs. 121, 122), Pecten groenlandicus, Sow., Crenella Icevigata, Gray, Crenella nigra, Gray, and others, some of them first brought by Captain Sir E. Parry from the coast of Melville Island, latitude 76 N. These were all identified in 1863 by Dr. Torell, who had just returned from a survey of the seas around Spitzbergen, where he had collected no less than 150 CH. XII.] AECTIC AND SOUTHERN FORMS 149 species of mollusca, living chiefly on a bottom of fine mud derived from the moraines of glaciers which protrude into the sea. He found that the fossil fauna of this Scotch glacial deposit exhibits Fig. 121. Fig. 122. Leda truncata, Brown. a. exterior of left valve. 6. interior of same. Nat. size. Tellina calcarea, Chem. a. outside of left valve. 6. interior of same. not only the species but also the peculiar varieties of mollusca now characteristic of very high latitudes. On the other hand, there are a few species of mollusca in the Pleistocene beds of this country which are now found living in much warmer cli- mates. Of these Corbicula (Cyrend} fluminalis, Mull. (fig. 123), a shell found living in the Nile at the present day, may be taken as a conspicuous example. The great majority of the Pleistocene mollusca are found also in the underlying Pliocene deposits. But there are some striking examples of forms occurring in the Pleistocene which are quite unknown in any earlier formation. Of these Tellina balthica, L. (fig. 124), a shell still common in the British seas, is an interesting example. Fig. 123. Fig. 124. Cyrcna (Corbicula) fluminalis, Tellina balthica, L. Nat. size. Mull. ; fossil, Grays, Essex, and living in the Nile, nat. size. While the mollusca of the Pleistocene strata are so similar to those of the present day on the one hand, and to those of the Newer Pliocene on the other hand, there are the most striking differences exhibited, not only in the geographical distribution, but likewise in the specific forms of the vertebrate forms of life which flourished at these different periods ; a part, and indeed often a very considerable part, of the mammalia found in Pleis- tocene deposits belonging to extinct species. Relative longevity of species in the mammalia 150 KELATIVE LONGEVITY OF SPECIES [CH. xn. mollusca. In 1830 1 attention was called to the fact which had not at that time attracted notice that the association in the Pleistocene deposits of shells, exclusively of living species, with many extinct quadrupeds, betokened a longevity of species in the mollusca far exceeding that in the mammalia. Subsequent researches seem to show that this greater duration of the same specific forms in the class mollusca is dependent on a still more general law, namely, that the lower the grade of animals, or the greater the simplicity of their structure, the more persistent are they in general in their specific characters throughout vast periods of time. Those organisms which are of more simple structure have varied at a slower rate than those of a higher and more complex organisation ; the Brachiopoda, for example, more slowly than the Lamellibranchiata, while the latter have been more persistent than either the Gastropoda or the Cephalopoda. In like manner the specific identity of the characters of the Foraminifera, which are among the lowest types of the in- vertebrata, has outlasted that of the mollusca in an equally decided manner. Teetb of Pleistocene mammalia. To those who have never studied comparative anatomy, it may seem scarcely credible that a single bone, or even the fragment of a bone, taken from any part of the skeleton, may enable a skilful osteologist to distinguish, in many cases, the genus, and some- times the species, of quadrupeds to which it belonged. Al- though few geologists can aspire to such knowledge, which must be the result of long practice and study, they will nevertheless derive great advantage from learning, what is comparatively an easy task, to distinguish the principal divisions of the mammalia by the forms and characters of their teeth. The figures on pages 151 to 153 represent the teeth of some of the more common species and genera founu in the alluvial and cavern deposits of the Pleistocene period. On comparing the grinding surfaces of the corresponding molars of the three species of elephants, fig. 125, it will be seen that the folds of enamel are most numerous in the Mam- moth ; fewer and wider, or more open, in E. antiques, Falc. ; and most open and fewest in E. meridionalis, Nesti, a Pliocene form. It will be also seen that the enamel in the molar of the Rhinoceros tichorhinus, Cuv. (fig. 127), is much thicker than in that of the Rhinoceros leptorhinus, Cuv. (fig. 126). When a comparison is made between the mammalia found in the Pleistocene deposits of different parts of the earth's sur- face and the forms of life now inhabiting the same areas, we find 1 Principles of Geology, 1st ed. vol. iii. p. 140. CH. xii.] TEETH OF EXTINCT MAMMALIA 151 Molar teeth of late Tertiary Elephants. A. Elephas primigenius, Blumenb. (or Mammoth). Molar of upper jaw, right side ; one-third of natural size. Pleistocene. a. grinding surface. It. side view. B. Elephas anliquns, Falc. Penultimate molar ; one-third of natural size. Pleis- tocene and Pliocene. 0. Elephas meridionals, Nesti. Penultimate molar ; one-third of natural size. Pliocene, 152 TEETH OF EXTINCT MAMMALIA [CH. xii. ample proof that the geographical distribution of these forms of terrestrial life was similar to what is found at the present day. This is well seen by the study of the bones which have been found in peat mosses and caves in Australia. No remains of any European or Asiatic animal have been Fig. 126. Fig. 127. Fig. 128. Rhinoceros leptorhinus, Cu- vier (R. megarhinus, Christol) ; fossil from freshwater beds of Grays, Essex ; penultimate mo- lar, lower jaw, left side ; two-thirds of nat. size. Pleistocene and Newer Pliocene. Fig. 129. Rhinoceros tichorhinus, Cu- Hippopotamus major, Nesti; vier ; penultimate molar, from cave near Palermo ; lower jaw, left side ; two- molar tooth, two-thirds of thirds of nat. size. Pleis- nat. size. Pleistocene, tocene. Living. Fig. 130. Horse. Equus caballus, L. (common horse) ; from the shell-marl, Forfarshire ; se- cond molar, lower jaw. Recent. o. grinding surface, two-thirds nat. size. fe. side view of same, half nat. size. Deer. Elk (Cervus alces, L.) ; recent ; molar of upper jaw. a. grinding surface. &. side view ; two-thirds of nat. size. found in those deposits ; the bones belong to those families of Marsupials, without exception, which are now existing in Aus- tralia. The animals were in some instances gigantic. The genera Macropus (Kangaroo), Peramales (Bandicoot), Phalan- ger, Dasyurus, and Phascolomys (Wombat), were represented TEETH OF EXTINCT MAMMALIA 153 by the remains of gigantic and small species, some of which are extinct, while others still exist. A huge animal called Fig. 131. Fig. 132. Ox (Bos taurus, L.). from shell marl, Forfarshire ; true molar, upper jaw ; two-thirds nat. size. Living. c. grinding surface. d. side view ; fangs uppermost. . canine tooth or tusk of bear ( Ursus spelceus, Blumenb.) ; from cave near Liege. 6. molar of left side, upper jaw, one- third of nat. size. Pleistocene. Fig. 133. Fig. 134. Tiger. c. canine tooth of tiger (Felis tigris, Hycena spelcea, Goldf. (variety of H. crocuta, L.). Living. \ nat. Zimm.); part of lower jaw. Kent's Hole, Torquay, Devonshire. One-third nat. size. Pleistocene. Living in Africa. d. outside view of posterior molar, lower jaw ; one-third nat. size. Recent. Teeth of A rvicola intermedius, E. T. Newton ; a vole, or field-mouse, from the Norwich crag. Newer Pliocene. a. grinding surface. b. side view of same. c. nat. size of a and &. Diprotodon from its great front teeth, another, the Nototherium, find also Protemnodon and Sthenurus were found, and all were 154 EXTINCT MAMMALIA [CH. xn. marsupials. Another marsupial called Thylacoleo, probably of carnivorous habit, also abounded. Of the same geological age as these breccias are the bogs and swamp beds of the valleys of South Australia and Queensland, which contain Diprotodon and other marsupial remains. It is very note- worthy that marsupials alone should have lived in Australia in the Pleistocene age, for they are the only mammalia truly in- digenous in that continent at the present time. It is one of the many instances of the persistence of a type on the same area, and it indicates long separation from other lands. This law of geographical relationship between the living and Pleistocene vertebrata is extremely interesting, and is not confined to the mammalia only. Thus, when New Zealand was first examined by Europeans, it was found to contain no indige- nous land quadrupeds ; but a small bird, wingless or with very rudimentary wings, abounded there, the smallest living re- presentative of the Ostrich family, called the Kiwi by the natives (Apteryx). In the remains of the Pleistocene period in the same island, there are numerous well-preserved specimens of gigantic birds of the Struthious or Ostrich order, belonging to genera called by Owen Dinornis and Palapteryx, which are entombed in superficial deposits. These genera comprehended many species, some of which were four, some seven, others nine, and others eleven feet in height! No contemporary mammalia shared the land with this population of gigantic feathered bipeds. Mr. Darwin, when describing the recent and Pleistocene mammalia of South America, dwelt much on the wonderful relationship of the extinct to the living types of that part of the world, inferring from such phenomena that the existing species are all related to the extinct ones which preceded them by the bond of common descent. In the Pampas of South America the skeletons of Mega- therium, Megalonyx, Mylodon, Glyptodon, Toxodon, Macrau- chenia, and other extinct forms, find their nearest analogues in the living Sloth, Armadillo, Cavy, Capybara, and Llama of that continent. The skeleton of one of these great extinct sloths is represented in fig. 136 on the opposite page. The fossil quadrumana, also associated with some of these forms in the Brazilian caves, belong to the Platyrrhine family of monkeys, now peculiar to South America. That the extinct fauna of Buenos Ayres and Brazil was not very ancient has been shown by its relation to deposits of marine shells, agreeing with those now inhabiting the Atlantic. Bones of great Carnivora have been found, and also of the Peccary. Moreover, human rQ- CH. xii.] OF AUSTRALIA AND SOUTH AMERICA 155 mains have been got from the Brazilian caves with these bones. It is interesting to note that the Opossum, which belongs to a marsupial family peculiar to America, is found in these cave 156 PALEOLITHIC IMPLEMENTS [CH. xn. breccias, and it is not associated with any Australian kinds, neither, on the other hand, is any Didelphys (Opossum) found in Australia (Note I, p. 603). The old natural history provinces of this Pleistocene period were limited by natural boundaries and had their characteristic fauna. There was no mixture of European types with the South American or Australian, and the animals of Asia did not roam to the south, into Australia. While we have no certain indication of the existence of human beings before the Pleistocene period, there is indisputable evidence that man existed on the earth at least during the latter- Fig. 137. Palaeolithic flint implements. A. Spear-head type. St. Acheul. . B. Oval-shaped type. Mautort, near Abbe- One-third of the original size. ville. One-half of the original size. portion of the Pleistocene, and that he was the contemporary of the remarkable forms of mammalia, many of them now ex- tinct, which we have been describing. In 1847 Boucher de Perthes observed in an ancient alluvium at Abbeville, in Picardy, the bones of extinct mammalia associated in such a manner with flint implements of a rude type, as to lead him to infer that both the organic remains and the works of art were referable to one and the same period. This inference was soon after confirmed by Professor Prestwich, who found in 1859 a flint implement in $itu in the same stratum at Amiens that contained the remains CH. XII.] NEOLITHIC IMPLEMENTS 157 of extinct mammalia. Since that time palaeolithic stone imple- ments have been found in many valley gravels on all the con- tinents. The flint implements found at Abbeville and Amiens (fig. 137) are different from those commonly called ' celts ' (fig. 138). These celts, so often found in the recent formations, have a more regular oblong shape, the result of grinding, by which also a sharp edge has been given to them. The Abbeville im- plements found in gravel at different levels, as in Nos. 3 and 4, fig. 112, p. Ill, in which bones of the Elephant, Rhinoceros, and other extinct mammalia occur, are always unground, having evidently been brought into their present form simply by the chipping off of fragments of flint by repeated blows, such as could be given by another stone. Some of them are oval, others of a spear- headed form, no two exactly alike, and yet the greater number of each kind are ob- viously fashioned after the same general pattern, which is world-wide. Their outer surface is often white, the original black flint having been discoloured and bleached by exposure to the air, or by the action of acids as they lay in the gravel. They are most commonly stained of the same ochreous colour as the flints of the gravel in which they are embedded. Occasionally their antiquity is indicated not only by their colour but by superficial incrustations of calcium carbonate, or by dendrites formed of oxide of iron and manganese (see figs. 73-75, p. 73). The edges also of most of them are worn, sometimes by having been used as tools, and sometimes by having been rolled in the old river's bed. In addition to the flints which have evidently been chipped or ground, so as to form implements of very definite shape, others of a much ruder type are found, which Professor Prest- wich and other geologists and archaeologists regard as marking a still more primitive condition of the human race. The flints in question are simply flat, irregular fragments which have been picked up to serve the purpose of scrapers, and they bear on their edges the marks of having been so employed. The Tas- manians and some other savage tribes are known to employ fragments of flint or similar hard materials in this way, without Neolithic polished celt found at Cotton, Cam- bridgeshire, 1863. One- half of the original size. 158 PEAT-MOSSES [CH. xn. any attempt at fashioning them into tools. The rude flints of this type are found scattered over the plateaux of our chalk districts, or embedded in gravels in this situation. Concerning the artificial origin of the fractures on the edges of some of this rude plateau-type of flint implements, which have been called ' Eolithic ' by some authors, doubt has, however, been expressed by many geologists and antiquaries. Representatives of tbe Pleistocene deposits in Britain and the adjoining portions of Western Europe. There is one very noteworthy distinction between the deposits of Pleis- tocene age and those which constitute the older geological systems. "While the latter are almost entirely represented by subaqueous accumulations, and have, indeed, for the most part been laid down on the sea-bottom and subsequently elevated, the Pleistocene formations are largely of terrestrial or at least of lacustrine or fluviatile origin, and comprise deposits that have not been washed away during the subsidence of the land, like nearly all similar accumulations of greater antiquity. We will proceed to consider the chief types of these Pleis- tocene deposits terrestrial, fluviatile, lacustrine, fluviomarine, marine, and glacial, as they are represented in this country and the adjoining parts of Europe. Among the terrestrial deposits of the Pleistocene we may call attention to the peat deposits which have yielded such valuable evidence concerning the events which took place in prehistoric times. These peat deposits have been especially studied in Den- mark, and many monuments of the early inhabitants of that country have been brought to light by the combined labours of the antiquary, the zoologist, and the botanist. The late geological age of these peat-mosses is demonstrated by the fact that not only the contemporaneous freshwater and land shells, but all the quadrupeds, found in the peat, agree specifically with those now inhabiting the same districts, or known to have been indigenous in Denmark within the memory of man. In the lower beds of peat (a deposit varying from 20 to 30 feet in thickness), weapons of stone accompany trunks of the Scotch fir, Pinus sylveslris, L. This peat may be referred to that part of the stone period known as ' Neolithic,' in contradistinction to a still older era, termed ' Palaeolithic.' In the higher portions of the same Danish bogs, bronze implements are associated with trunks and acorns of the common oak. It appears that the pine has never been a native of Denmark in historical times, and it seems to have given place to the oak about the time when articles and instruments of bronze super- seded those of stone. It also appears that, at a still later period, the oak itself became scarce, and was nearly supplanted by the beech, a tree which now flourishes luxuriantly in Denmark. Again, at the still later epoch when the beech-tree abounded, tools of iron were introduced, and were gradually substituted for those of bronze. CH. xii.] CAVERN-DEPOSITS 159 On the coasts of the Danish islands in the Baltic, certain mounds, called in those countries ' Kjokken-modding,' or ' kitchen- middens,' occur, consisting chiefly of the castaway shells of the oyster, cockle, periwinkle, and other eatable kinds of mollusks. The mounds are from 3 to 10 feet high, and from 100 to 1,000 feet in their longest diameter. They greatly resemble the heaps of shells formed by the Red Indians of North America along the eastern shores of the United States. In the old refuse-heaps, recently studied by the Danish antiquaries and naturalists with great skill and diligence, no implements of metal have ever been detected. All the knives, hatchets, and other tools are of stone, horn, bone, or wood. With them are often intermixed fragments of rude pottery, charcoal, and cinders, and the bones of quadrupeds on which the early people fed. These bones belong to wild species still living in Europe, though some of them, like the beaver, have long been ex- tirpated in Denmark. The only animal which they seem to have domesticated was the dog. As there is an entire absence of metallic tools, these refuse- heaps are referred to the Neolithic division of the age of stone, which immediately preceded in Denmark the age of bronze. It appears that a race more advanced in civilisation, armed with weapons of that mixed metal, invaded Scandinavia and ousted the aborigines. Cavern deposits containing: human remains and bones of extinct animals. In England, and in almost all countries where limestone rocks abound, caverns are found, usually consisting of cavities of large dimensions, connected with one another by low, narrow, and sometimes tortuous galleries or tunnels. These sub- terranean water-ways are usually filled in part with mud, pebbles, and breccia, in which bones may occur belonging to various animals. Some of these bones are referable to extinct and others to living species, and they are occasionally intermingled with implements of one or other of the great divisions of the stone age, and these are sometimes, though very rarely, accompanied by human bones. Each suite of caverns, and the passages by which they com- municate with one another, afford memorials to the geologist of successive phases through which they must have passed. First there was a period when the calcium carbonate was dissolved away gradually by drainage water containing carbon dioxide in solution ; secondly, an era when engulfed rivers or occasional floods swept organic and inorganic debris into the subterranean hollows thus formed ; and thirdly, a time when the formation of stalagmite took place on the floor, covering up the deposits. The quarrying away of large masses of Carboniferous and Devonian limestone, near Liege, in Belgium, has afforded the geologist magnificent sections of some of these caverns, and the former communication of cavities in the interior of the rocks with the old surface of the country, by means of vertical or oblique fissures, has been demonstrated in places where it would not otherwise have been suspected so completely have the upper extremities of these fissures been concealed by superficial drift, while their lower ends, which extended into the roofs of the caves, have been masked by stalactitic incrustations. The origin of the stalactite has been noticed (p. 24), and it may 160 KENT'S HOLE AND BRIXHAM CAVERNS [CH. xn. now be explained that it is when caverns have ceased to be in a line of active drainage, or to form underground conduits, that a solid floor of hard stalagmite is formed on the bottom. The late Dr. Schmerling examined forty caves near Liege, and found in all of them the remains of the same fauna, comprising the Mammoth, Tichorhine Rhinoceros, Cave-bear, Cave-hyaena, Cave- lion, Reindeer, and many others some of extinct and some of living species and also flint-implements. In four or five caves only, parts of human skeletons were met with, comprising sometimes skulls with a few other bones, sometimes nearly every part of the skeleton except the skull. In one of the caves, that of Engihoul, where Schmerling had found the remains of at least three human indivi- duals, they were mingled in such a manner with bones of extinct mammalia, as to leave no doubt in his mind of man having coexisted with them. The careful investigations carried on by Falconer, Pengelly, and others, in the Brixham cave and at Kent's Cavern, near Torquay, afforded evidence that flint knives were embedded in red earth underlying a floor of stalagmite, in such a manner as to prove that man had been an inhabitant of that region, when the Cave-bear and other members of the ancient Pleistocene fauna were also in existence. The following are the species which have been discovered in the English caves. Those which are extinct are Elephas primigenius, Blumenb., and E. antiquus, Falc., Rhinoceros tichorhinus, Cuv., B. leptorliinus, Ouv., Machairodus latidens, Ow., Ursus spelceus, Blu- menb.. Cervus megaceros, Hart., C. Brownii, Dawk., Bisonpriscus, Boj. The species still living in Africa are the Hippopotamus, Lion, and Hyaena. Antilope and Felis pardis, L. (Panther) are now Asiatic. Of species now living in North America we find the Grizzly Bear ; and of those occurring in N. Europe, the Elk, Reindeer, Lemming, and Glutton. Besides these, there are found many of the com- monest European species of mammalia. The absence of gnawed bones led Dr. Schmerling to infer that none of the Belgian caves which he explored had served as the dens of wild beasts ; but there are many caves in Germany and England which have certainly been so inhabited, especially by the extinct Hyaena and Bear. A fine example of a hyasna's den was afforded by the cave of Kirkdale, so well described by the late Dr. Buckland in his 'Reliquiae Diluvianae.' In that cave, about twenty-five miles NNE. of York, the remains of about 300 hyasnas, belonging to individuals of every age, were detected. The species (Hyana, spelcea, Goldf.) has been con- sidered by palaeontologists as extinct ; it was larger than the fierce Hyccna crocuta, Zimm., of South Africa, which it closely resembled, and of which it is regarded by Professor Boyd Dawkins as a variety. Dr. Buckland, after carefully examining the spot, proved that the hyaenas must have lived there ; a fact attested by the quantity of their dung, which, as in the case of the living hyama, is of nearly the same composition as bone, and almost as durable. In the cave were found the remains of the Ox, Mammoth, Hippopotamus, Rhinoceros, Horse, Bear, Wolf, Hare, Water rat, and several birds. All the bones have -the appearance of having been broken and gnawed by the teeth of the hyaenas ; and they occur confusedly CH. XII.] VALLEY GKAVELS 161 mixed in loam or mud, or dispersed through a crust ot stalagmite which covers it. In these and many other cases it is supposed that portions of herbivorous quadrupeds have been dragged into caverns by beasts of prey, and have served as their food an opinion quite consistent with the known habits of the living hyaena. Alluvial deposits of the Palaeolithic age. The alluvial deposits of the Palaeolithic age are the earliest in which any vestiges of man have yet been certainly detected, and they belong to a time Fig. 139. Section across the Valley of the Ouse, two miles WNW. of Bediu;d. 1. Oolitic strata. 2. Boulder clay, or marine northern drift, rising to about ninety feet above the Cuse, 3. Ancient gravel, with elephant bones, freshwater shells, and flint implements. 4. Modern alluvium of the Ouse. a. Biddenham gravel pit, at the bottom of which flint tools were found. when the physical geography of Europe differed in a marked degiee from that now prevailing. Since those deposits originated, changes of considerable magnitude have been effected in the depth and width of many valleys, as also in the direction of the superficial and sub- terranean drainage, and, as is manifest near the sea-coast, in the relative position of land and water. In the above diagram (tig. 139) is shown the relative position which the gravel, containing flint implements and the bones of extinct animals, bears to the older formations, out of which the valley has been formed. In fig. 140, a similar but ideal section is given, illustrating the different positions which the Pleistocene alluvial deposits occupy in many European valleys. The peat No. 1 (fig. 140) has been formed in a low part of the modern alluvial plain, in parts of which gravel No. 2 of the recent period is seen. Over this gravel the loam or fine sediment 2' has in Fig '40. Ideal section across a river vallev. many places been deposited by the river during floods which covered nearly the whole alluvial plain. No. 3 represents an older alluvium, composed of sand and gravel, 162 BRICK-EARTHS [CH. xn. formed before the valley had been excavated to its present depth. It contains the remains of fluviatile shells of living species associated with the bones of mammalia, in part of recent, and in part of extinct species. Among the latter, the Mammoth (Elcplias primigcnius, Blu- menb.) and the Hairy Rhinoceros (R. tichorhinus, Cuv.) are common. No. 3' is a remnant of the loam or brick earth by which No. 3 was overspread. No. 4 is a still older and more elevated terrace, similar in its composition to No. 3, and covered in like manner with its in- undation mud (4'). Sometimes some or all of the valley gravels of older date are missing. They usually occur at heights, above the present stream, varying from 10 to 300 feet, sometimes on the right, and sometimes on the left, side of and usually on exactly opposite sides of the valley. The upper deposit (5) is the gravel of the plateaux ; 4 is termed High-level, and 3 Low-level, gravel. Among the genera of quadrupeds most frequently met with in England, France, and Germany, the commonest remains in the high- and low-level river gravels (4 and 3) are, in England, the Mam- moth, Ancient Elephant (E.antiquus,Fsilc.), Hairy Rhinoceros, Lepto- rhine Rhinoceros, Horse, Boar, Great Hippopotamus, Bison, Primitive Ox (Bos primigenius, Boj.), Musk Ox, Reindeer, Irish Elk, Red Deer, Cave Lion, Cave Hyaena, Wolf, Grizzly Bear, and Otter. Some of these kinds of animals are extinct, others inhabit Africa and Asia, whilst some are only found within the Arctic circle. Two are N. American. A few kinds still exist on the area. In the peat (No. 1) and in the more modern gravel and silt (No. 2), works of art Fig. 141. Fig. 142. Fig. 143. Succinea oblonga, Pupa muscorum, Mull., Helix hupida, Hull., nat. size. Drap., uat. size. nat. size. of the ages of iron and bronze, and of the later or Neolithic stone period, already described, are met with. In the more ancient gravels (3 and 4), there have been found in several valleys in France and England as, for example, in those of the Seine and Somme, and of the Thames and Ouse, near Bedford stone implements of a rude type, termed ' Palaeolithic,' showing that man coexisted in those districts with the Mammoth and other extinct quadrupeds of the genera above enumerated. The loam or brick-earth of our English river-valleys presents many points of analogy with the ' loess ' of the Rhine and other European rivers. Although this loess of the Rhine is unsolidified, it usually ter- minates, where it has been undermined by running water, in a ver tical cliff, from the face of which shells of terrestrial, freshwater, and amphibious molluscs project in relief. These shells do not imply the permanent sojourn of a body of fresh water on the spot, for the most aquatic of them, the Succinea, inhabits marshes and wet grassy meadows. The Succinea o&Zon#rt, Drap. (fig. 141), is very characteristic both of the loess of the Rhine and of some other European river-loams. Among the land-shells of the Rhenish loess, Helix liispida, Mull., CH. xii.] LOESS 163 fig. 143, and Pupa muscorum, L., fig. 142, are very common. Both the terrestrial and aquatic shells are of most fragile and delicate struc- ture, and yet they are almost invariably perfect and uninjured. They must have been broken to pieces aad they been swept along by a violent inundation. Even the colour of some of the land-shells, as that of Helix nemoralis, Mull., is occasionally preserved. In parts of the valley of the Khine, between Bingen and Basle, the fluviatile loam or loess now under consideration is several hundred feet thick, and contains here and there throughout that thickness land and fresh water shells. As it occurs in masses fringing both sides of the great plain, and as, occasionally, remnants of it are found on eminences in the centre of the valley, and also forming hills several hundred feet in height, it seems necessary to suppose, first, a time when it slowly accumulated ; and secondly, a later period, when large portions of it were removed that is to say, when the original valley, which had been partially filled up with it, was re-exca- vated. The greatest altitude of the loess is at Fribourg (284 metres). Such changes may have been brought about by a great movement of oscillation, consisting of a general depression of the land, followed by a gradual re-elevation of the same. The amount of con- tinental depression which first took place in the interior must be imagined to have exceeded that of the region near the sea, in which case the higher part of the great valley would have its alluvial plain gradually raised by an accumulation of sediment, which would only cease when the subsidence of the land was at an end. If the direc- tion of the movement were then reversed, and, during there-elevation of the continent, the inland region nearest the mountains rose more rapidly than that near the coast, the river would acquire a denuding power sufficient to enable it to sweep away, gradually, much of the loess with which parts of its basin had been filled up. Terraces and hillocks of mud and sand would then alone remain to attest the various levels at which the river had thrown down and afterwards removed alluvial matter. High plateau gravels and loess. These are spread far and wide (see fig. 140, No. 5), and are sometimes very distinct in their character and at other times merge gradually into soil above and the parent rock below. In the first instance they often contain rock- fragments brought from a distance, and this and the circumstance that they cover up equally strata of many kinds are only explicable by the area of deposit having been at one time below slowly drifting water. It is in these deposits that the very rudely worked flints, described by Professor Prestwich, have been found. Inundation-mud of rivers. Brick-earth. Fluviatile loam, or loess. As a general rule, the fluviatile alluvia of different ages (Nos. 2, 3, 4, fig. 140, p. 161) are severally made up of coarse materials in their lower portions, and of fine silt or loam in their upper parts. For rivers are constantly shifting their position in the valley-plain, encroaching gradually on one bank, near which there is deep water, and deserting the other or opposite side, where the channel is growing shallower, being destined eventually to be con- verted into land. Where the current runs strongest, coarse gravel is swept along, and where its velocity is slackened, first sand, and then only the finest mud, is thrown down. A thin film of this fine sedi- ment is spread, during floods, over a wide area, on one, or sometimes LACrSTKlXfc JLSD ESITJLKISE BEPOSITS [cm. aJ^ of the Bum stream, oft as fcr as the hase the smiley. Of sA a is sui mcfy to of thoBSHafe of jwars ft hi f ^ - i ^ - - - -- - ^_ B0t BanBg Mea itacAut ay of not which OOCBPT the ales of oU 7 :i, L. I :: i- 1,- .:. ipih P m] MAEDTE DEPOSITS 165 vations. In other greater heights above the sea. Darwin traced thecn on the coast of Sooth America up to elevations of MOO feet above the sea-level, though human remains were only found in them op to a height of 85 feet, and shells up to a height of 6& ****-* It is only *m+mta^torlma**lmmy*tenmeri~*A&*,**l*t*fc printed out, the passage of atmospheric waters through masses of loose material upon the land wffl tend to the obliteration alike of the traces of organisms or articles of human workmanship. The beds of peat, often containing trunks of trees, which am found bdow the present sea-level, and are< low tides, are known as 'submerged forests.* Theyare often* by strata of shingle, sand, or sflt, and are only seen when latter have been seoored away by the action of the waves. They ,wifli yield many plant remains (leaves, seeds, and fruits), the wings and wing-cases of "***, sometimes preserved in ax On the other hand, we frequently find, as in the West In Florida, the Solomon Islands, Ac, great masses of coral rock deep-sea deposits, like the elevations up to and even eMeedmg 1,000 feet to make out a chronological tocene deposits, we are aided by the Europe and North America alike, we find dear evidence of a remarkable episode namely, the setting in for a time of a period of intense cold. According to die calculations of Pro- fessor Bonney, a lowering of the mean annual temperature of Western Europe and Eastern North Aiwrv^ by 16? to 18 F. would be required to produce the results of which we find The Glacial Epoch. Over a great part of Europe north of the 50th, and of North America north of the 39th parallel of ;of loose transported materials, lying indifferently upon all the aider formally and evidently up of fragments derived from th ** angular and rounded fragments of rock, of which large sixe. It is often wholly devoid of 50, 100, or even a greater number of stratified, especially in the higher parts of the series of dsoeeur with marine organisms. Tothe form of the deposit the name of All has long 1 whOt Hi I**' on ftvA 4*riw, cfap, n. 166 BOULDER CLAY [CH. xn. organic remains, except those washed into it from older formations, but in a few localities it contains marine shells, usually of northern or Arctic species, and nearly always in a very fragmentary state. The bulk of the till has usually been derived from the grinding down into mud of rocks in the immediate neighbourhood, so that it is red in a region of Bed Sandstone, as in Strathmore in Forf arshire ; grey or black in a district of coal and bituminous shale, as around Edinburgh ; and white in a chalk country, as in parts of Norfolk and Denmark. The stony fragments dispersed irregularly through the till usually belong, especially in mountainous countries, to rocks found in some part of the same hydrographical basin. But there are regions where the whole of the boulder clay has come from a distance, and huge blocks, or ' erratics,' as they have been called, many feet in diameter, have not unfrequently travelled scores or even hundreds of miles from their point of departure, or from the parent rocks from which they have evidently been detached, and have crossed over the water-partings between the valleys. These stones are commonly angular, and have often one or more of their sides polished and furrowed. The rock on which the boulder formation reposes, if it consists of granite, gneiss, limestone, or other hard material, capable of permanently retaining any superficial markings which may have been imprinted upon it, is usually smoothed or polished, like the erratics above described (see fig. 144). It exhibits parallel striae and furrows having a determinate direction. Such striae are found at great elevations, even up to 3,000 feet in the Scottish Highlands. The direction, both in Europe and North America, agrees generally in a marked manner with the course taken by the erratic blocks in the same district. Another form of glacial drift consists of beds of gravel and sand, which usually rest on the boulder clay when the two formations occur together. It is probable that the bulk of this drift had the same origin as the till and boulder clay, nut that subsequently the clay and sand have been washed out, and the stones and gravel spread out by currents of water are so worn that they rarely show scratches and polished surfaces. Like the boulder clay, this gravel rarely contains fossils, and when these do occur they are generally fragmentary and much waterworn. The boulder clay, when it was first studied, seemed in many of its characters so singular and anomalous, that geologists despaired of ever being able to interpret the phenomena by reference to causes now in action. In those exceptional cases, where marine shells of the same date as the boulder clay were found, nearly all of them were recognised as living species, but now flourishing in Arctic latitudes facts conspiring with the superficial position of the drift to indicate a comparatively modern origin and a great change in climate. The term ' diluvium ' was for a time the popular name of the boulder formation, because it was referred, by many, to the deluge of Noah, while others retained the name as expressive of their opinion that a series of diluvial waves raised by hurricanes and storms, or by earthquakes, or by the sudden upheaval of land from the bed of the sea had swept over the continents, carrying with them vast masses of mud and heavy stones, and forcing these stones over rocky surfaces so as to polish and imprint upon them long CH. xii.] GLACIAL STRIATION 167 furrows and striae. But geologists were not long in seeing that the boulder formation was characteristic of high latitudes, and that on the whole the size and number of erratic blocks increase as we travel* towards the Arctic regions. They could not fail to be struck with the contrast which the countries bordering the Baltic presented, when compared with those surrounding the Mediterranean. The multitude of travelled blocks and striated rocks in the one region, and the absence of such appearances in the other, were too obvious Fig. 144. Limestone, polished, furrowed, and scratched by the glacier of Rosenlaui in Switzerland. (Agassiz.) a a. White streaks or scratches, caused by small grains of flint frozen into the ice. b b. Furrows. to be overlooked. When the nature of glaciers, and their great, denuding and transporting powers were first studied, it was supposed that the terminal moraines, or the collections of mud and angular stones at the foot of the glacier, were analogous to boulder clay. This is not the case, but it is probable that where the glaciers terminate in the sea, as they do in the Arctic regions in many places, the moraine matter may assume many of the characters of boulder clay. During a recent North-Polar expedition, an unctuous clay was frequently found covering the sea floor near glaciers. Some authors maintain that a deposit similar to boulder clay may accumulate underneath glaciers, but of the existence of such a moraine profonde, as it has been called, no very satisfactory proofs have been adduced. Similar traces of glacial action to those now found in the higher mountain chains as smoothed and striated rock surfaces, often forming ' roches moutonnees,' heaps of moraine matter (sometimes damming up the ends of lakes), perched and erratic blocks are found in all the higher portions of the British Islands, North Wales, the 168 GLACIAL GKAVELS WITH MAKINE SHELLS [CH. xir. Lake District, the Scottish Highlands, and also on the more elevated ground in Ireland. Attention was first called to the occurrence of these traces of ancient glaciers in our own island by Agassiz in 1840 ; and since his time the labours of Darwin, Buckland, Kamsay, the brothers Geikie, and many others have greatly strengthened the conclusions which the great Swiss naturalist drew from their oc- currence as to the existence of a glacial epoch during Pleistocene times. While these relics of glaciers exist in the higher grounds, the lower lands of Britain, as far south as the valley of the Thames, exhibit great masses of boulder clay, sometimes alternating with sands, which are occasionally finely laminated aijd at other times much contorted (see fig. 11, p. 41) ; erratic blocks are found even as far south as the Thames Valley. Not unfrequently, in the glacial sands, and occasionally in the boulder clay itself, we find shells of marine mollusca usually belong- ing to species which now flourish in the Arctic seas. Such deposits of sand with Arctic shells have been found at Airdrie in Scotland up to elevations of 524 feet above the sea, at Three-Kock Mountain, and other localities in Dublin and Wexford, Ireland, up to 1,OOG or 1,200 feet above the sea ; at Mottram, east of Manchester, at 568 feet, at Vale Royal, near Macclesfield, between 1,100 and 1,200 feet, at Gloppa, Cheshire, at 1,200 feet, and at Moel Tryfaen in North Wales at over 1,300 feet above that level. By many geologists, the occurrence of these marine shells at such great elevations is accepted as evidence of great submergence during the Pleistocene period* Others, however, considering the very fragmentary condition and local occurrence of these shells, the fact that in the same band we find species that live at very different depths, and the other circumstances of their occurrence, maintain they may have been brought into their present elevated condition by the action of great glaciers or ice-sheets pushing up portions of the sea-floor, possibly in a frozen condition, to the hill slopes on which we now find them. Bridlington drift. The so-called crag at Bridlington, containing many mollusca of Arctic kinds, appears to be a portion of an old glacial deposit which has been torn up by stranding ice and em- bedded in boulder clay. Erratics near Chichester.The most southern memorial j of ice- action and of a Pleistocene fauna in Great Britain are found on the coast of Sussex. A marine deposit exposed between high and low tide occurs on both sides of the promontory called Selsea Bill, in which Mr. Godwin-Austen found thirty-eight species of shells, and the number has since been raised to a hundred and forty. These erratics and shells were probably brought by ice floating in the English Channel of those days. This assemblage is interesting because, while all the species are recent, they have on the whole a somewhat more southern aspect than those of the present British Channel. What renders this curious is the fact that the sandy loam in which they occur is over- lain by yellow clayey gravel with large erratic blocks which must have been drifted into their present position by ice when the climate had become much colder. These transported fragments of granite and greenstone, as well as of Devonian and Silurian rocks, may have from. th.e coast gf Normandy aiid Brittany, and are many of CH. xii.] LAKES IN GLACIATED DISTRICTS 169 them of such large size that we must suppose them to have been drifted into their present site by coast-ice. Connection of the predominance of lakes -with glacial action. Generally, wherethe winter cold is intense, as in Canada, Scandinavia, and Finland, even the plains and lowlands are thickly strewn with innumerable ponds and small lakes, together with some of larger size ; while in more temperate regions, such as Great Britain, Central and Southern Europe, the United States, and New Zealand, lake-districts occur in all such mountainous tracts as can be proved to have been glaciated in times comparatively modern, or since the geographical configuration of the surface bore a considerable resemblance to that now prevailing. In the same countries lakes abruptly cease beyond the glaciated regions. A large proportion of the smaller lakes are dammed up at their lower end by barriers of unstratified drift, having the exact character of the moraines of glaciers, and are termed by geologists ' morainic,' but a few of them are true rock- basins and would hold water even if all the loose drift now resting on their margins were removed. The late Sir Andrew Kamsay maintained that all the rock-basins now occupied by lakes, both great and small, were actually excavated by the glaciers which once occupied their beds. Most geologists, however, are now agreed that while the occupation of such rock-basins by ice for long periods of time may have retarded the work of filling them up by sediments, the rock-basins must themselves be regarded as due to differential movements in the earth's crust along lines of drainage. The study by Gilbert, Spencer, and others of the great lake-basins of the northern and western parts of the North American continent, and of the alterations which have been effected in the levels of the old beaches which surround them, has afforded remark- able confirmations of these views. One of the most serious objections to the exclusive origin by ice- erosion of wide and deep lake-basins arises from their capricious distribution ; thus, for example, in Piedmont, both to the eastward and westward of Turin, great lakes are wanting, although some of the largest extinct glaciers descending from Mont Blanc and Monte Eosa came down from the Alps, leaving their gigantic moraines in the low country. Here, therefore, we might have expected to find lakes of the first magnitude rivalling the contiguous Lago Maggiore in importance. A still more striking illustration of the same absence of lakes where large glaciers abound is said to be afforded by the Caucasus, whose loftiest peaks attain heights from 16,000 to 18,000 feet. The present glaciers of this mountain chain are equal or superior in dimensions to those of Switzerland, yet it is remarked by Mr. Freshfield that ' a total absence of lakes, on both sides of the chains, is the most marked feature. Not only are there no great 'subalpine sheets of water, like Como or Geneva, but mountain tarns, such as the Dauben See, on the Gemmi, or the Klonthal See, near Glarus, are equally wanting.' The Himalayas also are singularly free from lakes. Lakes contain a remarkable fauna ; the Crustacea have marine affinities, and in some lakes there are seals which cannot have passed in by the existing rivers. The great North American lakes have submerged canons on their floors. The grander lakes are old areas of denudation, depressed or raised above sea-level by earth move* 170 CLASSIFICATION OF PLEISTOCENE DEPOSITS [CH. xn. ments. They were not formed during the glacial epoch, but long before that period, though their forms may have been greatly modi- fied both during and since glacial times. The subdivision of the great system of strata constituting the Pleistocene period is facilitated, as we have seen, by the occurrence of the glacial episode. The epoch which preceded the coming in of the intense cold we call the j?re-glacial, and during this time we find many evidences of a gradual re- frigeration, which probably commenced before the end of the Pliocene. Some observers have thought not only that they could detect proofs that there was a gradual increment and de- crement of cold during the earlier and later stages of the glacial epoch, but that evidence of one or more intervals of marked amelioration in the severity of the climate can be traced in studying the deposits. These intervals of less intense cold, or, according to some observers, of moderately warm climate, have been called interglacial periods. The postglacial division of time was marked in its earlier portion by considerable rigour of climate, and possibly an abundant rainfall. During this epoch much of the soil may have remained frozen to a great depth, and the rivers appear to have been more swollen and torrential in their character. To this epoch the late Mr. A. Tylor proposed to apply the name of ' the pluvial period.' It probably corresponds approximately to the Charnplain period of North America. In favour of the use of the terms ' recent ' period and ' human ' period there is, as already pointed out, little to be said. But both archaeologists and geologists find it useful to dis- tinguish the several epochs at which implements of human work- manship, showing different stages in civilisation, were employed. Setting aside the very doubtful alleged discoveries of objects of human workmanship in Pliocene and Miocene formations, we have some very rude implements found in very high-level (plateau) gravels in the south of England. If the artificial origin of these be established beyond doubt by further research, they possibly constitute the oldest known relics of man or of tool-making animals upon the globe. The oldest undisputed types of implements, those figured on p. 156 as occurring in the valley-gravels and cavern-deposits are known as Palaeolithic, and the more carefully finished instruments occurring in France in association with remains of the reindeer are known as Newer Palaeolithic. The Older and Newer Palaeolithic are sometimes distinguished as the Age of the Mammoth and Reindeer respec- tively. The epoch at which more carefully finished, and often polished, stone implements were employed is known as Neo- CH. XIIL] AGE OF THE HUMAN KACE 171 lithic ; and then follow the Copper Age (of certain areas), the Bronze Age, and the Iron Age. This classification of epochs by the aid of the works of art which characterise them is of use to the archaeologist, however, rather than to the geologist. It is still uncertain if the so-called ' human ' period began after the Glacial, or overlaps, as some observers think, the Glacial and even the Pre-glacial. On the other hand, it is clear that the several ages of stone and metal implements belong to very distinct periods of time in different countries. Some savage races have not, even at the present day, advanced beyond their ' stone age ' (Note J, p. 603). For fuller details concerning the of Man,' Professor Prestwich's Pleistocene deposits, the student is papers on the river-gravels of the referred to Professor Dawkins's North of France and the South-east ' Cave and Cave Hunting,' to Pro- of England, and Sir John Evans's fessor James Geikie's ' Great Ice ' Ancient Stone Implements, Wea- Age'(8rded. 1895) and 'Pre-historic pons, and Ornaments of Great Europe,' and to Professor Wright's Britain.' Professor Prestwich's ' Es- ' Man and the Glacial Period.' On says on Controverted Questions in the earliest relics of the human race, Geology,' 1895, furnishes an account he may consult Lyell's ' Antiquity of the so-called ' Eolithic ' remains. CHAPTER XIII THE NEWER-TERTIARY STRATA (NEOGENE OR NEOCENE) Use of the Terms Miocene and Pliocene, Neogene and Neocene Mollusca of the Newer-Tertiary Strata Mammalia of the Newer Tertiaries The Newer-Tertiary Flora British Newer and Older Pliocene Strata Forest-bed of Cromer Chillesford and Aldeby beds Red Crag White or ' Coralline ' Crag Older Pliocene deposits of the North Downs and of St. Erth Relation of the Fauna of the Crag to that of the present day Proofs of denudation between the periods of deposi- tion of the British Older and Newer Tertiarios. Nomenclature and Classification of the Newer-Tertiary strata. The principle on which the original classification of the Tertiary strata was based has been explained in a previous chapter. Experience, however, has shown that, valuable as is this classification for many purposes, a grouping of the Lyellian subdivisions is necessary to form systems of strata comparable to the great divisions of the Mesozoic and Palaeozoic rocks. Alike in Eastern Europe and in Western North America, it has been found impracticable to separate the Pliocene and the Miocene as well-defined systems of strata; and hence they have been united as characterising one great period of the earth's history called by the Vienna geologists the ' Neogene,' and by the American geologists the ' Neocene,' A twofold division of the 172 SUBDIVISIONS OF THE NEWER TERTIARY [CH. xm. Tertiary strata in this country has long been in use, the names applied to the two portions being ' Older Tertiary ' and * Newer Tertiary ' respectively. We shall, therefore, in the present work treat the Tertiary strata which underlie the Pleistocene as forming two great systems the Newer Tertiary, of which the Pliocene and Miocene constitute the main divisions, and the Older Tertiary, including not only the Eocene of Lyell but also the strata overlying it, which are known as Oligocene, and those underlying it, which have been called the Paleocene. The Miocene, which is so well represented in the south-west of France (the Faluns of Touraine), is sometimes called the Falunian. In Switzerland the strata of this age are known as the Molasse. The Pliocene strata attain a great thickness and importance in Italy along the flanks of the Apennines, and are hence often called the Sub-apennine strata. In Eastern Europe the Miocene and Pliocene, as we have seen, form one great series, which is called the Neogene ; and the same is the case in the United States, where they are called Neocene. Of the two great divisions of the Newer-Tertiary strata, the Miocene and the Pliocene, only the latter has any representatives in the British Islands ; and the thin and scattered patches of shelly sand called ' crag ' occurring in East Anglia constitute a very insignificant representative of the Pliocene strata of Italy and Eastern Europe, which attain a thickness of many hundreds or even thousands of feet. In Belgium, too, we find a more com- plete representation of the Pliocene strata than in our own country. The Pliocene or ' Crag ' strata of East Anglia consist of the following members, beginning with the highest bed : The Forest-bed series, with many plant remains and bones and teeth of terrestrial mammalia. The Chillesford sands and clays with a Molluscan fauna, con- taining only a few extinct forms, but with a proportion of over 60 per cent, of these belonging to Northern types. The Norwich (or Fluvio-marine) Crag, containing an admixture of marine and freshwater shells and some Mammalian remains. Of the shells 93 per cent, are living forms, 14-6 per cent, being Northern types. The Red Crag, consisting of shelly sands with 93 per cent, of living forms, of which 1O7 per cent, are Northern types. The White (' Coralline ') 3 Crag, formed of sands with argillaceous bands containing Mollusca and Bryozoa. Of the former 54 per cent, are living forms, and only 5 per cent, are Northern species. 3 The term ' Coralline ' was which occur in such profuse abunh erroneously applied to the JJryozoa, dance in these CII. XIII.] NEWER-TERTIAKY MOLLUSCA 173 Characteristics of the fauna and flora of the Newer - Tertiary strata. While in the Pleistocene strata all the mollusca belong to living species, the Newer-Tertiary beds con- tain, side by side with those recent forms of life, many others which have never been found in the existing seas ; and such species must be presumed to have become extinct. In the younger (Pliocene) strata the proportion of these extinct species of shells is usually less than that of the living forms, while in the older (Miocene) beds the extinct forms are more numerous than the living ones. As examples of living forms of mollusca found in the Newer Tertiaries, we may cite the bivalves and univalves represented below (figs. 145-148). Fig. 145. Fig. 146. Nucula Cobboldice, Sow., nat. size. Fig. 147. Tellina obliqua, Sow., \ nat. size. Fig. 148. Trophon antiquum, Mull. (Fusus contrariiis, Sow.), J nat. size. Purpura tetragona, Sow., nat. size. It is noteworthy that the shell Trophon antiquum, Mull. (fig. 147), is in the existing seas almost always represented by ordinary or right-handed forms, while the opposite is found to be the case in the Neocene strata, where reversed or left-handed forms are much more common than the normal right-handed forms. As examples of Neocene shells, now extinct, we may take the following common forms (figs. 149-154). 174 EXTINCT AND LIVING FORMS [CH. xiu. Fig. 149. Asdtrte Omalii, Laj., nat. size ; species common to Upper and Lower Crag. Fig. 151. Fig. 150. Tig, 152. Valuta Lambertt, Sow., Mat. Variety characteristic of Fa- lunsof Touraiue. Miocene. Fig. 153. Valuta Lamberti, Sow., i nat. Variety characteristic of Suffolk Crag. Pliocene. Valuta Lamberti, Sow., young individual, Cor. and Red Crag. Fig. 154. Oliva flammulata., Lani. Mio-pliocene of Belgium, nat. size. a. front view ; 6. back view. Murex vayinatus, Jan. Miocene. CH. xiii.] OF NEWER-TERTIARY MOLLUSCA 175 The case of Valuta Lamberti, Sow., is interesting as showing that a species has sometimes undergone very marked varietal changes during the long periods of time covered by the Neozoic. The Neocene molluscaif studied in any particular area, like Britain, Italy, or the United States seem to indicate the existence of a climate somewhat warmer than that of the seas Fig. 155. Ptcten Jacobceus, L., J nat. of the area at the present day ; thus we find many tropical or sub-tropical forms like Cyprcea, Oliva, Murex, &c., in the strata of Britain, France, and Belgium. Nevertheless, there is often a marked relation between the Pliocene, and to a less extent the Miocene, shells in a particular area and those living in the neigh- bouring seas at the present day. Thus Pecten Jacobceus, L (fig. 155), which is abundant in the Pliocene (Sub-apennine) strata of Italy, is still found living in the Mediterranean. In the same way, we find some of the forms most abundant in the Newer Tertiaries of the Atlantic States of North America still living on the western shores of the Atlantic (see figs. 156, 157). Occasionally shells which were at first supposed to be extinct have been afterwards detected in the existing seas ; of this the shell Natica helicoides, Johnst. (fig. 158), is an example. 176 MAMMALIA OF THE [CH. XIII. Many forms of Corals, Echinodermata, &c., very similar to those of the existing seas, abounded during this period. Foraminifera are very numerous in some of the Miocene Pig. 15C. Fig. 157. Fulgwr canaliculatus, L., sp., \ nat. Maryland. Fig. 158. i Fusus quadricostatuS) Say, nat. Maryland. Fig. 159. Natica helicoides. Johnst., nat. Amphisteyina Hauerina, D'Orb. Upper Miocene strata, Vienna ; mag. 10 diains. deposits ; forms of NummuUna and of the closely allied Am- jyJiistegina still occurring, though in smaller numbers than in the Eocene. Among the Newer-Tertiary strata, freshwater and terrestrial deposits are less commonly preserved than in the case of the Pleistocene ; but they are so abundant and contain such well- preserved fossils that our knowledge of the plants, insects, and terrestrial vertebrates of the period is considerable. The land-mammalia of the Pliocene are very different from those of the Pleistocene and of the present day. Elephants appear towards the close of the period, but the great group of the Proboscidians is generally represented by the extinct Mastodon and Dinotherium, (see next page). CH. XIII.] NEWER-TERTIARY PERIOD 177 The Mastodons which occur in the Miocene pass up into the Pliocene, and in North America occur even in Pleistocene strata ; Fig. 160. Pig. 161. Mattndon arvernensis, Croiz. et Job., third milk molar, left side, upper jaw ; grinding surface, natural size. Norwich Crag, Postwick ; also found iu Eed Crag, see p. 187. the molar teeth of the Mastodon (fig. 160) are less specialised than those of the Elephant, and some of the forms had incisors (tusks) in both upper and lower jaws. The Dinotherium (fig. 1G1), with tusks in the lower jaw, had still less specialised molar teeth than the Mastodons and Elephants, and is found only in Newer-Tertiary strata (Note K, p. 604). Remarkable and less specialised ancestors of the horse are found in the Newer Tertiaries in the Hip- po therium or Hipparion of the Pliocene and the Anclntherium of the Miocene. And these were preceded by the still more gene- ralised type Orohippus (Hyraco- therium) of the Eocene (see fig. 162) (Note L, p. 604). Other Newer-Tertiary mam- mals appear similarly to represent ancestral forms of the rhino- ceros, hippopotamus, and camel. In the Sivalik Hills in the North of India, at Pikermi in Greece, and in the island of Samos, remarkable forms allied to the giraffes have been found (Helladotherium, Camelopardalis, Sivatherium (fig. 163), Sa- Dinotherium giganteum, Kaup. 178 ANCESTRY OF THE HORSE [CH. XIII motherium (fig. 164, &c.), as well as curious types which ap- pear to be links between the existing goats and antelopes. Illustrations of the Ancestry of the Horse (Equtis caballus, L.), after Marsh. la. Shows the highly specialised fore-foot of the horse, with a single fully developed digit, and others reduced to rudimentary ' splint-bones.' \b. An upper molar of the horse, with its complicated pattern of enamel. 2a. Fore-foot of Hipparion from the Pliocene, with lateral diaits more fully developed. 2b. Upper molar of Hipparion, showing simpler pattern of enamel. 3. Fore-foot of Anchitherium from the Miocene, with fuller development o~f three digits. 36. Upper molar of Anchitherium, with still simpler pattern of enamej. 4a. Fore-foot of Orohippus (Hyracotherium) from the Eocene, in which four digits are present. 46. Upper molar of Orohippus, with still simpler arrangement of the enamel. Deer, antelopes, oxen, sheep, and pigs are all represented in Newer-Tertiary times by many peculiar forms now extinct ; in- sectivores, rodents, and true apes are also found. The Carnivores are represented by types related to, but very different in many of the details of their structure from, the hyaenas, dogs, and bears of the present day. We thus see that the Newer Tertiaries include many forms of mammalia which are distinctly less specialised than those of the present day, but do not exhibit the striking generalised characters which we shall find to be so characteristic of the Eocene types (Note M, p. 604). The Newer-Tertiary flora is a very rich one, and has been carefully studied by Heer and other botanists at Oeningen, near Schaffhausen, in Switzerland, and other localities, where leaves, fruit, and even flowers are often extremely well preserved in the CH. xni.] AND OTHER LIVING MAMMALS 179 fine calcareous mud now indurated into a finely laminated stone. Many of the common European trees are represented such as Fig. 163. Sivatkerium giganteum, Falc. and Cautl. Skull with horns restored. From the Lower Pliocene, Sivalik Hills, India. ^ nat. size. Fig. 164. Samotherium Bois.rieri, Forsyth Major. Sk\ill and lower jaw. } nat. size. A giraffe-like ruminant from the Pliocene of Samoa. Ulmus (elm), Quercus (Oak), and Acer (maple). Of the latter, many species, and even varieties, can be recognised by their 180 THE FLORA OF THE [CH. xitt. leaves and fruit, the inflorescence being sometimes admirably preserved. With the forms of Acer (Maple) leaves of the Platanus (Plane) very similar to the Platanus occidentalis, L., of the North Pig. 165. Acer trilobatum, Ad. Brong., normal form. Hoor, Flora Tert. Helv., PI. 114, fig. 2. Size \ diani. (Part only of the long stalk of the original fossil specimen is here given.) Upper Miocene, Oeningen ; also found in the Oligocene of Switzerland. Fig. 166. Acer trttobatum, Ad. Brong. . Abnormal variety of leaf. Heer, PL 1 10, fig. 16. I. Flower and bracts, normal form. Heer, PL lll,fig. 21. c. Half a seed vessel. Heer, PI. Ill, fig. 5. American continent, have also been found at Oeningen. With the characteristic temperate forms of vegetation there also occur at Oeningen and many other places, where a Newer-Tertiary flora occurs, forms that resemble plants now characteristic of CH. XIII.] NEWER-TERTIARY PERIOD 181 more tropical climates, such as Cinnamomum, Oreodaphne, and Liquidambar. Monocotyledon ous plants, like Smilax, are also represented in the Newer-Tertiary floras with some leaves and fruits which Fig. 168. Platanus aceroides, Gbpp. Heer, PI. 88, figs. 5-8. ' Size nat. Upper Miocene, Oeniiigen. a. Leaf. 6. The core of a bundle of pericarps. c. Single fruit or pericarp, nat. size. Fig. 1G9. Cinnamomum polmyorphum, Ad. Brong. Upper and Lower Miocene. a. Leaf. 6. Flower, nat. size. Heer, PI. 93, fig. 28. c. Ripe fruit of Cinnamomum polymorphum, from Oeningen. Heer, PI. 94, fig. 14. d. Fruit of re- cent Cinnamomum Camphora of Japan. Heer, PI. 152, fig. 18. Fig. 170. Oreodaphne Heerii, Gaud. Liquidambar europceum, var. (rilobatum, A. Brong. ; some- Leaf, half nat. size. times 4-lobed and more commonly 5-lobed. fa. Leaf, half nat. size. c. Fruit, nat. size. b. Part of same, nat. size. d. Seed, do., Oeningen. have been referred by botanists to the Proteacese, an order now confined to Australia and South Africa. Conifers, like Seqiwia, 182 FLO WEES AND INSECTS PRESERVED [CH. xm. Taxodium, and Glyptostrobus (fig. 174), also occur in consider- able numbers with Ferns and still more lowly plants. In some finely divided sediments, like those of Oeningen,not only do we find leaves exhibiting their characteristic venation and sometimes with the fruits and even the flowers of the plant attached to them, as shown in the preceding figures, but insects, retaining all their peculiar markings, and even tbeir colours, are Fig. 17 Smilax sagittifera. Heer, PI. 30, fig. 7. Size, i diameter. a. Leaf. b. Flower magnified, one of the six petals wanting at d. Upper Miocene, Oeningen. c. Smilax obtusifolia. Heer, PI. 30, fig. 9, nat. size. Upper Miocene, Oeninge Fig. 172. Fruit of the supposed fossil and recent species of Hakea, a genus of Proteacefe. a. Leaf of fossil form. Hakea (?) salicina, Heer. Upper Miocene, Oeiiingen ; Heer PI. 97, tig. 29, diam. b. Impression of \voodj- fruit of same, showing thick stalk, diam. c. Seed of same, natural size. d. Fruit of living Australian species, Hakea saligna, R. Brown, \ diam. e. Seed of same, natural size. occasionally discovered. Fig. 173 illustrates an admirably pre- served specimen of one of the Hemiptera. British representatives of the Newer Tertiary system. It is in the counties of Norfolk, Suffolk, and Essex that we obtain our most valuable information respecting the British Pliocene strata. They have been termed * Crag,' from a provincial word which is applied to shelly sand in that district. Newer Pliocene. The old land surface upon which the glacial deposits collected was necessarily worn and much CH. xin.] IN NEWEK-TEBTIAKY DEPOSITS 183 denuded, and the result has been to destroy nearly every relic of the fauna and flora of the Pre-glacial or Upper Pliocene age in England. But on the eastern coast there are some remark- able deposits which underlie glacial beds, and one in particular at Cromer may be taken as the topmost member of the great formation which accumulated late in the Pliocene period, during which a gradual diminution of mean annual temperature took place culminating in the Glacial age. Fig. 173. Pig. 174. Harpactor maculipes, Heer. Upper Miocene, Oeningen. Glyptostrobus europ&us, Heer. Branch with ripe fruit. Heer, PI. 20, fig. 1. Upper Miocene, Oeningen. Cromer Forest-bed. Intervening between the glacial for- mations of Norfolk and the subjacent chalk lies what has been called the Cromer Forest-bed, near the base of a series of freshwater, estuarine, and marine formations. This buried forest has been traced from Cromer to near Kessingland, a distance of more than forty miles, being exposed at certain seasons between high and low water- mark. It is the remains of an old land- and estuarine deposit, containing the submerged stumps of trees, which appear to stand erect with their roots in the ancient soil. Associated with the stumps, and overlying them, are lignite beds, with land and fresh- water shells, of species still inhabiting England with two exceptions ; and the remains of the Water-lily, the Buckbean, and other plants that now live in marshes and ponds. Through the lignite and forest-bed are scattered cones of the Scotch and Spruce firs with the leaves of tbe white Water-lily, yellow Pond-lily, Buckthorn, Oak, and Hazel. The fauna is a very suggestive one, and should be compared with that of the river gravels and caves (pp. 159-162) of the Pleistocene age, and with that of the Pliocene of the Val d' Arno in Italy (p. 234). About fifty mammals, some Reptilia, Amphibia, Fish, and Birds, lived in the age of this pre-glacial deposit. The genera and species studied by 184 THE FOKEST-RED [CH. xm. Mr. E. T. Newton, of the Geological Survey, are Canis, Machairodus, Felis, Maries sylvaticus, Nils., Gulo luscus, L., Ursus spelceus, Blu- menb., U. fcrox, Geoff., Trichechus, Phoca, Equus caballus, L., E. Stenonis, Cocchi, Rhinoceros etruscus, Falc., R. megarhimis, Christol., Hippopotamus major, Nesti, Sus scrofa, L., Bos primigcnius, Boj., Caprovis, Cervus bovides, L., C.capreolus, L., C.elaphus,Li., C.mega- ceros, Hart., and nine other species of Deer, Antilope, Trogonthcrium, Castor europteus, Ow., Arvicola, Mus sylvestris, L., Talpa, Sorex, Myogale, Eleplias meridionalis, Nesti, E. antiquus, Falc., Balceno- ptera, Moiiodon, Delphinus, the common Snake and Viper, Toad, and Triton, the Pike, &c. It is doubtful if Elephas primigenius, Blu- menb., then existed. The forest-bed is evidently an old land surface, and whilst some geologists reduce it to a clay with rootlets in it, others insist that the stumps of trees found upon it lived and grew there. Mr. Searles Wood, jun., after a long study of the localities, believed that the forest-bed resting on the chalk near Cromer, and containing the important fauna just noticed, is of Crag age that is to say, is anterior to any glacial phenomena of importance. He considered that the Chillesford Clay has been worn into a valley at Kessing- land, and that the mammalian remains found there, associated with a clay containing rootlets, are newer than thojse of the Cromer forest-bed. Mr. C. Keid, of the Geological Survey, however, considers that all the tree-stumps are drifted specimens. He states that the deposit is covered by a freshwater, and this by a marine deposit This last contains Leda myalis, Couth., Trophon antiquum, Mull., Nucula Cob- boldice, Sow. The freshwater deposit has Unio, Paludina, Planorbis, Limnaa, Succinea, and Helix as genera, and Corbicula (Cyrena) fluminalis, Mull., and Paludestrina (Hydrobia} marginata, Michaud, which no longer live in the British area. Although the relative antiquity of the forest -bed to the overlying glacial till is clear, there is some difference of opinion as to its relation to the crag presently to be described. Chillesford and Aldeby beds.- At Chillesford, between Woodbridge and Aldborough, in Suffolk, and at Aldeby, near Beccles, in the same county, there occur stratified deposits which are com- posed of sands and laminated clays, with much mica, forming horizontal beds about twenty feet thick. In the upper part of the laminated clays at Chillesford a skeleton of a whale was found associated with casts of the characteristic shells, Nucula Cobboldice, Sow., Tellina obliqua, Sow., Astarte borealis, Chem. sp.,and Cyprina islandica, L. sp. The same shells occur in a perfect state in the lower part of the formation. Natica helicoides, Johnst. (fig. 158, p. 176), is an example of a species formerly known only as fossil, but which has now been found living in our seas. There are at Aldeby 70 species of mollusca, comprising the Chillesford species and some others. Of these about nine-tenths are recent. They are in a perfect state, and clearly indicate a cold climate, as two-thirds of them are now met with in Arctic regions. As a rule, the Lamellibranchiate molluscs have both valves united, and many of them, such as Mya arenaria, L., stand with the siphonal end upwards, as when in a living state. Tellina balthica, L., before mentioned (fig. 124, p. 149) as so characteristic of the glacial beds, in- C3. xiii.] THE NORWICH CKAG 185 eluding the drift of Bridlington, has not yet been found in deposits of Chillesford and Aldeby age, whether at Sudbourn, Easton Bavent, Horstead, Coltishall, Burgh, or where they overlie the Norwich Crag proper at Bramerton and Thorpe. Norwich or Fluvio-marine Crag 1 . The Norwich Crag is chiefly seen in the neighbourhood of Norwich, and consists of beds of incoherent sand, loam, and gravel, which are exposed to view on both banks of the Yare, as at Bramerton and Thorpe. As the beds contain a mixture of marine, land, and freshwater shells, with bones of fish and mammalia, it is clear that they have been accumu- lated at the bottom of the sea near the mouth of a river. The beds form patches rarely exceeding twenty feet in thickness, resting on chalk. At their junction with the chalk there invariably intervenes a bed called the ' Stone-bed,' composed of unrolled chalk flints, com- monly of large size, mingled with the remains of a land fauna, com- prising Mastodon arvernensis, Croiz. et Job., Elephas meridionalis, Nesti, Elephas antiquus, Falc., Hippopotamus major, Nesti, Rhino- ceros leptorhinus, Guv., Trogontherium Cuvieri, Fisch., and ex- tinct species of Deer and Horse. Eemains of the recent species of Otter and Beaver are found. The Mastodon, which is a species characteristic of the Pliocene strata of Italy and France, is the most abundant fossil, and one not found in the Cromer forest-bed just mentioned. When these flints, probably long exposed in the atmosphere, were submerged, they became covered with Barnacles, and the surface of the chalk was perforated by the Pholas crispata, L., each fossil shell still remaining at the bottom of its cylindrical cavity, now filled up with loose sand from the incumbent crag. This species of Pholas still exists, and drills the rocks between high and low water-mark on the British coast. The name of ' Fluvio-marine ' has often been given to this formation, as no less than twenty species of land and freshwater shells have been found in it. They are all of species which still exist ; at least, only one univalve, a Paludina, has any claim to be regarded as extinct. Of the marine shells, 111 in number, about 17 per cent, are extinct, according to the latest estimate given by Mr. Searles Wood in his Supplement to the Crag Mollusca ; but this percentage must be regarded only as provisional. Some of the Arctic shells, which form so large a proportion in the Chillesford and Aldeby beds, are more rare in the Norwich Crag, though many northern species such as Rhynchonella psittacea, Jer., Scalaria grcenlandica, Chemn., Astarte borealis, Chemn., Panopcea norvegica, Sow., and others still occur. The Nucula Cobboldics, Sow., and Tellina obliqua, Sow., are frequent in these beds, as are also Littorina littorea, L., Cardium edule, L., and Turrltella communis, Eisso, of our seas, proving the littoral origin of the beds. Red Crag*. Among the English Pliocene beds the next in antiquity is the Red Crag, wnich often rests immediately on the London clay, as in the county of Essex, illustrated in the diagram on the next page. In Suffolk it rarely exceeds twenty feet in thick ness, and sometimes overlies another Pliocene deposit called the Coralline Crag. It has yielded exclusive of 87 species regarded by Mr. Wood as derivative 248 species of mollusca, of which 92 per cent, are still living. Thus, apart from its order of superposition, its greater antiquity as a whole than the Norwich, and its still greater 186 THE RED CRAG [CH. xm. antiquity than the glacial beds already described, is proved by the increased difference of its fauna from that of our seas. It may also be observed that in most of the deposits of this Red Crag, the northern forms of the Norwich Crag, and of such glacial formations as Bridlington, are less numerous, while those having a more southern aspect begin to make their appearance. Both the quartzose sand, of which it chiefly consists, and the included shells, are most com- monly distinguished by a deep ferruginous or ochreous colour, whence its name. Many of the shells are littoral species. They are often rolled, sometimes comminuted, and the beds have the appearance of having been shifting sandbanks, like those now forming on the Doggerbank, in the sea, sixty miles east of the coast of Northumber- land. False-bedding, the result of currents, is frequently observable, the planes of the strata being sometimes directed towards one point of the compass, sometimes to the opposite, in beds immediately superposed. It has long been suspected that the different patches of Red Crag are not all of the same age, although their chronological relation cannot always be decided by superposition. Separate masses are characterised by shells specifically distinct or greatly varying in relative abundance, in a manner implying that the deposits con- taining them were separated by intervals of time. At Butley, Tunstall, Sudbourn, and in the Red Crag at Chillesford, the mollusca appear to assume their most modern aspect and indicate a colder climate than when the earliest deposits of the same period were formed. At Butley is found Nticula Cobboldice, Sow., so common in the Norwich and certain glacial beds, but unknown in the older parts Fig. 175. Red Crag. London Clay. Chalk. of the Red Crag. On the other hand, at Walton-on-the-Naze, in Essex, we seem to have an exhibition of the oldest phase of the Red Crag ; in which the percentage of extinct forms is almost as great as in the Coralline Crag, and where Purpura tetragona, Sow. sp. (fig. 148, p. 173), is very abundant. The Walton Crag also indicates a warme: climate, both by the absence of many characteristic Arctic shells that are common in newer portions of the Red Crag, and by a greater pro- portion of Mediterranean species. Valuta Lamberti, Sow., an extinct species, which seems to have flourished chiefly in the antecedent Coralline Crag period, is still represented here by individuals of every age (see figs. 151, 152, p. 174). The reversed Whelk (fig. 147, p. 173) is common at Walton, where the dextral form of that shell is unknown. Here also specimens of lamellibranchiate molluscs are sometimes found with both the valves united, showing that they belonged to this sea of the Upper Crag, and were not washed in from an older bed, such as the Coralline Crag ; had such been the case, the ligament would not have held together the valves, in strata so often showing signs of the boisterous action of the waves. Such specimens of united valves are, however, rare. Mr. Searles Wood, after a most assiduous search, only detected thirteen species in this perfect condition, and among these Mactra CH. xin.] THE WHITE CRAG 187 avails, Sow., alone is common. The true corals found in the Red Crag indicate a sea with a temperature higher than that of the present German Ocean. At and near the base of the Red Crag is a loose bed of brown nodules, first noticed by Professor Henslow as containing a large percentage of earthy phosphates. This bed of coprolites (as it is called, because they were originally supposed to be the faeces of animals) does not always occur at one level, but is generally in largest quantity at the junction of the Crag and the underlying formation. In thickness it usually varies from six to eighteen inches, and in some rare cases amounts to many feet. It has been much used in agriculture for manure, as not only the nodules, but many of the separate bones associated with them, are largely impregnated with calcium phosphate, of which there is sometimes as much as 60 per cent. They are not unfrequently covered with barnacles, showing that they were not formed as concretions in the stratum where they now lie buried, but had been previously consolidated. Amongst the remain? are those of Mastodon arvernensis, Croiz. et Job., Mastodon tapiroides, Cuv., Eleplias mcridionalis, Nesti, Rhinoceros Schleicr- maclieri, Kaup, Tapirus prisons, Kaup, Hipparion (a quadruped of the horse family), the antlers of a stag, Cervus anoceros, Kaup, Hycena antiqna, Lank., Felis pardoides, Ow., and a large portion of the skull of a marine animal of the genus Halitherium (Dugong), which was recognised by Sir W. Flower in the collection of the Rev. H. Canham, of Waldringfield, and were named by him II. Canhami. The tusks of a species of Walrus are also met with, together with the teeth of gigantic Sharks and the ear-bones and other portions of several species of Whales, Dolphins, and other Cetaceans. The phosphatic nodules often include fossil Crustacea and fishes from the Eocene London Clay. Organic remains also of the older Chalk and Lias have been found, showing how great must have been the denudation of previous formations during the Pliocene period. As the older White Crag, presently to be mentioned, contains similar phosphatic nodules near its base, those of the Red Crag may be partly derived from this and other sources, such as Miocene strata. White or Coralline Crag;. The lower or Coralline Crag is of very limited extent, ranging over an area about twenty miles in length, and three or four in breadth, between the rivers Stour and Aide, in Suffolk. It is generally calcareous and marly often amass of comminuted shells, and the remains of Bryozoa passing occa- sionally into a soft building-stone. At Sudbourn and Gedgrave, near Orford, this building-stone has been largely quarried. At some places in the neighbourhood the softer mass is divided by thin flags of hard limestone, and Bryozoa placed in the upright position in which they grew. From the abundance of these Molluscoida the lowest or White Crag obtained its popular name of 'Coralline Crag; ' but true corals, or Zoantharia, are very rare in this formation. The White Crag rarely, if ever, attains a thickness of thirty feet in any one section. Professor Prestwich, who has thrown more light than any other writer on the geology of the Crag, imagines that if the beds found at different localities were united in the probable order of their succession, they might exceed eighty feet in thickness; but since no continuous section of any such depth can be obtained, ppeculations as to the thickness of the whole deposit must be very 188 KELATIONS AND FAUNAS [CH. XIII. vague. A bed of phosphatic nodules, very similar to that before alluded to in the Bed Crag, with remains of mammalia, has been met with at the base of the formation at Sutton. Whenever the Red and Coralline Crag occur in the same district the Bed Crag lies uppermost ; and in some cases, as in the section Fig. 176. Fascicularia aurantium, Milne Edwards, J. Family, Tubuliporidce, of same author. Bryozoan of extinct genus, from the Coralline Crag, Suffolk. . Exterior. &. Vertical section of interior. c. Portion of exterior magnified. d. Portion of interior magnified, showing that it is made up of long, thin, straight tubes, united in conical bundles. represented in fig. 177, which was well exposed to view in 1839, it is clear that the older deposit or Coralline Crag b had suffered denuda- tion before the newer formation a was thrown down upon it. At D there was not only seen a distant cliff, eight or ten feet high, of Coralline Crag, running in a direction NE. and SW., against which the Bed Crag abuts with its horizontal layers, but this cliff occa- sionally overhangs. The rock composing it is drilled everywhere by Pholades, the holes which they perforated having been after- wards filled with sand, and covered over when the newer beds were thrown down. The older formation is shown by its fossils to have Ramsholt. . Red Crag. Section near Woodbridg'e, in Suffolk. b. Coralline Crag. c. London Clay. accumulated in a deeper sea, and contains very few of those littoral forms such as the Limpet (Patella), found in the Bed Crag. So great an amount of denudation could scarcely have taken place, in such incoherent materials, without some of the fossils of the inferior beds becoming mixed up with the overlying Bed Crag ; hence considerable difficulty must be occasionally experienced by the palaeontologist in deciding to which bed the species originally be^ longed. CH. XIII. OF THE CKAG DEPOSITS 189 Mr. Searles Wood estimated the total number of marine shell- bearing mollusca of the Coralline Crag at 316, of which 84 per cent, are known as living. No less than 130 species of Bryozoa have been found in the Coralline Crag, some belonging to genera believed to be now extinct, and of a very peculiar structure ; as, for example, that represented in rig. 176, which is one of several species having a Fig. 178. Fig 179. Fig. 180. Linrjula, Dumorticri, Nyst, nat. size. Suf- folk and Antwerp Crag. Purula reticulata, Lam. Coralline Crag, Rams- holt, nat. size. Temnechinus excavatus, Forbes ; Temnopleurus-excavatus, Wood ; nat. size. Cor. Crag, Ramsholt. globular form. Among the Mollusca the genus Astarte (see fig. 149, p. 174) is largely represented, no less than fourteen species being known, many of them being rich in individuals. There is an absence of genera peculiar to hot climates, such as Conus, Oliva, Fasciolaria, Crassatella, and others. The absence also of large cowries (Cyprccti) is remarkable, those found belonging exclusively to the section Trivia. The large Volute, called Valuta Lamberti, Sow. (see fig. 151, p. 174), may seem an exception ; but it differs in form from the Volutes of the torrid /one, and its nearest living ally, Valuta Junonia, Chemn., has been dredged up in the Gulf Stream in extra- tropical latitudes. The occurrence of a species of Lingula at Sutton (see fig. 178) is worthy of remark, as this genus of Bracliiopoda is now confined to more equatorial latitudes ; and the same may be said still more decidedly of a species of Pyrula, supposed by Mr. Wood to be identical with P. reticulata, Lam. (fig. 179), now living in the Indian Ocean. A genus also of echinoderms, called by Professor Forbes Temnechi- nus (fig. 180), is represented in the Bed and Coralline Crag of Suffolk. Its nearest analogue is in the warm eastern seas of Burma and of the Western Pacific Islands. Older Pliocene Deposits of the South of England. The coprolitic beds at the base of the Ked and White crags not unfre- quently contain waterworn fragments of sandstone, which sometimes include casts of shells. These sandstone-fragments are known as * box-stones,' and are the only relics in this country of an older Pliocene formation found in Belgium and known as the ' Diestien,' which overlies the ' black crag ' (see p, 228). At Paddlesworth and a number of other localities along the North Downs there are sandpipes in the chalk, into which portions of the Pliocene strata which once covered the Cretaceous beds have been let down and preserved. They have yielded to Professor Prest- wich, and subsequently to the officers of the Geological Survey, a number of casts of shells, which have put their Pliocene age beyond 190 CLIMATE OF THE PLIOCENE [CH. xiri. question. They probably belong to the oldest Pliocene the Diestien of Antwerp. Lastly, at St. Erth's in Cornwall, there is a small patch of marine clay which has yielded a great number of marine shells and foraminifera. These also belong to species characteristic of the oldest Pliocene. Climate of tbe Crag- Deposits. One of the most interesting conclusions deduced from a careful comparison of the shells of the British Pliocene strata and the fauna of our present seas was pointed out by Professor E. Forbes. It appears that during the Glacial period, an epoch intermediate, as we have seen, between that of the Crag and our own time, many shells, previously established in the temperate zone, retreated southwards to avoid an uncongenial climate, and they have been found fossil in the Newer Pliocene strata of Sicily, Southern Italy, and the Grecian Archipelago, where they may have experienced, during the era of floating icebergs, a climate resembling that now prevailing in higher European lati- tudes. Forbes gave a list of fifty shells which inhabited the British seas while the Coralline- and Red-Crag were forming, and which, though now living in our seas, were wanting, as far as was then known, in the glacial deposits. Some few of these species have subsequently been found in the glacial drift, but the general con- clusion of Forbes remains unshaken. This view was ably supported by Mr. Searles Wood in the concluding remarks of his Supplement to the Crag Mollusca, where he pointed out how the geographical changes produced by that sinking down of land which accompanied the Glacial period may have altered the coast line, shutting out a former connection with the Mediterranean and opening for a time a new one with the Scandinavian seas. The transport of blocks by ice, when the Eed Crag was being de- posited, appears to be evident from the huge size of some irregular, quite unrounded chalk flints, retaining their white coating, and 2 feet long by 18 inches broad, in beds worked for phosphatic nodules at Foxhall, four miles south-east of Ipswich. These must have been tranquilly drifted to the spot by floating ice. Mr. Prestwich also mentions the occurrence of a large block of porphyry at the base of the Coralline Crag at Sutton, which would imply that the ice-action had begun in our seas even in this older period. The me?n annual temperature gradually diminished from the time of the Coralline to that of the Norwich Crag, and the climate became more and more severe, not perhaps without some oscillations of temperature, until it reached its maximum in the Glacial period. Relation ot the Fauna of tbe Crag: to tbat of tbe recent Seas. By far the greater number of the marine species occurring in the several Crag formations are still inhabitants of the British seas ; but even these differ considerably in their relative abundance, some of the commonest of the Crag shells being now extremely scarce as, for example. Buccinum Dalei, Sow. while others, rarely met with in a fossil state, are now very common, as Mnrcx crinaccus,'L.i and Cardium echinatum, L. Some of the species also, the identity of which with living forms would not be disputed by any conchologist are nevertheless distinguishable as varieties, whether by slight deviations in form or a difference in average dimensions. Since Mr. Searles Wood first described the marine mollusca of the Crags, CH. xiv.] THE MIOCENE 191 the additions made to that fossil fauna have been considerable, but those made in the same period to our knowledge of the living mollusca of the British and Arctic seas and of the Mediterranean have been much greater. By this means the naturalist has been enabled to identify with existing species many forms previously supposed to be extinct. The recent careful deep-sea dredgings of the ' Challenger ' and other expeditions have led to the discovery of some fe\v Mediterranean species of shells as still living in the abysmal depths of the ocean, which were formerly regarded as extinct members of the Coralline-Crag fauna. But in spite of this resusci- tation, as it might be called, of a few fossil forms, geologists find that they scarcely produce any appreciable difference in the percentage before arrived at of forms unknown as living. Such gene- ralisations must, however, always depend on the limits assigned by different naturalists to the terms ' species ' and ' variety.' Of the strata of Miocene age, the next older division of the Tertiaries, we have no representatives whatever in this country. Between the period of the deposition of the Eocene and that of the Pliocene great movements of the land and extensive denudation must have taken place, for the small patches of Pliocene in all cases lie unconformably upon the Eocene and older rocks, while the so-called ' coprolite-beds ' and ' stone-beds ' at their base contain many water- worn fragments, evidently derived from the Eocene and older strata. A full discussion of the ques- ' Supplement to the Monograph of tions connected with the age and Crag Mollusca,' Palaeontographi- relationships of the various Plio- cal Society; and in the following eene deposits in this country will Memoirs of the Geological Survey, be found in Prestwich's Memoirs 'The Geology of Norwich,' by H.B. on 'The Structure of the Crag-beds Woodward, 'The Geology of Ips- of Suffolk and Norfolk,' ' Quart. wich,' &c., by W, Whitaker, and Journ. Geol. Soc.,' vol. xxvii. (1871), ' The Pliocene Deposits of Britain,' pp. 115, 325, 453 ; in Searles Wood's by C. Reid. CHAPTEE XIV THE OLDER TERTIARY (EOGENE OR EOCENE) Geographical Distribution of the Older-Tertiary Strata The London and Hampshire basins Foraminifera, corals, cchinodermata, and crusta- ceans rf the Older Tertiaries The Older-Tertiary Mollusca The fish, reptile*, birds, and mammals of the period The Older Tertiary flora. The British Older-Tertiary Strata. Hempstead Beds The Bembridge Series The Headon Series The Brockenhurst Marine Group The Barton Sands and Clay The Bracklesham Series The Bournemouth Beds The Plant-beds of Bovey Tracey and Mull The London Clay The Oldhaven beds and Woolwich and Reading Series The Thanet Sands. Nomenclature and Classification of the Older-Tertiary strata. Under the name of Older Tertiaries we include not only the Eocene proper of English, French, and German authors, but the strata above them, called by Beyrich Oligocene 192 DISTRIBUTION OF EOCENE STRATA [CH, (and in part included by Lyell in his Lower Miocene), and those which lie beneath the Eocene as originally denned, which are sometimes called Paleocene. The distribution of these strata in England and the adjoining parts of the continent of Europe is illustrated in the accompanying sketch-map. From this sketch-map it will be seen that the British Lower Tertiaries, with the exception of several small and outlying patches to be hereafter more particularly described, are confined to the south-east of England, where they occupy two areas known as the London and Hampshire basins respectively. Other similar areas of Older-Tertiary strata occur in Belgium and Northern France (the Paris basin), with some small scattered Pig. 181. Map of the principal Eocene areas of North-Western Enrope. Hypogene rooRs and strata older than the Devonian. Eocene areas denoted by oblique lines. N.B. The space left blank is occupied by fossiliferous formations from tr* Devonian to the chalk inclusive. outlying patches in Brittany. The correlation of the English Lower Tertiaries with those of Belgium and France is often a matter of great doubt and difficulty, notwithstanding their geographical proximity. This arises from one or other of the following circumstances : the former prevalence of marine conditions in one basin simultaneously with fluviatile or lacus- trine in the other, or the existence of land in one area causing a break or absence of all records during a period when deposits may have been in progress in the other basin. As bearing on this subject, it may be stated that we have unquestionable evidence of oscillations of level which are shown by the super- position of salt or brackish-water strata on fluviatile beds ; and THE LONDON BASIN 193 those of deep-sea origin on strata formed in shal- low water. Even if the upward and downward movements were uniform in amount and direction, which is very improbable, their effect in producing the conversion of sea into land, or land into sea, would be different accord- ing to the previous shape and varying elevation of the land and bottom of the sea. Lastly, denuda- tion, marine and subaerial, has frequently caused the absence of deposits in one basin, of corresponding age to those in the other; and this destructive agency has been more than ordinarily effective on account of the loose and unconsolidated nature of the sands and clays. Even in the case of the London and Hamp- shire basins (which were once united and are now separated by an anticlinal fold of the cretaceous rocks along which denudation of the tertiaries has taken place), it is often difficult to determine the exact equivalent of the strata in the two areas. The series is much more complete in the Hampshire basin than it is in the London basin, and the general order of succession in both areas is shown in the following table : 194 FORAMINIFERA AND CORALS [CH. XIV, MIDDLE OLIGOCENE LOWER OLIGOCENE UPPER EOCENE LOWER EOCENE HAMPSHIRE BASTX Hempstead Series (marine) ' Bembridge Series (estuarine ) Brockenhurst Series (marine) , Headon Series (estuarine) [Barton Sands (marine) Barton clay (marine) Bracklesham Series (marine) f Bournemouth Beds (estuarine) Bognor Beds (marine) ( Plastic clay (estuarine) LONDON BASIN Bagshot Beds (estuarine) London clay (marine) Woolwich and Reading Beds (estuarine) Thanet sands (marine) The Lower London Terfciaries include the Woolwich and Reading beds (fluvio-marine), the pebble beds (Oldhaven series), into which they locally pass upwards, and the Thanet sands (marine), which underlie them in parts of Kent and Surrey. The general relations of the Older Tertiaries to the under- lying rocks is shown in the accompanying section (fig. 182, p. 193). There is a great unconformity between the Tertiary and the Secondary Strata, and another between the Mesozoic and the Palaeozoic. Characteristics of the Older-Tertiary fauna and flora. Corals occur in considerable numbers in the Brockenhurst beds Fig. 183. Nummulites Puschi, D'Archiac, . Peyrehorade, Pyrenees. a. External surface of one of the nummulites, of which longitudinal sections are seen in the limestone. 6. Transverse section of same. of Hampshire (fig. 187), and reef-building forms abound in the Alpine Eocene strata. Among the Echinoderraata the great prevalence of bilaterally symmetrical types (Irregulares), which had already become common in the Cretaceous rocks, is very noteworthy. While the great majority of the species of mollusca in the Older-Tertiary strata are extinct, they nearly all belong to genera which still live in the existing seas. As a general rule, however, the genera represented in the Older Tertiaries of this country and CH. XIV.] OF THE OLDER TERTIARIES 195 of Western Europe are such as are now found most abundantly in subtropical or even tropical seas. We are well acquainted not only with the marine form of life of the period but also with the brackish-water and freshwater types, and even with the numerous terrestrial mollusca, the shells of which are found enclosed for the most part in beds of tufaceous limestone, like those of Bembridge and Headon, in the Isle of Wight. The Rotulia armatn, Peneroplis cylindraceus, Miliolina seminulum D'Orb. sp. Lam. sp. L. sp. a. Natural size. b. Magnified. general characteristics of the Oligocene and Eocene Mollusca will be understood from the figures given to illustrate the cha- racteristic fossils of the several divisions of the strata. The foraminifera of the Older Tertiaries are remarkable for the great development * of num- nmlites, which were often of large size, and occurred in such pro- digious numbers that many beds of limestone are almost made up of them. In Britain and Western Europe nummulites oc- cur in comparatively small num- bers, but in the Alpine regions, and in Asia and North Africa, the rocks of this age are so crowded with them that the Eocene of these regions is often spoken of as the ' Nummulitic Formation.' Other beds of lime- stone, of Older Tertiary age, are found to be made up of Orbi- toides or Miliolina (fig. 186), and many forms of Eotalia (fig. 184), Alveolina, Calcarina, Peneroplis (fig. 185), and other genera also occur. Among the Crustaceans of the period, the predominance of short-tailed or Crab-like forms (Brachyura) of the Decapods over the long-tailed or Lobster-like types (Macroura) becomes very marked. o 2 Solenastrcea cellulosa, Dune., nat. size. Brocken hurst. 196 OLDER-TERTIAKY FISH AND BIRDS [OH. xiv. The fish are represented by great numbers of sharks, of which the teeth, often of considerable size, are the only relics which remain (see p. 211). The ordinary bony fish (Teleostei), which appeared in considerable numbers in the Cretaceous, become much more numerous in the Older Tertiaries, while the Ganoids have almost wholly disappeared. Of the inhabitants of the land, during the Older-Tertiary period, we have numerous and interesting remains. Among reptiles we find lizards, tortoises and turtles, and crocodiles, all represented in the Older Tertiaries of the British Islands ; and the serpents (Ophidia) now make their first appearance (see p. 209). The few birds found do not offer very noteworthy points of distinction from living forms ; they do not belong to the remarkable synthetic types found in the Mesozoic rocks. One form. Odontopten/x (fig. 188), found in the London clay, Fig. 188. Odontopteryx foliapicus, Owen. Skull and beak restored. The mandibles are ser- rated, but there are no teeth in sockets as in the birds of the Cretaceous and Jurassic rocks. From the London Clay, I. of Sheppey. has tooth-like serrations on both jaws, like some Chelonians, but these are very different from the distinct teeth implanted in sockets found, as we shall hereafter see, in the birds of the Cretaceous and Jurassic periods. It is in the mammalian fauna of the Older Tertiaries that we meet with the most remarkable assemblage of extinct forms. Unlike the mollusca and other lower groups of animals, the mammals of the period exhibit the widest divergence from existing types. It is generally supposed that all the Mesozoic mammals were Aplacental (Monotremes and Marsupials), and these Aplacental forms, now confined to the Australian and Ame- rican continents, certainly existed in Europe during the Older- Tertiary period. But with these Aplacental mammals we have in the Older-Tertiary strata great numbers of the higher or Pla- cental mammals, nearly all of which were remarkable synthetic types that is, they combine many peculiarities which are now CH. XIV.] OLDER-TERTIARY MAMMALS 197 found only in distinct groups. Among the Perissodactyla or Ungulates with an odd number of toes we find the tapir-like forms known as Palceotherium and Lophiodon. The Artio- dactyla or Ungulates with an even number of toes are re- presented by many forms, such as Xiphodon (see fig. 189), Anoplotherium, Anthracotherium, Hyopotamus, &c. In the Older Tertiaries of the Western Territories of North America a remarkable assemblage of mammals has been made known to us by the labours of Leidy, Marsh, and Cope. These seem to unite many of the characters of the Ungulates and the Proboscidians. They are all remarkable for the small size of their brain-cavities, and some of them bore several pairs of horns. Among these remarkable forms may be mentioned Phenacodus, Dinoceras, Coryphodon, Brontotherium, Uinta- Fig. 189. Xiphodon gracilis, Cuvier. Restored outline. therium, &c. Some of them attained to an enormous size, A restoration of one of these remarkable gigantic mammals is shown on the following page (fig. 190). The Carnivora of the Older Tertiaries were as different from those of the present day as were the Ungulates. Synthetic types resembling in some respects the hyaenas and foxes have been referred to the genera Hycenodon and Protoviverra ; while bear- like forms have been called Amphicyon, Cynodon, &c. With these are other forms which osteologists find a difficulty in referring to any of the orders of living mammalia, so remarkably do we find united in their structures characters now confined to distinct groups of animals. The names given to many of these animals are intended to indicate their curiously blended characters. 198 OLDER-TERTIARY CETACEA [cu. xiv. Lemurs are known in the Older Tertiaries, but true monkeys do not make their appearance till the succeeding period. A number of forms of Cetacea are lound in the Eocene, some corresponding in the main features of their structure with the whales of the present day, but with these we find the remarkable toothed forms known as Zeuglodonts (fig. 191). The Zeuglodonts are much more abundant in North America than in Europe (see Notes K, L, M, p. 604). The Older-Tertiary flora shows an even closer agreement in its general characters with that of the present day than does CH.. XIV.] OLDER-TERTIARY STRATA 199 the flora of the Cretaceous rocks. There are many subtropical and tropical forms of ferns like Lastrcea (fig. 192), and Conifers, Fig. 191. Molar tooth, natural size. Zeuglodon cetoiiiex, 0\v. Sasilosatirus, Harlan. Vertebra, reduced. among the latter of which we may mention the Sequoias, now confined to the Eocky Mountains, but in Tertiary times very widely distributed from the Arctic to the Equa- torial zones. Palms like Sabal (fig. 193), Cliamce- rops, Phoenix, and Fla- bellaria abounded in Northern Europe, and even extended into the Arctic regions. The fruits of a palm closely resembling the Indian Nipa (Nipadites] abound in our London Clay. Among the Conifers, the Sequoias, which became so abundant in Newer Tertiary times, and now appear to be on the point of extinction, are re- presented by the widely distributed Sequoia Langsdorfii, Ad. Brong. (fig. 194). Proteaceae, now found chiefly in Australia and South Africa, are thought by many botanists to be repre- sented among the leaves and fruits found in Older-Tertiary Deposits of Europe (see fig. 195), but the correctness of these Lastrcea stiriaca, Ting. Natural size. Oligocene and Miocene, Switzerland. . Specimen from Monod, showing the position of the sori on the middle of the tertiary nerves. b. More common appearance, where the sori remain and the nerves are obliterated. 200 OLDER-TERTIARY FLORAS [CH. XIV. identifications has been doubted by other authorities. The chief distinction between the European Older-Tertiary flora and that of the present day is found in the prevalence of apetalous plants Fig. 193. Fig. 194. Sdbal major, TJnger sp. Veray, Oligocene. (Heer, PL 41.) Sequoia Langsdorfii, Ad. Brong., $ natural size. Bivaz, near Lausanne. Oligocene, Miocene, and Lower Pliocene, Val d' Arno. a. Branch with leaves, b. Young cone. Fig. 196. Fig. 195. a. Fruit of a fossil Banksia. b. Leaf of Banksia (?) Deickieana, Hr. Miocene of Switzerland. Cinnamomum Rossmassleri, Heer. Daphnogene cinnamomifolia, Un- ger. Oligocene and Miocene, Switzerland and G ermany. and the remarkable admixture of tropical forms like Cinna- momum (fig. 196), Ar alia, Ficus, Laurus, Magnolia, &c., with the plants still living in Northern Europe, like Acer, Platanus, Quercns, Ulmus, Carpinus, Populup. Salix, $ c. OH. xiv.] INSECTS OF OLDER-TERTIARY STRATA 201 The remains of insects are sometimes found in association with those of plants in the Older Tertiaries, as we have seen to be the case in the Newer Tertiaries. The Brown Coal of Radaboj in Croatia has been shown by Unger to contain more than two hundred species of plants, with a very rich insect fauna, including no less than ten species of Termites or White Ants, some of gigantic size, large dragon-flies with speckled wings, and also grasshoppers of considerable size. Even the Fig. 197. Mylothrites (Vanessa) Pluto, Heer, nat. size. Oligocene, Radaboj, Croatia. Lepidoptera (butterflies and moths) are not unrepresented, and in one instance a butterfly has been found in which the pattern on the wing has escaped obliteration, and has been faithfully transmitted to us. Arctic Eocene Flora. A rich terrestrial flora flourished in the Arctic regions in the Older- Tertiary period, many species of which are common to strata of the same age in North- West Europe. Professor Heer has examined the various collections of fossil plants that have been obtained in N. Greenland (lat. 70) , Iceland, Spitz- bergen, and other parts of the Arctic regions, and has determined that they indicate a temperate climate. Including the collections brought from Greenland later by Mr. Whymper, this Arctic flora now comprises 353 species, and that of Greenland 169 species, of which 69, or nearly two-fifths, were supposed to be identical with plants found in the Lower-Tertiary beds of Central Europe. Considerably more than half the number are trees, which is the more remarkable since at the present day trees do not exist in any part of Greenland even 10 farther south. More than 50 species of Coniferae have been found, including several Sequoias (allied to the gigantic Wellingtonia of California), with species of Thujopsis and Salis- buria (Gingko), genera now found in Japan. There are also beeches, oaks, planes, poplars, maples, walnuts, limes, and even a Mag- nolia, two fruits of which have recently been obtained, proving that this splendid tree not only lived but ripened its fruit within the Arctic circle. Many of the limes, planes, and oaks were large-leaved species, and both flowers and fruit, besides immense quantities of leaves, are in many cases preserved. Among the shrubs were many ever- 202 STRATA OF THE HAMPSHIRE BASIN [CH. xiv. greens, as Andromeda, and two extinct genera, Daphnogene and M'Clintockia, with fine leathery leaves, together with hazel, black- thorn, holly, logwood, and haw- thorn. Potamogeton, Sparganium, and Menyanthes grew in the swamps, while ivy and vines twined around the forest trees, and broad- leaved ferns grew beneath their shade. Even in Spitzbergen, as far north as lat. 78 56', no less than 179 species of fossil plants have been obtained, including Taxodium of two species, hazel, poplar, alder, beech, plane-tree, and lime. Such a vigorous growth of trees within 12 of the Pole, where now a dwarf willow and a few herbaceous plants form the only vegetation, and where the ground is covered with almost perpetual snow and ice, is truly remarkable. Professor Heer believes that the temperature of North Greenland must have been at least 80 higher than at present, while an addition of 10 to the mean temperature of Central Europe would probably be as much as was required. The chief locality where this wonderful flora is preserved is at Atanekerdluk in North Greenland (lat. 70), on a hill at an elevation of about 1,200 feet above the sea. There is here a considerable succession of sedi- mentary strata pierced by volcanic rocks. Fossil plants occur in all the beds ; and the erect trunks as thick as a man's body, which are sometimes found, together with the abundance of specimens of flowers and fruit in good preserva- tion, sufficiently prove that the plants grew where they are now found. At Disco Island and other localities on the same part of the coast, good tertiary coal is abundant, interstratified with beds of sandstone, in some of which fossil plants have also been found, similar to those at Atanekerdluk. A rather different flora was found under glacial marine drift, 1,000 feet above the present sea- level of Robeson Channel, N. lat. 81 45', long. W. 64 45'. Twenty- six species were noticed, and eighteen had been found in the Older Tertiary deposits of Spitz- bergen and Greenland. The Conifers?, with Taxodium dis- tich inn, Rich., are abundant, this last being found in a state of bloom. Pinus abies, Heer, occurred, whose extreme limit is now N. lat. 69 30', but it spreads over 25 degrees -of latitude. It was only Arctic in the Older- Tertiary times. Large reeds, poplar, birch, hazel, elm, and water- lily occurred ; but the large-leaved plants like Magnolia were not discovered. The similarity of these Tertiary Arc*tic floras to those of the Eocene of North America and of Bournemouth, Mull, and Antrim, has led to their being placed in the Older Tertiary series rather than in the Miocene as was done by Heer. British representatives of the Older-Tertiary strata. We have already stated that Miocene sedimentary formations do not exist in the British Islands ; but lower strata, now recog- nised as the equivalents of the Oligocene series of the Con- tinent, are known in Hampshire and in the Isle of Wight. So far as is known, there is little or no unconformity between these strata and the underlying true Eocene deposits. They have been termed the Fluvio-marine Series by Forbes. An important marine deposit, found in sinking wells, opening brickyards, and making railway-cuttings in the district of the New Forest, in Hampshire at Brockenhurst, Roydon, Lynd- hurst, and other places has yielded a very large and interesting marine fauna, including many tropical forms of Lower-Oligocene mollusca. This has been called the Brockenhurst Series. In the Isle of Wight, however, this purely marine type is either CH. XIV.] HEMPSTEAD BEDS 208 wholly wanting or is represented only by brackish-water beds. Some difference of opinion has arisen concerning the portion of the Isle-of- Wight estuarine series which represents the marine Brockenhurst beds of the New Forest. Hempstead (or Hamstead) Beds. Of the series of strata so well exposed in the cliffs of the Isle of Wight the uppermost or Corbula-beds consist of marine sands and clays, and contain Valuta Rathieri, Heb., a characteristic Oligocene shell ; Corbula pisum, Sow. (fig. 198), a species common to the Upper Eocene clay of Fig. 198. Fig 199. Corbula pisum, Sow. Hempstead Beds, Isle of Wight. Cyrena semistriata, Desh., nat. Hempstead Beds. Barton ; Cyrena scmistriata, Desh. (fig. 199), several Ccrithia, and other shells peculiar to this series. Next below are freshwater and estuarine marls and carbonaceous clays, in the brackish-water portion of which are found abundantly Cerithium plicatum, Lam. (fig. 200), C. elegans, Desh. (fig. 201), and C. tricinctum, Broc. ; also Rissoa CJiastelii, Nyst (fig. 202), a very Fig 200 Fi- 201 common Klein-Spauwen shell, which occurs in each of the four subdivisions of the Hempstead series down to its base, where it passes into the Bembridge beds. In the freshwater portion Fig. 203. Cerithium plicatum, Cerithium elegans, Rissoa Chastelii, Nyst Paludina lento, Lam., nat. size. Desh., nat. size. sp. Hempstead, Is' e Brand , . Hem p- Hempstead. Hempstead. of Wight. stead Beds. of the same beds Paludina lenta, Brand, (fig. 203), occurs ; a shell identified by some conchologists with a species now living, P. unicolor, Lam. ; also several species of Limncea, Planoi'bis, and Unio. The next series, or middle freshwater and estuary marls, are dis- tinguished by the presence of Melania fasciata, Sow., Paludina lenta, Brand, and clays with Cypris ; the lowest bed contains Cyrena scmistriata, Desh. (fig. 199), mingled with Ccrithia and a Panopaa. The lower freshwater and estuarine marls contain Melania cos- tata, Sow., Melanopsis, &c. The bottom bed is carbonaceous, and called the ' Black band,' in which IHssoa Chastelii, Nyst (fig. 202), 204 BEMBRIDGE BEDS [CH. XIV. before alluded to, is found. This bed contains a mixture of Hempstead shells with those of the underlying Bembridge series. The mammalia, among which is Hyopotamus bovinus, Ow., differ, so far as they are known, from those of the Bembridge beds. The Hyo- potamus belongs to the hog tribe, or the same family as the Anthra- cotherium, of which last, seven species, varying in size from the hippopotamus to the wild boar, have been found in Italy, and in other parts of Europe, associated with the lignites of the Oligocene period. The seed-vessels of Chara medicaginula, Brong., and C. helic- teres, Brong., are characteristic of the Hempstead beds generally. Bembridge series. These beds are about 120 feet thick, and lie immediately under the Hempstead beds near Yarmouth, in the Isle of Wight. They consist of marls, clays, and limestones of fresh- water, brackish, and marine origin. Some of the most abundant shells, as Cyrena semistriata, Desh. var., and Paludina lento, (fig. 203), are common to this and to the overlying Hempstead series ; Fig. 204. Fig. 205. Mflania turritissima, Fragment of carapace of Trionyx. Forbes. Bembridge. Bembridge Beds, Isle of Wight. but the majority of the species are distinct; The following are the subdivisions described by the late Professor Forbes : a. Upper marls, distinguished by the abundance of Melania turritissima, Forbes (fig. 204). b. Lower marls, characterised by Cerithium mutabile, Lam., Cyrena pulchra, Sow., &c., and by the remains of Trionyx (see fig. 205). c. Green marls, often abounding in a peculiar species of oyster, and accompanied by Cerithium, Mytilus, Area, Nucula, &c. d. Bembridge limestones, compact cream-coloured tufaceous limestones alternating with shales and marls, in all of which land- shells are common, especially at Sconce, near Yarmouth, as described by Mr. F. Edwards. The Bulimus ellipticus, Sow. (fig. 206) , and Helix occlusa, F. Edw. (fig. 207), are among its best-known land-shells. Paludina orbicularis, Sow. (fig. 208), is also of frequent occurrence. One of the bands is filled with a little globular Paludina. Among the freshwater pulmonifera, Limncea fusiformis, Sow. (fig. 210), and Planorbis discus, F. Edw. (fig. 209), are the most generally distributed : the latter represents or takes the place of the Planorbis euomplialus, Sow. (see fig. 213), of the more ancient Headon series. Chara tubcr- culata, Lyell (fig. 211), is the characteristic Bembridge ' gyrogonite ' or seed-vessel. From this formation on the shores of Whitecliff Bay, Dr. CH. xiv.] TEKKESTKIAL AND FRESHWATER 205 Mantell obtained a fine specimen of a fan palm, Flabellaria Lama- nonis, Brong., a plant first obtained from beds of corresponding age in the suburbs of Paris. The well-known building-stone of Binstead, near R y de > a limestone with numerous hollows caused by Cyrence, which have dis- appeared and left the moulds of their shells, belongs to this subdivision of the Bembridge Fig. 207. Fig-. 208. Bulimut ellipticus, Sow. Bembridge Limestone, * uat. size. Helix ooclusa, F. Edw., nat. size. Bembridge Limestone, Isle of Wight. Fig. 210. Pdludina orbicularly, Sow., J. Bembridge. PlanorUs discus. F. Edw. Bembridge, \ diam. Limncea fmiformis, Sow., nat. size. Chara tuber culata, Lyell, seed-vessel mag. Bembridge Limestone, I. of Wight. Fig. 215 series. In the same Binstead stone Mr. Pratt and the Rev. Darwin Fox first discovered the remains of mammalia characteristic of the gypseous series of Paris, such as Palceotherium magnum, Cuv., P. medium, Cuv., P. minus, Cuv., P. curium, Cuv., P. crassum, Cuv. ; also Anoplotherium commune, Cuv. (fig. 212), A. secundarium, Cuv., Dichobune cervinum, Ow., and Ghc&ropotamus Cuvieri, Ow. The Paleo- there, above alluded to, resembled the living tapir in the form of the head, and in having a short proboscis, but its molar teeth were more like those of the rhinoceros. Palao- Lower molar tooth, .... ,-v t A T * i nat, size. thenum magnum, Cuv., was of the size of a Anoplotherium commune, small horse, about four or five feet in height. Cuv. Binstead, Isle of At the base of the Bembridge series there Wight. is another group of strata of fresh- and brackish-water origin, and very variable in mineral character and thickness. Near Ryde, it 206 HEADON BEDS [CH. XIT. supplies a freestone much used for building, and called by Professor Forbes the Nettlestone grit. In one part ripple-marked flagstones occur, and rocks with fucoidal markings. This series of rocks was called by Professor Forbes 'the Osborne and St. Helen's series,' but its fossils do not appear to be so distinct in character from those of the Bembridge series as to necessitate a special designation for the group of beds. Headon series. These beds are well seen both in Whitecliff Bay and at Headon Hill ; that is, at the east and west extremities of the Isle of Wight. The upper and lower portions are freshwater, and in Fig. 213. Planorbis euomphalus, Sow. Headon Hill, diam. Helix IdbyrintMca, Say. Headon Hill, Isle of Wight ; and Hordwell Cliff, Hants also recent. the middle a few brackish -water beds occur. Everywhere Planorbis euomphalus, Sow. (fig. 213), characterises the freshwater deposits, just as the allied form, P. discus, F. Edw. (fig. 209), does the Bembridge limestone. The brackish-water beds contain Potamomya plana, Sow. sp., Cerithium mutabile, Lam., and Potamides cinctus, Sow., and Venus (or Cytherea) incrassata, Desh., a species common to the Fig. 216. Fig. 215. Neritina concata, Sow., nat. size. Headon series. Limncea caudatit, F. E. Edw., i- Headon series. Cerithium concavum, Sow. 3. Headon series. Limbourg beds and the Gres de Fontainebleau, of the Oligocene series. Among the shells which are widely distributed through the Headon series are Neritina concava, Sow. (fig. 215), Limncea caudata,F. Edw. (fig. 216), and Cerithium concavtim, Sow. (tig. 217). Helix laby- rinthica, Say. (fig. 214), a land-shell now inhabiting the United States, was discovered in this series by Mr. Searles Wood in Hordwell Cliff. It is also met with in Headon Hill, in the same beds. At Sconce, in the Isle of Wight, it occurs in the Bembridge series. The lower and middle portion of the Headon series is also met with in Hordwell CH. xiv.] BROCKENHURST BEDS 207 Cliff (or Hordle, as it is often spelt), near Lymirigton, Hants. The chief shells which abound in this cliff are Paludina lenta, Brand., and various species of Limncea, Planorbis, Melania, Cyclas, Unio, Potamomya, Dreissena, &c. Among the chelonians we find a species of Emys, and no less than six species of Trionyx ; among the saurians an alligator and a crocodile ; among the ophidians two species of land-snakes \Paleryx, Owen) ; and among the fish Sir P. Egerton and Mr. Wood have found the jaws, teeth, and hard shining scales of the genus Lepi- dosteus, or Bony Pike of the American rivers. The same genus of freshwater ganoids has also been met with in the Hempstead beds in the Isle of Wight. The bones of several birds have been obtained from Hordwell, and the remains of quadrupeds of the genera PalcBo- tlierium (P. minus, Cuv.), Anoplotheriwn, Dichodon, Dichobune, Hyracotherium, Microchcsrus, Lophiodon, Hyopotanms, and Hyce- nodon. From another point of view, however, this fauna deserves notice. Its geological position is considerably lower than that of the Bembridge or Montmartre beds, from which it differs almost as much in species as it does from the still more ancient fauna of the Eocene beds. It therefore teaches us what a grand succession of distinct assemblages of mammalia flourished on the earth during the Tertiary period. Many of the marine shells of the brackish-water beds of the above series, both in the Isle of Wight and Hordwell Cliff, are common to the underlying Barton Clay; and, on the other hand, there are some freshwater shells, such as Cyrena semistriata, Desh., which are common to the Bembridge beds. The Brockenlmrst Marine Group In the New Forest, at about the same horizon probably as the Headon beds of the Isle of Wight, we find a series of sands and clays, often crowded with marine shells, belonging to forms found only in tropical seas, with many corals. The beds are concealed by gravels, and can only be studied in artificial openings, such as brickyards and railway cuttings. The rich fauna of this important marine formation was studied by the late Mr. F. E. Edwards, and the valuable collection of shells made from it is now in the British Museum. There is still some difference of opinion among geologists as to the exact correlation of these marine strata of the New Forest with the Fluvio-marine beds of the Isle of Wight. Baron von Koenen has pointed out that no less than 46 out of the 59 Brockenhurst shells, or a pro- portion of 78 per cent., agree with species occurring Fig. 218. in the Tongrian formation in Belgium. Barton Clay. The top of the Eocene series is formed in the Isle of Wight and Hampshire by a series of sands, which in the latter locality contain an admixture of Oligocene and Eocene forms ; and this is underlaid by the celebrated Barton Clay. The latter formation consists of grey, greenish, and brown clays, with bands of sand. It is seen vertical in Alum Bay, Isle of Wight, and nearly horizontal in the cliffs soL,T of the mainland near Lymington. The thickness is 300 feet at Barton Cliff, where it is rich in marine fossils. Usually, the fossils are beautifully preserved, and Chama squa- mosa, Sol. (fig. 218), is very characteristic. 208 BARTON CLAY [CH. xiv. The foraminifera called Nummulites begin, when we study the _ertiary formations in a descending order, to make their Appearance in great numbers in these beds Nummulites elegans, Sow., and a small species called Nummulites variolarius, Lam. (fig. 227), are found both on the Hampshire coast and in beds of the same age in White- cliff Bay, in the Isle of Wight. Several marine shells, among which is Corbula pisiim, Sow. (fig. 198, p. 203), are common to the Barton Fig. 219. Fig. 220. Fig. 221. Mitra scdbra, Sow., nat. size. Fig. 222. Valuta ambigua, SoL, J. Fig. 223. Typhis pungens, Brand, nat. size. Fig. 224. Valuta athleta, Sol., |. Barton Terebellum fusiforme, Lam., Terebellum sopitum, and Bracklesham. nat. size. Barton and Bracklesham. Brand. Nat. size. Fig. 225. Fig. 226. Cardita sulcata, Brand., . Barton. NummuWes variolarius> Lam. Middle Eocene, Crassafella sulcata, Sow., \. Bracklesham Bay. Bracklesham and Barton, a, Nat. size. 6. Magnified. beds and the higher Hempstead series, and a still greater number are common to the Headon series. Bracklesham Beds. Beneath the Barton Clay we find in the north of the Isle of Wight, both in Alum and Whitecliff Bays, a great series of various-coloured sands and clays for the most part un- fossiliferous, and probably of estuarine origin. As some of these beds contain Cardita planicosta, Lam. (fig. 228), they have been identified CH. XIV.] BRACKLESHAM SERIES 209 with the marine beds much richer in fossils seen in the coast section in Bracklesham Bay, near Chichester in Sussex, where the strata consist chiefly of green clayey sands with some lignite. Among the Bracklesham fossils, besides the Cardita, occurs the huge Ceri- Fig. 228. Cardita ( Venericardia) planicosta, Lain. Fig. 229. Nummulites Icevigatus, Lam. Bracklesham, nat. size. a. Section of the nummulite. b. Group, with an individual showing the exterior of the shell. Fig. 230. Palceophis typhoeus, Owen, ; an Eocene sea-serpent. Bracklesham. a, 6. Vertebra, with long neural spine preserved. c. T wo vertebrae articulated together. thium giganteum, Lam., so conspicuous in the Calcaire grossier of Paris, where it is sometimes two feet in length. Nummulites Icevi- gatus, Lam. (see fig. 229), also characteristic of the lower beds of the P 210 BOURNEMOUTH STRATA [CH. XIV. Calcaire grossier in France, where it sometimes forms stony layers, as near Compiegne, is very common in these English beds, together with N. variolarius, Lam. Out of 193 species of mollusca procured from the Bracklesham beds in England, 126 occur in the Calcaire grossier in France. It was clearly, therefore, coeval with that part of the Parisian series more nearly than with any other. According to tables compiled from the best authorities by Mr. Etheridge, the number of mollusca now known from the Bracklesham beds in Great Britain is 3 1 J3, of which no less than 240 are peculiar Fig. 231. Fig. 2:!2. Defensive spine of Ost radon, ^. Bracklesham. to this subdivision of the British Eocene series, while 70 are common to the older London Clay, and 140 to the newer Barton Clay. The volutes and cowries of this formation, as well as the Bryozoa and corals, favour the idea of a warm climate having prevailed, which is borne out by the discovery of the remains of a serpent, Palaopliis typhfevs, Ow. (see fig. 230), exceeding, according to Professor Owen, twenty feet in length, and allied in its osteology to the Boa, Python, Coluber, and Hydrophis. The compressed form and dimi- nutive size of certain caudal ver- tebrae indicate so much analogy with the HydropliidcB as to induce Professor Owen to pronounce this extinct ophidian to have been marine. Amongst the com- panions of this sea-snake of Bracklesham were an extinct crocodile (Gavialis Dixoni, Palatal or dental plates of Myliobafis Edwardsi, Dix., i. Bracklesham Bay. Owen) and numerous fish, such as now frequent the seas of warm latitudes, as the Ostracion of the family Balistidfe, of which a dorsal spine is figured (see fig. 231), and gigantic Kays of the genus MyliobaUs (see fig. 232). The teeth of sharks also, of the genera Carcharodon, Otodus, Lamna, Galeocerdo, and others, are abundant. (See figs. 233-236.) Bournemouth Beds. The sands and clays which intervene between the equivalents of the Bracklesham Beds and the London Clay or Lower Eocene, are well seen in the vertical beds of Alum Bay in the Isle of Wight and eastwards of Bournemouth on the south coast of Hampshire. There are some very interesting leaf-beds which underlie the marine strata of the Bracklesham clays at this locality. None of the beds are of great horizontal extent, and there is much cross-stratification or false bedding in the sands, with many pebble beds, and in some places black carbonaceous seams and lignite. In the midst of a leaf-bed at the base of the Bournemouth strata in Studland Bay, Dorsetshire, shells of the genus Unio attest the fresh- water origin of the white clay. No less than forty species of plants are mentioned by MM. De la XIV.] AND THEIR FLORA. 211 Harpe and Gaudin from this formation in Hampshire, among which plants referred to the Proteacese (Dryandra, &c.) and the fig tribe are abundant, as well as the cinnamon and several other laurineee, with some papilionaceous plants. It appears from the researches of Mr. Starkie Gardner that the leaves, fruits, and seeds were deposited close to where they once Pig. 233. Fig. 234. Fig. 235. Lamna elegans, Agass., - nat. size. Carcharodon angusiidens, Agass., Otodus obliquus, Agass., Galeocerdo latidens t nat. size. nat. size. Aga^s., nat. size. Teeth of Sharks from Bracklesham Beds. grew. The fruit Nipadites, closely allied to that of the existing Nipa Palm, was found with the rind and pulp more or less preserved. Tufts of leaves of Proteacca, branches of Conifers, seeds of Higlitca minima, Bow., and Anona were observed. A small patch at the Vig. 237. Marine Shells of Bracklesham Beds Fig. 238 Fig. 239. Fig. 240. Fig. 241. + Lucina serrata, Sow. Conus a. Magnified. Lamarckii, b. Nat. size. F. B. Edw. base of the cliffs was crowded with seeds of Hightea, Cucumites, and Pctrophiloides. Pinnaa of an Osmunda were present. There is a fine Irartea palm-leaf in the British Museum from this locality. Heer has mentioned several species which are common to this flora and that of Monte Bolca, near Verona, so celebrated for its P2 212 PLANT-BEARING BEDS OF [CH. xiv- fossil fish, and where the strata contain nummulites and other Middle Eocene fossils. He has particularly alluded to Aralia primigenia, Hr. (of which genus a fruit has since been found by Mr. Mitchell at Bournemouth), Daphnogene veronensis, Hr., and Ficus granadilla, Hr., as among the species common to and characteristic of the Isle-of- Wight and Italian Eocene beds. The American types found among these Eocene plants have been noticed by the same authors. Lignites and Clays of Bovey Tracey, Devonshire. Sur- rounded by the granite and other rocks of the Dartmoor district in Devonshire, is a formation of kaolin (China-clay), sand, and lignite, long known to geologists as the Bovey-Coal formation, respecting the age of which, until late years, opinion was greatly divided. This deposit is situated at Bovey Tracey, a village distant eleven miles from Exeter in a south-west, and about as far from Torquay in a north-west, direction. The strata extend over a plain nine miles long, and they consist of the materials of decomposed and worn- down granite mixed with vegetable matter, and have evidently filled up an ancient hollow or lake-like expansion of the valleys of the Bovey and Teign. The lignite is of bad quality for economical purposes, having a great admixture of iron pyrites, and emitting a sulphurous odour ; it has, however, been successfully applied to the baking of pottery, for making which some of the fine clays are well adapted. Mr. Pen- gelly has confirmed Sir H. De la Beche's opinion that much of the upper portion of this old lacustrine formation has been removed by denudation. At the surface is a dense covering of white clay and gravel with angular stones probably of the Pleistocene period, for in the clay are three species of willow and the dwarf birch Betula natia, L., indicating a climate colder than that of Devonshire at the present day. Below this are Middle-Eocene strata about 300 feet in thickness, in the upper part of which are twenty-six beds of lignite, clay, and sand, and at their base a ferruginous quartzose sand, varying in thickness from two to twenty- seven feet. Below this sand are forty-five beds of alternating lignite and clay. No shells or bones of mammalia, and no insect, with the exception of one fragment of a beetle (Bupestris) in a word, no organic remains, except plants have as yet been found. These plants occur in fourteen 01 the beds ; namely, in two of the clays, and the rest in the lignites. Amongst the species are a number of terns Lastr tea stiriaca, Ung., Pecoptcris lignitum, Heer ; conifers, Sequoia Couttsitz, Heer, the matted debris of which forms a lignite bed. There are also remains belonging to the genera Cinnamomum, Eucalyptus, Quercus, Salix, Laurus, Anona, Palmacites, leaves of evergreen oaks, spindle trees, figs, water- lily, and the seeds of two species of vine. The crozier-like vernation of some of the young ferns is very perfectly shown, and was at first mistaken, by collectors, for shells of the genus Planorbis. On the whole, the vegetation of Bovey implies the existence of a subtropical climate in Devonshire in the Middle-Eocene period. Scotland. Isle of Mull. In the sea-cliffs, forming the head- land of Ardtun, on the west coast of Mull, in the Hebrides, several bands of tertiary strata containing leaves of dicotyledonous plants were discovered in 1851 by the Duke of Argyll. From his description CH. xiv.] BOVEY TKACEY, MULL, AND ANTRIM 213 it appears that there are three leaf -beds, varying in thickness from 1^ to 2| feet, which are interstratified with volcanic tuff and trap, the whole mass being about 130 feet in thickness. ,A sheet of basalt of later age, 40 feet thick, covers the whole ; and another columnar bed of the same rock, 10 feet thick, is exposed at the bottom of the cliff. One of the leaf-beds consists of a compressed mass of leaves unaccompanied by any stems, as if they had been blown into a marsh where a species of Equisetum grew, of which the remains are plentifully embedded in clay. It is supposed by the Duke of Argyll that this formation was accumulated in a shallow lake or marsh in the neighbourhood of a volcano, which emitted showers of ashes and streams of lava. The materials in which the fossils are embedded may have fallen into the lake from the air as volcanic dust, or have been washed down into it as mud from the adjoining land. Even without the aid of Tertiary fossil plants, we might have decided that the deposit was newer than the chalk, for chalk flints containing cretaceous fossils were detected by the Duke in the principal mass of volcanic ashes or tuff. The late Edward Forbes observed that some of the plants of this formation resembled those of Croatia, described by Dr. Unger ; and his opinion has been confirmed by Professor Heer, who found that the conifer most prevalent was the Sequoia Langsdorfii, A. Brong. (fig. 194, p. 200), also Corylus grosscdentata, Heer, an Oligocene species of Switzerland and of Menat in Auvergne. There is likewise a plane tree, the leaves of which seem to agree with those of Platanus ace- roides, Gopp. (fig. 167, p. 181), and a fern, Filicites hebridica, Forbes (which is as yet peculiar as a European fossil to Mull, but which is considered by Dr. Newberry to be identical with a living American species, Onoclea sensibilis, L.), and a species of Gingko. It is thought probable that these beds may belong to a somewhat similar horizon to that of Bovey Tracey and Bournemouth, and, according to Mr. Starkie Gardner, they may be of Eocene age. Ireland. These interesting discoveries in Mull have led to the suspicion that the basalt of Antrim and of the Giant's Causeway, in Ireland, may be of the same Eocene age. It must be remembered, however, that the evidence of fossil plants must be accepted with considerable caution ; not only is the determination of leaves by their forms and venation open to great question, in the opinion of many eminent botanists, but certain forms like Acer, Seq^loia, Gingko, &c., had certainly a very wide range in time as well as in space. The volcanic rocks that overlie the chalk, and some of the strata associated with, and interstratified between masses of basalt, contain leaves of dicotyledonous plants, somewhat imperfect, but resembling the beech, oak, and plane, and also some coniferas of the genera Pinus and Sequoia. These old land surfaces are exceedingly interesting. Bag-shot Beds In the London basin the highest strata known are the sands of Bagshot, which contain bands of pipeclay and layers of flint pebbles, but only very rarely yield traces of fossils. These strata not improbably represent the Bournemouth beds of the Hamp- shire basin. In the upper and middle Bagshots a few casts of marine fossils have been found in green glauconitic sandy clays, but no fossils &re known from the lower Bagshots, The Bagshot beds are seen oq 214 THE LONDON CLAY [CH. xiv. the top of Hampstead Hill, and cover extensive tracts in the south- east of the London basin, where they form wide, sandy heaths. London Clay. This formation sometimes attains a thickness of 500 feet, and consists of tenacious brown and bluish-grey clay, with layers of concretions called septaria, and is found in the London basin. In the Hampshire basin the more sandy Bognor beds are of the same age, and, like the London clay, they are essentially marine. The London clay was probably deposited on a sea-floor close to the entry of a large estuary and river, and the strata were formed at different depths, and some in shallow water. Several zones of fossils have been discovered by Professor Prestwich, the deepest and most marine being to the east, and the uppermost containing a terrestrial vegetation, mammalian, fish, and reptilian remains. The following genera of plants have been noticed by Bowerbank, Ettingshausen, and Gardner : Pinus, Collitris, Sequoia, Musa, Sabal, Nipa, Elais, Agave, Quercus, Liqnidambar, Nysa, Magnolia, Juglans, Eucalyptus, Amygdalus, and Fig. 242. Banksia (?). Mr. Bowerbank, in a valuable publi- cation on these fossil fruits and seeds, has described fruits of palms of the recent type Nipa, now found only in the Molucca and Philippine Islands, and in Bengal. (See fig. 242.) In the delta of the Ganges, Sir J. Hooker observed the large nuts of Nipa fruticans, Thunb., floating in such numbers in the various arms of that great river as to obstruct the paddle- wheels of steam-boats. These plants are allied to the cocoa-nut tribe on the one side, and on the other to the Pan- mpaditeseiiipticus.-Bow.,$. danus > r screw-pine. There are also Fossil fruit of palm, from Sheppey. met with three species of Anona, or custard apple ; and cucurbitaceous fruits (of the gourd and melon family), and fruits of various species of Acacia. Besides fir-cones or fruit of true Coniferag there are cones sup- posed to belong to the Proteaceae ; and the celebrated botanist, Kobert Brown, pointed out the affinity of these to the New Holland types Petrophila and Isopogon. Of the first there are about 50 and of the second 30 described species now living in Australia. Baron von Ettingshausen and Mr. Carruthers, having examined the original specimens now in the British Museum, state that all these cones from Sbeppey (see fig. 243) may be reduced to two species, which have an undoubted affinity with the two existing Australian genera above mentioned. Other botanists, however, think that the supposed Proteaceous fruits may be referred to Alnus and other non-Proteaceous genera. The contiguity of land may be inferred not only from these vegetable productions, but also from the teeth and bones of crocodiles and turtles. Of turtles there were numerous species referred to extinct genera. These are, for the most part, not equal in size to the largest living tropical turtles. A sea-snake, thirteen feet in CH. XIV.] AND ITS FOSSILS 215 Fig. 243. length, PalceopTiis toliapicus, Ow., has been described by Owen from Sheppey, and the species differs from that of Bracklesham. A croco- dile, Crocodilus toliapicus, Guv. et Ow., has been described by the same palaeontologist, and a form nearly allied to the Gangetic Gavial also. The relics of several birds have been found belonging to the genera Lithornis, Argillornis, and Halcy- ornis. The first was a Vulturine, the second an Albatross, and the third a Kingfisher. Moreover, Odontopteryx (see fig. 188, p. 196) re- presented the birds whose bony jaw margins are produced as denticu- lations. The Mammalian remains are very rare; Hyracotherium, an supposed Eocene Proteaceous Fruit, odd -hoofed herbivore, and Lopliiodon Peti-ophiioides Richardson i. Bowcrb. allied to "the modern Tapir, have London clay Sheppey. Natural size. a. Cone. b. Section of cone sbovdue the position of the seeds. ng been found at the base of the for- mation, with a part of a jaw of a Didelpliys (Opossum), discovered by Charlesworth, and the tooth of a Bat (Vespertilio). The species Coryphodon eoccenus, Ow., most probably came from the underlying Woolwich beds. Neverthe- Fig. 244. Shells of the London Clay Fig. 246. Fig. 245. Valuta nodosa, I'horus f.rlensus, Sow., J. Highgate. Sow., \. Highgate. Fig. 247. Rostellaria (Hippocrenes) ampla, Brand., Nautilus centralis, Sow., ^. Highgate. J of nat. size ; found also in the Bar- ton clay and Brockenhurst beds. less, this scanty fauna of a Herbivore, a Marsupial and an Insecti- vorous Bat is not without its interest. All seem to have inhabited 216 LONDON-CLAY FOSSILS [CH. XIV. the banks of the great river which floated down the Sheppey fruits. This fauna was long antecedent to the present aspect of nature in Europe and Asia, for the Alps and Himalayas were not elevated till later Oligocene times. The marine shells of the London clay confirm the inference derivable from the plants and reptiles in favour of a high temperature. Thus Conns and many species of Valuta occur, a large Cyprcea, C. ori- formis, Sow., a very large Rostellaria (fig. 246), a species of Cancellaria, Fig. 248. Fig. 249. Aturia ziczac, Bronn. (syn. Nautilus ziczaf, Sow.) London clay, Sheppey, . Belosepia sepioidea, De Blainv., nat. size. London clay. Sbeppey. Fig. 250. Fig. 251. Fig. 252. Leda amygdaloides, Sow.,. Highgate. Cryptodon (Axinus) angulatum, Sow., nat. size. London clay, Hornsey. Astropccten crispafus, E. Forbes, . Sheppey. six species of Nautilus (figs. 247, 248), besides other Cephalopoda of extinct genera, one of the most remarkable of which is the Belosepia (fig. 249). Among many characteristic bivalve shells are Lcda amygdaloides, Sow. (fig. 250), and Cryptodon angulatum, Sow. (fig. 251), and among the Radiata a star-fish, Astropectcn (fig. 252). Nearly 100 species of fish are known, amongst which there are a sword-fish (Tetrapterus priscus, Ag.), about eight feet long, and a saw- fish (Pristis bisulcatus, Ag.), about ten feet in length, both now foreign to the British seas. The Crustacea were abundant, and most of them belonged to the short-tailed tribe ; one species may have belonged to the true crabs. The other genera found are Xanthopsis, Xantholites, and Grajwis. One of the Anomura, with a moderately long abdomen, was Dromi- lites, allied to the Sponge-crab. The Oldhaven Beds form the upper portion of the Woolwich and Beading series, but only occur in Kent and portions of Surrey. They consist almost entirely of rolled flint pebbles in a sandy matrix. Although only twenty to thirty feet thick, 150 species of fossils have been yielded by them, consisting of marine and estuarine shells and plant remains. The flora, so far as it goes, is interesting, and contains Ficus, Cinnamomum, and Conifer, and appears to be without the American and Australian types which were so dominant in later times, CH. xiv.] WOOLWICH AND BEADING BEDS 217 Woolwich and Reading- series. This formation is apparently of the same age as the Plastic clay of the Hampshire basin, which resembles a clay used in pottery (Argile plastique) in the French series. This formation, when studied in the basins of London, Hamp- shire, and Paris, presents very variable characters ; but typically the beds consist, over a large area, of mottled clays and sand, with lignite, and with some strata of well-rolled flint-pebbles, derived from the chalk, varying in size, but occasionally several inches in diameter. These strata may be seen in the Isle of Wight or at Bognor, in contact with the chalk; or in the London basin, at Reading, Blackheath, and Woolwich, covering the Thanet sand. In the lowest beds banks of oysters are observed, consisting of Ostrea bellovacina, Lam., common also in France. In these beds at Bromley, Buckland found a large pebble to which five full-grown oysters were affixed, in such a manner as to show that they had commenced their first growth upon it, and remained attached through life. In several places, as at Woolwich on the Thames, at Newhaven in Sussex, and elsewhere, a mixture of marine and freshwater mollusca distinguishes this member of the series. Among the latter, Melania inquinata, Defr. (see fig. 254), and Cyrena cuneiformis, Sow. (see fig. Fig. 253. Fig. 254. Melania (Melartatria) Cyrena cuneiformis, Sow. Natural size. inquinata, Defr. . Woolwich clays. Woolwich clays. 253), are very common, as in beds of corresponding age in France. They clearly indicate points where rivers entered the Eocene sea. We usually find a mixture of brackish-water, freshwater, and marine shells, and sometimes, as at Woolwich, proofs of the river and the sea having successively prevailed on the same spot. At New Charlton, in the suburbs of Woolwich, De la Condamine discovered in 1849 a layer of sand associated with well-rounded flint pebbles, in which numerous individuals of the Cyrena cuneiformis were seen standing endwise with both their valves united, the sipnonal extremity of each shell being uppermost, as would happen if the mollusks had died in their natural position. Traced eastward towards Herne Bay, the Woolwich beds become sandy and assume a more decidedly marine character ; while, in an opposite or south-western direction, the beds are more uniformly clayey, and in some places, as near Chelsea, they assume 218 THANET SANDS [CH. XIV. freshwater characters, and contain Unio, Paludina, and layers of lignite. Hence the land drained by the ancient river seems clearly to have been to the south-west of the present site of the metropolis. Plants of the genera Fields, Grevillea, and Laurus, and leaves of the plane, poplar, and willow have been found, and the flora has affinities both with the cretaceous and the tertiary. Mr. Newton, of the Geological Survey, has described Coryphodon, a remarkable mammal, allied to those discovered in North America, from these beds. Thanet sands. The Woolwich or Plastic clay above described may often be seen in the Hampshire basin in actual contact with the chalk, constituting in such places the lowest member of the British Eocene series. But at other points another formation of marine origin, characterised by a somewhat different assemblage of organic remains, has been shown by Professor Prestwich to intervene be- tween the chalk and the Woolwich series. The sand is micaceous, Fig. 255 Fig. 256. Pholadomya cuneata. Sow., Aporrhais Sowerbyi, Mant., Cyprina Morrtsii, Sow., . Thanet sands. nat. size. Thanet sands. J. Thanet sands. and was derived from a granitic district. It rests on a denuded surface of the chalk, and is not found in the Hampshire basin. For these beds he has proposed the name of ' Thanet sands,' because they are well seen in the Isle of Thanet, in the northern part of Kent, and on the sea-coast between Herne Bay and the Eeculvers, where they consist of sands with a few concretionary masses of sandstone, and contain, among other fossils, Pholadomya cuneata, Sow. (fig. 255), Cyprina Morrisii, Sow. (fig. 257), Corb 7 a ref/nl- Uensis, Mor., Scalaria Bowerlanld, Mor., Ajwrrhais Sowerbyi, Mant. (fig. 256). That the Eocene strata of the London and Hampshire basin are unconformable to the underlying chalk is shown by the over-lap (or ' over-step ') of the Tertiary beds on the several zones of the Creta- ceous. The eroded surface of the chalk with the band of green- coated flints usually seen at the base of the Thanet Sand is due to the action of percolating water in dissolving away the upper layers of the chalk. Fuller details concerning the British Eocene and Oligocene strata will be found in Prof. Prestwich's various memoirs on these forma- tions in the ' Quart. Journ. Geol. Soc. 'for 1847, 1850, 1851, 1852, 1853, 1854, 1855, 1857, 1888, and the fol- lowing memoirs of the Geological Survey : ' The Tertiary Fluvio- marine Formation of the Isle of Wight,' by Edward Forbes (1856) ; ' Geology of the Isle of Wight,' by II. W. Bristow, C. Keid, and A. Straban (2nd ed.), 1889; the 'Geo- logy of London ' (1889), and other memoirs by W, Whitalr^r- CH. XV.] 219 CHAPTER XV FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE CAINOZOIC OF THE BRITISH ISLES Tertiaries of France and Belgium Montian Argile plastique Calcaire grossier Gypsum of Montmartre Mammals of Oligocene of Northern and Central France Faluns of Touraine and Bordeaux Pliocene of Northern France and Belgium Tertiaries of Central Europe Lower Brown Coal and Amber deposits Mayence Basin Pliocene of Eppelsheim Tertiaries of Alps and Switzerland Flysch and Num- mulitic formations Lower, Middle, and Upper Molasse Plants and insects of Oeningen Tertiaries of Italy Oligocene and Miocene Subapennine strata Newer Pliocene of Sicily and the Val d'Arno Tertiaries of Eastern Europe Oligocene of Croatia Miocene (Leithakalk and Sarmatian) of Vienna Basin Pliocene (Congeria) strata Tertiaries of India Sind and Sivalik strata Post-pliocene deposits of Northern Europe and the Alps Scandinavia and Russia Central Europe Alps and Jura Older and Newer Palaeolithic periods Lake-dwellings Post-pliocene of India, New Zealand, and Australia Tertiaries of North America Eocene and Neocene of Eastern States Mammals and Plants of Tertiaries of the Western Territories American Post-pliocene deposits Glacial and Champlain periods Tertiary Zones in Europe. IT is a remarkable circumstance that the capitals of nearly all the great European States London, Paris, Brussels, Eome, Vienna, Berlin, &c. are situated on strata of Tertiary age. In most cases these great cities stand in the midst of ' basins ' of Tertiary strata that is, of isolated tracts of sediments which have been preserved by synclinal folding of the strata, preceding the denu- dation which has removed the Tertiary rocks from the inter- vening anticlinals. In this way have been formed the well-known London Basin, Paris Basin, Berlin Basin, Vienna Basin, &c. It is doubtless owing to the circumstance of their proximity to great cities with universities that the Tertiary strata and their fossils have attracted so much attention, and have had so much study devoted to them. The Lower Eocenes of France and Belgium can be fairly well correlated with those of our own London and Hampshire basins by the assemblages of fossils contained in the several beds, though the strata themselves often differ in a very marked manner in their mineral characters from their equivalents in thi,s country. 220 PALEOCENE STRATA [CH. XV. CAINOZ01C STRATA OF FRANCE AND BELGIUM The general succession of the Lower Tertiary (Eocene and Oli- gocene) strata of France and Belgium is shown in the following table : Paris Basin UPPER OLIGOCENE (absent in England) MIDDLE OLIGOCENE (Hempstead and Bembridge Beds of England) LOWER OLIGOCENE (Brockenhurst and Headon Beds of England) Freshwater limestone of \ g, , "SaSSSST 6 ( Cassel, Bundle. Fontaine,,,,*,, sandstone { ****?!*** UPPER EOCENE (Barton Beds of England) MIDDLE EOCENE (Bracklesham Beds and London clay) LOWER EOCENE (Woolwich and Reading Beds and Thanet Sands) Gypsum and marls of Montmartre, with mam- malian remains Freshwater limestones of St. Ouen, marine sand of Beauchamp Coarse marine limestone \ known as the ' Calcaire j- grossier ' and sands of Guise ) Plastic clay and lignite of \ Soissons, sand of Bracheux, h marl of Meudon ) Lower Tongrian and clays of Egeln Wenimelian sands and clays Laekenian, Bruxellian, Paniselian, Ypresian Landenian, Heersian, Montian The Miocene is well developed in Southern France and Switzer- land ; but in France and Western Europe generally, as in England, the Pliocene is represented only by thin and comparatively insignifi- cant deposits, and it is necessary to go to Italy and the Vienna basin to find the full development of the Pliocene System. Paleocene Beds ('Mon- tian '). In the coarse limestone of Mons in Belgium and in the Marls of Meudon in the Paris basin we have strata which are perhaps older than any in the British Islands. The Belgian beds contain a few Cre- taceous Echinodermata, and some authors have proposed to rank these oldest known Tertiary strata of Europe as a distinct group, to which they apply the name of Paleo- cene. The Calcaire grossier of Mons is lower than the horizon of the Thanet Sands, and fills a depression in the chalk, being 300 feet thick. Upwards of 400 species of fossils have been obtained from it. Vast numbers of Gastropoda, Lamelli- branchiata, Bryozoa, Forami- nifera (Quinqueloculina), and cal- careous Algae are found. Some limestones, sands, and marine con- glomerates at Rilly, beneath the Meudon conglomerate, are the lowest members of the French Eocene, and are older than the Thanet Sands, but slightly younger than the Calcaire de Mons at the base of the Belgian Eocene. The conglomerates rest on the chalk, and their fauna is marine and tertiary. The Heersian of Belgium is also slightly older than tLe Thanet sands, and contains the flora of Gelinden. In this flora we find many species of Dryophylhtm, a genus somewhat resembling that of the modern American Chestnut Oak. But the flora as a whole has no satisfactory alliance with the Eocene flora of America. Sables de Bracbeux. The marine sands called the Sables de Bracheux (a place near Beauvais) were considered by M. Hebert to be older than the Lignites and Plastic clay, and to coincide in age with the Thanet Sands of England. At La Fere, in the department of Aisne, in a deposit of this age, a fossil skull has been found of u CH. XV.] ARGILE PLASTIQUE 221 quadruped called by Blainville Arctocyon primcevus, Blain., and supposed by him to be related both to the Bear and to the Kinkajou (Cercoleptes). This creature is one of the oldest known tertiary mammals. The Lower Landenian of Belgium resemble and are of the same age as the Thanet Sands. liigrnites of Soissoniiais and Argile plastique. At a slightly higher horizon in the Paris basin are extensive deposits of sands, with occasional beds of clay used for pottery. Fossil oysters (Ostrea bellovacina, Lam.) abound in some places ; and in others there is a mixture of fluviatile shells, such as Cyrena cuneiform-is, Sow. (fig. 253, p. 217), Melania inqui- nata, Defr. (fig. 254, p. 127), and others, frequently met with in the Woolwich beds of the London basin. Layers of lignite are also intercalated. In the year 1855, the tibia and femur of a large bird, equalling at least the ostrich in size, was found at Meudon, near Paris, at the base of the Plastic clay. This bird, to which the name of Gas- trornisparisiensis, Heb., has been assigned, appears, from the Me- moirs of MM. Hebert, Lartet, and Owen, to belong to an extinct genus. Professor Owen refers it to the class of wading land birds rather than to an aquatic species. That a formation so much ex- plored for economical purposes as the Argile plastique around Paris, and the clays and sands of corre- sponding age near London, should never have afforded any vestige of a feathered biped previously to the year 1855, shows what diligent search and what skill in osteological interpretation are required before the existence of birds of remote ages can be established.. The Ypresian and Paniselian of Belgium represent the English London clay. Iiower Eocene. There is no exact equivalent of the London clay in the Paris basin, and the next strata, above the Argile plastique, are the Sables de Guise. Sables de Guise. These are of considerable thickness, es- pecially at Cuise-uaii ^^, near Compiegne, and other localities in the Soissonnais, about fifty miles N.E. of Paris, from which about 300 species of shells have been obtained, many of them common to the Calcaire grossier and the Bracklesham beds of England, and many peculiar. Nummulites planulatus, Lam., is very abundant, and the most characteristic shell is Fig. 258. Nerita couoidea, Lam. the Nerita conoidea, Lam., a fossil which has a very wide geographical range ; for, as M. d'Archiac remarks, it accompanies the Nummulitic formation from Europe to India, having been found in Cutch, near the mouths of the Indus, associated with Nummulites scabra, Lam. No less than 33 shells of this group are said to be identical with shells of the London clay proper. It was believed by Professor Prestwich that the sands of Guise are probably newer than the London clay, and perhaps older than the Bracklesham beds of England. The Middle Eocene is composed of the Calcaire grossier, formed of limestones, and siliceous limestones, with sandy glauconitic beds at the base, all highly fossiliferous. Xiower Calcaire grossier, or Glauconie grossiere. The lower part of the Calcaire grossier, which often contains much green earth, is characterised at Auvers, near Pontoise, to the north of Paris, and still more in the environs of Compiegne, by the abundance of nummulites, consisting chiefly of N. Itevigatus, Lam., N. scabra, Lam., and N. Lamarcki, D'Orb., which constitute a large proportion of some of the stony strata, though these same foraminifera are want- ing in beds of a similar age in the immediate environs of Paris. The upper division of this group consists in great' part of beds of 222 CALCAIRE GROSSIER [CH. XV. compact, fragile limestone, with some intercalated green marls. The shells belong to such varied genera as Cerithium, Corbula, Limncea, Paludina, Cyclostoma, &c. In the green marls the bones of reptiles and mammalia (PaLceo- iherintn and Lophiodon) have been found. The middle division, or Calcaire grossier proper, consists of a coarse limestone, often passing into sand. It contains the greater number of the fossil shells which characterise the Paris basin. No less than 400 distinct species have been procured from a single spot near Grignon, where they are em- bedded in a calcareous sand, chiefly formed of comminuted shells, in which, nevertheless, individuals in a perfect state of preservation, both of marine, terrestrial, and fresh- water species, are mingled together. Some of the marine shells may have lived on the spot; but the shells of Cyclostoma and Limncea, being land and freshwater forms, must have been brought thither by rivers and currents. Nothing is more striking in this assemblage of fossil mollusca than the great proportion of species referable to the genus Cerithium. There occur no less than 187 species of this genus in the Paris basin, and almost all of them in the Calcaire grossier. Most of the living Cerithia inhabit the sea near the mouths of rivers, where the waters are brackish, so that their abundance in the marine strata now under consideration is in harmony with the hypothesis that the Paris basin formed a gulf into which several rivers flowed. In some parts of the Calcaire grossier round Paris, certain beds occur of a stone used in building, and called by the French geologists ' Miliolite limestone.' It is almost entirely made up of millions of microscopic shells, of the size of minute grains of sand, which belong to the Foraminifera. The Brux- ellian and Laekenian of Belgium represent the French Calcaire grossier. The Middle Eocene of Belgium approximates to the English type, and the Upper or Wemmelian series is full of Nummulites variolarius, Lam. Upper Eocene. The strata of this age in the Paris basin are continuous with the lower Oligo- cene. They are the marine gypseous series, yellow and greenish marls, with Cerithium tricarinatum, Desh., and Pholadomya ludensis, Desh. Beneath these are the ' Sables moyens,' with green sands, over- lying the nearly freshwater lime- stone of St. Ouen. They rest on the Grcs de Beauchamp, a marine sandstone with corals, sharks' teeth, and Nummulites variolarius, Lam. lacustrine gypseous se- ries of IVIontmartre. These strata, commencing with white marls and blue marls at the top, and having the important gypsum beds below, represent the Lower Oligocene, and are most largely developed in the central parts of the Paris basin, among other places in the hill of Montmartre, the fossils of which were first studied by Cuvier. The gypsum, there quarried for the manufacture of plaster of Paris, occurs as a granular crystal- line rock, and, together with the associated marls, contains land and fluviatile shells, and the bones and skeletons of birds and quadru- peds. Several land-plants are also met with, among which are fine specimens of Flabellaria. The remains also of freshwater fish, and of crocodiles and other reptiles, occur in the gypsum. The skele- tons of mammalia are usually isolated, often entire, the most delicate extremities being pre- served, as if the carcases, clothed with their flesh and skin, had been floated down soon after death, and while they were still swollen by the gases generated by their first de- composition. The few accompany- ing shells are of those light kinds which frequently float on the surface of rivers, together with wood. In this formation the relics of about fifty species of quadrupeds, including the genera Palceo- therium, Anoplotherium, and CH. XV.] GYPSUM OF MONTMARTRE 223 others, have oeen found, all extinct, and nearly four-fifths of them belonging to the Perissodactyle or odd-toed division of the or-der Ungulata. The Anoplotheridce form a family intermediate between pachyderms and ruminants, and belong to the even-toed group of Ungulates. With these Ungulata were associated a few carnivorous animals, among which were Hycenodon and a species allied to the dog (Cynodictis parisiensis, Gerv. sp.). Of the Bodentia was found a squirrel-like form; of the Cheiro^itera, a bat; while the family Didelphidce of the Mar- sitpialia, now confined to America, are represented by a true Opossum (Didelphys). Of birds, about 17 species have been discovered, five of which are still undetermined. The skeletons of some are entire, but none are referable to existing species. Crocodiles, and tortoises of the genera Emys and Trionyx, are found. Fossil footprints. Amongst the numerous interesting remains of this series are footprints of animals, which occur at six dif- ferent levels. M. Desnoyers dis- covered large slabs, which are now in the Museum at Paris, where, on the upper planes of stratification, the indented footmarks were seen, while corresponding casts in relief appeared on the lower surfaces of the strata of gypsum which were immediately superimposed. Upper Oligocene of Northern France. The Cal- caire de la Beauce constitutes a large tableland between the basins of the Loire and the Seine. It is associated with marls and other deposits, such as may have been formed in marshes and shallow lakes in the newest part of a great delta. Aquatic plants (Chara) left their stems and seed - ves- sels, which ,are now found em- bedded both in marl and flint, together with freshwater and land shells. Some of the siliceous rocks of this formation are used ex- tensively for millstones. The flat summits or platforms of the hills round Paris, and large areas in the forests of Fontainebleau, as well as the Plateau de la Beauce, already alluded to, are chiefly composed of these freshwater strata. Next to these, in the descending order, are marine sands and sandstone, com- monly called the Gres de Fontaine- bleau. Next in succession, forming the Middle Oligocene, are the Sables d'Etampes with ferruginous sands at Paris, resting on marls with Ostrea cyathula, Lam., and Cor- bula subpisum, D'Orb. These cover the Calcaire de Brie, which overlies clay and green marl with Cerithium plicatum, Lam., and Cyrena convex a, Lam. Oligoceue of Central France. Lacustrine strata, be- longing, for the most part, to the same age as the Calcaire de la Beauce, are again met with further south, in Auvergne, Cantal, and Velay. They appear to be the monuments of ancient lakes, which, like some of those now existing in Switzerland, once occupied the de- pressions in a mountainous region. The study of these regions possesses a peculiar interest, for we are presented in Auvergne with the evidence of a series of events of astonishing magnitude and grandeur, by which the original form and features of the country have been greatly changed, yet never so far obliterated but that they may still, in part at least, be restored in imagination. Great lakes have disappeared, and lofty volcanic mountains have been formed by the reiterated emission of lava, preceded and followed by showers of sand and scoriae. Deep valleys have been subsequently furrowed out through masses of lacustrine and volcanic origin ; and at a still later date, new cones have been thrown up in these valleys, new lakes have been formed by the damming up of rivers, and several assemblages of quadrupeds, birds, and plants, Eocene, Oligocene, Miocene, and Pliocene, have fol- lowed in succession. Yet the region has preserved from first to last its geographical identity ; and we can still picture to our minds its external condition and physical structure, 224 LOWER TERTIARIES OF AUVERGNE [CH. xv. before these wonderful vicissitudes began, or while a part only of the series of changes had been com- pleted. There was a first period when the spacious lakes, of which we still may trace the boundaries, lay at the foot of mountains of moderate elevation, unbroken by the bold peaks and precipices of Mont Dore, and unadorned by the picturesque outline of the Puy-de- Dome, or of the volcanic cones and craters now covering the granitic platform. During this earlier scene of repose, deltas were slowly formed ; beds of marl and sand, several hun- dred feet thick, deposited ; siliceous and calcareous rocks precipitated from the waters of mineral springs ; shells and insects embedded to- gether with theremains of the croco- dile and tortoise, the eggs and bones of water-birds, and the skeletons of quadrupeds, most of them of genera and species characteristic of the period. To this tranquil condition of the surface succeeded the era of volcanic eruptions, when the lakes were drained, and when the fertility of the district was probably en- hanced by the igneous matter ejected from below, and poured down upon the more sterile granite. During these eruptions, which appear to have taken place towards the close of the Miocene epoch, and which continued during the Pliocene, various assemblages of quadrupeds successively inhabited the district, amongst which are found the genera Mastodon, Rhi- noceros, Elephas, Tapir, Hippo- potamus, together with the ox, various kinds of deer, the bear, the hyaena, and many beasts of prey which ranged the forest or pastured on the plain, and were occasionally overtaken by a fall of burning cinders, or buried in flows of mud such as accompany volcanic erup- tions. Lastly, these quadrupeds became extinct, and gave place in their turn to the species now exist- ing. There are no signs, during the whole time required for this series of events, of the sea having inter- vened, or of any denudation which may not have been accomplished by rivers and floods accompanying repeated earthquakes. Auvergne. The most northern of the freshwater groups is situated in the valley plain of the Allier, which lies within the department of the Puy-de-D6me, being the tract which went formerly by the name of the Limagne d'Auvergne. The principal divisions into which the lacustrine series may be separated are the following : 1st, Sandstone, grit, and conglomerate, including red marl and red sand- stone ; 2ndly, Green and white foliated marls ; Srdly, Limestone, or travertin, often oolitic in struc- ture ; 4thly, Gypseous marls. The whole rest on granite. It seems that, when the ancient lake of the Limagne first began to be filled with sediment, no volcanic action had yet produced lava and scoriae on any part of the surface of Auvergne. No pebbles, there- fore, of lava, were transported into the lake no fragments of volcanic rocks embedded in the con- glomerate. But at a later period, when a considerable thickness of sandstone and marl had accumu- lated, eruptions broke out, and lava and tuff were deposited, at some spots, alternating with the lacus- trine strata. It is not improbable that both cold and hot springs, holding different mineral in- gredients in solution, became more numerous during the successive convulsions attending this develop- ment of volcanic agency, and thus deposits of calcium carbonate and sulphate, with silica, and other sub- stances were produced. The sub- terranean movements may then have continued until they altered the relative levels of the country, and caused the waters of the lakes to be drained off, and the further accumulation of regular freshwater strata to cease. Oligrocene mammalia of the liimagrne. It is scarcely possible to determine the age of the oldest part of the freshwater series of the Limagne large masses both of the sandy and marly strata being devoid of fossils. Some of the lowest beds may be of Upper Eocene date, although, according to Pomel, only one bone of a Palieotherium has been discovered OH. XV.] OLIGOCENE OF BELGIUM 225 in Auvergne. But in Velay, in. strata containing some species of fossil mammalia common to the Limagne, no less than four species of Palceotlierium have been found by Aymard, and one of these is generally supposed to be identical v/ithPalcEOtherium magnum, Cuv., an undoubted Upper Eocene fossil, of the Paris gypsum the other three being peculiar to the Limagne. Not a few of the other mammalia of the Limagne belong undoubtedly to genera and species elsewhere proper to the Oligocene. Thus, for example, the Caino- therium of Bravard, a genus not far removed from the Anoplo- therium, is represented by several species. A small species of rodent, of the genus Titanomys of Meyer, is common to the Oligocene of Mayence and the Limagne d'Au- vergne, and a remarkable car- nivorous genus, the Hycenodon to the duck, stork, and many of the swallow tribe; also several kinds of pheasants and species of trogon and parrot, birds which are now confined to Asia and the tropics of both hemispheres . Oligocene of Belgium: Tongriar and Rupelian. These strata are marine and fluvio- marine, and are well developed near Tongres, in Limbourg. The Middle Oligocene, or Rupelian, includes the Marine series of the Bolderberg and Argile de Boom (so called from the village of Boom) and of Rupelmonde, south of Ant- werp, which cover a fluvio-marine group with Ccrithium and Pcctun- culus and the Argile de Henis. The lower part of the Tongrian includes the sands in the neighbour- hood of Tongres, and is the con- tinuation of the Lower Oligocene, or Egeln series of Germany, corre- sponding with the upper part of Fig. 259. Zeda Deshayesiana, Duch.. nat. size. of Laizer, is represented by more than one species. The same genus has also been found in the marls of Hordwell Cliff, Hampshire, just below the level of the Bembridge Limestone, and therefore in a formation of about the same age as the gypsum of Paris. Several spe- cies of opossum (Didelpliys) are met with in the same strata of the Limagne. The total number of mammalia enumerated by Pomel as appertaining to the Oligocene fauna of the Limagne and Velay, falls little short of a hundred, and with them are associated some large crocodiles and tortoises, and some ophidians and batrachians. The birds of the Limagne and those of the Mayence basin are, according to Milne Edwards, almost identical. Among those of the Limagne are extinct species related the Gypseous series of Montmartre, and with the Headon series of England. Having this base, it is not diffi- cult to comprehend the extension of the overlying middle Oligocene, The Argile de Henis is equivalent to the green clays with GyrencL of the Mayence basin, with the de- posits at Bembridge in the Isle of Wight, and with the upper Mont- martre green marls which overlie the Gypseous series. The deposits of Klein- Spauwen, a village to the west of Maestricht, which are above the Henis clay, are of the same age as the Gres de Fontainebleau and as our Hemp- stead series. The Upper, or Marine division of the Middle Oligocene of Belgium, with the Argile de Boom and the Bolderberg sands, is younger than Q 226 MIOCENE OF FRANCE [CH. xv. the Geptarien-Thon of Germany but older than the Upper Lacus- trine series of the Calcaire de la Beatice of France. Halitherium is found in the Middle Oligocene, and the teeth of Carcharodon, Myliobates, Lam- na, and other sharks are common to it and the Lower Oligocene, or Tongrian. Many small Crustacea are found in the Middle series, and a fossil lobster (Homarus). The Nau- tilus (Aturia ziczac, Sow.) occurs in the upper deposit, and many Gastropoda are found, some being Lower and others Upper Oligocene forms. LedaDeshayesiana, Duch. (fig. 259), is common to the Lower and Middle series, and Cerithium plicatum, Lam., is found in the Middle series. The Miocene strata of France. Faluns of Tou- raine. The strata which we meet with next in the ascending order are those which have no represen- tatives in the British Islands, and were called by some geologists 'Middle Tertiary;' in 1838 the name of Miocene was proposed for these strata, the ' faluns ' of the valley of the Loire in France being selected as a type. The name ' faluns ' is given provincially by French agri- culturists to shelly sand and marl spread over the land in Touraine, just as the similar shelly deposits called Crag were formerly much used in Suffolk to fertilise the soil, before the coprolitic or phosphatic nodules came into use. Isolated, masses of such faluns occur from near the mouth of the Loire, in the neighbourhood of Nantes, to as far inland as a district south of Tours. They are also found at Pontlevoy, on the Cher, about seventy miles above the junction of that river with the Loire, and thirty miles SJE. of Tours. Deposits of the same age also appear under diffe- rent mineral conditions near the towns of Dinan and Rennes, in Brittany. The scattered patches of faluns are of slight thickness, rarely exceeding fifty feet; and between the district called La Sologne and the sea they repose on a great variety of older rocks ; being seen to rest successively upon gneiss, clay-slate, various secondary formations (including the chalk), and lastly, upon the upper fresh- water limestone of the Parisian tertiary series, which, as before mentioned, stretches continuously from the basin of the Seine to that of the Loire, and which is of Oligocene age. Fragments of this limestone are included in the ' faluns.' At some points, as at Louans, south of Tours, the shells are stained with ferruginous matter, not unlike those of the Red Crag of Suffolk. The species are, for the most part, marine, but a few of them belong to land and fluviatile genera. Remains of terrestrial quadrupeds are here and there intermixed, belonging to the genera Dinotherium (fig. 161, p. 177), Mastodon, Rhinoceros, Hippopo- tamus, Cheer opotamus, Dichobune, Cervus, and others, and these are accompanied by Cetacea of extinct species. The molluscan fauna of the faluns indicate a moderate depth of water and a climate warmer than that of Europe at the present time. Thus it contains seven species of Cyprcea, some larger than any existing cowry of the Mediterranean, several species of Oliva, Ancillaria, Mitra, Terebra, Pi/rula, Fasciolaria, and Conns. The genus Nerita, and many others, are also represented by individuals of a type now charac- teristic of equatorial seas, and wholly unlike any Mediterranean forms. These proofs of a more elevated temperature ?eem to im- ply the higher antiquity of the faluns as compared with the Suffolk Crag, and are in perfect accordance with the fact of the smaller pro- portion of mollusca of recent species found in the faluns. The principal grounds for re- ferring the French faluns to tho Miocene epoch is the fac that the recent species are in a decided minority as compared with the fossil forms; and most of the falunian shells of living species are now inhabitants of the Medi- terranean, the coast of Africa, and PLIOCENE OF FKANCE 227 the Indian Ocean ; in a word, these falunian shells present a less northern character, and point to the prevalence of a warmer climate. They indicate a state of things farther removed from the present condition of Central Europe in physical geography and climate, and doubtless, therefore, receding farther from our era in time. The Miocene strata of Bordeaux and South of France. A great extent of country between the Pyrenees and the Gironde is overspread by tertiary deposits of various ages and chiefly of Miocene date. Some of these, near Bordeaux, co- incide in age with the faluns of Touraine, alr-ady mentioned, but many of the species of shells are peculiar to the south. The suc- cession of beds in the basin of the Gironde implies several oscillations of level by which the same wide area was alternately converted into sea and land or into brackish- water lagoons, and finally into fi'eshwater ponds and lakes. Among the freshwater strata of this age near the base of the Pyrenees are marls, limestones, and sands, in which the eminent comparative anatomist, M. Lartet, obtained a great number of fossil mammalia common to the faluns of the Loire and the Miocene beds of Switzerland, such as Dino- therium glyanteum, Kaup., and Mastodon angustidens, Cuv. More recently M. Gaudry has enumerated 16 species of vertebrata from strata of this age at Mont Leberon in Vaucluse, among which are Ma- chairodus cultridens, Cuv. , Rhino- ceros Sclileiermacheri, Kaup , Di- notherium giganteum, Kaup., and the gigantic ruminant Helladothe- rium Duvernoyi, Gaud, et Lart., rivalling the Giraffe in stature. This herbivore had a wide range over Europe and Asia, its remains having been found in Greece and India. But the most remarkable of all the remains found in the Miocene strata of the South of Prance were the bones of Quadru- mana, or of the ape and monkey tribe, which were discovered by M. Lartet in 1837. They were referred by MM. Lartet and de Blainville to a genus closely allied to the Gibbon, to which they gave the name of Pliopithecus. In 1856, M. Lartet described another species of the same family of long-armed apes (Hylobates), which he obtained from strata of the same age at Saint-Gaudens in the Haute- Garonne. The fossil remains of this animal consisted of a portion of a lower jaw with teeth and the shaft of a humerus. It is supposed to have been a tree-climbing fru- givorous ape, equalling Man in stature. As the trunks of oaks are common in the lignite beds in which it lay, it has received the generic name of Dryopitfacus. Pliocene of Prance. There is some difficulty in distinguishing the scattered beds of this age in France from those of the Miocene ; but in some instances there is un- conformity between the two series. Some of the deposits of Pliocene age are marine, but the majority are of freshwater and terrestrial origin. At Dixmerie, in Brittany, there is a sandy deposit in which are fossil shells of species found in the British Crag, but mixed with a preponderance of Miocene forms. In Roussillan a marine deposit is found containing similar shells. The sands of Landes appear to be of Pliocene age. In the Cotentin there are marls with marine shells and bones of Halitherinm. These are all deposits of the age of the Crags, but, owing to the localities being more to the south, the north- ern element of the molluscan fauna does not predominate in them. The mammalian fauna of the period was part of a very important continental assemblage of animals, and whilst some of the deposits are of the age of the Forest-bed and Norwich Crag, others are older, and approach the Miocene. At Saint-Prest, near Chartres, the characteristic Pliocene elephant (Elephas meridionaiis, Nesti) is found, with liliinoceros etruscus, Falc., and Trogontherium, asso- ciated with Hippopotamus major, Nesti. At Montpellier a marine deposit overlies sand with a fossil monkey Q2 228 OLIGOCENE OF GERMANY [CH. XV. Semnopithecus monspessulanus, Gerv., Mastodon, Rhinoceros me- garhinus, Christol., Tapirus, Hyce- na, Felis, Lutra, Lagomys, Sus, Cervus, Antilope, and Hy&narctos. In the Auvergne, numerous species of deer, a few antelopes, and Elephas, Hippopotamus, Hycena, Hipparion, and Machairodus have been found. In the valley of the Saone, deposits contain Elephas me- ridionalis, Nesti, E. antiquus, Falc., Mastodon arvernensis, Croiz. et Job., M. Borsoni, Hays, Equus stenonis, Cocchi; and in the Li- magne the same Mastodons were accompanied by Rhinoceros, Ma- chairodns, Tapirus, and Antilope. Count Saporta has examined the flora of the Older Pliocene of Maximieux, near Lyons, and found the genera Bambusa, Liquidam- bar, Liriodendron, Acer, Glypto- strobus, Magnolia, Populus, and Salix There was a marked abun- dance of evergreens, which gives the flora a southern aspect ; but with a diminishing mean tempera- ture, the flora became transitional between that of the Miocene and the present day. The Pliocene deposits of Belgium, as now limited by Mour- lon, consist of a lower division the Diestien, at the base of which are sands with great quantities of bones of Cetaceans with excessive elongation of the head (Hetero- cetacece). On the ferruginous sands of this system rest sands with Isocordia cor, L., covered by others with Fusus contrarius, Sow. (Trophon antiquum, Mull.). These two last groups compose the Scaldisian system, and contain a vast quantity of Cetacean remains, with those of fish and also shells. The base of the Diestien is the Black Crag, or Antwerp Crag, which is considered to be a passage bed between the Miocene and Pliocene formations. It is rich in Cetacean bones (see p. 189). CAINOZOIC STRATA OF CENTRAL EUROPE Oligocene of Germany. The division of the Oligocene was first established by the study of strata in North Germany. Pro- fessor Beyrich has made known to us the existence of a long succession of marine strata in North Germany, which lead, by an almost gradual transition, from beds of Lower Oligocene age to others of the age of the Upper Miocene. Although some of the German lignites called Brown Coal belong to the upper parts of this series, others of them are referred to the Lower Miocene and many to the Lower and Middle Oligocene. Professor Beyrich con- fines the term ' Miocene ' to those strata which agree in age with the f aluns of Touraine, and he proposed the term ' Oligocene ' for the older formations of the district, including some formerly classed with the Upper Eocene as well as those called Lower Miocene by earlier authors. Oligocene beds of marine and freshwater origin occupy depres- sions and detached areas which present very distinct faunas and lloras. The Lower Oligocene is marine above. The marine beds of Egeln, with corals and mollusca, cover an amber-bearing glauconitic sand the amber containing many beau- tifully preserved insects and at the base of all are conglome- rates and clays and pitch-coal the Lower Brown Coal series. The flora is largely composed of Conifers, Oaks, Laurels, Magnolia, Dryandroides, Ficus, with Sabcu, Flabellaria, and other Palms. The facies is subtropical and North American, with some Indian and Australian types. The Middle Oligocene is the Sep- tarien-Thon, with Leda Deshaye- siana, Duch. (see fig. 259, p. 225), and in some places plants are found forming local Brown coals. The upper deposits are Brown coals, found in the Lower "Rhine district, and the flora contains the genera Acer, Cinnamomum, Juglans, Nyssa, Pinites, Quercus, having a sub-tropical American facies. Some CH. XV.] MAYENCE BASIN 229 marine beds contain Terebratula grandis, Blumenb. Mayence basin. An elaborate description has been published by Dr. F. Sandberger of the Mayence tertiary area, which occupies a tract from five to twelve miles in breadth, extending for a great distance along the left bank of the Rhine from Mayence to the neighbourhood of Mannheim, and is also found to the east, north, and south-west of Frankfort-on-Main. Hilmersdorf. the lowest is the marine sand of Weinheim. The Miocene of the Mayence Basin. This under- lies the bone-bed of Eppelsheim, and is fluviatile, estuarine, and terres- trial in its nature. The beds contain a fauna which differs from that of Eppelsheim, none of the genera being identical ; Dinotherium, Pa- Iceomeryx, Micro tlierium, Hippo- therium occur in them. Amongst the shells are Drcissena, Mytilus, New Brandenburg, Section through the basin of Berlin, conite sands, &c.). c. Middle Olig Oligocene (sands). /. Miocene (Lignite or Berendt and Kayser.) . Older Kpcks. 6. Lower Oligocene (Glau- me (Septariaclay). d. Stettin sand. e. Upper Brown Coal d Coal deposits), g. Drift. (After De Koninck, of Liege, has pointed out that the purely marine portion of the deposit contained many species of shells common to the Klein-Spauwen beds and to the clay of Rupelmonde, near Antwerp. The deposits underlie the sand- stones with leaves and the Ceri- thiuni limestones of the Miocene, and may be divided into three groups. The upper is a Cyrena marl with Cyrena semistriata, Desh., and Cerithium plicatum, Lam.; the middle is a clay with Leda Deshayesiana, Duch. ; and and Littorinella. Among the plants are Sabal and Cinnamomum. The Miocene rests on the Cyrena marl of Oligocene age. German Pliocene. At Ep- pelsheim, near Worms, there is a group of sands and gravel with lignite, containing mammalian re- mains, overlying a freshwater for- mation of later Miocene or Older Pliocene age. The mammalia be- long to the genera Dinotherium, Mastodon, Rhinoceros, Hippo- therium, Sus, Felis, and Cervus. CAINOZOIC STRATA OF THE ALPINE DISTRICTS AND SWITZERLAND Nummulitic Formation of Southern Europe, Asia, &c. In the Alps and Southern Europe generally, the Lower Eocene is represented by the sandy and argillaceous strata known asFlysch and Macigno, which contain few fossils except impressions of fu- coids ; while the upper part of the Eocene is developed on a grand scale, and contains beds of lime- stones crowded with Nummulites. To these strata the name of Nummulitic is given. 230 THE MOLASSE OF SWITZERLAND [CH. xv. Of all the rocks of the Eocene period, no formations are of such great geographical importance as the Upper and Middle Eocene, or Nummulitic. Separate groups of strata are often characterised by distinct species of Nummulites. The nummulitic limestone of the Swiss Alps rises to more than 10,000 feet above the level of the sea, and attains here and in other mountain-chains a thickness of several thousand feet. It may be said to play a far more conspicuous part than any other Tertiary group in the solid framework of the earth's crust, whether in Europe, Asia, or North Africa. It occurs in Algeria and Morocco, and has been traced from Egypt, where it was largely quarried of old for the building of the pyramids, into Asia Minor, and across Persia by Bagdad to the mouths of the Indus. It has been observed not only in Cutch, but in the mountain-ranges which separate Sind from Persia, and which form the passes leading to Cabul; and it has been followed still farther eastward into India, as far as Eastern Bengal amongst the Himalayas, and the frontiers of China. Dr. T: Thomson found Nummu- lites in Western Thibet at an ele- vation of no less than 16,500 feet above the level of the sea. One of the species, which occurs very abun- dantly on the flanks of the Pyrenees, in a compact crj r stalline marble, is called by M. d' Archiac Nummulites Puschi (fig. 183, p. 194). The same is also very common in rocks of the same age in the Carpathians. When we have once arrived at the conviction that the Nummulitic formation occupies a middle and upper place in the Eocene series, we are struck with the compara- tively modern date to which some of the greatest revolutions in the physical geography of Europe, Asia, and Northern Africa must be re- ferred. All the mountain-chains, such as the Alps, Pyrenees, Car- pathians, and Himalayas, into the composition of whose central and loftiest parts the Nummulitic strata enter bodily, could have had no such altitude till after the Middle Eocene period. During that period the sea mainly prevailed where these chains now rise, for the Nummulites were unquestionably inhabitants of salt water. The Lower Ittolasse of Switzerland (Aquitanian). In Switzerland the Nummulitic formation is covered by great de- posits of Oligocene, Miocene, and Pliocene age. These strata, which are of great thickness and include deposits of marine, brackish-water, and freshwater origin, are called by the Swiss geologists Molasse. The Miocene or Molasse Forma- tion of Switzerland consists of the following members : 1. The Upper Freshwater Mo- lasse, including the Lacustrine Marls of Oeningen. 2. The Marine Molasse corre- sponding in age with the faluns of Touraine, 3. The Lower Freshwater Molasse. Nearly the whole of this Lower Molasse is freshwater ; yet some of the inferior beds contain a mixture of marine and fluviatile shells, the Ceritliium plicatum, Lam., a well-known Oligocene fossil, being one of the marine species Notwithstanding, there- fore, that some of these Oligocene strata consist of old shingle beds several thousand feet in thickness, as in the Rigi near Lucerne, and in the Speer near Wesen, forming mountains 5,000 and 7,000 feet above the sea, the deposition of the whole series must have begun at or below the sea-level. The conglomerates, as might be expected, are often very unequal in thickness in closely adjoining dis- tricts ; since in a littoral formation accumulations of pebbles would swell out in certain places where rivers entered the sea, and would thin out to comparatively small dimensions where no streams, or only small ones, came down to the coast. For ages, in spite of a gradual depression of the land and adjacent sea-bottom, the rivers continued to cover the sinking area with their deltas ; until finally, the subsidence being in excess, the sea of the Middle Molasse gained upon CH. XV.] MIDDLE AND UPPER MOLASSE 231 the land, and marine beds were thrown down over the dense mass of freshwater and brackish-water deposit, called the Lower Molasse, which had previously accumulated. Flora of the lower Molasse. In part of the Swiss Molasse which belongs exclusively to the Oligocene Period, the number of plants has been esti- mated at more than 500 species. The series may best be studied on the northern borders of the Lake of Geneva between Lausanne and Vevay. The strata there consist of alternations of conglomerate, sandstone, and finely laminated marls with fossil plants. The flora contains, according to Heer, 103 species, of which 81 pass up into the flora of Oeningen. The proofs of a warmer climate and the excess of arborescent over herbaceous plants and of evergreen trees over deciduous species, are characters common to the whole flora, which are intensified as we descend to the inferior deposits. Among the Conif erse the Seq^lo^a is common at Rivaz, and is one of the most universally distributed Elants in the Oligocene of Switzer- ind. Lastrcea stiriaca, Unger, has a wide range in the Tertiary period from strata of the age of Oeningen to the lowest part of the Swiss Molasse. In some specimens, as shown in fig. 192, p. 199, the fructi- fication is distinctly seen. Among the laurels several species of Cinnamomum are very conspicuous. Besides C. Poly- morphum, Ad. Brong., before figured (fig. 168, p. 181), another species also ranges from the Lower to the Upper Molasse of Switzer- land, and is very characteristic of different deposits of Brown Coal in Germany. It has been called Cin- namomum Rossmassleri, Heer. (See fig. 196, p. 200.) American character of the flora. If we consider not merely the number of species but those plants which constitute the mass of the Oligocene vegetation, we find the European part of the fossil flora very much less prominent than in the Oeningen beds, while much more conspicuous are Ameri- can forms, such as evergreen oaks, maples, poplars, planes, Liquid- ambar, liobinia, Sequoia, Taxo- dium, &c. There is also a much greater fusion of the characters now belonging to distinct botanical pro- vinces than in the Miocene flora, and we find this fusion still more strikingly exemplified when we go back to the antecedent Eocene and Cretaceous periods. Middle or Marine Molasse of Switzerland ( Helvetian '). Some of the beds of the marine or middle series reach a height of 2,470 feet above the sea. A large number of the shells are common to the f aluns of Touraine, the Vienna basin, and other Upper Miocene localities. The terrestrial plants play a subordinate part in the fossi- liferous beds, yet more than ninety of them are enumerated by Heer as belonging to this Falunian division, and of these more than half are common to the subjacent lower molasse, while a proportion of about 45 in 100 are common to the overlying Oeningen flora; 26 of the 92 species are peculiar. Remains of an ape (Dryopithecus) have been found in these beds. Upper Miocene fresh- water Molasse. Thisformation is best seen at Oeningen, in the valley of the Rhine, between Con- stance and Schaffhausen, a locality celebrated for having produced in the year 1700 the supposed human skeleton called by Scheuchzer ' homo diluvii testis,' a fossil after- wards demonstrated by Cuvier to be a reptile, or aquatic salamander, of larger dimensions than even its great living representative the salamander of Japan. The Oeningen strata consist of a series of sandstones, marls, and limestones, many of them thinly laminated, which appear to have slowly accumulated in a lake pro- bably fed by springs holding cal- cium carbonate in solution. All the fossil-bearing strata of Oeningen were evidently formed with txtreme slowness. Although they aic in the aggregate not more than a few yards in thickness, and have only baen examined in a small 232 ITALIAN TEKTIARIES [CH. XV. area, they give us an insight into the state of animal and vegetable life in part of the Miocene period, such as no other region in the world has presented us with. In the year 1859, Prof. Heer had already determined no less than 475 species of plants and more than 800 insects from these Oeningen beds. He sup- posed that a river entering a lake floated into it some of the leaves and land insects, together with the car- cases of quadrupeds, among others that of a great Mastodon. Occa- sionally, during tempests, twigs and even boughs of trees with their leaves were torn off and carried for some distance so as to reach the lake. Springs, containing calcium carbonate, seem at some points to have supplied calcareous matter in solution, and to have thus formed a tufaceous limestone or travertine. The Upper Miocene flora of Oeningen contains some tropical forms, like Palms, Cinnamomum, and Vines, with leaves and fruits of trees like Acer, and Platanus ; the cones and leaves of pines such as Glyptostrobus ; and forms re- ferred by many botanists to the Proteacese. The conclusions drawn from the insects are for the most part in perfect harmony with those derived from the plants, but they have a somewhat less tropical and less American aspect, the South Euro- pean types being more numerous. On the whole, the insect fauna is richer than that now inhabiting any part of Europe. No less than 844 species were enumerated by Heer from the Oeningen beds alone, the number of specimens which he examined being 5,080. Nearly all the species belong to existing genera. Almost all the living families of Coleoptera are repre- sented ; but, as we might have an- ticipated from the preponderance of arborescent and ligneous plants, the wood-eating beetles play the most conspicuous part, the Bupres- tidse and other long-horned beetles being particularly abundant. The patterns and some remains of the colours both of Coleoptera and Hemiptera are preserved at Oenin- gen (fig. 173, p. 183). CAINOZOIC STRATA OF THE ITALIAN PENINSULA It is in the Italian peninsula and in Sicily that we find the grandest development of the Newer- Ter- tiary strata. Olig ocene of Italy. In the hills of which the Superga, near Turin, forms a part, there is a great series of Tertiary strata which pass downwards into the Oligocene. Even in the Superga itself there are some fossil plants which, accord- ing to Heer, have never been found in Switzerland so high as the marine Molasse. In several parts of the Ligurian Apennines, as at Dego and Carcare, the Oligocene appears containing some Nummu- lites, and at Cadibona, north of Savona, freshwater strata of the same age occur, with beds of lig- nite enclosing remains of the An- thracotheriiim magnum, Cuv., and A. minus, Cuv., besides other mammalia enumerated by Gas- taldi. In these beds a great num- ber of the Oligocene plants of Switzerland have been discovered. The Marine Oligocene is of great importance, containing only few Nummulites, but a most interesting reef-building coral fauna. Miocene strata of Italy. We are indebted to Signer Miche- lotti for a valuable work on the Miocene shells of Northern Italy. Those found in the hill called the Superga, near Turin, have long been known to correspond in age with the faluns of Touraine, and they contain so many species common to the Miocene strata of Bordeaux as to lead to the conclusion that there was a free communication between the northern part of the Mediterra- nean and the Bay of Biscay during the Miocene Period. In the adjoin- ing hills to the Superga, these Tertiary strata pass down into the Oligocene. Older Pliocene of Italy. Subapennine strata. The Apennines as is well known, are composed chiefly of Secondary or Mesozoic rocks, forming a chain OH. XV.] SUBAPENNINE STRATA 233 which branches off from the Li- gurian Alps and passes down the middle of the Italian peninsula. At the foot of these mountains, on the side both of the Adriatic and the Mediterranean, are found low hills occupying the space between the older chain and the sea. Their strata belong both to older and newer members of the tertiary series. The strata, for example, of the Superga, near Turin, are Miocene ; those of Asti and Parma Older Pliocene, as is the blue marl of Siena ; while the shells of the overlying yellow sand of the same territory approach more nearly to the recent fauna of the Mediterra- nean, and may be Newer Pliocene. We have seen that most of the fossil shells of the Older Plio- cene strata of Suffolk which are of recent species are identical with mollusca now living in British seas, yet some of them belong to Mediterranean species, and a few even of the genera are those of warmer climates. We might there- fore expect, in studying the fossils of corresponding age in countries bordering the Mediterranean, to find some species and genera of warmer latitudes among them. Accordingly, in the marls belonging to this period at Asti, Parma, Siena, and parts of the Tuscan and Roman territories, we observe the genera Conus, Cyprcea, Strombus, Py- rula, Mitra, Fasciolaria, Sigare- tus, Delphinula, Ancillaria, Oliva, TerebeUum, Terebra, Perna, Pli- catula, and Corbis, some charac- teristic of tropical seas, others represented by species more numerous or of larger size than those now proper to the Medi- terranean. Older Pliocene flora of Italy. The Val d'Arno blue clays, with some subordinate layers of lignite, exhibit a richer flora than the overlying Newer Pliocene beds, and one receding farther from the existing vegetation of Europe. They also comprise more species common to the antecedent Miocene period. Among the genera of flowering plants, M. Gaudin enumerates pine, oak, ever- green oak, plum, plane, alder, elm, fig, laurel, maple, walnut, birch, buckthorn, hiccory, sumach, sarsa- parilla, sassafras, cinnamon, Glyp- tostrobus, Taxodium, Sequoia, Persea, Oreodaphne, Cassia, and some others. This assemblage of plants indicates a warm climate, but not so subtropical a one as that of the Upper Miocene period. Newer Pliocene strata of Sicily. At several points north of Catania, on the eastern sea-coast of Sicily as at Aci-Castello, Trezza, and Nizzeti for example marine strata, associated with volcanic tuffs and basaltic lavas, are seen, which belong to a period when the first igneous eruptions of Mount Etna were taking place in a shallow bay of the Mediterranean. They con- tain numerous fossil shells, and out of 142 species that have been collected, all but eleven are identi- cal with species now living. Some few of these may possibly still linger in the depths of the Medi- terranean. There is probably no part of Europe where the several Pliocene formations enter so largely into the structure of the earth's crust, or rise to such heights above the level of the sea, as Sicily. They cover nearly half the island, and near its centre, at Castrogiovanni, reach an elevation of 3,000 feet. Seguenza has divided the deposits into three groups, the oldest or Zanclean being composed of marls and lime- stones. Many tropical shells are found, and out of 504 species about 17 per cent, only are found living in the Mediterranean. Large ti'opical shells and many littoral and deep-sea corals and fo- raminifera are found in this series. On the top of the Zanclean are blue clays followed by Ostian yellow sands. The Zanclean is Older Plio- cene, and the superincumbent strata are Newer Pliocene. South of the plain of Catania is a region in which the tertiary beds are intermixed with volcanic matter, which has been for the most part the product of submarine eruptions. It appears that, while the Newer Pliocene strata were in course of deposition at the bottom of the sea, volcanoes burst out 234 VAL D'ARNO BEDS [OH. xv. beneath the waters, like that of Graham Island in 1831, and these explosions recurred again and again at distant intervals of time. Volcanic ashes and sand were showered down and spread, by the waves and currents, so as to form strata of tuff, which are found intercalated between beds of lime- stone and clay containing marine shells, the thickness of the whole mass exceeding 2,000 feet. No shell is more conspicuous in these Sicilian strata than the great scallop, Pecten Jacobceus, L., (fig. 155, p. 175), now so common in the neighbouring seas. The more we reflect on the preponderating number of this and other recent shells, the more are we surprised at the great thickness, solidity, and height above the sea of the rocky masses in which they are entombed, and the vast amount of geographical change which has taken place since their origin. Newer Pliocene strata of tlie Upper Val d' Arno. When we ascend the Arno for about 10 miles above Florence, we arrive at a deep, narrow valley, called the Upper Val d'Arno, which appears to have been a lake at a time when the valley below Florence was an arm of the sea. The horizontal lacustrine strata of this upper basin cover an area 12 miles long and 2 broad. The depression which they fill has been excavated out of Eocene and Cretaceous rocks, which form everywhere the sides of the valley and exhibit highly inclined stratification. The thick- ness of the more modern and un- conformable beds is about 750 feet, of which the upper 200 feet consist of Newer-Pliocene strata, while the lower are Older Pliocene. The newer series are made up of sands and a conglomerate called ' sarsino." Cocchi has found a Macacus in them, and a second species has been discovered by Forsyth Major, and these are amongst the last fossil Monkeys of Europe. Among the embedded fossil mammalia are Mastodon arvernensis, Croiz. et Job., Elephas meridionalis, Nesti, Rhinoceros etruscus, Falc., Hippopotamus major, Nesti, and remains of the Bear, Hyaena, and of Felidae, nearly all of which occur in the Cromer forest bed. In the same upper strata are found, according to Graudin, the leaves and cones of a Glyptostrobus closely allied to one now inhabiting the north of China and Japan. This conifer had a wide range in time, having been traced back to the Oligocene strata of Switzerland and being common at Oeningen in the Upper Miocene. CAINOZOIC STRATA OF EASTERN EUROPE In Eastern Europe and the Vienna basin we find great deposits of sandstone and shale, of Eocene age, known as the Flysch. It is poor in fossils, but sometimes con- tains enormous numbers of erratic blocks of granite, gneiss, and other old rocks, which appear to have come from Central Europe. The beds are believed by some geologists to have in part at least a glacial origin. Oligocene beds of Croatia. The Brown Coal of Radaboj, near Agram, in Croatia, not far from the borders of Styria, is covered, says Von Buch, by beds containing the marine shells of the Vienna basin, The strata correspond in age to the Middlj Oligocene of Belgium. They have yielded more than 200 species of fossil plants, described by the late Pro- fessor Unger. These plants are well preserved in a hard marlstone, and contain several palms ; among them Sabal (fig. 198, p. 200), and another genus allied to the date- palm. The only abundant plant among the Radaboj fossils which is characteristic of the Miocene period is the Populus mutabilis, Heer, whereas no less than fifty of the Radaboj species are common to the more ancient flora of the Lower Molasse or Oligocene of Switzer- land. The insect fauna is very rich CH. XV.] VIENNA BASIN 235 and, like the plants, indicates a more tropical climate than do the fossils of Oeiiingen already men- tioned. There are ten species of Termites, or White ants, some of gigantic size, and large dragon-flies with speckled wings, like those of the Southern States in North America ; there are also grass- hoppers of considerable size, and even the Lepidoptera are not un- represented. (See fig. 197, p. 201.) In the Vienna basin we find strata possibly as old as the Upper Oligocene, with Cerithiuni pli- catum, Lam., in the marine layers and Melania in the freshwater de- posits. Miocene beds of the Vienna basin. In South Ger- many the general resemblance of the Vienna basin the remains of several mammalia have been found, and among them a species of Dino- therium, a Mastodon of the Trilo- phodon division, a Rhinoceros (allied to R. megarhinus, Christol), also an animal of the hog tribe, (Listriodon, Von Meyer), and a carnivorous animal of the canine family. The Helix turonensis, Desh. (fig. 53, p. 56), the most common terrestrial shell of the French faluns, accompanies the above land animals. The flora of the Vienna basin exhibits some species which have a general range through the whole Miocene period. There are two main divisions in the Miocene tertiaries of the great area called the Vienna basin. Fig. 261. Vienna Diagrammatic Section through the Vienna Basin. (After Karrer and Kayser.) a. Crystalline rocks of the Leitha Mountains, b. Flysch (Eocene) of the Vienna Hills, c. Marine Miocene (Mediterranean series), d. Brackish-water, Upper Miocene (Sarmatian series), e. Pliocene (Congeria Beds). /. Drift, g. Allu- vium. the shells of the Vienna tertiary basin to those of the faluns of Touraine has long been acknow- ledged. According to Professor Suess, the most ancient and purely marine of the Miocene strata in this basin consist of sands, conglomerates, limestones, and clays, and they are inclined inwards, or from the borders of the trough towards the centre, their outcropping edges rising much higher than the newer beds, whether Mio- cene or Pliocene, which overlie them, and which occupy a smaller area at an inferior elevation above the sea. Dr. Homes has described no less than 500 species of Gastro- poda, of which he identifies one- fifth with living species of the Mediterranean, Indian, or African seas. In the lowest marine beds of Sandstones, limestones, and clays, with Cerithia, and vast quantities of a few species of Tapes, Mactra, Murex, &c. Corals and Bryozoa are rare. This Sarmatian division covers a marine group with lime- stones crowded with corals the Leithakalk which was deposited at a period when a subtropical climate prevailed. Pliocene strata of the Vienna basin. The Congerian strata, which contain vast numbers of Congeria subglobosa, Partsch, are sands with the bones of large ani- mals overlying a clay of 300 feet in thickness. The fossils indicate Cas- pian conditions, rather than those of an open sea, and show that there was an inland gulf, with its water gradually becoming brackish and fresh. As might be expected, deposits 286 PIKERMI- AND SIVALIK-BEDS [en. xv. of rock-salt, gypsum, and anhy- drite occur in this formation, the result of evaporation of the old sea. The mammalia belonged to the genera Dinotherium, Mastodon, Acerotherium, Rhinoceros, Hippo- therium, Machairodus, Hycena, Cervus, and Antilope, The flora includes conifers of the genera Sequoia, Pinus, Glyptostrobiis ; with dicotyledons, like the Birch, Alder, Oak, Beech, Chestnut, Horn- beam, Liquidambar, Plane, Laurel, Cinnamon, and forms referred to the Asiatic genus Parrotia, and the Australian Hakea. Older Pliocene formations of Greece. At Pikermi, nar Athens, Wagner and Roth have described a deposit in which they found the remains of a splendid fauna. This fauna attests the former extension of a vast expanse of grassy plains, where we have now the broken and mountainous coun- try of Greece ; and these plains were probably united with Asia Minor, spreading over the area where the deep .ZEgean Sea and its numerous islands are now situated, and ex- tending into Africa. We are in- debted to Gaudry for a treatise on the fossil bones of Pikermi, showing how many data they contribute to the theory of a transition from the mammalia of the Pliocene and Pleistocene to those of living genera and species. For example, he recognised such synthetic types as an Ape (Mesopithecus) intermediate between the living genera Semno- pithecus and Macacus ; a carnivore intermediate between the hyaena and the civet; a pachyderm (Hipparion) intermediate between the Anchitherium and the horse : and a ruminant intermediate between the goat an.d the antilope. The Carnivora belong to the genera Machairodus, several species of Felis, Hycena, Hycenictis, Limno- cyon, Mustela, Ictitherium, Pro- mephitis ; Rodents, Hystrix ; Edentata, Ancylotherium ; Pro- boscidaea, Mastodon, Dinotherium ; Perissodactyla, several species of Rhinoceros, Acerotherium, Lepto- don, Hipparion ; Artiodactyla, Sus, Camelopardalis, Helladotherium, Antilope, Gazella, Palceoryx, Pa- Iceoreas, Dromotherium. A turtle, a Saurian, birds of the pheasant tribe, and a crane have also been found. This remarkable assemblage is characterised by a stron^ African facies. CAINOZOIC STRATA OF INDIA Eocene, Oligocene, and Miocene of India. In Sind we find strata of Miocene age resting upon an important Oligocene series called the Nari series, which contains a characteristic fauna of reef-build- ing corals, and very flat Echinolam- pads, with a few Nummulites. The Oligocene strata rest on the Num- mulitic. Pliocene of India. In the Sind area, the succession of Eocene, Oligocene, and Miocene marine strata is covered by freshwater and terrestrial deposits of great thick- ness, called Manchhar beds. These last are the geological equivalents of the conglomerates, sands, marls, and gravels which flank the Hima- layas on the south, and which are called the Sivalik strata. The latter are terrestrial and freshwater de- posits, and are the results of the denudation of the country during the time when the Himalayas gra- dually rose into a great mountain mass. In the Manchhars the following genera of Vertebrata have been discovered: Amphicyon, a carni- vore; Proboscidaea, Mastodon (three species), Dinotherium ; Perisso- dactyla, Rhinoceros, Acerotherium; Artiodactyla, Sus, Hemimeryx, Sivameryx, Chalicotherium, An- thracotherium, Hyopotamus, Hyo- therium, Dorcotherium ; Edentata, Manis ; Reptilia, Crocodilus, Che- Ionia, Ophidia, &c. The mollusca of the Sivfilik strata, now that the recent forms of India have been studied, turn out to be identical with living forms, or to be closely allied. The genera of Vertebrata are Quadrumana, Ma- cacus, Semnopithecus ; Carnivora, Felis, Machairodus, Pseudaleu- rus, Ictitherium, Hycena, Cani$ en. xv.] GLACIAL DEPOSITS OF NORTHERN EUROPE 237 (vulpes), Amphicyon, Ursus,'Hy(e- narcius, Mellivora (meles), Lutra, Enhydriodon-, Proboscidea, Ele- phas, Mastodon ; Perissodactyla, Bhinoceros, Acerotherium, Lis- triodon, Equus, Hipparion ; Artiodactyla, Hijipopotamus, Hip- popotamodon, Tetrocondon, SHS, Hippohyas, Chalicotherium, Me- rycopotamus, Cervus, Dorca- therium, Camelopardalis, Siva- therium, Hydaspitherium, Bos, Bison, Bubahis, Peribos, Amphi- bos, Hemibos, Antilope, Capra, Ovis, Camelus ; Rodentia, Rhi- zomys, Rystrix ; Eeptilia, Croco- dilus, Ghavialis, Emys, Colosso- chelys. Some of the Siva"lik fauna lived on during the Pleistocene age, and their remains have been found in the river gravels of the Nerbudda and Godaveri, accompanied by im- plements of man's making. POST-PLIOCENE DEPOSITS OF NORTHERN EUROPE AND THE ALPINE DISTRICTS Post-Pliocene deposits of glacial origin, more or less similar to those of Western Europe, which we have described, are found all over North- ern and Central Europe, and even in the mountainous parts of the south of the continent. As far south as the Harz Mountains and the Riesengebirge we find masses of boulder-clay and glacial sands, of varying thickness up to 400 feet. They are full of erratic blocks of granite, gneiss, and other rocks, some of which can be' identified as having come from Scandinavia, while others are of more local origin. On the moun- tains of Central Europe these erratic blocks are sometimes found at, heights of from 1,200 to?- 1,500 feet. The rocks on which these glacial deposits lie are often much striated ; they present great pot- holes ('giant kettles') formed by glacier mills, and there are easily recognisable moraines which are of great length and height. Beds of clay and sand containing marine shells, sometimes of very arctic character (Yoldia or Leda clays), are found, with others containing more temperate forms which are believed to represent pre-glacial or interglacial deposits. The German geologists classify their Post-Plio- cene deposits as follows : Post-Glacial. Upper sands. Neiver Glacial. Upper Boul- der Clay (yellow). Interglacial. Middle sands (with mammalian remains), con- taining intercalated bands of calca- reous tufa. Older Glacial. Lower Boulder Clay (blue) Pre-GlaciaL Stratified Sands and Clays, sometimes containing marine shells. The Glacial Deposits of Scandinavia and Russia. Ins? large tracts of Norway and Sweden, where there have been no glaciers in historical times, the signs of ice-action have been traced as high as 6,000 feet above the level of the sea. These signs con- sist chiefly of polished and furrowed rock - surfaces, of? /noraines and erratic blocks. The direction of the erratics, like that of the furrows, has usually been conformable to the course of the principal valleys ; but the lines of both sometimes radiate outwards in all directions from the highest land, in a manner which is only explicable by the hypothesis of a general envelope of continental ice, like that of Greenland. Some of the far- transpoited blocks have been carried from the central parts of Scandinavia towards the Polar regions; others southwards to Denmark ; some south-westwards, to the coast of Norfolk in England ; other south-eastwards, to Ger- many. In the immediate neighbour- hood of Upsala, in Sweden, there occurs a ridge of stratified sand and gravel, in the midst of which. occurs a layer of marl, evidently formed originally at the bottom of the Baltic, and containing the mussel, cockle, and other marine shells of living species, intermixed 238 FORMER GREATER EXTENT [CH. XV. with some proper to fresh water. The marine shells are all of dwarfish size, like those now inhabiting the brackish waters of the Baltic ; and the marl, in which many of them are embedded, is raised more than 100 feet above the present level of the Gulf of Bothnia. Upon the top of this ridge repose several huge erratics, consisting of gneiss, for the most part unrounded, from 9 to 16 feet in diameter, which must have been brought into their present position since the time when the neighbouring gulf was already characterised by its peculiar fauna. Here, therefore, we have proof that the transport of erratics continued to take place, not merely when the sea was inhabited by the existing mollusca, but when the North of Europe had already assumed that remarkable feature of its physical geography which separates the Baltic from the North Sea, and causes the Gulf of Bothnia to have only one-fourth of the saltness belonging to the ocean. In Den- mark, also, recent shells have been found in stratified beds, closely associated with the boulder clay. The geologists of Sweden and Norway have classified their Post- Pliocene deposits as follows : Post-Glacial. Bedded sands formed during the retreat of the glaciers. Newer Glacial. Upper Boul- der Clay (yellow). Interglacial. Bedded sands and clays with remains of the dwarf birch, &c., best seen in Scania, Southern Sweden. Older Glacial. Lower Boulder Clay (blue). The Asar, corresponding to our eskers or kames, are great ridges composed of sand and pebbles, which run, often in sinuous lines, across the country for many miles, and are sometimes more than 100 feet in height. By some authors they are regarded as being moraines, by others as being accumulated by the waters flowing from the ice- sheets during their retreat. Drift Deposits of Moun- tain Districts. In the higher regions of mountains, where the amount of snow that falls in winter so far exceeds the loss in summer, through melting and evaporation, an indefinite thickness would accumulate if it were not pre- vented by the formation of neve. This becoming gradually converted into ice, the glaciers are fed, and they glide down the principal valleys. On the glaciers' surface, are seen long lines or heaps of sand and mud, with angular frag- ments of rock, which fall in quantities from the steep slopes or precipices on either side, where the rocks are daily exposed to great changes of temperature. These deposits, being arranged along the sides cf the glacier, are termed lateral moraines. When two glaciers meet, unite, and continue their course, the right lateral moraine of the one and the left of the other meet together in the centre of the joint glacier, forming what is called a medial moraine. These surface moraines finally fall, or are dropped at the lower end or foot of the glacier, and form the terminal moraine. Besides the blocks thus carried down on the top of the glacier, many fall, through fissures in the ice, to the bottom, where some of them become firmly frozen into the mass, and are pushed along the base of the glacier, abrading, polishing, and grooving the rocky floor below, as a diamond cuts glass, or as emery powder polishes steel, and the larger blocks are reciprocally grooved and polished by the rocky floor on their lower sides. Stones which have been frozen into the bottom of the glacier scratch the adjacent rocks, produ- cing long striae. The striae and the deep grooves thus made are recti- linear and parallel to an extent never seen in those produced on loose stones or rocks, where shingle is hurried along by a torrent, or by the waves on a sea-beach. At the same time a stream of water, pro- duced by the melting of the ice, issues from beneath the glacier charged with mud, derived, not only from the atmospheric waste of the rocks above, but in part also from*the crushing of the fragments of stone, which have reached the bottom of CH. XV.] OF THE ALPINE GLACIERS 239 the glacier, and the abrasion of its rocky floor. In addition to these polished, striated, and grooved surfaces of rock, another proof of the former action of a glacier is afforded by the 'roches moutonnees,' or pro- jecting eminences of rock which have been smoothed and worn into the shape of flattened domes by the glacier as it passed over them. They have been traced in the Alps to great heights above the present glaciers, and also to great distances below and beyond them. If the glacier is greatly diminished by melting, large angular fragments, which are called 'perched blocks,' are left behind. Alpine blocks on the Jura. The moraines, erratics, polished surfaces, domes, perched blocks, and striaa, above described, are observed in the great valley of Switzerland, fifty miles broad ; and on the Jura, a chain which lies to the north of this valley. The average height of the Jura is about one- third that of the Alps, and it is now entirely destitute of glaciers ; yet it also presents moraines, and polished and grooved surfaces. The erratics, moreover, which are upon it even to a height of 2,500 feet, present a phenomenon which has astonished and perplexed the geologist for nearly a century. No conclusion can be more incontest- able than that these angular blocks of gneiss and other crystalline for- mations came from the Alps, and that they have been brought for a distance of fifty miles and upwards, across a wide and deep valley; so that they are now lodged on hills composed of sedimentary forma- tions. The great size and angularity which the blocks retained, after a journey of so many leagues, has justly excited wonder ; for many of them are as large as cottages ; and one in particular, composed of gneiss, celebrated under the name of Pierre a Bot, rests on the side of a hill about 900 feet above the Lake of Neufchatel, and is no less than 40 feet in diameter. The manner in which these erratics were conveyed from the Alps to the Jura was formerly the subject of considerable contro- versy. Venetz proved that the Alpine glaciers must formerly have extended far beyond their present limits, and it was argued that the blocks now found on the Jura had been transported by their agency. Other writers, on the contrary, conjectured that the whole country had been submerged, and that the moraines and erratic blocks must have been transported by floating icebergs, as it was held that the difference in height between the two mountain ranges was not suf- ficient to have allowed the glaciers to flow from the Alps across the wide valley to the Jura. But the definite order in which the Alpine erratics are arranged, and the total absence of marine shells, have gone far to disprove this last hypo- thesis. Besides, we have no right to assume that the relative heights of the Alps and Jura have remained unaltered since the era of the transportation of the erratics ; still less that the change of level which last took place was uniform over a great district, either in amount or direction. The Palaeolithic Period in Western Europe. Of post- glacial deposits with the remains of man we find many examples in Southern Europe. River-gravels and peat-deposits, like those al- ready described in France and Denmark, occur in Switzerland, Italy, and Southern Germany. Many of these seem to be only a little younger than the glacial for- mations, while certain deposits con- taining human relics are believed by some geologists to be inter- glacial or even pre-glacial in age. On the other hand we have interesting remains in the South of France of a race of men, which, though certainly pre-historic, was younger than the race of which the relics are found in most of our river-gravels and caverns when the mammoth abounded, though this animal, as we shall see, had not entirely disappeared from Southern Europe during the later of the Palaeolithic periods. Newer Palaeolithic Age Reindeer Period. There are 240 OLDER AND NEWER PALEOLITHIC [CH. xv. some caves in the departments of Dordogne, Aude, and other parts of the South of Prance, the contents of which accumulated late in the Palaeolithic period. They are said to belong to the ' reindeer-period,' because vast quantities of the bones and horns of that deer have been met with. In some cases separate plates of molars of the mammoth, and several teeth of the great Irish deer, Cervus megaceros, Hart., and of the cave-lion, F. 267. Ventnculites titfundfbulifermfa, S. Woodw. A siliceous and hexacti- nellid sponge. White Chalk. CH. xvi.] ECH1NODEKMS OF THE CHALK 257 and fine sand like that of a beach, has been shown by Mr. Grodwin- Austen to be inexplicable except by the agency of floating ice. If we consider that icebergs now reach 40 north latitude in the Atlantic, and several degrees nearer the equator in the southern hemisphere, we can the more easily believe that, even during the Cretaceous epoch, assuming that the climate was milder, fragments of coast-ice may have floated occasionally as far as the south of England. Fossils of the several divisions of the Chalk. Among the fossils of the Upper Chalk, echinoderms are very numerous ; and Fig. 268. Echinocorys vulgaris, Breyn. (Ananchytes ovatus, Leske), J. Upper Chalk a. Side view. b. Base of the shell on which both the oral and anal apertures arc placed ; the anal being at the smaller end. Fig. 269. Fig. 270. Fig. 271. Micraster cor-anguinum. Echinoconus conicus, Breyn. Leske, i. Upper White Chalk. (Galerites albogalenis.Ij&mS), J. Upper White Chalk. Fig. 272. Fig. 273. Marsupites Milleri, Mant, |. Upper White Chalk. Fig. 275.' Terebratvlina striata. Wahlenb" nat. size, Upper White Chalk. Rhynchonella octo- plicata, Sow., \. (Var. of R. plicatilis, Sow.) Upper White Chalk. Terebratula carnea, Sow., \. Upper White Chalk of Norwich. some of the genera, like Echinocorys (Ananchytes, see fig. 268), are ex- clusively Cretaceous. Among the Crinoidea, the Marsupites (fig. 271) is a characteristic genus. Among the mollusca, the Cephalopoda are 258 BRACHIOPODA AND [CH. XVI. represented by Ammonites, Baculites (fig. 292, p. 261), and Belemni- tella (fig. 442, p. 327). Although there are eight or more species of Ammonites and six of them peculiar to it, this group is much less fully represented than in each of the other subdivisions of the Upper Cretaceous. * Fig. 27 Fig. 27 Trrebralula Crania parisien.tis, biplicata, Defr., f. Inferior, Brocchi, J. or attached valve. Upper " ' Upper Chalk. Cretaceous. Fecten Beaveri, Sow., J nat. Lower Chalk, Spondylus spinosus, Sow. sp., i. Upper Chalk. Among the Brachiopoda in the White Chalk, the Terebratulte are very abundant (see figs. 276, 298). With these are associated some forms of oyster (fig. 280), and other bivalves (figs. 278, 279). Fig. 280. Fig. 281. Ostrea vesicularis, Lam., . Upper Chalk and Upper Greensand. Inoceramus Lamarckii, Park., . White Chalk. Among the bivalve mollusca, no form marks the Cretaceous era in Europe, America, and India in a more striking manner than the extinct genus Inoceramus (fig. 281), the shells of which are dis- tinguished by a fibrous texture, and are often met with in fragments having probably been extremely friable. CH. xvi.] OTHER MOLLUSCA OF THE CHALK 259 The singular order called Rudistes by Lamarck, hereafter to be mentioned as extremely characteristic of the chalk of Southern Europe, has species (figs. 282-285) in the Chalk of England. Fig. 283. Radiolites Mortoni, Mant. Houghton, Sussex. White Chalk. Diameter one-seventh nat. size. Fig. 282. Two individuals deprived of their upper valves, adhering together. 283. Same seen from above. 284. Transverse section of part of the wall of the shell, magnified to show the structure. 285. Vertical section of the same. On the side where the shell is thinnest, there are one external furrow and corre- sponding internal ridge, , 6, figs. 282, 283 ; but they are usually less prominent than in these figures. The upper or opercular valve is wanting. The general absence of univalve mollusca in the Chalk of England is very marked. Of Bryozoa there is an abundance, such as EscJiara and Escharina (figs. 286, 287). These and other organic bodies, especially sponges, such as Ventriculites (fig. 267), are dispersed in- Fig. 286. Eschara disticha, Goldf. White Chalk. . Natural size. 6. Portion magnified. differently through the soft chalk and hard flint, and some of the flinty nodules owe their irregular forms to enclosed sponges, such as fig. 288 a, where the hollows in the exterior are caused by the branches of a sponge (fig. 288, 6), seen on breaking open the flint. s 2 260 FISH-REMAINS AND COPROLITES [CH. xvi. The remains of fishes of the Upper Cretaceous formations consist chiefly of teeth belonging to the shark family, of the genera Lamna and Otodus, for instance. Some of the genera are common to the Tertiary formations, but many are distinct. To the latter belongs Fig. 287. scharina oceani, D'Orb. a. Natural size. 6. Part of the same magnified. White Chalk. A branching sponge in a flint, from the White Chalk. From the collection of Mr. Bowerbank. Fig. 290. Palatal tooth of Ptychodus decurrens, . Lower White Chalk. Cestration Phillippi, Cur. ; recent. Maidstone. Port Jackson. Buckland. Bridgwater Treatise, PI. 27, d. CH. xvi.] THE MIDDLE AND LOWEK CHALK 261 Pig. 291. the genus Ptychodus (fig. 289), which is allied to the living Port- Jackson shark, Cestracion Phillippi, Cuv., the anterior teeth of which (see fig. 290, a) are sharp and cutting, while the posterior or palatal teeth (6) are flat. The Teleostean division, to which most of the living bony fishes belong, appears to have been fairly represented in Creta- ceous times, and it is represented by species of Beryx, a genus still exist- ing in the Atlantic and Pacific Oceans^ But we meet with no bones of land animals, nor any terrestrial or fluviatile shells, nor any plants, except seaweeds, and here and there a piece of drift-wood. All the appearances concur in leading us to conclude that the Chalk was the product of an open sea of great depth. The collector of fossils from the Chalk was formerly puzzled by meeting with cer- tain bodies which were called ' larch-cones,' which were afterwards recognised by Dr. Buckland to be the excrement of fish (fig. 291). They are composed in great part of phosphate of lime. Certain bands in the Chalk are found largely converted into calcium phosphate, and constitute a valuable manure. Of these phosphatic chalks we have an example in this country at Taplow. The Middle and Lower Chalk have yielded twenty-five species of Ammonites, of which one half are peculiar to these divisions. The genera Baculites, Hamites, Scaphites, Turrilites, Nautilus, and Belemnitella, are also represented. At the top of the Middle Chalk is found the hard cream-coloured Chalk-rock. Below this is a thick- ness of at least 220 feet of Chalk, with some few flint bands near the top. This part is full of fragments of shells, and may be divided into three zones the zone of Holaster planus, Mant., at the top, that of Terebratulina gracilis, Schloth., next below, and at the bottom the zone Fig. 292. Coprolites of fish, from the Chalk. Baculites anceps, Lam., J. Lower Chalk of Inoceramus vobiatus. Schloth. These zones rest upon the gritty and nodular chalk known as the Melbourn Rock, which forms the base of the series. It contains the Inoceramus just noticed, with Ectiinoconus subrotundzis, Mant., &c. The Middle Chalk is the equivalent of the continental Turonian, and there is a considerable palaeontological break between it and the underlying Cenomanian or Lower Chalk with Chalk-marl. The fairly weM defined bed of yellowish gritty chalk, referred to under the name of ' Melbourn Rock,' may be seen, in some places in the south-east of England, in cliff sections lying below the Chalk without flints. It contains Actinocamax (Belemnitella) plenus, Blain. sp., and Radiolites Mortoni, Mant. It merges downwards into the Grey Chalk, which is somewhat grey in colour and destitute of flints. The Lower Chalk contains several zones of fossils, of which that 262 CHALK MARL [CH. xvr. just mentioned may be considered the top. Beneath the zone with Actinocamax (Belemnitella] plemis, Blain. sp., is that of Holaster subglobosus, Ag., and Discoidea cylindrica, Ag. ; and this zone con- tains Ammonites (Acantkocera,*') rothomagensis, Defr., A. (Schloen- lacMa) variarts, J. Sow., and Pecten Beaveri, Sow. Fig. 293. Ammonite?, (Acanthoceras) rothomagemis, Defr., J. Chalk-marl. Back and side view. In the neighbourhood of Dunstable, the hard Totternhoe stone lies at the base of the Grey Chalk, and both overlie the true Chalk - marl. Chalk-marl. This is an argillaceous chalk, and it forms with the next division the base of the true Chalk formation. It is seen . 294. Fig. 295. Scaphites cequalis, Sow., i. Chloritic marl and sand^ Dorsetshire. Turrilites costatus, Lam., . Lower Chalk and Chalk-marl. a. Segment, showing the foliated border of the sutures of the chambers. at Folkestone and in the Isle of Wight, and contains amongst the common fossils Scaphites cequalis, Sow. (fig. 295), Turrilites costatus, Lam. (fig. 294), Ammonites (Acanthoceras) Mantelli, Sow., and Lima globosa, Sow. CH. XVI.] UPPEE GEEENSAND 263 Chloritic Marl. This yellow or whitish chalky marl contains grains of glauconite, and phosphatic nodules. It yields Ammonites (AcantJioceras) Mantelli, Sow., A. (Schloenbachia) varians, J. Sow., A. (Acanthoceras) laticlavius, Sharpe, Nautilus Icevigatus, D'Orb., Terebratula biplicata, Sow., &c. The Greensand of Cambridge, a bed about a foot thick, lying at the base of the chalk of Cambridge, is a glauconitic marl with phos- phatic nodules, rolled fossils, and erratic blocks. It is the equivalent of the Chloritic marl, forms the base of the Chalk-marl, and rests unconformably on the Gault, from the denudation of which its rolled fossils have been derived. Numerous reptilian and other bones have Fig. 296. Ostrea (E.rogyra) columba, Lam., . Upper Greensand. Fig. 298. Ostrea carinata, Lam., . Clialk marl and chloritic sand. Pig. 300. Terebrirostralyra, Sow., Pecten (Janird) \. Upper Greensand quinquecostatus, Sow., i. Lower Chalk Lima (Plagiostoma) ffoperi, Sow. i. Svn. Lima Hoperi. White Chalk. been found in this deposit belonging to Chelonians, Lacertilia, Croco- diles (Polyptycliodon], Dinosauria, Mosasauria, many species of Pterodactylus, small and large, and species of Plesiosaurus and Ichthyosaurus. Two species of true birds occur (belonging to the genus Enaliornis), and Professor H. G. Seeley considers them to have been swimming birds. The Eed Chalk of Hunstanton is probably of about the same geological age as the Cambridge Greensand, its colour being due to oxidation of the glauconite. Upper Greensand. The sandy strata of this age, often greenish in colour, are not very readily separable from the Chalk-marl. The for- mation may be divided into an upper zone with Pecten asper, Lam., 264 THE GAULT [ CH . xvi. and a lower with Ammonites (Schloenbachia) rostratus, J. Sow. But there is a mixture cf Chalk-marl, Chloritic-marl, and even of Gault species in this series, so that it is very debateable ground. It is well developed in Devon and Somerset. The Warminster beds con- tain Micrabacia coronula, Edw. and H., Ostrea carinata, Lam. (fig. 297), Pecten asper, Lam., Terebratula biplicata, Sow., and the Black- down beds have Trigonia altzformis, Park., Pecten (Janira) qidn- quecostatus, Sow. (fig. 299), and Exogyra conica, Lam. The development of this series is slight in the Kentish area, but it is well seen in the Isle of Wight and again in Antrim. Among the characteristic species of the Upper G-reensand may be mentioned Terebrirostra lyra, Sow. (fig. 298), Pecten qitinqiie- costatus, Sow. (fig. 299), and Ostrea columba, Lam. (fig. 296). The Cephalopoda are abundant : 40 species of Ammonites are now known, 10 being peculiar to this subdivision, and the rest common to the beds immediately above or below. Gault. The lowest member of the Upper Cretaceous group, usually about 100 feet thick in the S.E. of England, is provincially termed Gault. It is a stiff dark-blue marl, sometimes intermixed Fig. 301. Ancyloceras spinigerum, D'Orb. Syn. Hamites spiniger, Sow. Near Folkestone. Gault. with greensand. Messrs. De Kance and Price have shown that one of the best sections is at Copt Point, near Folkestone, where the upper and lower divisions of the series can be seen. The upper division contains Ammonites (Schloenbachia) rostratus, Sow., Kingena lima, Defr., Scaphites ccqualis, Sow., Ammonites (Schloenbachia) cristatus, De Luc., and nearly half of its species pass up into the superincum- bent beds. The lower division rests on Lower Cretaceous rocks, over- laps them, and lies in turn on the various beds of the Jurassic system, showing the physical break between the Lower and Upper Cretaceous formations. About one-eighth only of the fossils pass from the Lower into the Upper Gault. The lower division contains Ammonites (Hoplites) auritus, Sow., A. (Hoplites) lautus. Sow., Solarium monili- ferum, Mich., Ancyloceras spinigerum, D'Orb. (fig. 301), numerous corals and crabs, and species of Crioceras and Hamites. The great break between the Upper and Lower Cretaceous is shown by the remarkable unconformity and overlap (overstep) of the chalk on all the other strata. (See fig. 109, p. 101). The researches of M. Barrois, Mr. Price, and other authors have shown that the English Upper Cretaceous consists of a number of sub- CH. XVI.] ZONES OF THE CRETACEOUS 265 divisions or zones, each characterised by a peculiar assemblage of fossils. These zones are as follows (see Note P, p. 605) : Zone Senonian (Upper Chalk) Cenomanian (Lower Chalk and/ U. Greensand) Zone Zone Zone of Belemnitella mucronata, Schloth. Belemnitella quadrata,Deh. Marsupites Micraster cor-anguinum, Forbes , , Micras ter cor- tesludinarium, Goldf. Holaster planus, Mant. (Chalk Rock) of Tcrebratulina gracilis, Schloth. Inoceramus labiatus, Schloth., and Rhynchonella Ctivieri, Sow. of Belemnitella (Actinocamax) plena, Blain. (Melbourn Rock) Ammonites (Acanthoceras) rothomagensis, Defr. Holaster subglobosus, Ag. Ammonites (Schloenbachia) varians, Sow., and Rhyn- chonella Martini, Mant. Plocoscyphia meandrina, Rom. sp. Pecten asper, Lam. of Ammonites (Schloenbachia) inftatus, Sow. ,, Kingena lima, Defr. Ammonites (ScJiloenbachia) varicosus, Sow. Ammonites (Schloenbachia) \ cristatus, De Luc. [Junction ,, Ammonites (Hoplites) auritus, j bed Sow. J dena- Lower Gault Upper Gault Albian (Gault) / " " ,, ,, Ammonite s(Hoplites)lautus, Sow. ,, ,, Ammonites (Schloenbachia) Delaruei, D'Orb. ,, Crustacea Ammonites (Hoplites) au- ritus, Sow. var. Ammonites (Hoplites) inter- ruptus, Brug. The Lower Greensand (Upper Neocomian) of the South of England represent the upper part of a great system of strata, attaining a great development on the continent. The sands which crop out beneath the Gault in Wiltshire, Surrey, and Sussex are sometimes in the uppermost part pure white, at other times of a yellow, green, or brown colour, and some of the beds con- 266 LOWER GREENSAND [CH. XVI. Fig. 302. tain much ferruginous matter. At Hythe they contain layers of calcareous rock and chert, and at Maidstone and other parts of Kent the limestone called Kentish Rag is intercalated. This somewhat sandy and calcareous rock forms strata two feet thick, alternating with quartzose sand. The total thickness of these sandstone and calcareous beds is less than 300 feet, and the Hythe beds are seen to rest immediately on a grey clay, to which we shall presently allude as the Atherfield clay. Among the fossils of the Hythe beds we may mention Nautilus plicatus, Sow. (fig. 302), Ancyloceras (Scaphites) gigas, D'Orb. (fig. 303) which has been aptly described as an Ammonite more or less uncoiled Trigonia caudata, Ag. (fig. 305), Gervillia anccps, Desh. (fig. 304) a bivalve genus allied to Avicula and Tcrebratula sella, Sow. (fig. 306). In ferruginous beds of the same age in Wilt- Pig. 303. Nautilus plicatus, Sow., }, in Fitton's Monog. Ancylocerai gigas, D'Orb, Fig. 304. Fig. 305. Gervillia anceps, Desb.., Upper Neocomiari. Trigonia caudata, Ag., J. Upper Neocomian. shire is found the remarkable shell called Diceras Lonsdalii, Sow. (fig. 307), belonging to the Chamidae, which abounds in the Upper and Middle Neocomian of Southern Europe. M. Barrois and other authors regard the Folkestone beds or ' Car- stone,' which form the upper member of the Lower Greensand, as being closely connected with the Gault, and they believe that an CH. XVI.] ATHERFIELD CLAY 267 unconformity accompanied by a great change in fossils exists between the Folkestone beds and the underlying members of the 'Lower Greensand.' If this view be correct, the Folkestone beds will have to Fig. 306 Fig. 307. Terebratula sella, Sow., Upper Neocomian. Diceras Lonsdalii, Sow., i. Upper Neocomian, Wilts. a. The bivalve shell. b. Cast of one of the valves enlarged. be removed from the Neocomian and grouped with the Upper Cretaceous. Atherfield clay. We mentioned before that the Hythe series rests on a grey clay. This clay is only of slight thickness in Kent and Surrey, but is better developed at Atherfield, in the Isle of Wight. The difference, indeed, in mineral character and thickness of the Upper Neocomian formation near Folkestone, and the corre- sponding beds in the south of the Isle of Wight, about 100 miles dis- tant, is truly remarkable. In the latter place we find no limestone answering to the Kentish Rag, and the entire thickness from the bottom of the Atherfield clay to the top of the Neocomian, instead of Fig. 308. Pema Mulled, Desh., nat. size. a. Exterior. 6. Part of hinge-line of upper or right valve. being less than 300 feet as in Kent, is given by the late Professor E. Forbes as 843 feet, which he divides into sixty-three strata, forming three groups. The uppermost of these consists of ferruginous sands ; the second of sands and clay, and the third or lowest of a brown clay abounding in fossils. Pebbles of quartzose sandstone, jasper, and flinty slate, together with grains of chlorite and mica, occur in the Lower Greensand of Surrey; and fragments and water-worn fossils of the Jurassic rocks speak plainly, as Mr. Godwin-Austen has shown, of the nature of the pre-existing formations, by the wearing down of which the Neocomian beds were formed. The land consisting of such rock? 268 SPEETON AND TEALBY BEDS [CH. XVI. Fig. 309. was doubtless submerged before the origin of the Chalk, a deposit which was formed in a more open and probably deeper sea, and in clearer waters. Among the shells of the Atherfield clay the most characteristic, perhaps, is the large Pcrna Mulleti, Desh., of which a reduced figure is here given (fig. 308). Speeton clay On the coast beneath the Chalk of Flam- borough Head, in Yorkshire, an argillaceous formation crops out, called the Speeton clay. It is several hundred feet in thickness, and its palaeontological relations have been worked out by Professor Judd, and later by Mr. Lamplugh, and it has been shown that it is separable into three divisions, the uppermost of which, 150 feet thick, and containing 87 species of mollusca, decidedly belongs to the Atherfield clay and associated strata of Hythe and Folkestone, already described. It is characterised by the Perna Mulleti Desh. (fig. 308), and Terebratula sella, Sow. (fig. 306), and by Ammonites (Hoplites) Deshayesii, Leym. (fig. 309), a well-known Hythe and Ather- field fossil. Remains of skeletons of the genera Ptesiosaurus and Teleosaurus have been obtained from this clay. At the base of this upper division of the Speeton clay there occurs a layer of large Septaria, formerly worked for the manufacture of cement. This bed is crowded with fossils, especially Ammonites, some of which are gf great size. The Speeton Clay is represented in Russia by the upper portion of the great Volga formation, which contains many fossils in common with those of Speeton. Tealby series At Tealby, a village in the Lincolnshire Wolds, there occur, beneath the White Chalk, some non-fossiliferous Ammonites (Hoplites) Deshayesii, Leym., . Upper Neocomian. Fig. 310. Fig. 311. Peclen cinctus, Sow. (P. crassitesta, Rom.) Middle Neocomian, England : Middle and Lower Neocomian, Germany. uat. size. Anculoceras (Crioceras") Duvallii, Leveille. Middle and Lower Neucomian. \ nat. size. OH. xvi.] IE WEALDEN FORMATION 269 ferruginous sands about twenty feet thick, beneath which are beds of clay and limestone about fifty feet thick, with an interesting suite of fossils, among which are Pecten cinctus, Sow. (fig. 310), from 9 to 12 inches in diameter, Ancyloceras Duvallii, Leveille (fig. 311), and some 40 other shells, many of them common to the Middle Speeton clay, about to be mentioned. As Ammonites (Placenticeras) clypei- formis, D'Orb., and Terebratulahippopus, Rom., found in these beds, characterise the Middle Neocomian of the Continent, it is to this stage that the Tealby series must be assigned. The middle division of the Speeton clay, occurring at Speeton below the cement-bed, before alluded to, is 150 feet thick and contains about 39 species of mollusca, half of which are common to the overlying clay. Among the shells are Ancyloceras Duvallii, Leveille (fig. 311) and Pecten cinctus, Sow. (fig. 310). lower Neocomian. In the lower division of the Speeton clay, 200 feet thick, 46 species of mollusca have been found, and three divisions, each characterised by its peculiar ammonite, have been noticed. Fig. 312. The central zone is marked by Ammo- nites (Hoplites) noricus, Schloth. (see fig. 312). On the Continent these beds are well known by their corresponding fossils, the Hils-thon and conglomerate of the North of Germany agreeing with the Middle and Lower Speeton; the latter of which, with the same mineral characters and fossils as in Yorkshire, is also found in the little island of Heligoland. Ammonites (Hopliles') noricus, Wealden Formation Beneath Schloth., nat. size. Lower the Atherfield clay or Upper Neocomian Neocomian. Speeton. of the S.E. of England, a freshwater or delta formation is found, called the Wealden, which, although it occupies a small horizontal area in Europe, as compared with the Chalk and the marine Neocomian beds, is nevertheless of great geological interest, since the embedded remains give us some insight into the nature of the terrestrial fauna and flora of the Lower Cretaceous epoch. The name of Wealden was given to this group because it was first studied in parts of Kent, Surrey, and Sussex, called the Weald ; and we are indebted to Dr. Mantell for having shown, in 1822, in his ' Geology of Sussex,' that the whole group was of fluviatile origin. In proof of this he called attention to the entire absence of Ammonites, Belemnites, Brachiopoda, Echinoderrnata, Corals, and other marine fossils, so characteristic of the Cretaceous rocks above, and of the Oolitic strata below, and to the presence of Paludina?, Melanin, Cyrenas, and various fluviatile shells, as well as the bones of terrestrial reptiles and the trunks and leaves of land-plants. The evidence of so unexpected a fact as that of a dense mass of purely freshwater origin underlying a deep-sea deposit (a phenome- non with which we have since become familiar) was received, at first, with no small doubt and incredulity. But the relative position of the beds is unequivocal ; the Weald clay being distinctly seen to pass beneath the Atherfield clay in various parts of Surrey, Kent, and 270 RELATIONS OF THE WEALDEN BEDS [CH. xvi. Sussex, and to reappear in the Isle of Wight at the base of the Cre- taceous series, being, no doubt, continuous far beneath the surface, as indicated by the dotted lines in the annexed diagram (fig. 313). w.s w. Fig. 313. E.N.B. Isle of Wiyhi South Dcwns. 1. Tertiary. 2. Chalk c.nd Gault. 3. Upper Neocomian (or Lower G-reensand). 4. Wealden (Weald Clay and Hastings Sand). They are also found occupying the same relative position below the chalk in the peninsula of Purbeck, where, as we shall see in the sequel, they rest on strata referable to the Upper Oolite. Weald Clay. The upper division, or Weald clay, 1,000 feet thick, is, in great part, of freshwater origin, but in its highest portion contains beds of oysters and other marine shells which indicate fluvio-marine conditions. The uppermost beds are not only con- formable to the inferior strata of the overlying Neocomian, but are of similar mineral composition. To explain this, we may suppose Fig. 314. Fig. 315. Fig. 314. a. b, Tooth of Iguanodon Mantelli, Meyer, nat. size. Fig. 315. a. Partially worn tooth of young individual of the same. b. Crown of tooth in adult, worn down. (Mantell.) that, as the delta of a great river was tranquilly subsiding, so as to allow the sea to encroach upon the space previously occupied by fresh water, the river still continued to carry down the same sediment into the sea. In confirmation of this view it may be stated, that CH. XVI. J WEALDEN REPTILES 271 the remains of the Iguanodon Mantelli, Meyer, and also species of the genera Hypsilophodon, Pelorosatirus, Ornitliopsis, and Hylczosaurus, gigantic terrestrial reptiles, belonging to the order Dinosauria, have been discovered near Maidstone, in the overlying Kentish Rag, or marine limestone of the Upper Neocomian, and in the Isle of Wight and elsewhere. Hence we may infer that some of the Reptilia which inhabited the country of the great river which formed the Wealden delta, continued to live when part of the district had become submerged beneath the sea. Thus, in our own times, we may suppose the bones of large crocodiles to be frequently entombed in recent freshwater strata in the delta of the Ganges. But if part of that delta should sink down so as to be covered by the sea, marine formations might Fig. 316. Ijuanodon Bernissartemis, Boulenger. Almost complete skeleton About T V nat. size. From the Wealden of Belgium. begin to accumulate in the area in which freshwater beds had previously been formed ; and yet the Ganges might still pour down its turbid waters in the same direction, and carry seaward the carcases oithe same species of crocodile ; and in this case their bones might be included in marine as well as in subjacent freshwater strata. Complete skeletons of Iguanodon have been found in Belgium, one of which showing the general structure of these remarkable extinct reptiles is shown in fig. 316. Occasionally bands of limestones called Sussex Marble occur in the Weald clay, almost entirely composed of a species of Paludina, closely resembling the common P. vivipara, L., of English rivers. 272 THE HASTINGS SAND Shells of the Cypridea (fig.317), a genusof Ostracoda before mentioned as abounding in lakes and ponds, are also plentifully scattered through the clays of the Wealden, sometimes producing, like plates of mica, a thin lamination (see fig. 318). Fig. 317. Fig. 318. Cypridea spinigera, Fitton. Weald Clay, with Cypriote. Hastings Sands. This lower division of the Wealden consists of sand, sandstone, alciferous grit, clay, and shale ; the argillaceous strata, notwithstanding the name, predominating somewhat over the arenaceous, as will be seen by reference to the following section, drawn up by Messrs. Drew and Foster, of the Geological Survey of Great Britain : Names of Subordinate Formations. /Tun bridge Sand Hastings Sand Wells | Mineral Composition of the Strata. Sandstone and loam Thickness in Eeet. [Blue and brown shale and \ clay with a little calc-grit Hard sand with some beds of calc-grit AshburnhamBedsl Mott J' d ' white and red clay [ with some sandstone Wadhurst Clay Ashdown Sand 150 100 160 330 The picturesque scenery of the ' High Bocks ' and other places in the neighbourhood of Tunbridge Wells is caused by the steep natural cliffs, to which a hard bed of white sand, occurring in the upper part of the Tunbridge Wells Sand, mentioned in the above table, gives rise. This bed of ' rock-sand ' varies in thickness from 25 to 48 feet. Large masses of it, which were by no means hard or capable of making a good building-stone, form, nevertheless, pro- jecting rocks with perpendicular faces, and resist the degrading action of the river because, says Mr. Drew, they present a solid mass without planes of division. The calcareous sandstone and grit of Tilgate Forest, near Cuckfield, in which the remains of the Iguano- don and Hylceosaurus were first found by Dr. Mantell, constitute an upper member of the Tunbridge Wells Sand, while the ' sand-rock ' of the Hastings cliffs, about 100 feet thick, is one of the lower members of the same The reptiles, which are very abundant in this division, consist partly of marine saurians, among which we find the Megalosaurus and Plesiosaurus. The Pterodactylus is also met with in the same strata, and many remains of Chelonians of the genera Trionyx and Emys, now confined to tropical regions. The fishes of the Wealden are chiefly referable to the Ganoid and Placoid orders. Among them the teeth and scales of Lepidotus CH. XVI.] WEALDEN FOSSILS 273 are most widely diffused (see fig. 319). These ganoids were allied to the Lepidosteus, or Gar-pike, of the American rivers. The whole body was co red with large and very thick rhomboidal scales, Fig. 319. Lepidotus Mantelli, Ag. Wealden. a. Palate and teeth. 6. Side view of teeth. c. Scale. Fig. 320. Unio valdensis, Mant., J. Isle of Wight and Dor- setshire ; in the lower beds of the Hastings Sands. Underside of slab of sandstone about one yard in dia- meter, showing casts of ' suri- cracks.' Stammer- ham, Sussex. having the exposed part coated with enamel. Most of the species} of Lepidotus are supposed to have been either river-fish, or inhabi- tants of the sea at the mouth of estuaries. 274 THE PUNFIELD BEDS [CH. XVI. At different horizons in the Hastings Sand we find again and again slabs of sandstone with strong ripple marks, and between these slabs are beds of clay many yards thick. In some places, as at Stammerham, near Horsham, there are indications of this clay having been exposed so as to dry and crack before the next layer was thrown down upon it. The open cracks in the clay have served as moulds, of which casts have been taken in relief, and which are, therefore, seen on the lower surface of the sandstone (see fig. 321). Near the same place a reddish sandstone occurs in which are in- numerable remains of a fern, apparently a Sphcnopteris, the stems and fronds of which are disposed as if the plants were standing erect on the spot where they originally grew, the sand having been gently deposited upon and around them ; and similar appearances have been remarked in other places in this formation. In the same division also of the Wealden, at Cuckfield, is a bed of gravel or con- Pig. 322. Sphenopteris gracilis, Fitton. Prom the Hastings Sands near Tunbridge Wells. a. Portion of the same magnified. Vicaryn Lnjani, De Verneuil. Wealden, Punfield. . Nearly perfect shell, b. Vertical section of smaller specimen, showing continuous ridges, as in Nerineea. glomerate, consisting of water-worn pebbles of quartz and jasper, with rolled bones of reptiles. These must have been drifted by a current, probably in water of no great depth. From such facts we may infer that, notwithstanding the great thickness of this division of the Wealden, the whole of it was a delta deposit, in water of a moderate depth, and often extremely shallow. This idea may seem startling at first, yet such would be the natural consequence of a gradual and continuous sinking of the sea-bottom in an estuary or bay, into which a great river discharged its turbid waters. By each foot of subsidence, the fundamental rock would be depressed one foot farther from the surface ; but the bay would not be deepened, if newly deposited mud and sand should raise the bottom one foot. On the contrary, such new strata of sand and mud might be frequently laid dry at low water, or overgrown for a season by a vegetation proper to marshes. Punfield beds, brackish and marine. These lie between the Lower Greensand and the Wealden proper. The shells of the Wealden belong to the genera Mclannpsis, Afelania, Paludina, Cyrcna, Cyclas, Unio (see fig. 320), and others, which inhabit estu- aries, rivers or lakes ; but as shown by Godwin- Austen, and E. Forbes CH. xvii.] THE JURASSIC SYSTEM 275 at 1'unfield, in Dorsetshire, the genera Corbvla, Mytilus, and Ostrea occur, indicating a brackish state of the water ; and in some places this bed becomes purely marine, containing some well-known Neocomian fossils, among which Ammonites (Hoplites) Dcshaycsii, Leym (fig. 309) may be mentioned. Others are peculiar as British fossils, but very characteristic of the Upper and Middle Neocomian of the North of Spain, and among these the Vicarya Lnjani, De Verneuil (fig. 323). a shell allied to Nerinaa, is conspicuous. The middle Neocomian beds of Spain, in which this shell abounds, attain at Utrillas a thickness of 520 feet, and contain ten beds of coal, lignite, or jet, which are extensively worked. As the Wealden strata commence with the brackish -water Purbeck, of Upper Jurassic age, and end with the brackish- water Punfield beds of Middle Neocomian age, they probably represent the base of the Neocomian, the gap between the Jurassic and Neocomian and part of the Upper Jurassic. The classification of the Ore- fication may be applied to English taceous strata into zones dis- strata in his ' Recherches sur le tinguished by groups of charac- Terrain Cietace Superieur de teristic fossils has been brought PAngleterre etdel'Irlande' (1876). about by the labours of the French Valuable information on the Cre- geologists D'Orbigny, Heberfc, taceous strata will be found in and Barrois. The last-mentioned the Memoirs of the Geological author has shown how this classi- Survey. CHAPTER XVII THE JURASSIC SYSTEM Classification c f Jurassic strata Foraminifera, Sponges, Corals, Echino- dermata, Brachiopoda, Lamellibranchiata, Gastropoda, and Cephalo- poda of Jurassic rocks Fishes, reptiles, birds, and mammals of the Jurassic rocks Terrestrial Flora of the Jurassic period Purbeck strata Purbeck mammals Dirt beds Portlandian Kimeridge Clay Coralline Oolite Oxford Clay Cornbrash Forest? Marble Great Oolite Stonesfield Slate with its Mammalia Inferior Oolite Upper Lias sand and clay Marlstone and Middle Lias White Lias and Rhastic beds. Nomenclature and classification of the Jurassic strata. The name of this great system of stratified rocks is derived from the Jura Mountains, where the formations are admirably developed, and were carefully studied by Marcou. In England, and also in France, the system is usually divided into two members, the upper of which is called ' Oolite,' from the preva- lence in it of limestones of oolitic structure, and the lower 'Lias,' a provincial term applied to the finely laminated beds of clay and limestone of which it is chiefly made up. In Germany the system is usually divided into three members, which bear the names of White Jura or Malm, Brown Jura or Dogger, and 276 SUBDIVISIONS OF THE JURASSIC [CH. xvn. Black Jura or Lias. The German Dogger is the equivalent of the Lower Oolite of England. It was while studying the Jurassic rocks that William Smith was first able to establish the important principle that strata can be identified by their organic remains ; and the chief sub- divisions of the Oolitic and Liassic rocks still bear the names (often of provincial origin) first applied to them by Smith. The general order of succession and approximate thicknesses of the beds of this system, as seen in the south-west of England, are given in the following table : Upper or [Purbeck beds, 300 feet. Portland j Portland Oolite and sand, 200 feet. Oolites. ( Kimeridge clay, 600 feet. Middle or /'Coral-rag and calcareous grit (coralline Oolite Oxford or Corallian), 250 feet. Oolites. 1 0xford clay, with Kellaways rock, 600 feet I (Callovian). / Cornbrash, 25 feet. Lower or Forest marble with Bradford clay, 150 feet. Bath J Great Oolite with Stonesfield slate, 120 feet. Oolites. ] Fuller's earth, 150 feet. Inferior Oolite (including ragstones, freestones, V the oolite-marl, pea-grit and sands), 270 feet. Upper j Midford sand, 200 feet. Lias. 1 Upper Lias clays, 200 feet. Middle j Marlstone rock-bed, 400 feet. Lias. Middle Lias clays, 400 feet. Lower r Lower Lias clays, 800 feet. Lias. < Lower Lias limestone and shale, 800 feet. White Lias and Khaetic or Avicula-contorta beds. It should be noted that the terms Great Oolite and Inferior Oolite are used in the sense of principal Oolitic limestone and lower Oolitic limestone. As in the case of the Cretaceous rocks, latinised names of the local designations have been adopted in France, and are not unfrequently employed in this country. They are given on pages 325, 326. . It will be seen that, speaking generally, the Jurassic strata of the south-west of England may be regarded as made up of three great masses of limestone, or sandstone, alternating with others of blue clay or shale. The hard limestones and sand- stones form well-marked escarpments, which appear in succes- sion beyond that of the chalk, and traverse the country from CH. XVII.] JURASSIC FAUNA AND FLORA 277 N.E. to S.W., as illustrated in the following diagrammatic section. N.w. S.E. London Chalk. Clay. Lias. Oxford Clay. Kim. Clay. Gault. Characteristics of the Jurassic fauna and flora. It appears doubtful if any species of British fossil, whether of the vertebrate or invertebrate class, is common to the Jurassic and Cretaceous. But there is no similar break or discordance as we proceed downwards, and pass from one to another of the several leading members of the Jurassic group, there being often a considerable proportion of the mollusca, sometimes as much as a fourth, common to such divisions as the Upper and Middle Oolite. Between the Lower Oolite and the Lias there is a some- what greater break, for out of 256 mollusca of the Upper Lias of Britain thirty-seven species only pass up into the Inferior Oolite. It is in the Jurassic system of strata that we find the most perfect illustration of the Mesozoic marine fauna in the British Islands. Many of the limestones are largely made up of the remains of Foraminifera ; and siliceous sponges (Lithistidae and Hexactinellidae) are also found. Corals of the order Hexa- coralla, both compound and reef-building, like Isastrcea (fig. 350, p. 298), Thecosmilia (fig. 358, p. 294), Tliamnastraa (fig. 359, p. 294), and simple forms like Montlivaltia also abound, and many rocks, like the Coral Eag, are almost entirely made up of the remains of these organisms. Echinoderms are represented by Apiocrinus (fig. 366, p. 297) and Ptntacrinus (fig. 408, p. 307) among the stalked forms (Crinoidea), and by many sea-urchins like Cidaris, Echinobrissus, Holectypus, &c., and some Star-fishes. The Brachiopoda show the same abundance and variety as in the Cretaceous system. In addition to the Terebratulidae (fig. 388, p. 302) and Bhynchonellidae (fig. 389, p. 302), we find a few surviving forms of the Palaeozoic Spiriferidse (fig. 409, p. 307). The Lamellibranchiata are represented by abundant species of Oysters (figs. 353, 356, 360, 386, 396), and also the forms known as Exogyra and GryplicBd (fig. 403, p. 306), together with many species of Lima (fig. 405), Pecten, Modiola, Gervillia, and Cardium (fig. 354, p. 293). Of Trigonia many very interesting forms occur in different divisions of the Jurassic system (fig. 351, p. 293) ; and the same is true of the genera Pholadomya (fig. 390, p. 302), Goniomya, and Myacites, Among the Gastro- 278 JURASSIC CEPHALOPODA [CH. XVII. poda some of the most abundant genera are Pleurotomaria (figs. 391, 392, p. 303), Nerincea (fig. 361, p. 294), Pterocera, and Cylindrites (fig. 371. p. 298). It is by the abundance and richness of its Cephalopod fauna that the Jurassic rocks are best characterised. Ammonites belonging to the genera Aspidoceras, Perisphinctes, Cosmoceras (fig. 364, p. 295), Macrocephalites, Steplianoceras (figs. 368, 394, 395), Harpoceras, Amaltheus (fig. 398), Phylloceras^fjoceras (fig. 401), and Arietites (fig. 400), are particularly abundant and characteristic. The persistent genus Nautilus is still represented Fig. 325. 6. Scales of JEchmochis Leachii, Ag. a. dZchmodus. Restored outline. 0, Fig. 326. c. Scales of Dapeclius monilifer, Ag. Scales of Lepidotus ffigas, Ag. a. Two of the scales detached. by many forms ; and Belemnites of many varieties, some Bhort and stout, and others very slender and several feet in length, are found in nearly all the beds ; some of the Belemnites still retain in their fossil state the ink-bag, the contents of which were employed to darken the waters, so that they might escape from their enemies. In a few finely laminated rocks, like the Stonesfield slate, the septaria of the Lias, and the lithographic limestone of Solenhofen, abundant remains of Crustaceans, both long-tailed and short-tailed, have been found. CH. XVII.] JURASSIC FISH 279 Fish remains are very numerous in some of the Jurassic strata ; Ganoids, for the most part with homocercal tails, abound (figs. 325, 326), as do also Selachians like Hybodus (fig. 328) and Acrodus (fig. 327). The palatal teeth and fin-spines (ichthyodorulites) of Selachian fish are found in many of the Oolitic and Liassic strata. Fig. 327. Acrodus nobilis, Ag. (tooth) ; commonly called ' fossil leech.' Lias, Lyme Regis and Germany, uat. size. ffyboditx reticuJatus, Ag. Lias, Lyme Regis. a. Part of fin, commonly called an Ichthyodorulite. b. Tooth. The manner in which the ichthyodorulite supports the fin is illustrated by the following sketch of a living Selachian. Fig. 329. Chimwra mon.i(roia, L. a. Spine forming anterior part of the dorsal fin. Ordinary bony fishes (Teleosteans) are almost unknown. The highest organisms found in the Jurassic seas were the remarkable and gigantic reptiles belonging to the orders Plesio- sauria (fig. 331) and Ichthyosauria (fig. 330). Of these extinct reptiles, some of which are between twenty 280 JUKASSIC KEPTILES [CH. XVII. and thirty feet in length, we find skeletons illustrating every stage of development. Occasionally even the outer integument of the animals has been preserved, with the contents of their stomachs and tneir excrement (coprolites). CH. XVII.] DLNOSAUK1A 281 Of the freshwater and terrestrial fauna and flora we have less perfect but very interesting evidence. In the Purbecks and other similar beds, intercalated with the Jurassic marine series, we find many characteristic forms of freshwater mollusca, crustaceans, and fish. Insects in great variety and of remark- able forms occur in some of the fine-grained deposits. Reptilia belonging to the living orders Lacertilia, Crocodilia, and Che- Ionia abound ; and with these occur many remarkable types now extinct, some of which attained enormous dimensions. Among these were the Dinosauria, a great extinct reptilian order, exhibiting, as Professor Huxley and other comparative anatomists have pointed out, very remarkable affinities with birds. Some of the earlier representatives of the order were of moderate size and were covered with a bony armour, while they exhibit less bird-like characters than later forms. Of this type is the Scelidosaurus of the Lias (fig. 332). Fig. 332. Scelidosaurus ffarrisoni, Ow. Restored Skeleton (^ nat. size). Lower Lias of Lj-me Ilegis, Dorsetshire. From the A Dinosaurian reptile with its shoulders, back, and tail covered with thick bony scutes or spines. In addition to the three toes on the hind foot, found in the later Dinosaurs, there is a fourth rudimentary one present in this ancient form. The later bird-like forms were often of gigantic dimensions, like the Megalosaurus (fig. 333) and Iguanodon (figs. 314- 316, 282 GIGANTIC DINOSAUBIA [CH. pp. 270, 271). They appear to have walked on their hind legs, and to have left trifid impressions like those of birds. In the Western Territories of North America, and also in this country, remains are found of a remarkable group of Dinosauria, which are remarkable for their great size and the smallness of their skulls. The reptiles of this group (Atlantosauridae) did not in all probability assume the erect habit of the Dinosauria before noticed (see fig. 334). CH. XVII.] OF NORTH AMERICA 283 Flying reptiles (Pterosauria) have been found in many Ju- rassic deposits. In the celebrated lithographic stone of Solen- hofen, in Bavaria, which is of about the same geological age as our Kimeridge Clay, we find not only the delicate hollow bird-like bones preserved, but also impressions of the mem- 284 JUEASSIC PTEROSAURIA [CH. XVII. branes that formed the wings and rudder-like tail. These have enabled Professor Marsh to make the following interesting restoration of the animal. Pig. 335. Rhamphorhynchus Muensteri, Goldf. (Restored by Marsh.) ^ nat. size. From the Lithographic Stone, Eichstadt, Bavaria. Woodcut furnished by Prof. 0. C. Marsh. Fig. 336 1'terodactylus antiquus, Sbmmerring. Almost complete skeleton, | nat. size. From Lithographic Stone of Eichgtadt, Bavaria. /, y, h, i. Modified digits of fore-arm, supporting wing-membrane, e. Other digits of fore-arm forming claw, s, r. Hind leg with feet (m, /) OH. XVII.] JUEASSIC BIRD 285 In the same stone of Solenhofen two examples have been met with of a true bird with teeth, almost entire, and having even the feathers so well preserved, that the vanes as well as the shaft are seen. It has been called by Professor Owen Archceopteryx macrura. Although anatomists agree that it is a true bird, yet they also find that in the length of the bones of the tail, and some other minor points of its anatomy, it approaches more nearly to Fig. 337. Tail and feather of Archceopteryx, from Solenhofen, and tail of living bird for comparison. A. Caudal vertebra of Archceopteryx macrura, Ow. ; with impression of tail feathers, \ nat. size. B. Two caudal vertebrae of same, nat. size. C. Single feather, found in 1861 at Solenhofen, by Von Meyer, and called Archieo- pleryx lithographica. Nat. size. D. Tail of recent vulture (Gyps bengalensis, Q-m.), showing attachment of tail-feathers in living birds. \ nat. size. K. Profile of caudal vertebrfe of same, J nat. size, e, e. Direction of tail-feathers when seen in profile. /. Ploughshare bone or broad terminal joint (seen also in/, D). reptiles than does any living bird. In the living representa- tives of the class Aves, the tail-feathers are attached to a coccy- geal bone, consisting of several vertebrae united together; whereas in the Archceopteryx the tail is composed of twenty vertebrae, each of which supports a pair of quill feathers. (See fig. 337.) The first specimen of this oldest known and most remarkable bird is preserved in the British Museum. A second specimen of Archceopteryx, which has been discovered at Solenhofen and is 286 JUKASSIC PLANTS [CH. xvn. preserved in the Berlin Museum, shows the skull ; and the jaws are seen to be armed with conical teeth which are set in sockets, like those of the Cretaceous birds already described, fig. 338. In the Jurassic rocks a number of lower jaws and a few other bones have been found belong- ing to mammalia of the primitive order Allo- theria, with others that have been referred by zoologists to the Mar- supialia. They are all of small size, indicating ArauKoptei-yx macrura, Ow. Skull with teeth. , . . , nat. size. From Solenhofeu. the existence of animals with dimensions between those of rats and rabbits. They have been chiefly found in the Stonesfield slate and Purbeck beds of England, the Solenhofen stone of Bavaria, and the Upper Jurassic of North America. In the Jurassic flora we miss the numerous flowering plants (Phanerogamia) of the Cretaceous and Tertiary ; but Conifers, Cycads (fig. 346, p. 291), and also Cryptogams occur in great abundance. The researches of the late Professor Neumayr have proved that there existed in Jurassic times not only a distribution of the forms of marine and terrestrial life in geographical provinces (similar to, but quite distinct from those of the present day), but that also, as in the case of existing fauna and flora, the influence of climate upon this distribution can be distinctly traced at this early period of the earth's history. British Representatives of tbe Jurassic System. It was in the British Isles tha.t the subdivisions of the Jurassic strata were first worked out by William Smith, and many of the names still applied to these strata are taken from English localities. The Oolitic strata of the south of England consist of deposits of shelly and often oolitic limestones alternating with beds of clay, marl, and sand. Tlie Upper Oolite. This division consists of the estuarine Purbeck, with the marine Portlandian and Kimeridge beds below. Purbeck beds. These strata, which we class as the uppermost member of the Jurassic, are of limited geographical extent in Europe, but they acquire importance when we consider the succession of three distinct sets of fossil remains which they contain. Such repeated changes in organic life must have reference to the history of a vast lapse of ages. The Purbeck beds are finely exposed to view in Durdlestone Bay, near Swanage, Dorsetshire, and at Lulworth Cove and the neighbouring bays between Weymouth and Swanage. CII. XVII.] PURBECK STRATA '287 The highest of the three divisions is purely freshwater, the strata, about fifty feet in thickness, containing shells of the existing genera Paludina, Physa, Limncea, Planorbis, Valvata, Cyclas, Unio, with Cypridce. and fish. All the species seem peculiar, and among them the Cypridce are very abundant and characteristic. The freshwater limestone called ' Purbeck Marble,' formerly much used in ornamental architecture in the old English cathedrals of the southern counties, is exclusively procured from this division. Next in succession is the Middle Purbeck, about thirty feet thick, the uppermost part of which consists of freshwater limestone, with cypridae, turtles, and fish of different species from those in the preceding strata. Below the limestone' are brackish-water beds full of Cyrena, and traversed by bands abounding in Corbula and Helania. These are based on a purely marine deposit, with Pecten, Modiola, Avicula, and Thracia. Below this, again, come limestones Fig. 339. O&trra distorta, Sow., nat. size. Ciuder-bed, Middle Purbeck. Hemicidaris purbcckensis, E. Forbes, nat. size. Middle Purbeck. Fig. 341. and shales, partly of brackish and partly of freshwater origin, in which many fish, especially species of Lepidotus and Microdon radiatus, Ag., are found, and a crocodilian reptile named Macro- rlnjnclms. Among the molluscs a remarkable ribbed Melania, of the subgenus Chilina, occurs. Immediately below is a great and conspicuous stratum, twelve feet thick, formed of a vast accumulation of shells of Ostrea distorta, Sow. (fig. 339), long familiar to geologists under the local name of ' Cinder- bed.' In the uppermost part of this bed Professor Forbes discovered a species of Hemicidaris (fig. 340), a genus characteristic of the Oolitic period. It was accompanied by a species of Perna. Below the Cinder- bed, freshwater strata are again seen, filled in many places with species of Cypridea and with Valvata, Paludina, Planorbis, Limncca, Physa (fig. 341), and Cyclas, all different from any occurring higher in the series. Thick beds of chert occur in the Middle Purbeck filled with mollusca and Cypridce of the genera already enumerated, in a beau- tiful state of preservation, often converted into chalcedony. Among these Professor Forbes met with Gyrogonitcs (the spore-vessels of Chara), plants never before discovered in rocks older than the Eocene. i Bristovii, E. Forbes. Middle Purbeck. 288 PURBECK MAMMALIA [CH. XVII. About twenty feet below the ' Cinder-bed ' is a stratum two or three inches thick, in which the fossil mammalia presently to be mentioned occur ; and beneath this is a thin band of greenish shales, with marine shells and impressions of leaves like those of a large Zostera ; it forms the base of the Middle Purbeck. Fossil Mammalia of the Middle Purbeck. In 1852, after alluding to the discovery of numerous insects and air-breathing Fig. 342. Fig. 34?. Pre-molar of the recent Australian HypKiprymnus (Potorons), show- ing 7 grooves at right angles to the length of the jaw, magnified 3 diameters. Third and largest pre-molar (lower jaw) of Plagiaulax Becklesii, Falc., magnified 5^ diameters, showing 7 grooves arranged diagonally to the length of the jaw. mollusca in the Purbeck strata, Lyell pointed out that, although no mammalia had then been found, ' it was too soon to infer their non- existence from mere negative evidence.' Within the next six years Mr. W. R. Brodie and Mr. S. H. Beckles succeeded in detecting a great number of bones, chiefly lower jaws, in the dirt-bed of the Middle Purbeck, an old terrestrial surface. These have been referred by Professor Owen to twenty-five species belonging to eleven genera. Fig. 344. Pl(igiaula.r Recklexii, Falc. Middle Pnrhcck. Right ramus of lower jaw, magnified two diameters. a. Incisor. b, c. Line of vertical fracture behind the pre-molars. d. Throe pre-molars, the third and last (much larger than the other two taken together) being divided by a crack. e. Sockets of two missing molars. The largest pre-molar (see fig. 343) in one fossil genus exhibits seven parallel grooves, producing by their termination a serrated edge in the crown ; but their direction is diagonal and not vertical as in the living Hypsiprymnus a distinction, says Dr. Falconer, which is ' trivial, not typical.' As these oblique furrows form so marked a character of the majority of the teeth, Dr. Falconer gave to the fossil the generic name of Plagiaulax. The shape and relative size of the incisor, a, fig. 344, exhibit a no less striking CH. xvii.] PURBECK: MAMMALIA 289 similarity to Hypsiprymnus. Nevertheless, the more sudden upward curve of this incisor, as well as other characters of the jaw, indicates a great deviation in the form of Plagiaulax from that of the living Hijpsiprymnus or Kangaroo-rat. There are two fossil specimens of lower jaws of this genus evidently referable to two distinct species extremely unequal in size and otherwise distinguishable. The Plagiaulax Becklesii, Falc. (fig. 344), was about as big as the English squirrel or the flying phalanger of Australia (Peiaurus aiistralis, Waterhouse). The smaller fossil, having only half the linear dimensions of the other, was probably only l-12th of its bulk. It is of peculiar geological interest, because, as shown by Dr. Falconer, its two back molars bear a decided resemblance to those of the Triassic Microlestcs, one of the most ancient of known mammalia, of which an account will be given further on. Up to 1857 all the mammalian remains discovered in Secondary rocks had consisted solely of single branches of the lower jaw, but in that year Mr. Beckles obtained the upper portion of a skull, and on the same slab the lower jaw of another quadruped with eight molars, a large canine, and a broad and thick incisor. It has been named Triconodon from its three-coned teeth, and is supposed to have been a small insectivorous mammal, about the size of a hedgehog. Other jaws have since been found, indicating a larger species of the same genus. To the largest of these Professor Owen has given the name of Triconodon major. It was a carnivorous marsupial, rather larger than the Polecat, arid equalling probably in size the Dasyurus viverrinus, Shaw, of Australia. Between forty and fifty mandibles, or sides of lower jaws, with teeth, have been found in the Purbecks ; and it is remarkable that with these there were no examples of an entire skeleton, or of any considerable number of bones in juxtaposition. When we endeavour to account for the absence of other bones, we are almost tempted to indulge in speculations like those once suggested by Dr. Buckland, when he tried to solve the enigma. ' The corpses,' he said, ' of drowned animals, when they float in a river, distended by gases during putrefaction, have often their lower jaw hanging loose, and sometimes it has dropped off. The rest of the body may then be drifted elsewhere, and sometimes may have been swallowed entire by a predaceous reptile or fish, such as an Ichthyosaurus or a Shark.' Beneath the thin marine band, the base of the Middle Purbeck, some purely freshwater marls occur, containing species of Cypris, Valvata, and Limncea, different from those of the Middle Purbeck. This is the beginning of the inferior division, which is about 80 feet thick. Below the marls at Mewps Bay, more than 30 feet of brackish- water strata are seen, abounding in a species of Serpula, allied to, if not identical with, Serpula coacervata, Blum., found in beds of the same age in Hanover. There are also shells of the genus Rissoa (of the subgenus Hydrobia), and a little Cardium of the subgenus Protocardium, in these beds, together with Cypridfc. Some of the Cypridea-bearing shales are strangely contorted and broken up, at the west end of the Isle of Purbeck. The great dirt-bed or vegetable soil containing the roots and stools of Cycadece, to be presently described, underlies these marls, and rests upon the lowest freshwater - - U 290 DIET-BED OF PORTLAND [CH. XVII. limestone, a rock about eight feet thick, containing Cyclas, Valvata, and Limmea, of the same species as those of the uppermost part of the Lower Purbeck, or above the dirt-bed. The freshwater limestone in its turn rests upon the top beds of the Portland stone. Dirt-bed or ancient surface soil. A stratum called by quarry- men ' the dirt,' or ' black dirt,' was evidently an ancient vegetable soil. It is from 12 to 18 inches thick, is of a dark brown or black colour, and contains a large proportion of earthy lignite. Through it are dispersed rounded and sub-angular fragments of stonej from 3 to 9 inches in diameter, in such numbers that it almost deserves th? name of gravel. Many silicified trunks of coni- ferous trees, and the remains of plants allied to Zamia and Cycas, are buried in this dirt-bed, and must have become fossil on the spots where they greAV. The stumps of the trees stand erect for a height of from one to three feet, and even in one instance to six feet, with their roots attached to the soil, at about the same distances from one another as the trees in a modern forest. The carbonaceous matter is most abundant immediately around the stumps, and round the remains of fossil Cycadeae. The fragments of the prostrate trees are rarely more than three or four feet in length; but by joining many of them together, trunks have been restored, having a length from the root to the branches of from 20 to 23 feet, the stems being undivided for 17 or 20 feet, and then forked. The diameter of these near the root is usually about one foot, but sometimes as much as 3i feet. Root-shaped cavities were observed by Professor Hen slow to descend from the bot- tom of the dirt-bed into the subja- cent freshwater stone, which, though now solid, must have been in a soft and penetrable state when the trees grew. The thin layers of calcareous shale (fig. 345) were evidently de- posited tranquilly, and would have been horizontal but for the protru- sion of the stumps of the trees, around the top of each of which they form hemispherical concre- tions. There is also at Portland a smaller dirt-bed, six feet below the principal one, six inches thick, con- sisting of brown earth with upright Cycads of the same species (Man- tel lia nidiformis, Brong., fig 34(>) as those found in the upper bed, but no Coniferte. The weight of the in- cumbent strata squeezing down the compressible dirt-bed has caused the Cycads to assume that form Which has led the quarrymen to call them 'petrified birds' nests,' which suggested to Brongniart the specific name of nidiformis. The annexed figure shows one of these Purbeck specimens, in which the original cylindrical figure has been less distorted than usual by pres- sure, and a figure of the living Cijcas is added (fig. 347) that the student may have an idea of a form so predominant in Mesozoic vegeta- tion. The dirt-bed is by no means confined to the island of Portland, where it has been most carefully studied, but is seen in the same relative position in the cliffs east of Lulworth Cove-, hi Dorsetshire, where, as the strata have been dis- turbed, and are now inclined at an angle of 45, the stumps of the trees are also inclined at the same angle in an opposite direction a beautiful illustration of a change in the position of beds originally hori- zontal (see fig. 348). From the facts above described we may infer, fust, that those beds of the Upper Oolite, called 'the Portland,' which are full of marine shells, were overspread with nuvia- tile mud, becoming dry land, covered by forest, throughout a portion of the space now occupied by the South of England, the climate being such as to permit the growth of the Zamia and Cycas ; secondly, this land at length sank down and was submerged with its forests beneath a body of fresh water, from which sediment was thrown down enveloping fluviatile shells; thirdly, the regular and uniform preservation ofi his thin :H. xvn.J AND ITS PLANT EEMAINS Fig. 345. 291 Freshwater calcareous shale. Dirt-bed and ancient forest. Lowest freshwater beds of the Lower Purbeck, Portland stone, marine. Section in Isle of Portland, Dorset. (Buckland ami De la Beche,) Fig. 347. Mg. 346. Mantellia nidiformis, Brongn. The upper part shows the woody stem ; the lower part the bases of the leaves. Fig. 348. Cycas circinalis, L. Living in the East Indies. Freshwater calcareous shale. Dirt-bed, with stools of trees. Freshwater. Portland stone, marine. Section of cliff east of Lulworth Cove. (Bucklaud and De la Beche.) U2 292 PORTLAND OOLITE [CH. Xvit. bed of black earth, over a distance of many miles, shows that the change from dry land to the state of a freshwater lake or estuary, was not accompanied by any violent de- nudation, or rush of water, since the loose black earth, together with the trees which lay prostrate on its surface, must inevitably have been swept away had any such violent catastrophe taken place. The forest of the dirt-bed was neither the first nor the last which grew in this region. Besides the lower bed containing upright Cy- cadese, just mentioned, another has sometimes been found above it, which implies oscillations in the level of the same ground, and its alternate occupation by land and water more than once. The plants of the Purbeck beds, so far as our knowledge extends at present, consist chiefly of Ferns, Conifers, and Cycads, without any Dicotyledonous Angiosperms ; the whole being more allied to the Jurassic than to the Cretaceous vegetation. The same affinity is indicated by the vertebrate and in- vertebrate animals. Mr. Brodie has found the remains of insects of the orders Coleoptera, Diptera, Or- thoptera, Hemiptera, and Neuro- ptera, and these orders have modern species, some of which now live on plants,, while others hover over the surf ace of rivers (see Note Q, p. 605). Remains of Chelonia, of the genus Platemys, of a Crocodile (Goniopholis) , and Ganoid fish have also been found in the strata. Fig. 349. Portland Oolite and Sand. The Portland Oolite has already been mentioned as forming, in Dorsetshire, the foundation on which the freshwater limestone of the Lower Purbeck reposes. An in- terval of time and some change in the physical geography of the area occurred after the deposition of the Portland stone, for it was upheaved and worn and depressed before the Purbecks were deposited upon it. The well-known building- stone of which St. Paul's and so many of the principal edifices of London are constructed is Portland free-stone. About fifty species of mollusca occur in this formation, among which are some Ammonites of large size, such as Ammonites (Perisphinctes) giganteus, Sow. A. (Perisphinctes) biplex, Sow., also occurs. The cast of a spiral univalve called by the quarry-men the ' Portland Screw ' (a, fig. 349), is common ; the shell of the same (b) being rarely met with. Also Trigonia gibbosa, Sow. (fig. 351) and Cardium dibsimile, Sow. (fig. 352). This upper member rests on a dense bed of sand, called the Portland sand, containing similar marine fossils, such as Ostrea expansa, Sow. (fig. 353), below which is the Kimeridge clay. Corals are rare in the Portlandian, although one species is found plentifully at Tisbury, Wiltshire, in the Portland sand, converted into flint or chert, the original cal- careous matter being replaced by silica (fig. 350). The Kimeridge Clay consists, in great part, of a blue shale, Co-it hium portlandicum, Sow. sp., . a. Cast of shell known as 'Portlaud screw.' fe. The shell itself. CH. XVIT.] KIMERIDGE CLAY 293 sometimes becoming highly carbonaceous, and passing into a coaly material (Kimeridge coal). This carbonaceous matter is probably of animal rather than of vegetable origin. Among the fossils, amounting Fig. 350. to nearly 100 species, may be mentioned Cardium striatulum, Sow. (fig. 354), and Ostrea del* toidea, Sow. (fig. 355), the latter Fig. 351. Isastrcca oblonga, M. Ethv. aud J. Haimc, mag. 2 diams. Converted into cliert from the Portland Sand, Tisbury, Trigonia yiblosa, Sow. \ nat, size. a. The hinge. Portland Stone, Tisbury, Fig. 352. Fig. 353. Cardium dissimile, Sow. \ nat. size. Portland Stone. Fig. 354. Ostrea expansa, Sow. Portland Sand* Fig. 355. Fig. 356. Cardium strialulum, Sow., \. Ostrea deltoidea, Sow. Exogyra virgula, Defr., , Kimeridge Clay, Kimeridge Clay. nat. size. Kimeridge Clay. Hartwell. found in the Kimeridge clay throughout England and the North of France, and also in Scotland, near Brora. Many Foraminifera occur, and many forms of Ammonites. The Exogyra virgula, Defr. 294 CORAL RAG [CH. XVII. Fig. 357. (fig. 356), also met with in the Kimeridge clay near Oxford, is so abundant in the Upper Oolite of parts of Prance, as to have caused the deposit to be termed ' marnes a virgules.' The Aptychi of Ammonites (fig. 357) are also widely dispersed through this clay. Middle Oolites. These consist of the Coral- line Oolite beds of limestone, in some places con- taining many corals above, and the thick mass of blue clays, known as Oxford clay, below ; the base of the series being formed by the sandy stone known as the Kellaways rock. Coral Rag. One of the limestones of the Middle Oolite has been called the ' Coral Rag,' because it consists, in part, of beds of fossil corals, some of them retaining the position in which they grew at the bottom of the sea. In their forms they frequently resemble the reef -building corals of the Pacific. Aptychus. Kimeridge Clay. Fig. 358. Fig. 359. Thecosmilia annularis, Flem., Coral Rag, Steeple Ashton. Fig. SCO. Thamnastraa arachnoides,Tai'k. sp. Coral Rag, Steeple Ashton. Fig. 361. Ostrea gregaria, Sow., \. Coral Rag, Steeple Ashton. Nerincea Goodhallii, Sow. J nat. size. Coral Rag, Weymouth. The number of species is small. They belong chiefly to the genera ThecQsmilia (fig. 358), Protoscris, and Thamnastrcea (fig. 359), and sometimes form masses of coral fifteen feet thick. Echinodermata are numerous, Cidaris florigcmma, Phil., -yvith species of Pygiirv.s, CH. XVII.] OXFORD CLAY 295 Pygaster, and Hemicidaris, being characteristic. These coralline strata extend through the calcareous hills of the north-west of Berk- shire and north of Wilts, and again recur in Yorkshire, near Scar- borough. The Ostrea gregaria, Sow. (fig. 360), is very characteristic of the formation in England and on the Continent. Fig. 302. Fig. 36:5. One of the limestones of the Jura, referred to the age of the English coral rag, has been called ' Nerinaean limestone ' (Calcaire a Nerinees), Nerin&a being an extinct genus of uni- valve shells (fig. 361), much resembling Cerithium in ex- ternal form, and common in the Jurassic rocks. The annexed section shows the curious and continuous ridges on the colu- mella and whorls. Oxford Clay. The coralline limestone, or ' coral rag,' above Fig. 364. Beieoutitet hastatui, Blain., Oxford Clay. Belemnitps Puzosianits, D'Orb., J. Oxford Clay, Christian Malford. Section of the shell projecting from the phragmacone. b-c. External covering to the ink-bag and phrag- macone. c, d. Osselet, or guard, or that portion commonly called the belemnite. e. Conical chambered body called the phragma- cone. /. Position of ink-bag beneath the shelly co- Ammonites (Cosmoccras) Jcuon, Reinecke. (A. Elizabethcc, Pratt.) Oxford Clay, Christian Malford, Wiltshire. described, and the accompanying sandy beds, called ' Calcareous grits,' of the Middle Oolite, rest on a thick bed of clay, called the ' Oxford clay,' sometimes not less than 600 feet thick. In this there are no corals, but great abundance of Cephalopoda belonging to the Ammo> 296 GKEAT OOLITE [CH. xvn. nites and Belemnites. In some of the finely laminated clays, Ammo- nites are very perfectly preserved, although somewhat compressed, and they are frequently found with the lateral lobe extended on each side of the aperture into a horn-like projection. (See fig. 364.) In the same clays the soft parts of the Belemnite, including the ink-bag, are also found (fig. 363). Remains of the Reptilian genera Ichthyosaurus, Pliosaurus, Plesiosaurus, Megalosaurus, and RhamphorJiynchus, are found in the Oxford clay. Kellaways Bock. The arenaceous limestone which passes under this name is generally grouped as a member of the Oxford clay, in which it forms, in the south-west of England, lenticular masses, 8 or 10 feet thick, containing at Kellaways, in Wiltshire, numerous casts of Ammonites, and other shells. But in Yorkshire this calcareo- arenaceous formation thickens to about 30 feet, and constitutes the lower part of the Middle Oolite, extending inland from Scarborough in a southerly direction. The lower Oolites consist, in the South of England, of a somewhat variable series of deposits which generally retain the names given to them by William Smith. In Yorkshire, however, they are represented by a thick series of sands and clays, with some thin beds of coal, the whole being evidently of estuarine origin, and yielding many interesting remains of land-plants. Cornbrash and Forest Marble. The upper division of this series, which is more extensive than the preceding or Middle Oolite, is called in England the Cornbrash, as being a brashy, easily broken rock, good for corn land. It consists of sandy limestone and clay, which pass downwards into the Forest-marble, an argillaceous lime- stone, abounding in marine fossils. Brachiopods are very abundant, and the Echinoidea, Ecliinobrissus clunicularis, Llhwyd, E. orbi- cularis, Phil, sp., and Holectypus depressus, Lam. sp., and also the bivalve Avicula cchinata, Sow., are common. In some places, as at Bradford, near Bath, this limestone is replaced by a mass of clay. The sandstones of the Forest-marble of Wiltshire are often ripple- marked and filled with fragments of broken shells and pieces of driftwood, having evidently been formed on a coast. In the same stone the claws of crabs, fragments of Echini, and other signs of a neighbouring beach, are still observed. Great (or Bath) Oolite. Although the name of ' coral-rag ' has been appropriated, as we have seen, to the highest member of the Middle Oolite before described, some portions of the Lower Oolite are equally entitled in many places to be called coralline limestones. Thus the Great Oolite near Bath contains various corals, among which Calamophyllia radiata, Lam. (fig. 365), is very conspicuous, single individuals forming masses several feet in diameter, and having probably occupied much time in growing, like the large existing Brain-coral (Meandrina) of the tropics. Different species of Crinoids, or stone-lilies, are also common in the same rocks with the corals ; and, like them, must have lived on a firm bottom, where their base of attachment remained undisturbed, for years (c, fig. 366). Such fossils, therefore, are almost confined to the limestones; but an exception occurs at Bradford, near Bath, in the Forest-marble series, where they are enveloped in clay some- times sixty feet thick. In this case, however, it appears that the solid CH. XVII.] BRADFORD CLAY 297 upper surface of the ' Great Oolite ' had supported, for a time, a thick submarine forest of these beautiful crinoids, until the clear and still water was invaded by a current charged with mud, which threw down the ' stone-lilies,' and broke most of their stems short off Fig. 365. Calamophyllia radiala, Lamouroux. a. Section transverse to the tubes. b. Vertical section, showing the radiation of the tubes. c. Portion of interior of tubes magnified, showing striated surface. near the point of attachment. The stumps still remain in their original position ; but the numerous ossicles, once composing the stem, arms, and body of the encrinite, were scattered at random through the argillaceous deposit, in which some now lie prostrate. These appearances are represented in the section 6, fig. 366, where Apiucnnus rotundus, Mill., or Pear Encrinite. Fossil at Bradford, Wilts. a. Stem of Apivcri>/us, and one of the articulations, natural size. b. Section at Bradford of Great Oolite and overlying clay, containing the fossil encrinites. c. Three perfect individuals of Apiocrinns, represented as they grew on the surface of the Great Oolite. d. Body of the Apiocnnus rotundas, Mill. Half nat. size. the darker strata represent the Bradford clay. The upper surface of the calcareous stone below is completely incrusted over with a continuous pavement, formed by the stony roots or attachments of the Crinoidea ; and, besides this evidence of the length of time they 298 BATH OOLITE [CH. XVII. had lived on the spot, we find great numbers of single joints, of the stem and body of the encrinite, covered over with Serpulce. Now Fig. 367. Fig. a. Single plate of body of Apiocrinus, overgrown \viti\Serpulce and Bryozon. Natm-al size. Bradford Clay. b. Portion of the same 'magnified, showing the bryozoan Diastopora diluvutna, M. Edw., covering one of the Serpulce. these Serpulce could only have begun to grow after the death of some of the ' stone-lilies,' parts of whose skeletons had been strewed over the floor of the ocean before the irrup- tion of argillaceous mud. In some instances we find that, after the parasitic Serpulcs were full grown, they had become incrusted with a bryozoan, called Diastopora diluviana, M. Edw. (see b, fig. 367), and many genera- tions of these molluscoicls had succeeded each other in the pure water, before the whole became fossil. The calcareous portion of the Great Oolite consists of several shelly limestones, one of which, called the Bath Oolite, is much cele- brated as a building-stone. In parts of Gloucestershire, especially nerr Minchin- hampton, the Great Oolite, according to Lycett, Ammonites (StepJianoceras) mact-ocephaluK. Schloth. nat. size. Great Oolite and Oxford Clay. Fig. 370. Fig. 369. Fig. 371. Terebratula diyona. Sow., nat. size. Bradford Clay. Purpuroidea nodulata, Y.& B. sp., \ nat. size. Great Oolite, Minchin- hampton, Cylindrites acutus, Sow. Syn. Actceon acutus, nat. size. Great Oolite, Minchuihampton. CH. XVII.] STONESFIELD SLATE 299 ' must have been deposited in a shallow sea, where strong currents prevailed, for there are frequent changes in the mineral character of the deposit, and some beds exhibit false stratification. In others, heaps of broken shells are mingled with pebbles of rocks foreign to the neighbourhood, and with fragments of abraded corals, dicotyle- donous wood, and crabs' claws. In such shallow-water beds, shells of the genera Patella, Ncrita, Rimula, and Cylindrites are common (see figs. 370 to 374) ; while cephalopods are rare, and, instead of Fig. 372. Fig. 373. Fig. 374. 1 'atell 'a rugosa. Sow., Great Oolite. Nerita costulata, Desh., mag. 2 diams. Great Oolite. Simula (Emarginula) clathi'uta, Sow., mag. 3 diams. Great Oolite Ammonites and Belemmtes, numerous genera of carnivorous gastro- pods appear. Stonesfield Slate: Mammalia The slate of Stonesfield was shown by Lonsdale to lie at the base of the Great Oolite. It ds a slightly oolitic shelly limestone, forming large lenticular massestem- bedded in sand ; it is only six feet thick, but very rich in organic remains. The remains of Belemnitcs, Trigonice, and other marine remains, with fragments of wood, are common, and Fig. 375. impressions of ferns, Cycadeae, and Conifers. Portions of insects, also, among which are Fig. 376. Tupaia Tana, Raff. Right ramus of lower jaw.. Natural size. A recent insectivorous pla- cental mammal, from Sumatra. the wings of a butterfly, and the elytra or wing-covers of beetles are perfectly preserved (see fig. 375), some of the latter approaching the genus Buprestis. The remains, also, of many genera of rep- tiles, such as Ichthyosaurus, Pliosaurus, Plesiosaurus, Cetiosaurus, Teleosaurus, Mcgalosaurus, and Rhamphorhynchus, have been dis- covered in the same limestone. There have also been discovered no less than ten specimens of lower jaws of marsupial mammiferous quadrupeds, belonging to four different genera, for which the names of Amphitherium (figs. 381, 382), Amphilestes, Phascolothcrium, and Stereognathus have been adopted, 300 MAMMALS OF STONESFIELD SLATE [CH. xvn. The second mammiferous genus discovered in the same slates was named originally by Mr. Broderip Didelphys Bucklandi (see fig. 383), and has since been called Phascolotherium by Owen. Fig. 377. Fig. 378. Fig. 379. Fig. 380. Part of lower jaw of Tupaiti Tana, Raff. Twice natural size. Fig. 377. End view seen from behind, showing the very slight inflection of the angle at c. Fig. 378. Side view of same. Part of lower jaw of Didelphys Azara?< Temm. ; recent, Brazil. Natural size. Fig. 379. End view seen from behind, showing the inflection of the angle of the jaw, c, d. Fig. 380. Side view of same. a. Coronoid process. Pr 'fat Marshii, Sow. i nat. Middle and Lower Oolite. The Upper lias. The lower portion of the Jurassic system is known as the Lias, and it consists of three divisions. The Upper Lias consists of dark blue clays, containing some septaria, and passes upwards into beds of sand, and downwards into harder nodular bands, which sometimes contain the remains of fish and insects. The blue clays are sometimes highly pyritous, and in York- Fig. 397. Ammonites (Httdocerax) kifr-nns, Brng. A. Walcotti, Sow., |. Upper Lias shales. shire were formerly used for the manufacture of alum ; they also contain masses of wood converted into jet. The most common fossils of the Upper Lias are Ammonites (StepJianoceras) communis, Sow., Am. (Hildoceras) bifrons, Brug. (fig. 397), Am. (Harpoceras) serpentimis, Kein., Am. (Phylloceras) heterophyllus, Sow., and Leda ovum, Sow. The Middle lias consists of a ferruginous limestone full of shells, known as the Marlstone rock-bed ; this rock sometimes passes into an ironstone. The valuable iron- ores of Cleveland, in the North OH. XVII.] MIDDLE LIAS 305 of Yorkshire, are of this age, and consist of oolitic limestones which have been more or less completely converted into masses of ferrous carbonate. Fossils are very abundant in the Middle Lias, amongst the most characteristic being Ammonites (Amaltheus) spinatus, Brug., Am. (Amaltheus) margaritatus, Montf. (fig. 398), Am. (JEgoceras) Henleyi, Sow., Am. (sEgoceras) capricornus, Fig. 399. Schloth., with Pecten aqui- valvis, Sow., and Rhynchonella tetrahedra, Sow. Among the most beautiful of the fossils of this division we may in- stance the fine Ophiurid (Brittle Starfish) Palceocoma tenui- brachiata, E. Forbes (fig. 399). Fig. 398. Ammonites (Amaltheus') margarita- tus, Montf. Syn. A.Slokesii, Sow. ; A. Clecelandicus, Y. and B. Middle Lias. i. Palceocoma (< >phioderma) tenuibrachiata, K Forbes, sp. Middle Lias, Seatown, Dorset. The tower Lias consists in its upper part of thick beds of shale, and in its lower of numerous alternations of shale and shelly limestone, the latter being replaced at the base of the series by cam- pact argillaceous limestones, which are largely employed in the manufacture of hydraulic cements. In North Lincolnshire, at Scunthorpe and Froddingham, the shelly limestones of the Lower Lias are found to be converted into ferrous carbonate, which is worked as an iron ore. In all the divisions of the Lias and Oolite we are able to recognise the existence of a succession of Zones, each of whioh is distinguished by a characteristic assemblage of fossils. These zones, although so clearly recognisable by their fossil contents, appear usually to pass insensibly into one another, and are not necessarily distinguished by any changes in the mineral characters of the strata. Each zone is named after one of the most striking of the fossils which it contains, and in the case of the Mesozoic rocks, species of Ammonites are usually selected for the purpose. In the Lower Lias the succession of zones is especially distinct and well marked. The commonest Ammonites of the Lower Lias are Arietites obtusns, Sow., A. Turncri, Sow., A. ftemicostatw, Y. and B., A. 306 THE LOWEK LIAS Ten. xvn. Fig. 401. Ammonites (Arietites) BticMandi, Sow. (A. bimlcatus, Brug.) diameter of original. a. Side view. Z>. Front view, showing mouth and bisulcated keel. Characteristic of the Lower Lias of England and the Continent. Fig. 402. Am. (sEgoceras) plftnorbis, Sow. A diameter of original. From the base of the Lower Lias of England and the Continent. Fig. 403. Nautilus truncatus, Sow. Lias. & nat. size. Fig. 404. Gryphcea incurva, Sow. (G. arcuala, Lam.) i. Lias. Hippopodium ponderosum, Sow. i diameter. Lias. Cheltenham. Lima gigantea, Sow., \. Lias. if. XVII.] AND ITS FOSSILS 307 Bucklandi, Sow. (fig. 400), with JEgoceras angulatus, Sow., and ^37. planorbis, Sow. (fig. 401). Belemnites of many species abound, and examples of the persistent type Nautilus are not rare (fig. 402). Among other very common fossils of the Lower Lias are Gryplicea arcuata, Lam. (G. incurva, Sow.) (fig. 403), Lima gigantea, Sow. (fig. Fig. 40(5. vicitla iiueqitivalvis, Sow., . Lower Lias. Fig. 408. Avicula cygnipes, Phil.,i. Lower Lias, Gloucestershire and Yorkshire, a. Lower valve. 6. Upper valve. Fig. 409. Spirlferina Walcotli, Sow., A. Lower Lias. Ej:tracrinus (Ptmtctcrinus) Briareus, Lept&na Moard, Dav. Mill., i natural size. "Upper Lias, Ilminster, (Body, arms, and part of stem.) Lower Lias, Lyrne Regis. 405), Avicula inaquivahis, Sow. (fig. 406), and A. cygnipes, Phil, (fig. 407), Hippopodium ponicrosiim, Sow. (fig. 404), with Spiriferina Walcotti, Sow. (fig. 409), and the minute Lcpt&na Moorei, Dav. (fig. 410). The ossicles of the beautiful crinoid Pentacrinus, of which a very perfect example is represented above, also abound in the Lower Lias. x2 308 RHJETIO BEDS [CH. XVII, "White Xiias and Rhaetic Strata. Beneath the Lower Lias just described, we find at certain localities beds of a cream-coloured lime- stone (called the White Lias by William Smith), under which occur black pyritous shales and sandstones with an interesting assemblage .of marine mollusca, some of the most characteristic of which are represented below. Fig. 411. Fig. 412. Fig. 413. Cardium rhceticum, Mercian. Nat. size. Bhsetic Beds. Pecten valoniensis, Dfr. \ nat. size. Portrush, Ireland, &c. Bhsetic Beds. Avicula contorta, Portlock. Porbrush, Ireland, &o. Nat. size. Rhaetic Beds. In the midst of these black shales is found a band almost made up of the bones and teeth of fish and saurians. Some of the fish remains awe identical with forms found in the Trias of Germany. Fig. 414. Fig. 415, Fig. 416. ffybodus plicatilis, Ag. Teeth, Bone-bed. Aust and Axmouth. Saurichthys apicalis, Ag. Tooth ; natural size and magnified. Axmouth. Gyrolepis tenuistriatus, Ag. Scale : nat. size and magnified. Axmouth. These strata, which are of insignificant thickness and are known by the names of the Zone of Avicula contorta, Portl., the Infra Lias, and the Penarth beds, are of great interest as representing what in the Alpine district constitutes a great formation, several thousands of feet in thickness, known as the Khastic system, which appears to completely bridge over the interval between the Jurassic and Triassic systems. In England, Germany, and North America, teeth of a minute mammal, nearly the oldest as yet known, have been found. The British and German form is known as Microlestes (fig. 417), and the American as Dromatheriwn. Freshwater and terrestrial deposits of Jurassic age are found in CH. XVII.] KH^TIC MAMMALS 309 this country represented by the Purbecks of the South of England, the sandstones and shales with thin beds of coal of Lower Oolite age of Yorkshire, and various estuarine and freshwater beds which Fig. 417, Microlestes antiquus, Plieninger. Molar tooth, magnified. Rhgetic Diegerloch, near Stuttgart, WUrtemberg. a. View of inner side* ? 6. Same, outer side ? c. Same in profile. d. Crown of same. alternate with marine strata, from the Lias to the Upper Oolite inclusive, on the east coast of Sutherland. Similar strata attain a great thickness on the west coast of Scotland and the Inner Hebrides. Even in the English Lias, at the base of the Upper and Lower divisions respectively, we find beds crowded with the remains of insects, small crustaceans, and fish with occasional marine brackish -water and even freshwater shells which have pro- Fig. 418. Wing of a neuropterous insect, from the Lower Lias, Gloucestershire. (Rev. P. B. Brodie.) The line below the figure indicates the length of the object. bably been formed in shallow- water lagoons close to the land. The exquisite preservation of some of the insect remains discovered and described by the Eev. P. B. Brodie is illustrated by the accompanying figure. The classification of the Jurassic strata of this country was esta- blished on a sound basis by William Smith in 1815. The labours of Marcou in France, of Oppel and Queiistedt in Germany, and of Dr. Wright in this country have shown how widespread and distinctive are the various zones in this system of stratified rocks. The Jurassic rocks of Yorkshire have been described in the 'Geology of Yorkshire' of the late Professor John Phillips, and those of the South of England in the 'Geology of Oxford' of the same author. The correlation of the northern and southern types of Jurassic rocks in this country has been discussed in the Geological Survey Memoir on Rutland (1875). More recently, the Geological Sur- vey has published a series of Memoirs dealing with the same subject, entitled ' The Jurassic Rocks of Britain,' by C. Fox-Strang- ways and H. B. Woodward. 310 TRIASSIC SYSTEM [en. xvm. CHAPTER XVIII THE TRIASSIC SYSTEM Subdivisions of the Trias in England Corals, Echinodermata, Braehiopoda, Lamellibranchiata, Gastropoda, and Cephalopoda of the Trias Fish, Amphibians, and Reptiles Terrestrial Flora of the Trias Triassic Mammalia The Keuper and its Reptilia The Dolomitic Conglomerate Elgin Sandstones The Bunter Formation of Red Sandstones and Clays Rock-salt, Gypsum, &c. Nomenclature and Classification of the Triassic Strata. The name of Trias was first given to this great division of the Geological Series by the Germans, from the circumstance that, in Central Europe, the system consists of three members. The uppermost of these is called the Keuper (from the name given in Coburg to a kind of particoloured cloth), the middle is known as the Muschelkalk (shelly limestone), while the lowest receives the name of Bunter (variegated). The term ' Trias ' is now almost universally employed for the strata of this age, though the French sometimes apply to it the name of ' Saliferous,' owing to its containing important deposits of rock-salt. In the British Islands, the Triassic strata bear so close a general re- semblance to those of Permian age, which underlie them, that the older writers grouped these two formations together as 4 New Bed Sandstone,' the name being given in recognition of the fact that the coal-bearing strata are underlain by red and variegated beds of Devonian age (Old Red Sandstone) and over- lain by others of Permian and Triassic age (New Red Sandstone). Conybeare and De la Beche proposed to designate the whole of the New Red Sandstone as Poikilitic, on account of the varie- gated tints of its strata. In Britain and the greater part of France the middle division of the Trias the Muschelkalk is absent, and the system consists only of two members, the Bunter (or Gres bigarre of French authors), and the Keuper (or Marnes irisees of the French). The general order of succession in the Trias of Britain and the comparison of its subdivisions with equivalent strata on the continent of Europe are shown in the following table. CH. XVIII.] SUBDIVISION OF TRIAS 311 NOMENCLATURE OF TRIAS German Keuper Muschelkalk French Marnes irisees Muschelkalk, ou cal- caire coquillier Bnnter-Sandstein. Gres bigarre English Ked and grey sali- ferous and gyp- seous shales and sandstone, with rock salt. Dolomitic conglo- merate. , Wanting in Eng- 1 land. Ked sandstone and f pebble beds and J quartzose con- glomerate. Soft v red sandstones. Characteristics of the Triassic Fauna and Flora. 111 Britain and Central Europe generally, the marine fauna of the Trias is almost entirely unrepresented. The strata of that area appear to have been deposited for the most part in great salt- water lakes like the Caspian, and the Mollusca, when preserved, are few and often dwarfed. Even in the Muschelkalk, which contains great numbers of individuals, the variety of forms re- presented is not very great. It is necessary to go to the Alpine Trias of the South of Europe in order to form an idea of the rich and varied character of the marine fauna and to study the curious relations which it has with that of the Jurassic on the one hand, and that of the Permian and Carboniferous on the other. The red sandstones may be of desert origin. . Corals are very abundant in some of the strata of the Alpine Trias, and by some authors the formation of the great calcareous masses which are now converted into dolomite, and form such conspicuous mountains in the Tyrol is believed to be due to the action of reef-building corals of the period. The Echinoderms resemble those of the other Mesozoic rocks, the genus Encrinus being very well represented (fig. 419). Star-fish of Mesozoic types also occur (fig. 420). The Echini are of Mesozoic types, but are all regular forms ; the irregular forms, so abundant in the Jurassic and Cretaceous, appear not to have made their appearance in Triassic times. The Brachiopods are very abundant, but do not show in Triassic times that predominance over the Lamellibranchiata which is so distinctive of Palaeozoic faunas. Some of the genera are related to those of the Palaeozoic, others to those of the Mesozoic, while a few, like Koninckia (fig. 421), are confined to the Trias. 312 ECHINODEEMS AND BRACHIOPODS [CH. xvui. Among the very varied Lamellibranchiate fauna certain genera are very conspicuous, such as Gervillia (fig. 422), Myophoria (the precursors of the Jurassic and Cretaceous Trigonice), Halobia, Daonella, Megalodon, &c. Fig. 419. Fig. 420. Encrinus liliiformis, Schloth., . Body, arms, and part of stem. a. Section of stem. Muschelkalk. Aspidura loricata, Ag. Upper side. b. Lower side. Muschelkalk. Fig. 421. Koninckia Leonhardi, Wissmann. a. Ventral view. Part of ventral valve removed to snow the vascular im- pressions of dorsal valve. I. Interior of dorsal valve, showing spiral processes restored. c. Vertical section of both valves. Part shaded black showing place occu- pied by the animal, and the dorsal valve following the curve of the ventral. Gastropoda are very abundant in the Trias, and among them also we find an admixture of Palaeozoic types, like Murchisonia, Scoliostoma (fig. 423), and Loxonema, with Jurassic forms, such CH. XVIII.] AMMONOIDEA OF TEIAS 313 as Cerithium, Emarginula, &c. A few genera, like Platystoma (fig. 424), are peculiar to the Trias. Fig. 422. Fig. 423. Fig. 424. Gervillia (Avicula) socialis. Schloth., nat. size. Found in the Muschelkalk and Keuper. Scoliostoma, St. Cassian. Platystoma Sufss-ii, Homes. From Hallstadt. Fig. 425. The Cephalopoda of the Trias are particularly interesting. The persistent genus Nautilus is well represented, and we find with it the last repre- sentatives of the Pa- laeozoic Orthoceras. The most important representative of the Ammonoidea in the Trias is the charac- teristic Ceratites (fig. 425), but many re- markable genera of true Ammonites also occur. Among the Ammonites of the Trias some exhibit curiously foliated septa (see fig. 426, d, e,f), Ceratites nodosus, Schloth., J. Muschelkalk, Germany. Side and front views, showing the peculiar forms of the septa dividing the chambers. Fig. 426. a, b, c. Trachyceras Aon, Miinst. An Ammonite with very simply foliated sutures. d, e,f. Arcestes multilobatus, Brown. An Ammonite exhibiting sutures with very complicated foliations. 314 LABYRINTHODONTIA [CH. XVIII. while others in the simplicity of the foliation of the sutures approach the Ceratites (see fig. 426, a, b, c). Many of the Am- monoidea are peculiar to the Trias, but others lived on into the Jurassic and Cretaceous. In Atractites and Aulacoceras (fig. 427) we have interesting forerunners of the great group of the Belemnites. Fig. 427. Fig. 428. Aulacoceras sulcatum, Hau., \ nat. Exterior of shell, b. Longitudinal section showing septa, c. Cross-section. The siplmncle. which is very thin, lies on the edge of the septa. Fig. 429. Tooth of Labyrinthodon ; nat. size. Warwick sandstone. Transverse section of upper part of tooth of Labyrinthodon Jaegeri, Ow. ( Mastodon - saurus Jaegeri Meyer) ; natural size, and a segment magnified. a. Pulp cavity, from which the processes of pulp and dentine radiate. The Fish of the Trias include both Ganoids and Selachians. Among the former we find a great number with heterocercal tails like those of Palaeozoic times mingled with others with homocercal tails like those of the Jurassic. The remarkable CH. XVIII.] REPTILIA OF TRIAS 315 Dipnoid genus Ceratodus, which is still living in Queensland, is represented in the Trias ; and we also find the first representa- tives of the Teleostei or bony fishes of our modern seas. Fig. 4 30. Fig. 431. Equisetum arena ceum, Schimp. Frag- ment of stem, and a small portion of same magnified. Keuper. Fig. 482. a. Volt 2 in heterophylla. Brong. b. Portion of same magnified to show fructification. Sulzbad. Buuter-Sandstein. Lanosaurus Balsami, Curioni. Skeleton i nat. size. From the Muschelkalk, Lake Como, Italy. Amphibians were very abundant in the Trias, and are re- j'erred to the group of the Stegocephala or Labyrinthodoritia (figs. 428, 429). With these remarkable amphibians we find representatives of marine reptiles like Lariosaurus (fig. 430), which appear to 316 TERRESTRIAL REPTILES AND [CH. xvm. have been the precursors of the gigantic Enaliosauria (Ich- thyosauria and Plesiosauria), so abundant in the Jurassic period. The Terrestrial flora of the Trias consists of Conifers and Cycads, the former being represented by Voltzia (fig. 432), Albertia, &c., and the latter by Pterophyllwn, Zamites, Pseu- dozamites, Podozamites, Otozamites, &c. Ferns are abundant, and a true Equisetum is also found (fig. 431). Land Reptiles are represented by numerous forms in the Trias. Many of these belonged to the Rhynchocephalia. Crocodiles and Dinosaurs are found in the Trias, but no true Pterosauria, Lacertiha, Ophidians, or Chelonians. Many of the Reptilia of Fig. 434. Fig. 433. Palatal teeth of Placodus gigas, Tritylodon Fraasii, Lydekker. Upper true molar Ag. Muschelkalk. tooth. The two central figures of the natural size, the others enlarged ( x 3). o shows the triconodont character of the crown the Trias (Theriomorpha) % $ JJjfiJ 1 ' ". h lateral > ' ** curiously simulate the 'From the Upper Trias of Strasburg. Mammalia in the forms of their skulls, the position of their eyes, the differentiation of their teeth, and other characters. To this remarkable group of Reptiles are referred not only the Dicynodon and many other forms with large teeth, like canines, but also the Placodus, with its flat palatal crushing teeth, which was formerly regarded as a fish. Birds were unknown in the Trias, but mammals are repre- sented by the Tritylodon (see fig. 435) from South Africa and some similar lowly forms (Prototheria or Allotheria), which have been found in the highest portion of the Trias and in the over- lying Rhaetic. The remarkable ' triconodont ' teeth which characterised these oldest known mammals are illustrated in fig. 434. CH. XVIII.] MAMMALS OF THE TEIAS 317 Professor Seeley regards Tritylodon as possibly being not a true mammal but a synthetic type intermediate between the Fig. 435. Tritylodon longcevus, Owen. Skull with one side restored, nat. size. a. Palatal view of skull, showing molars and broken canines. b. Upper surface of skull. From the Trias of Basutoland, South Africa. mammal-like reptiles (Theriodontia) and the lowest Mammalia (Allotheria) (Note R, p. 606). British Representatives of the Tr lassie System. The British Triassic strata consist of a succession of variegated sands and clays containing very few fossils, which it is often difficult to separate from the underlying Permian strata. The Keuper. This upper division is of great thickness in Lancashire and Cheshire, attaining 3,450 feet in the last-mentioned county, and it covers a large extent of country between Lancashire and Devonshire, but it thins out rapidly to less than half its thickness in Staffordshire. It consists of New Ked Marl at the top, with red and grey shales and marls, rock-salt and gypsum being important minerals in it ; and it rests on thinly laminated micaceous sandstones and waterstones, with a base of calcareous conglomerate or breccia. In Worcestershire and Warwickshire, in sand- stone belonging to the uppermost part of the Keuper, the bivalve crustacean Estheria mimita, Albei ti sp., occurs. The member of the English ' New Ked ' containing this shell, in those parts of England, is, according to Sir Eoderick Mur- chison and Mr. Strickland, 600 feet thick, and consists chiefly of red Fig. 436. Esthe Alberti sp. Mag. 2 diams. 818 FOSSIL REPTILES [CH. xvin. marl or shale, with a band of sandstone. Spines of Hybodus, and teeth of other fishes, and footprints of reptiles were observed by the same geologists in these strata. The remains of four saurians have been found. One called Rhynchosaurus occurred at Grinseil, near Shrewsbury, and is characterised by having a small bird-like skull and jaws which Fig. 437. Hyperodapedon Gordoni, Huxley. Left palate, maxillary. (Showing the two rows of palatal teeth on opposite sides of the jaw.) a. Under surface. b. Exterior right side. were produced into a beak. The other three, Telerpeton, Hyperoda- 2>edon (fig. 437), and the crocodilian reptile Stagonolepis, were brought to light near Elgin, in strata formerly supposed to belong to the Old Red Sandstone, but now recognised as Upper Triassic. The Itypero- dape&on was afterwards discovered in beds of about the same age, in the neighbourhood of Warwick, and also in South Devon, and remains of the same genus have been found in Central India and Southern Africa, in rocks believed to be of Triassic age. There has been discovered more recently, near Elgin, an almost complete skeleton of Hypcrodapcdon, which has been described by the late Professor Huxley, and the study of this remarkable specimen has brought out very clearly the points of resemblance of these Old Triassic reptiles to the living New Zealand lizard, Sphenoclon (Hatteria), the sole survivor at the present day of the great order Rhynchocephalia, which is also represented in the Permian strata of Central Europe. The remarkable skull with beaks, and the denti- tion, of Hyperodapedon are represented on the opposite page. The discovery of a living reptile in New Zealand so closely allied to this supposed extinct division of the Reptilia seems to afford an illustration of a principle pointed out by Mr. Darwin of the survival in insulated tracts, after many changes in physical geography, of orders, of which the congeners have become extinct on continents where they have been exposed to the severer competition of a larger and progressive fauna. Still more recently, Mr. E. T. Newton has described some other forms of remarkable reptiles to which he has given the names of Gordonia, Elginia, &c. These are allied to the Dicynodontia and other Triassic reptiles of South Africa. CH. XVTII.] OF ELGIN AND BRISTOL 319 Dolomitic Conglomerate of Bristol. Near Bristol, and on the flanks of the Mendips, in Somersetshire, and in other counties bordering the Severn, the lowest strata belonging to the Trias consist of a conglomerate or breccia, resting unconformably upon the Old Red Sandstone and on different members of the Carboniferous Pins. Hyperodapedon Gordoni, Huxley. Skull and lower jaw (\ nat. size). From Triassic Sandstone, Lossiemoutli, near Elgin. A. Upper surface of skull, showing the orbits, S the supratemporal fossa, S' the lateral temporal fossa, N the anterior nares, and Pmx the pre- maxillary. n. Palate, with teeth PI and maxillary bones MX. (.'. Under side of front of lower jaw, with mandibles Md. The peculiar dentition of the Khynchosauria is well exhibited in this specimen, the rows of closely set conical palatal teeth acting against one another, as the jaw works backwards and forwards. rocks, such as the Coal Measures, Millstone Grit, and Mountain Limestone. This mode of superposition will be understood by reference to the section of Dundry Hill (fig. 114, p. 134), where No. 4 is the dolomitic conglomerate. Such breccias may have been partly the result of the subaerial waste of an old land-surface which gradually sank down and suffered littoral denudation in proportion 320 BUNTEB, BEDS [CH. xvm. as it became submerged. The pebbles and fragments of older rocks which constitute the conglomerate are cemented together by a red or yellow base of dolomite, and in some places the Encrinites, Corals, Brachiopoda, and other fossils derived from the Mountain Lime- stone are so detached from the parent rocks that they have the deceptive appearance of belonging to a fauna contemporaneous with the dolomitic beds in which they occur. Layers of Keuper are noticed between masses of the breccia. The embedded fragments are both rounded and angular, some consisting of Carboniferous lime- stone and Millstone-grit being of vast size, and many weighing nearly a ton. Fractured bones and teeth of saurians which are probably of contemporaneous age have been found in the lower part of the breccia, and two of these, called Thecodontosaurus (from the manner in which the teeth were implanted in the jawbone) and Pal&osaurus, obtained great celebrity because the patches of red conglomerate in which they were found at Durdham Down, near Bristol, were originally supposed to be of Permian or Palaeozoic age, and they were, therefore, considered the only representatives in England of vertebrate Tooth of Thecodonto- animals of so high a type in rocks of such saurus ; 3 times antiquity. The teeth of Thecodontosaurus are magnified. After CO nical, compressed, and with finely serrated bur^Dotomiticcm;- edges (see fig. 438) ; both Thecodontosaurus and glomerate. Durd- Palceosaurus were referred by Professor Huxley ham Down, near to tne Dinosauria. The basement beds of the Keuper rest, with a slight unconformability, upon an eroded surface of the ' Bunter,' next to be described. In these basement beds Professor W. C. Williamson has described the footprints of a labyrinthodont which has been called ' Cheirotherium ' similar to those presently to be mentioned in the Bunter beds ; this Keuper form, however, is peculiar in exhibiting a scaly surface. Lower Trias, or Bunter. The lower division or English representative of the ' Bunter ' attains, according to Sir A. Kamsay, a thickness of 1,500 feet in the Midland counties. Besides red and green shales and red sandstones, it comprises mu Jh soft white quartzose sandstone, in which the trunks of silicified trees have been met with at Allesley Hill, near Coventry. Several of them were a foot and a half in diameter, and some yards in length, the wood being coniferous and showing rings of annual growth. Impressions, also, of the footsteps of animals have been detected in Lancashire and Cheshire in this formation. Some of the most remarkable occur a few miles from Liverpool, in the whitish quartzose sandstone of Storton Hill, on the Cheshire side of the Mersey. They bear a close resemblance to tracks first observed in this member of the Upper New Eed Sandstone, at the village of Hesseberg, near Hildburghausen, in Saxony. For many years these footprints have been referred to a large unknown quadruped, provisionally named Cheirotherium by Professor Kaup, because the marks both of the fore and hind feet resembled impressions made by a human hand (see figs. 440, 441). The footmarks at Hesseberg are partly concave and partly in relief ; the former, or the depressions, are seen upon the upper surface of the CH. XVIII.] FOOTPRINTS OF TRIAS Fig. 440. sandstone slabs, but those in relief are only upon the lower surfaces, being in fact natural casts, formed in the subjacent footprints as in moulds. The larger impressions, which seem to be those of the hind foot, are generally 8 inches in length and 5 in width, and one was 12 inches long. Near each large footstep, and at a regular distance (about an inch and a half) before it, a smaller print of a fore foot. 4 inches long and 3 inches wide, occurs. The footsteps follow each other in pairs, each pair in the same line, at intervals of 14 inches from pair to pair. The large as well as the small steps show the great toes alternately on the right and left side ; each step makes the print of five toes, the first or great toe being bent inwards like a thumb. Though the fore and hind foot differ so much in size, they are nearly similar in form. As neither in Germany nor in England had any bones or teeth been met with, in the same identical strata as the footsteps, anatomists indulged for several years in various conjectures respecting the mysterious animals from which they might have been derived. Fig. 441. Single footstep of ' Chdrotherium. Buiiter-Sandstein, Saxony. One-eighth of natural size. Line of footsteps on slab of sandstone. Hildburghausen, in Saxony. But M. Link conceived that some of the four species of animals of which the tracks have been found in Saxony might have been gigantic Batrachians ; and when it was afterwards inferred that the Labyrinthodon was an Amphibian, it was suggested by Professor Owen that it might be one and the same as the Clieirotherium. Origin of Red Sandstone and Rock Salt. In Cheshire and Lancashire there are red clays of the age of the Trias containing gypsum and salt ihrough a thickness of from 1,000 to 1,500 feet thick. In some places, lenticular masses of pure rock-salt nearly 100 feet thick are interpolated between the argil- laceous beds. At the base of the formation beneatb Ohe rock-salt occur the Lower Sandstones and Marl, called provincially in Cheshire 'water-stones,' which are largely quarried for building. They are often ripple-marked, and are im- pressed with numerous footprints of reptiles. As in various parts of the world red and mottled clays and sand- stones, of several distinct geo- logical epochs, are found associated with salt, gypsum, and magnesiaii limestone, or with one or all of these substances, there is, in all likelihood, a general cause for such a coincidence. Nevertheless, wo must not forget that there are dense masses of red and variegated sand- stones and clays, thousands of feet in thickness, and of vast horizontal extent, wholly devoid of saliferous or gypseous matter. There are also deposits of gypsum and of common salt, as in the blue clay formation of Sicily, without any Y ORIGIN OF BOCK-SALT [CH. xvni. accompanying red sandstone or red clay. These red deposits may possibly be accounted for by the decomposi- tion of gneiss and mica schist, which in the Eastern Grampians of Scot- land has produced a mass of detritus of precisely the same colour as the New Ked Sandstone. It is a general fact, and one not yet very satisfactorily accounted for, that scarcely any fossil remains are ever preserved in stratified rocks in which red oxide of iron abounds; and when we find fossils in the New or Old Red Sandstone in England, it is in the grey, and usually calcareous beds that they occur. Beds of rock-salt are generally attributed to the evaporation of lakes or lagoons communicating at intervals with the ocean. Sir A. Ramsay has remarked in regard to the Trias that it was probably a Continental Period with many inland lakes and seas, the Keuper marls of the British Isles having been deposited in a great lake, fresh or brackish, at the beginning, and afterwards rendered salt by eva- poration. ' Were the rainfall,' he observed, ' of the area drained by the Jordan to increase gradually, the basin of the Dead Sea would by degrees fill with water, and successive deposits of sediment would gradually overlap each other on the shelving slopes of the lake basin in which solid salts had previously been deposited. There are examples of this kind of over- lap in the New Red Marl of England, in Somerset, Gloucester, Herefoi-d, and Leicester. Sir A. Ramsay suggests that the red per- oxide of iron of the sands and clays may in itself be an indication of la- custrine conditions, for each grain of sand and mud is encrusted with a thin pellicle of peroxide of iron, which he thinks could not have taken place in a wide and deep sea. Major Harris, in his 'Highlands of Ethiopia,' describes a salt lake called the Bahr Assal, near the Abyssinian frontier, which once formed the prolongation of the Gulf of Tadjara, but was afterwards cut off from the gulf by a broad bar of lava. ' Fed by no rivers, and ex- posed in a burning climate to the unmitigated rays of the sun, it has shrunk into an elliptical basin seven miles in its transverse axis, half filled with smooth water of the deepest cserulean hue, and half with a solid sheet of glittering snow-white salt, the offspring of evaporation.' ; If,' says Hugh Miller, ' we suppose, instead of a barrier of lava, that sand-bars were raised by the surf on a flat are- naceous coast during a slow and equable sinking of the surface, the waters of the outer gulf might occasionally topple over the bar, and supply fresh brine when the first stock had been exhausted by evaporation.' The Runn of Cutch, as has been shown elsewhere, is a low region near the delta of the Indus, equal in extent to about a quarter of Ireland, which is neither land nor sea, being dry during part of every year, and covered by salt water during the monsoons. Here and there its surface is encrusted over with a layer of salt caused by the evaporation of sea-water. A subsiding movement has been witnessed in this country during earthquakes, so that a great thickness of pure salt might result from a continuation of such sinking For further information on the Triassic rocks of this country, the student may consult the Geological Survey Memoir on ' The Triassic and Permian Rocks of the Midland Counties of England, 1 by E. Hull. The Elgin Sandstone and its fossils have been discussed in memoirs by Murchison, Harkness, Lyell, Hux- ley, Judd, and E. T. Newton. CH. xix.] FOEEIGN TEIASSIC STKATA 323 CHAPTER XIX FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE MESOZOIC STRATA OF THE BRITISH ISLES Secondary Strata of Central Europe Keuper, Muschelkalk, and Bunter The Black, Brown, and White Jura Planer and Quader Beds Chalk of Maestricht and Faxoe Freshwater Strata Wealden of Hanover Strata of Aix-la-Chapelle Secondary Strata of the Alpine Regions Hallstadt and St. Cassian Beds Alpine Jurassic, Tithonian, and Neocomian Hippurite Limestones Secondary Strata of Russia, India, and South Africa Secondary Strata of North America Newark Formation Strata of the Eastern States and of the Western Terri- tories. IT is a remarkable fact that, although the base of the Triassic rocks in Europe is not always readily separable from the Palaeo- zoic formation beneath, there is a vast palaeontological break between them. The Mesozoic age commenced when the first deposits of the Trias accumulated, and many hundreds of species common in the lower rocks ceased to exist, whilst a great marine fauna soon prevailed of an almost totally different kind from that which previously existed. The plants of the Mesozoic age were foreshadowed in the Palaeozoic, and some genera persisted into the Trias, but the majority ceased to exist. Some of the lower forms of invertebrate life persisted through the great change in the physical geography of the world which commenced at the close of the Carboniferous age. Of the corals, not a genus or species lived on, and many new genera are found in the marine Trias. Nautilus and Orthoceras lived on as genera, and Am- monites, which commenced in Permian times, attained an enormous development. A great change occurred in the Mol- luscaand Crustacea. The Labyrinthodontia persisted, and many genera of Ganoid fish. But the number of genera of all kinds, animals and plants, which passed from the Palaeozoic to the Mesozoic was small. The marine faunas of the Mesozoic era attained their fullest development in Jurassic times, and during the Cretaceous periods begin to show signs of decadence, and of their replacement by the forms of life so characteristic of the Cainozoic. The terrestrial flora which characterises the Mesozoic rocks had, however, to a great extent been replaced by the Cainozoic flora lon before the end of the Cretaceous Period. 324 THE GERMAN TRIAS [OH. XIX. The oldest of the three great Mesozoic systems is much better represented on the Continent of Europe than it is in this country. In the Alps we have at St. Cassian in the south, and at Hallstadt on the north, thick masses of marine strata crowded with abun- dant and well-preserved fossils. MESOZOIC STRATA OF CENTRAL EUROPE Trias of Germany. In Germany, as before noticed, the Trias first received its name as a Triple Group, consisting of two sandstones with an intermediate marine calcareous formation, which last is wanting in England. The succession of strata in the great German Triassic basin is Upper Trias or Keuper, with red marls, plant-beds, gypsum, and rock-salt, overlying the Letten Kohle, with Voltzia, Estheria minuta, Alb., the Labyrinthodont Mastndontosaurus, and the fish Ceratodus. Then comes the Musch- elkalk, with limestones, contain- ing Myopharia-, Ceratites, and Encrinus liliiformis, Lam., fol- lowed by Bunter red and green marls and coarse sandstones with Voltzia, Estheria, and Myophoria. The plants of the Trias belong partly to Coniferse, the genus Voltzia, with its cypress-like twigs, being characteristic. The genus Albertia is also represented. Ferns were numerous : genera, Pecopteris, Cyclopteris, Anomopteris, Acrosti- chites, Clathropteris, Sageno- pteris, Tceniopteris, &c. Cycads, Pterophyllum, Zamites, Pseudo- zamites, Podozamites, and Oto- zamites. These last prevailed ex- tensively, and have given the term 1 Age of Cycads ' to the Trias. A true Equisetum exists, and an ally, the genus Schizoneura. Xeuper of Germany. The sandstones of the Keuper of Germany, like those of England and France, contain the remains of plants, reptilia, and very few marine organisms. Muschelkalk. This con- sists chiefly of a compact greyish limestone, but includes beds of dolomite in many places, together with gypsum and rock-salt and clays. This limestone a forma- tion wholly unrepresented in Eng- land abounds in fossil shells, as the name implies. Among the Cephalopoda there are no Belem- nites, and no Ammonites with com- pletely foliated sutures, as in the Lias and Oolite and the Hallstadt beds ; but the genus Ceratites is present, in which the lobes of the sutures seen on the shell are denticulated or crenulated, whilst the ' saddles ' are simply rounded. Among the bivalve Crustacea, the Estheria minuta, Alb. (fig. 436), is abundant, ranging through the Keuper and Muschelkalk ; and Gervillia socialis (fig. 422), having a similar range, is found in great numbers in the Muschelkalk of Germany, France, and Poland. The abundance of the heads and stems of lily encrinites, Encri- nus liliiformis, Schloth. (fig. 419), shows the slow manner in which some beds of this limestone have been formed in clear sea water. The star-fish called Aspidura loricata, Ag. (fig. 420), is as yet peculiar to the Muschelkalk. In the same forma- tion are found the skull and teeth of the genus Placodus (see fig. 433). Perfect specimens enabled Pro- fessor Owen, in 1858, to show that this fossil was a Saurian, which probably fed on shell-bearing molluscs, and used its short and flat teeth, so thickly coated with enamel, for pounding and crushing the shells. Bunter - Sandstein. The Bunter-Sandstein consists of va- rious-coloured sandstones, dolo- mites, and red clays, with some beds, especially in the Hartz, of calcareous pisolite or roe-stone, the CH. XIX.] ZONES OF THE JURASSIC 325 whole sometimes attaining a thick- ness of more than 1,000 feet. The sandstone of the Vosges is proved, by its fossils, to belong to this lowest member of the Triassic group. At Sulzbad (or Saultz-les- Bains), near Strasburg, on the flanks of the Vosges, many plants have been obtained from the ' Bun- ter,' especially conifers of the ex- tinct genus Voltzia, of which the fructification has been preserved. (See fig. 432.) Out of thirty species of ferns, cycads, conifers, and other plants, enumerated by M. Ad. Brongniart, in 1849, as coming from the ' Gres bigarre,' or Bunter, nob one is common to the Keuper. The footprints of Labyrintho- don observed in the clays of this formation at Hildburghausen, in Saxony, have already been men- tioned. Some idea of the variety and importance of the terrestrial vertebrate fauna of the three mem- bers of the Trias in Northern Ger- many may be derived from the fact that in the great monograph by the late Hermann von Meyer on the reptiles of the Trias, the remains of no less than eighty distinct species are described and figured. Jurassic strata of Central Europe. These have been studied with great care, especially by Marcou in the Jura, and by Quenstedt and Oppel in Suabia. While the general parallelism of these strata with those made out by William Smith in England is very striking, the local differences are of unmistakable character. A very great number of palasontological horizons, each characterised by a species of Ammonite or other fossil, have been defined under the name of zones ; and, over all the districts referred to, these zones may be traced more or less continuously, though some horizons are represented by thick masses of sediments, while others appear only as thin and insignificant bands, and some are, over considerable areas, altogether absent. Throughout Central Europe the succession cf Jurassic strata represented in this country can be clearly followed, although the mineral characters of some of the horizons differ very widely from the British representatives. This succession of zones, with the group-names applied to them, is given in the following table : Portlandian Zone of Kimeridgian J > t ;; Corallian -[ " ( " Oxfordian Callovian I Bathonian \ " Bajocian Trigonia gibbosa, Sow. Discina latissima, Sow. Eocogyra virgula, Defr. Ammonites alternans, V. Buch. Astarte supracorallina, D'Orb. Ostrea deltoidea, Sow. Ammonites (Perisphinctes) plicatilis, Sow. ,, (Aspidoceras) perarmatus, Sow. ,, (Cardioceras) cordatus, Sow. (Cosmoceras) ornatus, Schloth. ,, (Stephanoceras) macrocephahis, Schloth. (Oppelia) asphidioides, Opp. (Parkinsonia) ferrugineus, Opp. ,, (Parkinsonia) Parkinsoni, Sow. ,, (Stephanoceras) Hiimphreisia- nus, Sow. ,, (Stephanoceras) Sauzei, D'Orb. ,, (Hammatoccras) Soiverbyi, Mill. (Ludivigia) Murchisonice, Sow. ,, (Ludwigia) opalinus, Rein. 326 SHALLOW-WATER CRETACEOUS DEPOSITS [CH. xix. Zone of Ammonites (Lytoceras) jurensis, Ziet. (Hildoceras) bifrons, Brug. (Harpoceras) serpentinus, Rein. ,, (Amaltheus} spinatus, Brug. (Amaltheus) margaritatus, De Montf. (^Egoceras) capricornus, Schloth. ,, (Amaltheus) ibex, Quenst. (JEgoceras) Jamesoni, Sow. ,, (Arietites) raricostatus, Ziet. (Oxynoticeras) oxynotus, Querist. (Arietites) obtusus, Sow. (Arietites) semicostatus, Y. & B. (Arietites) Bucklandi, 8ow. ,, (Schlotheimia) angulatus, Schloth. (JEgoceras) planorbis, Sow. In some cases geologists and palaeontologists have found i'j necessary to establish still smaller subdivisions than these zones in describing the succession of the Jurassic strata. Such minor sub- divisions have been called by Mr. S. Buckman ' hemare.' Toarcian Liasian ( Charm ou- thian) Sinemnrian fiettangian In France and Germany the succession of strata representing these life-zones has been studied in great detail, and the parallelism of the different horizons throughout Western Europe is sufficiently ob- vious. There are many interesting local variations in the sequence of beds, in the degree of representa- tion of different horizons, and even in the fossils which characterise the different zones, which are worthy of the closest attention as indicating the varying conditions under which the strata of this age were accumu- lated. In Germany the Lias is usually called the ' Black Jura,' the Lower Oolites 'Dogger, or Brown Jura,' and the Middle and Upper Oolites the ' Malm,' or ' White Jura.' Shallow - water Repre- sentatives of the Chalk. Chalk strata are found all through Central Europe and passing into Asia In the south of Russia the Upper Cretaceous is represented by beds of chalk, which have been proved by deep borings to be nearly 2,000 feet in thickness. Although the great mass of the chalk was evidently deposited in moderately deep water, we can in places trace the shores of the sea in which the beds were formed. Thus in the north-east of Ireland beds of conglomerate containing chalk fos- sils are seen resting on the old metamorphic rocks of the district; and in Bohemia and Saxony the calcareous beds of the chalk are found passing into masses of sandstone (Quader Sandstone) and into marls (Planer Marls), evi- dently fonned under littoral con- ditions. At some points, as at Aix-la-Chapelle, the Western Isles of Scotland, and the east coast of Greenland, beds of freshwater origin are found intercalated with the Upper Cretaceous marine beds, and these freshwater beds have yielded a very interesting series of plant-remains. Freshwater Strata of Cretaceous Age of Central Europe. The Wealden or fresh- water representatives of the Lower Cretaceous are found extending into the north of France ; and strata of about the same age occur in Hanover. Freshwater beds con- taining a terrestrial fauna have been studied at Aix-la-Chapelle. These strata are of the same age as our Upper Chalk ; they con- sist of white sands and laminated clays, and attain a thickness of 400 feet. With the exception of a few bands containing marine shells, all these strata are of freshwater CH. xix.] FRESHWATER CRETACEOUS DEPOSITS 327 origin, and they contain a series of plant-remains which deserve par- ticular attention. Nearly 100 spe- cies are recorded by Debey in the lists of the 'Geologie de la Belgique ' (M. Mourlon, 1881). Of fourteen genera of ferns, three are still existing namely, Gleichenia, now inhabiting the Cape of Good Hope and New Holland; Lygodiiim, now spread extensively through tropical regions, but having some species which live in Japan and North America; and Aspleniufn, a living cosmopolite form. The genus Pteridoleimma is represented by no less than 22 species, or nearly one half of the whole flora of ferns. Among the phanerogamous plants, the Conifers are abundant, the most common belonging to the genus Sequoia (or Wellingtonia),of which both the cones and branches are preserved. The silicified wood of this plant is very plentifully dis- persed through the white sands in the pits near Aix. In one silicified trunk 200 rings of annual growth have been counted. The Monoco- tyledons there are very peculiar types. No Palms have been recog- nised with certainty, but a species of Pandanus, or Screw-pine, is found. But the number of the Dicotyledonous Angiosperms is the most striking feature in this an- cient flora. Among them we find five spe- cies of Dryophyllum, an oak-like genus very American in its affini- ties. The resemblance of the flora of Aix-la-Chapelle to the tertiary and living floras is considerable, but the angiospermous Dicotyledons did not commence with the Tertiary age, but long before. We can now affirm that these Aix plants flourished before the rich reptilian fauna of the secondary rocks had ceased to exist. The Ichthyosaurus, Pterodactylus, and Mosasaurus were of coeval date with the plant Dryophyllum. Speculations have often been hazarded respecting a connection between the rarity of Exogens in the older rocks and a peculiar state of the atmosphere. A denser air, it was suggested, had in earlier times been alike adverse to the well-being of the higher order of flowering plants, and of the quick-breathing animals, such as mammalia and birds, while it was favourable to a cryptogamic and gymnospermous flora, and to a predominance of reptile life. But we now learn that there is no in- compatibility in the co-existence of a vegetation like that of the present globe, and some of the most re- markable forms of the extinct rep- tiles of the age of gymnosperms. In Bohemia a flora belonging to the base of the upper chalk contains the Dicotyledonous genera Acer, Alnus, Salix, and Credneria. The Youngest Cretaceous Strata. At Maestricht in Hol- land, Faxoe in Denmark, in Scania, the southern part of Sweden, and at Meudon in France, we find strata of Upper Cretaceous age overlying the equivalents of the youngest Chalk beds in the British Islands. Some of the fossils of these youngest Cretaceous strata are represented in the following figures. In these beds we find Ammonites and Belemnites of Cretaceous types mingled with species of such Tei- tiary Gastropoda as Voluta, Fas- ciolaria, Cypr&a, Oliva, Mitra, and Trochus. Some of the beds of 'Danian Chalk abound with Bryozoa. Fig. 442. Belemnifdla miicronata, Schloth., |. Maestricht, Faxoe, aud White Chalk. b. Osselet or guard, showing vascular impressions on outer surface, with charac- teristic slit, and mucro. . Section of same, showin place of phragmacone. 828 TRIAS OF THE ALPS Ten. Fig. 443. V V" A 1 Portion of Baculites Favjasii, Sow. Maestricht and Faxoe beds aud White Chalk. Fig. 444. Nautilus danicus, Schloth. Faxoe, Denmark. Maestricht, &c. MESOZOIC STRATA OF THE ALPS AND SOUTHERN EUROPE The Alpine Trias. The richness of the fauna of the Alpine Trias has been already referred to. Mojsisovicshas shown that the strata on the south of the Alps (St. Cas- sian, &c.) belong to a different life- province from that in which those shown that each is distinguished by assemblages of peculiar Am- monites and other shells. Triassic strata, closely related to those of the Alps, are found extending through Southern Europe and Asia into the Indian peninsula, 445. Map showing the position of St. Cassian and Hallstadt areas. on the north side of the same chain (Hallstadt beds) were deposited. These life-provincea he has named the Mediterranean and Juvavian provinces respectively ; and he has and through Siberia into Japan. Strata with similar fossils reappear in the western territories of the United States, in Alaska and British Columbia, CH. XIX.] AND ITS SUBDIVISIONS 329 The Trias is grandly developed in the Eastern Alps. Including the Rhsetic beds, which link the Trias and the Lias, the following is the succession of the great groups of strata. The Rhtetic group, consisting of marine limestones, dolomites, and (rarely) shales : 1. Kb'ssen beds and Azarolla beds, with corals, Brachio- poda, and Lamellitranchiata, such as Gervillia. 2. Dachstein lime- stone, with large forms of Megalo- don or the Dachstein bivalve, numerous corals, and Brachiopoda. 3. Dolomites. A pale, well-bedded. finely crystalline rock, usually without fossils. Upper Trias: 1. Cardita beds and Raibl beds, shales, marls with plants, Crustacea, Cephalopoda, and fish. 2. Hallstadt limestone and Esino beds, red and mottled marbles and limestones, with many Cephalopoda and large Gastro- poda. The Schlern Dolomite, 3,820 feet thick, forming picturesque mountains. 3. Lunz beds, contain- ing coal with plants, and forming the only freshwater group. 4. Zlam- bach coral beds. 5. St. Cassian beds calareous marls of South Tyrol, with Ammonites, Gastro- poda, Lamellibranchiata, Brachio- poda, Crinoidea, Echinoidea, and Corals. 6. Halobia-Lommeliibeds. Then comes the Lower Trias. 7. Alpine Muschelkalk, limestones, and dolomites, with lower strata containing Ceratites, which are equivalent to the Upper Division of the Bunter. Other Alpine deposits of marine origin occur in Southern Europe, of which the great masses of lime- stone of Hallstadt, north of the Alps, are the type. Huge Ammo- nites characterise these deposits. On the south of the Tyrol the St. Cassian beds were forming a little earlier, and the fauna was rich in the extreme. The following are characteristic genera Scoliostoma (fig. 423), Platystoma (fig. 424), and Koninckia (fig. 421). Ammonites and Orthoceratites occur in the St. Cassian and Hall- stadt beds. As the Orthocerata, which are common in some palaeo- zoic rocks, had never been met with in the Muschelkalk or the Lower Trias, much surprise was felt that seven or eight species of the genus should appear in the Hallstadt beds of the Upper Trias. Some are of large dimensions, and are associated with large Ammo- nites with foliated lobes, a form never seen before so low in the Mesozoic series. Centhium, so abundant in tertiary strata, and which still exists, is represented by no less than fourteen species. A rich fauna, comprising 225 species, of which about one-fourth are identical with those of St. Cas- sian, has been brought to light at D'Esino, in Lombardy, and has been admirably illustrated by Pro- fessor Stoppani. He described 65 species of the genus of spiral uni- valve Chemnitzia, reminding us by its abundance of the Cerithia of the Paris basin, while the enormous size of some specimens would almost bear comparison with the Ceritheum giganteum, Lam., of that Eocene formation. The study of the rich marine fauna of Hallstadt and St. Cassian of the Upper Trias or Keuper convinces us that when the strata of the Triassic age are better known, especially those belonging to the period of the Bunter Sand- stone, the break between the Palaeo- zoic and Mesozoic Periods will to a great extent disappear. Jurassic strata of the Alps. In the Alpine region the thick limestones of Jurassic age Tig. 446. Terebratula (Pygope) diphya, Col. contain representatives of a number of zones, which can only be com- pared generally with the divisions of the strata in Central Europe. These Alpine strata graduate upwards through the Tithonian into the 330 HJPPUEITE LIMESTONES [CH. XIX. Neocomian, and downwards through the Rheetic (or Dachstein and Kb'ssen beds) into the Trias. The Tithonian are a remarkable series Fig. 447. diphyoid Terebratulse (Pygope), fig. 446. By many authors the Titho- nian strata are classed as Upper Jurassic. Fig. 448. Radiolites foliaceus, D'Orb. Syn. Sphcerulites agarici- formis, Blainv. White Chalk of France. a. Radiolites radiosa, D'Orb. &. Upper valve of same. ffippurites orgnnisans, Desmoulins. Upper Chalk : Chalk marl of Pyrenees. a. Young individual ; when full grown they occur in groups adhering laterally to each other. Z>. Upper side of the upper valve, showing a reticulated" structure in those parts, b, where the external coating is worn off. c. Upper end or opening of the lower and cylindrical valve. d. Cast of the interior of the lower conical valve. of strata containing many peculiar forms of Cephalopoda and other shells, among the most character- istic of which are the singular The Cretaceous strata of Southern Europe. The general succession of Cretaceous strata in Western Europe, as we have seen, CH. xix.] TRIAS OF INDIA AND SOUTH AFEICA 331 closely resembles that of our own country, but in Southern France and Switzerland the Lower Cre- taceous or Neocomian becomes greatly developed, containing a wonderfully rich and varied fauna, and being divisible into a number of distinct zones. The Neocomian strata of the Alps are several thousands of feet in thickness and have a very rich fauna, including many remark- able forms of Ammonites and Belemnites. They graduate down- wards into that other thick series of limestones referred to, the Titho- nian, a great system of strata not represented by marine beds in the British Islands, which appears to bridge over the interval between the Cretaceous and the Jurassic Periods. In the south of France and the Alpine districts of Southern Europe the Upper Cretaceous is represented by thick masses of calcareous and other strata. These contain a fauna differing in many respects from the fauna of the Cretaceous of Central Europe. Many of the beds of lime- stone of Upper Cretaceous age (Hippurite limestones) are almost entirely made up of the shells of the large and remarkable bivalves be- longing to the extinct group of the Eudistes. Some of the chief forms of these Rudistes are shown in the figures on the opposite page. MESOZOIC STRATA OF OTHER PARTS OF THE EASTERN HEMISPHERE Trias of India and South Africa. There is a marine Trias in the Himalaya and in Baluchis- tan with Muschelkalk and St. Cas- sian species of somewhat different types from those of Europe. But the great development is in the penin- sula, where the terrestrial remains of the period form vast coal-beds and shales and clays with plants and animals belonging to the same periods. In Australia, in New South "Wales, Victoria, and Queens- land are important coal-bearing strata . Below these Triassic (Panchet) beds are still thicker series of coal- bearing strata which are referred to the Permo-Carboniferous and even older periods (Ranigaiij and Talchir Series). In South Africa extensive beds (the Karoo beds) containing simi- lar plant remains and many re- markable forms of terrestrial rep- tiles, which have been made known to us by the labours of Professors Owen and Seeley, cover an enor- mous area and attain a great thick- ness. The Triassic rocks of Southern Europe and Asia have been shown by the labours of Neumayr and Mojsisovics to have been accumu- lated in two life-provinces, to which the last-named geologist gave the names of the Juvavian pro- vince (lying north of the present Alps) and the Mediterranean pro- vince (lying south of that chain). In each of these provinces shallow- water and deep-water facies have been distinguished, and the effects of climate in influencing the distri- bution of life-form can be distinctly traced. In Siberia, Spitzbergen, and Japan, and the western portions of the North American continent, Triassic strata cover wide areas. In spite of the presence of many characteristic genera of Triassic Cephalopods, Mojsisovics points out that there are many remarkable differences between the fauna of these Asiatic strata and those of Europe, and he regards them as constituting another life-province, the ' Arcto-Pacific.' Jurassic strata of Russia and the Arctic Regions. In Russia, and in the northern parts of Europe, Asia, and North America, we find very widely distributed a series of strata possessing a number of features in common with the Upper and Middle Oolites, but at the same time offering many very striking differences from the typical Jurassic. The same formations, characterised by the peculiar genus Aucella, bv the presence of manv 332 THE NEWARK SYSTEM [OH. distinctive types of Ammonites and Belemnites, as well as by the absence of many forms found in the typical Jurassic, also extends southwards into the western terri- tories of the United States. It is a remarkable circumstance that not a few Ammonites and other fossils of Jurassic types are found as far north as the east coast of Greenland, and in the adjoining islands of the Arctic Ocean. Cretaceous Strata of Greenland, due. A Cretaceous flora has been discovered in Green- land of Cenomanian age at 70 N.L., and its genera resemble those of the Dakota group (p. 335) of the Cretaceous Formation of the West- ern Territories of the United States. There are Ferns, and a great assem- blage of Dicotyledons, including many evergreens and conifers. A second flora, which is proba- bly of Lower Cretaceous age, is remarkable for having only yielded one Dicotyledonous (Angiosper- mous) species, but numerous Coni- fers, many Cycads, and a few Monocotyledons. (See p. 885.) ME SO ZOIC S TEAT A OF NORTH AMERICA The Newark System of the Eastern States. While in the Old World it appears to be gene- rally possible to divide the Meso- zoic strata into the three great systems Triassic, Jurassic, and Cre- taceous, such is certainly not the case in the New World. When we cross the Atlantic, we find in the Eastern States of North America a series of strata (the Newark system) upwards of 4,000 feet in thickness, which are clearly homotaxial with the Jurassic and Triassic taken together, but in which it is impossible to establish any exact parallelism with the several divisions of those systems. the remains of Cycads and Equi- setum occur. With the plant re- mains are found many Crustaceans, including forms of Estkeria, simi- lar to those so abundant in the European Trias, and many Ganoid fish. In the Connecticut Valley strata of red sandstone occur which often exhibit great numbers of tracks formerly regarded as those of birds, but now believed to have been made by Labyrinthodoiits and Dinosaurs. The footprints of, it is supposed, no less than 50 species of animals have been detected in these rocks. The tracks have been found in Fig. 450. Footprints of a Dinosaur (?). Turner's Falls, Valley of the Connecticut. They consist of reddish sandstones and shales, with a few thin beds of limestone and coal. In the Eastern United States, the Triassic, Rhsetic, and part of the Jurassic system appear to be represented by this ' Newark System,' which, in great part at least, seems to have been of fresh- water origin. Near Richmond, in Virginia, beds of coal made up of more than twenty places scattered through an extent of nearly 80 miles from north to south, and they are repeated through a succession of beds attaining at some points a thickness of more than 1,000 feet. Yet no traces of bones or teeth have ever been detected in the beds. In North Carolina the teeth of a small mammal (Droinatheriuiti) CH. XIX.] VIRGINIAN COALFIELD 333 have been found in the same strata. It is closely related to the European Microlestes. The formation covers an im- mense area, and may be divided into the Eastern and Western series. The former are freshwater and terrestrial accumulations, and the latter are marine deposits. In the eastern area the valley of the Connecticut river offers a type. In a depression of the granitic or hypogene rocks in the States of Massachusetts and Connecticut, strata of red sandstone, shale, and conglomerate are found, occupying an area more than 150 miles in length from north to south, and about 5 to 10 miles in breadth, the beds dipping to the eastwards formation among its contorted rocks. Coal-field of Richmond, Virginia. In the State of Virginia, at the distance of about 13 miles eastward of Richmond, the capital of that State, there is a Coal-field, occurring in a de- pression of the granite rocks and occupying a geological position analogous to that of the New Red Sandstones, above mentioned, of the Connecticut Valley. It extends 26 miles from north to south, and from 4 to 12 from east to west. The plants consist chiefly of Zamites, Equlsetacece, and ferns, and were considered by Heer to have the nearest affinity to those of the European Keuper. Fig. 451. a. Eittheria ovata. Lea sp. b. Young of same. c. Natural size of a. d. Natural size of b. Triassic coal-shale, Richmond, Virginia. at angles varying from 5 to 50 degrees. Having examined this series of rocks in many places, Lyell concluded that they were formed in shallow water, and for the most part near the shore, and that some of the beds were from time to time raised above the level of the water, and laid dry, while a newer series composed of similar sediment was forming. The age of the Connecticut beds cannot be proved by direct superposition, but may be pre- sumed from the general structure of the country. That structure shows them to be newer than the movements to which the Appa- lachian or Alleghany chain owes its flexures, and this chain includes the ancient or palaeozoic Coal- The horsetails are very commonly met with in a vertical position, more or less compressed. It is clear that they grew in the places where they are now found, and were buried in strata of hardened sand and mud. They maintain their erect attitude, at points many miles apart, in beds both above and between the seams of coal. In order to explain this fact we must suppose such shales and sandstones to have been gradu- ally accumulated during the slow and repeated subsidence of the whole region. The fossil fish are Ganoids, some of them of the genus Catopterus, others belonging to the Liassic genus Tetragonolepsis (jEchmodus) (see fig. 325, p. 278). Amongst the 334 MESOZOIC MAMMALIA OF AMERICA [CH. xix. Crustacea, two or more species of Entomostraca called Estheria are in such profusion, in some shaly beds, as tcr divide them like the plates of mica in micaceous shales. (See fig. 451.) These Virginian Coal-measures are composed of grits, sandstones, and shales, closely resembling those of palaeozoic date in America and Europe ; and the measures rival those of the last-named continent in the thickness of the coal-seams. One of these, the main seam, is in some places from 30 to 40 feet thick, and is composed of bitumi- nous coal. The Dromatherium, before allu- ded to, is at least as ancient as the Microlestes of the European Rhaetic, described p. 308 ; arid the fact is highly important, as proving that a certain low grade of marsupials had not only a wide range in time, from the Trias to the Purbeck of Europe, but had also a wide range in space, namely, from Europe to North America, in an east and west direction, and, in regard to lati- tude, from Stonesfield, in 52 N., to North Carolina, in 35 N. . A somewhat similar form (Tritylo- don), has been found in the Triassic beds of South Africa, and others also in the Cretaceous of the United States (Note R, p. 606). If the three localities in Europe where the most ancient mammalia have been found Purbeck, Stones- field, and Stuttgart had belonged all of them to formations of the same age, we might well imagine so limited an area to have been peopled exclusively with pouched quadru- peds, just as Australia 1 now is, while other parts of the globe were inhabited by placental, or ordinary Mammalia. But the great diffe- rence of age of the strata in each of these three localities seems to indicate the predominance through- out a vast lapse of time (from the era of the Upper Trias to that of the Purbeck beds) of a low grade of 1 Australia now supports one hundred and sixty species of mar- supials, while the rest of the con- tinents and islands are tenanted by about seventeen hundred species of Mammalia ; and there must also have been a vast extension in geo- graphictJ area of the marsupials during that portion of the Secondary or Mesozoic era which has been called the Age of Reptiles. The predominance of these Mammalia of a low grade during the whole of the Mesozoic, and the absence of the higher forms of Mammalia, are strongly suggestive of a pro- gressive development of life-forms. It is also a very significant circum- stance, which has been pointed out by Professor Seeley and others, that the Triassic reptiles of South Africa exhibit, in the differentiation of their teeth and many other peculiarities of their structure, very curious affinities with the mammals. While the Jurassic strata are very imperfectly exhibited in North America, being only recognisable as possibly represented in the Newark formation, the Cretaceous are certainly present in the same area, but exhibit many striking diffe- rences from their European equi- valents. We find in the State of New Jersey a series of sandy and argillaceous beds wholly unlike in mineral character to our Upper Cretaceous system of Europe;' which we can, nevertheless, recog- nise as referable, palaeontologically, to the same division. That they were about the same age generally as the European Chalk and Neocomian was the con- clusion to which Dr. Morton and Mr. Conrad came alter their in- vestigation of the fossils in 1834. The strata consist chiefly of green- sand and green marl, with an over- lying coral limestone of a pale yellow colour, and the fossils, on the whole, agree most nearly with those of the Upper European series, from the Maestricht beds to the Gault inclusive. Among sixty shells from the New Jersey de- posits, five were found as early as Mammalia, of which only forty-six are marsupial, and these are of a different family from the marsupials of Australia namely, the opossums of North and South America. CH. XIX.] AMEKICAN CEETACEOUS 335 1841 to be identical with European species Ostrea larva, Lam., 0. vesicular is, Lam., Gryphcea cos- tata, Sow., Pecten quinquecos- tatus, Sow., Belemnitella mucro- nata, Schloth. As some of these have the greatest vertical range in Europe, they might be expected more than any others to recur in distant parts of the globe. Even where the species were different, the generic forms, such as Bacidites and certain genera of Ammonites, as also the Inoceramus and other bivalves, have a decidedly Cre- taceous aspect. Fish of the genera Lamna, Galeus, and Carcharodon are common to New Jersey and the European Cretaceous rocks. So also is the genus Mosasaurus among reptiles. Hadrosaurus and Dryptosaiirus occur amongst the Dinosaurs. Professor O. C. Marsh has described several species of birds from the Green sand of New Jersey. It appears from the labours of Dr. Newberry and others, that the Cretaceous strata of the United States, east and west of the Appalachians, are characterised by a flora decidedly analogous to that of the Upper Cretaceous of Central Europe, and having considerable resemblance to the vegetation of the Tertiary Period. Cretaceous rocks are grandly developed in the South- We stern States, in Texas, Wyoming, Utah, and Colorado. They are found to the north in Manitoba, and reach to the mouth of the Mackenzie, and into Northern Greenland. In Texas there are limestones with Hippurites and Orbitolites ; but northwards the strata become arenaceous, and were partly de- posited in the sea and partly on land. The following are the prin- cipal groups. The highest, or Laramie the Lignitic is a ter- restrial deposit, containing brackish water and some marine fossils, and a vast flora. The vegetation is remarkable for the number of Dicotyledons, showing that this great section of the vegetable kingdom was in existence before the Tertiary age. The Reptilia found in the deposits are mostly Mesozoic in their affinities, and there are no mammalian remains. Ammonites and Inoceramus have been found. The deposit is 5,000 feet thick on the Green River. The researches of the United States geologists and palaeontologists point to the conclusion that the Laramie formation was deposited, at least in part, during the vast period repre- sented by the great break between the Cainozoic and Mesozoic epochs. The second, or Fox- Hills group, consists of sandstones, some terrestrial and others marine, with Belemnitella, Nautilus, Ammo- nites, Bacidites, and Mosasaurus It is from 3,000 to 4,000 feet thick. Thirdly, the Colorado group, with Cretaceous fossils ; and fourthly, the Dakota group, ^ i.th a re- markable flora. The flora of the Dakota group (Cenomanian) contains ferns of the genera Lygodium, Sphenopteris, Pecopteris, Gleichenia, and Todea. Amongst the Gymnosperms, the genera Pterophyllum, Sequoia, Araucaria, Glyptostrobus, &c. ; and Flabellaria amongst the palms. There are 167 species of Angiospermous Dicotyledons, of which about one-half are still represented by living species. The order Proteaceae has three genera Proteoides, Embothrium, and Aristolochites ; and among the Lauracese are Laurus, Persea, Sassafras, Cinnamomum, Oreo- daphne ; whilst Magnolia and Liriodendron are amongst the Polycarpieaa. This flora should be carefully noticed, in order that we may not be deceived by the sup- position that Dicotyledons of the above-mentioned genera are neces- sarily of Tertiary age. The flora would at the present time be normal in a climate like that of the South of Europe of from 35 to 40 N. lat. Probably one-half of the Dicotyledons are allied to recent American forms. According to the most recent researches of Dr. C. A. White, the Cretaceous strata of North America, which are of enormous thickness, belong to an Upper and Lower Cretaceous division, but these 336 TABLE OF MESOZOIC STRATA [en. xix. CORRELATION OP THE MESOZOIC BOCKS IN DIFFERENT AREAS BRITAIN AND WESTERN EUROPE THE ALPS RUSSIA NORTH AMERICA ,Danian . Highest beds of _ _ Freshwater Chalk in strata of Scandinavia Western and France Territories Senonian Upper Chalk Hippurite lime- Chalk Strata stones of Southern Turonian Middle Chalk Russia en Cenofiianian. Lower Chalk Glauconite Greensands p and Upper limestones of New &l Greensand Jersey *\ Albian . Qttolt ! Aptian . Lower Green- Orbitolite and 63 sand Requienia PH limestones O Rhodanian & Tealby series Limestones and Limestones of Barremian Speeton Clay, Marls with the Balkan &c. Belemnites Provinces and Aptychi and the Neocomian Crimea (proper) Portlandian . Portland and Tithonian Purbeck beds Kimeridgian Kimeridge Clay Stramberg limestones Coralliau Coral Rag and Diphya lime- Volga beds Calcareous stones Grit Zone of Am- monites alternans Oxfordian . Oxford Clay Massive lime- Zone of Am- o Callovian Kelaways Rock stones monites cordntus CO Bathonian . Great Oolite Clavs of Sam- Newark on:scus comptus, Ag. Ganoid scale, magnified. Marl-slate. Fig. 470. Palceoniscus elegans, Sedgw. Under surface of ganoid scale, magnified. Marl-slate. Fig. 471. Palceoniscus glaphyrus, Ag. Under surface of ganoid scale, magnified. Marl-slate. Fig. 472. Ctflacanthus granulatus, Ag. Granulated surface of scale, magnified. Marl-slate. Fig. 473 Fig. 474. Pygopterus mandibularis, Ag. Marl-slate. a. Outside of scale, magnified. b. Under surface of same. Acrolepis Sedgwickii, Ag. Outside of scaie, magnified. Marl-slate. first pointed out by Agassiz, that the heterocercal form, which is confined to a small number of existing genera, is universal in the M'agnesian Limestone and all the more ancient formations. It, characterises the earlier periods of the earth's history, whereas CH. xx. LOWER PERMIAN 347 in the secondary strata, or those newer than the Permian, the homocercal tail greatly predominates. In Professor King's monograph on the Permian fossils a full de- scription has been given by Sir Philip Egerton of the species of fish characteristic of the marl-slate ; and figures of the ichthyolites. which are very entire and well preserved, will be found in the same memoir. Even a single scale is usually so characteristically marked as to indicate the genus, and sometimes even the particular species. They are often scattered through the beds singly, and may be useful to the geologist in determining the age of the rock. Two species of Proteromurus, a genus of reptiles, have been dis- covered in the marl-slate, one representative of which, P. Speneri, Meyer, has been celebrated ever since the year 1810 as characteristic of the Kupferschiefer or Permian of Thuringia. Remains of a Labyrin- thodont, Lepidotosaurus Duffi, Hancock and Howse, have been met with in the same slate near Durham ; and a quarry in the Permian sandstone of Kenilworth has yielded the skull of another species, called by Professor Huxley L. Dasyceps, on account of the roughness of the surface of the cranium. Xiower Permian. The principal development of the British Lower Permian is found in the north-west of England, where the Penrith sandstone, as it has been called, and the associated brec- cias and purple shales are estimated by Professor Harkness to attain a thickness of 3,000 feet. Organic remains are generally wanting, though footprints and worm-tracks are occasionally met with, and the leaves, cones, and wood of coniferous plants have been fo ind in beds considered by Professor Harkness to be the equivalent of the marl- slate which overlies the Penrith sands at Hilton. In the red sand- stones of this age at Corncockle Muir, near Dumfries, very distinct foot- prints occur in great number and variety. No bones of the animals which they represent have yet been discovered, but a cranium of Dasyceps has been found further south. Angular Breccias in Lower Permian. A striking feature in these beds is the occasional occurrence, especially at the base of the formation, of angular and sometimes rounded fragments of Car- boniferous and older rocks of the adjoining districts. These are included in a red matrix. Some of the angular masses are of huge size. These brecciated conglomerates are well seen in the Abberlej Hills, where they are 400 feet thick. Sir A. Ramsay refers the angular form and large size of the frag- ments composing these breccias to the action of floating ice in the sea. The angular masses of rock, sometimes weighing more than half a ton, and lying confusedly in a red unstratified marl, like stones in boulder drift, appear in some cases to be polished, striated, and furrowed like erratic blocks in the moraine of a glacier. They can be shown, in some instances, to have travelled from the parent rocks, thirty or more miles distant, and yet not to have entirely lost their angular shape. The monograph on Permian palaeopbytologists. Dr. Waagen's fossils, by the late Prof. King, memoir on the Salt Range fossils, contains figures and descriptions of published by the Geological Survey the chief British Permian forms of of India, must be referred to for life. The plants bave been de- figures and descriptions of the very scribed by Von Gutbier and other remarkable marine forms. 348 CARBONIFEROUS SYSTEM [CH. xxi. CHAPTER XXI THE CARBONIFEROUS SYSTEM Succession of Strata in the Carboniferous System Carboniferous Foramini- fera and Corals, Echinodermata, Brachiopoda, Lamellibraiichiata, Gastropoda and Cephalopoda of the Period Carboniferous Fishes and Amphibians The Carboniferous Flora Peculiarity in Mode of Growth of the Cryptogams of the Period Ferns, Calamites, Lepidodendra, &c. Land-shells and Insects of the Carboniferous Period Carboniferous Strata of Britain Coal-measures Millstone Grit Carboniferous Limestone and Yoredale Series Tuedian Series Scottish Carbonife- rous Calciferous Sandstone Series Mode of Formation of the Car- boniferous Strata Coal-seams, Ironstones, &c. Marine and Fresh- water Strata of Carboniferous. Nomenclature and Classification of the Carboniferous strata. This system of strata has received its name from the circumstance that, in Western Europe and the United States, most of the productive coal-seams occur among deposits of this age. The beds of coal vary in thickness from an inch, or even less, up to thirty feet or more, and alternate with various thicker strata of sandstone and shale, with occasional bands of limestone and argillaceous ironstone. Such assemblages of coal-bearing strata are called by the old English miners' name of ' coal-measures.' The coal-measures are usually found occupying basin-shaped hollows, owing to their having been thrown into synclinal curves by great earth movements. The intervening anticlinals having been removed by denudation, we often find the coal-bearing strata forming isolated patches, which are known a? ' Coalfields ; ' but these must not be mistaken for lake-like depressions in which deposition has taken place. The coal-measures some- times contain marine fossils, at other times brackish-water forms, and sometimes purely freshwater ones. The remains of land-plants occur in the coal itself, and in the sandstones, shales, and ironstones alternating with the coal ; in the same strata we occasionally find the remains of freshwater amphibians, fish, crustaceans, land-shells, and even of insects. The coal-measures alternate, however, with thick deposits of limestone, shale and sandstone, which abound with purely marine types of life. In South Wales and the Bristol and Somersetshire coal-fields the general succession of strata in the Carboniferous system is as follows : CH. xxi.] SUBDIVISION OF THE CARBONIFEROUS 349 (Upper series of sandstones, shales, aud 26 coal-seams. Pennant grit and 15 coal-seams. Lower coal-measures with iron- stone and 34 coal-seams. / A coarse quartzose sandstone used 2. Millstone grit -I for millstones. 400 feet shale ( called ' Farewell Rock.' {A calcareous rock containing marine shells, corals, and encrinites. Thickness variable: 2,000 feet. Lower limestone shale 400 feet. This threefold division of the Carboniferous is a purely local one, however. When we pass southwards into Devonshire, we find thick masses of strata (the Culm-measures) containing carbonaceous matter but no workable seams of coal. As we proceed northwards we find that, though the general distinction between the arenaceous division in the middle of the series (Millstone grit) and the strata above and below it respectively can still be recognised, the distribution of productive coal-seams is remarkably different. In Yorkshire thin seams of coal are found in the midst of the Millstone-grit series, and in the Lanca- shire and West Yorkshire district the Carboniferous Limestone Series is broken up into alternations of limestones, shales, and sandstones with some beds of coal, known as the Yoredale Series. Further north, in the parts of Northumberland near the Scottish Border, we find coal-beds present right down to the base of the Carboniferous Limestone division, forming the Tuedian Series, while in Scotland itself similar coal-seams are found from top to bottom of the Carboniferous system, and even in the great masses of sandstone (Calciferous Sandstone) which there underlie the representatives of the Carboniferous Lime- stone. It is worthy of notice that the highest strata of the Carbo- niferous (the Coal-measures) in the coal-fields of the West of England (Warwickshire, Staffordshire, and Coalbrook Dale) rest directly upon the older rocks, no representatives of the Carbo- niferous Limestone division having been deposited in that area. In Ireland, on the other hand, the lower portion of the Carbo- niferous series (Carboniferous Limestone and Carboniferous Slate) cover wide areas, while the upper members (the Coal- 1 It will be seen that the term member, as some of the coarse ' coal-measures ' is used not only for sandstones were at one time used any assemblage of coal-bearing for millstones, and the name ' moun- strata, but as a distinctive name tain limestone ' to the lowest mem- for the highest member of the Car- ber, from its forming the mountains boniferous series. The term ' mill- of Derbyshire and the West Riding stone grit ' is given to the middle o f Yorkshire. 850 CAKBONIFEROUS FORAMlNIFEftA [CH. xxi. measures) weje either never deposited, or have been almost entirely removed by denudation. In some areas the coal-fields are bordered by great faults, in others they are buried under masses of Permian and other younger strata. Where these covering strata are not of too great thickness, the coal-beds are reached by sinking shafts through them. A great ridge of ancient rocks including infolded basins of Coal-measures extends under the Mesozoic strata of the East and South-east of England, and has been reached by borings at several points (Burford, Northampton, Harwich, and Dover) ; the same strata reappear in the North of France and the South of Belgium, the coals being worked by shafts put down through the Chalk or other Mesozoic formations. Characteristics of the Carboniferous Fauna and Flora. The marine fauna of the Carboniferous, which is found espe- cially in the thick limestones in the lower part of the series, is a very rich one. Among the Foraminifera we find that living genera like 47g Textularia and Nodosaria are mingled with extinct ones like Endothyra and Fusulina. The forms of the last-mentioned genus are par- ticularly noteworthy ; they are fusiform bodies, a bout the size of a grain of barley, which some- Magnified 3' diam. times build up massive limestone rocks in Sestone. Bussia, Asia Minor, Japan, and North America. These Fusulina-limestones of the Carboniferous resemble the Nummulite-limestones of the Eogene; and, indeed, the two genera Nummulina and Fusulina have close affinities with each other. The Corals were, during this period, represented only by forms of the now extinct Tetracoralla (Kugosa), the Hexa- coralla of the Mesozoic and Tertiary rocks and of our modern seas having come into existence in later times than the Car- boniferous. Many remarkable examples of these Rugose corals are found in the Mountain Limestone, among which may be mentioned Amplexus, Lonsdaleia (fig. 479), Litkostrotion (fig. 478), Zaphrentis, &c. Other coral-like forms of this system are referred to the extinct order of the Tabulata (Monticuliporida), which appear to have relations both with the Bryozoa and the Actinozoa. Among the Echinodermata many forms of Crinoidea abound, their ossicles often making up great beds of limestone (Entro- chial-limestone). These Palaeozoic Crinoids, like Actinocrinus, Cyathocrinus (figs. 480, 481), Platycrvnus, &c., differ in many CH. XXI.] CARBONIFEROUS CORALS 351 Fig. 476. Palceozoic type of lamelliferous cup-shaped Coral. a. Vertical section of Cyathophyllum fexuosum, Goldf. ; ^ nat. size ; from the Devonian of the Eifel. The septa are seen around the inside of the cup : the walls consist of cellular tissue ; and large transverse plates, called tabtilce, divide .the interior into chambers. b. Arrangement of the septa in Polycoelia pro f unda, Germar, sp. ; nat. size ; from the Mag nesian Limestone, Durham. This diagram shows the quadripartite arrangement of the primary septa, characteristic of palaeozoic corals, there being 4 principal and 8 inter- mediate lamella?, the whole number in this type being always a multiple of four. c. Stauria out ranformis, Milne-Edwards. Young group, nat. size. Silurian, Gothland. The lamella? or septa in each cup are divided by four prominent ridges into four groups. Order TETRACORALLA Fig. 477. Neozoic type of lamelliferous cup-shaped Coral. Order HEXACOUALLA. a. Parasmilia centralis, Mantell, sp. Vertical sec- tion ; nat. size. Upper Chalk, Gravesend. In this type the lamellw extend to the columella composed of loose cellular tissue, and there are no tab nice. b. Caryophyllia Bowerbankii, Ed. and H. Transverse section, enlarged. Gault, Folkestone. In this coral the primary septa are a multiple of six. The six primary and six secondary septa reach the columella, and between each pair of long septa there is a tertiary septum with a quaternary on either side, in all forty-eight. The short inter- mediate plates which proceed from the columella are called pall. c. Fungia patellnris, Lam. Recent ; very young state. Diagram of its six primary and six secon- dary septa, magnified. Fig. 478. Fig. 479. Lithoslrotion basaltifornw, i'hil. sp., England ; Ireland , Russia ; Iowa, and westward of the Mississippi, United States. Lonsdaleiafloriformis, Mart, sp., \. a. Young specimen, with buds or coral- lites on the disk, illustrating calicular gemmation, b. Part of a full-grown compound mass. 352 CARBONIFEROUS ECH1NODERMATA [CH. xxi. important points of their structure from the living Crinoids and those of the Cainozoic and Mesozoic times. A remarkable extinct order of Echinodermata, the Blastoidea, is represented by many forms of Pentremites (fig. 482), Grana- tocrinus, &c., while the ordinary Echini are replaced by the Fig. 480. Fig. 481. dyathocrin us planus, Miller. Body and arms. Mountain Limestone. Cyathocrinus caryocrinoides, M'Coy. a. Surface of one of the joints of the stem. b. Pelvis or body ; called also calyx or cup. c. One of the pelvic plates. curious extinct order Palaechinoidea, including Palcechinus (fig. 483), Archceocidaris and Melonites, in which we find the ambulacra not separated by two rows of plates, as in all the living and Mesozoic forms, but by five or more rows. Bryozoa are very abundant in some portions of the Carbo- niferous limestone of England and Scotland. The most Tig. 482. Fig. 483. Pcntremites ellipticus, Sow., . Carb. Limestone, Derby- shire, &c. Palcechinus gigas, M'Coy, Reduced one-third. Carboniferous Limestone. Ireland. abundant genera are the extinct ones, Fenestella, Polypora, Diastopora, and Glauconeme. Among the Brachiopoda, Productus (fig. 484). is the most abundant genus in the Carboniferous ; but many forms of Spiri- fera, both ribbed and smooth types (figs. 485, 486), occur with species of Chonetes, Orthis, Athyris, Ehynchonella, &c. The CH. XXI.] CARBONIFEROUS MOLLUSCA 353 oldest form of Terelratula (fig. 487), a genus so characteristic of the Mesozoic, is found in the Carboniferous. The Lamellibranchiata, which are present in much smaller Fig. 484. Fig. 485. Prodnctus semireticulatus, Mart, sp., J. Carboniferous Limestone. England ; Russia ; the Andes, &c. Pig. 486. Spirifera triyonaiis, Mart, sp., nat. size. Carboniferous Limestone. Derbyshire, &c. Fig. 487. Spirifera glabra, Mart, sp., i. Carboniferous Lime- stone. Fig. 488. Terebratula hastata, Sow., , with radiating bands of colour. Carboniferous Lime- stone. Derbyshire ; Ireland ; Russia, &c. Fig. 489. Fig. 490. Aviculopecten papyraceus, Goldf., . (Pecten papyraceus^ Sow.) Aviculopecten sublobatus, Phill., nat. size. Carboniferous Limestone. Derbyshire ; Yorkshire. Pleurotomaria carinata, Sow., . (P.flammigera, Phill.) Carboniferous Limestone. Derbyshire, &c. numbers than in younger rocks, are principally represented by purely PalEeozoic genera like Aviculopecten (figs. 488, 489), Cono- cardium, Cardiomorpha, Edmondia, Anthracosia, Cypricar- dinia, Posidonomya, &c. A A 354 CARBONIFEROUS GASTROPODA [CH. XXI. With a few Mesozoic types of Gastropoda, Natica, Pleura- tomaria (fig. 490), Chiton, &c., we find great numbers of extinct Palaeozoic types, which have usually entire (holostomatous) apertures, such as Macrocheilus, Loxonema, &c., and also the Euomphalus pentangulatus, Sow., . Mountain Limestone. a. Upper side. &. Lower or umbilical side. c. View, showing mouth, which is less pentagonal in older individuals. d. View of polished section, showing internal chambers. Fig. 493. Fig. 492. Bellerophon costatus, Sow., nat. size. Mountain Limestone. Portion of Orthoceras laterale, Phill., \, Mountain Limestone. remarkable Euomphalus (fig. 491), a Gastropod with its shell divided by imperforate septa, and Bellerophon, which probably belonged to the Heteropoda. 'The Cephalopoda of the Carboniferous are of great interest. CH. xxi.] CARBONIFEROUS CEPHALOPODA 855 and present a remarkable contrast to those of the Mesozoio rocks. The persistent Nautiloidea are found, but represented by various subgenera with channelled or tuberculated shells, while the forms of straight or slightly curved forms, Orthoceras (fig. 493), Cyrtoceras, &c., are very abundant. True Ammonites Fig. 494. Fig. 495. Goniatites crenistria, Phill., . Mountain Limestone. N. America ; Britain ; Germany, &c. a. Lateral view. b. Front view, showing the mouth. Goniatites Listeri, Mart., Coal-measures, Yorkshire and Lancashire. are quite unknown, but the group of the Ammonoidea is repre- sented in the Carboniferous by many forms of Goniatites (figs. 494, 495), some of which are of simple structure and allied to the older Devonian types, while others begin to show the greater complication of lobes indicative of the later Ceratites. Fig. 496. Fig. 497. Aficroconchu* (Spirorbis) carbonarius, Murch. Nat. size and magnified. b. Variety of sama. Leperditia inflata, Murch. sp. Nat. size and magnified. A Carboniferous Ostracod. The Vermes are represented by the minute Microconchus (Spirorbis), which Is often found attached to fragments of the vegetation of the period which had floated in the ocean. Among the Arthropods, a few Decapod Crustaceans are found in the Carboniferous, with numerous Limulids (Prsstwichia, Belinurus, &c.), and some Phyllopods, Isopods, and Ostracods, Many limestone bands in the Carboniferous system are 356 CAKBONIFEROUS FISH [CH. xxi. found to be completely made up of the remains of these minute bivalve Crustaces, and the surfaces of many of the shales are covered with them. Among the genera which occur in the greatest abundance are Leperditia (fig. 497), Carbonia, Bey- richia, Cytherella, and Kirkbya. Fig. 498. Fig. 499. Psammodus porosus, Ag. Bone-bed, Mountain Limestone. Bristol ; Armagh. The. sole survivors of the abundant Trilobites of the older systems are the genera Pkillipsia, Griffitliides, and Bracliyme- topus, the first-mentioned of which survived to Permian times. Of the fish of the period we find numerous remains in the fin-spines (ichthyodorulites), scales, and palatal teeth of the Sela- chians : Psammodus (fig. 498), Cocliliodus (fig. 499), Orodus, Gyracanthiis, Ctenacanthus, &c. Heterocercal ganoids, like Palceoniscus, Acrolepis, &c., abound, while forms of the Dipnoi (Ctenodus) are also found. Amphibians are represented by a few early types of Stego- cephala, a group which attained such a remarkable development in Permian and Triassic times, while Eeptilia and all higher groups are unknown in the Car- boniferous. It is, however, the Terrestrial flora and fauna of the Carboni- ferous which are of such great geological importance. Interest attaches to the Coal-measure flora not only on account of the remarkable differences which it presents, alike from the Mesozoic, the Cainozoic, and the existing floras, but from the circumstance that it is the oldest assemblage of land plants of which at present we have any knowledge. Although many of the Carboniferous plants are only repre- sented by casts and impressions or thin coaly films, often showing the outer markings on bark or leaves with great fidelity, yet the internal structure of many of these ancient plants has been very Cochliodus contortus, Ag. Bone-bed, Mountain Limestone. Bristol ; Armagh. CH. xxi.] CARBONIFEROUS FLORA 357 fully investigated by the late Professor W. C. Williamson and other botanists. The manner in which this is accomplished is by making thin sections of the ' coal-balls,' or masses of woody tissue impregnated with calcium carbonate, which are found in some beds of coal, and comparing these with sections of living plants under the microscope. One of the chief difficulties in studying the ancient plants of the Carboniferous period arises from the circumstance that we only find detached portions of many of them, and it is now known that the bark, leaves, root, stem, pith, and fructification of the same coal-plants have often Fig. 500. Fig. 501. Tricfonocarpnm ovatum, Lindl. and Hutt. Peel Quarry, Lancashire. Fig. 502. Fragment of coniferous wood, Da- doxylon, Endl., fractured longi- tudinally ; from Coalbrook Dale. W. C. Williamson. a. Bark. 6. Woody zone or fibre (pleuren- chyma). c. Medulla or pith. T rigonocarpum olivceforme, Lindl., d. Cast of hollow pith or 'Stern- with its fleshy envelope. Felling bergia.' Colliery, Newcastle. / been referred to as many distinct genera. Such mistakes can only be rectified when we have the rare good fortune to find these various parts of the plant united in the same specimen. So far as is at present known, there were no forms of vege- table life present in the Carboniferous higher in the scale than the Conifers. Numerous varieties of w r ood like the Dadoxylon (fig. 500), Araucarioxylon, &c., are found having the characteristic exogenous structure of the Conifers, with the peculiar pitted vessels of that group. The fossil fruits so abundant in some partsof the Coal-measures, andknown as Trigonocarpon, have the very closest analogy too with the fruit of the Ginkgo (Salisburia), 358 CARBONIFEROUS GYMNOSPERMS [CH. xxz. (a remarkable genus of the Yew 'tribe having some affinities with the Gnetacese), of which the leaves are found in the Permian. Other forms of Gymnosperms found in the Carboniferous are more doubtful in their relationships. In some silicified fruits occurring in the Carboni- ferous of France, Brongniart showed that we have a curi- ous combination of charac- ters now found in the groups of the Cycads, the Conifers, and the Gnetaceee ; and it is not improbable that the earliest plants of this group were synthetic types, in which the differentiation of characters found in existing """SSS? SSkSS p&JSr ""' f 8 did not exist - Cordaites with large simple parallel-veined leaves is referred to an extinct order which is believed to have been related to the existing Cycads. With the exception of these aberrant Gymnosperms, the plants of the Carboniferous period appear to have belonged to the Crypto- gams, or plants propagated by means of spores. Some of them were homosporous (like Lycopodium), and others were hetero- sporous (like Selaginella), but nearly all of them presented a re- markable peculiarity in their mode of growth which distinguishes them from all living Cryptogams. While all living Cryptogams are ' Acrogens ' and grow by additions to their summit only, the Cryptogams of Carboniferous times were able to increase by an c. Pith. b. Wood. e, e, e. Medullary rays. exogenous growth like the Conifers and the higher dicotyledonous plants. In all of these we have a bark and pith, united by plates of cellular tissue known as ' medullary rays,' between which the wedge-like masses of woody tissue are developed. Each year a layer of ' cambium ' is formed between the woody CH. xxi.J CARBONIFEROUS CRYPTOGAMS 359 axis and the bark, and thus an addition is made to the diameter of the trunk. It has now been fully demonstrated that this mode of growth was followed by the Cryptogams of the Car- boniferous period, and in consequence of this peculiarity they were able to assume the characters and dimensions of great forest-trees. The exact analogies of many of the gigantic Cryptogams of the Carboniferous are still very doubtful. Many ferns un- doubtedly existed, some of which attained the size of the largest Fig. 506. Living tree-ferns of different genera. (Ad. Brong.) Fig. 505. Tree-fern from Isle of Bourbon. Fig. 506. Cyathea glauca, Borg., Mauritius. Fig. 507. Tree-fern from Brazil. tree-ferns. In the Carboniferous ferns we find the character* istic venation of the leaves or fronds; similar arrangements of the spore-cases upon them ; and the same vernation, or mode of rolling up of the leaves, which are found in existing forms. The trunks of these ferns are often marked by scars, as in Caulopteris (fig. 509), and they frequently exhibit aerial root-like stems, like Psaronius. Some at least among them appear to have had the same exogenous mode of growth that distinguished the other primitive Cryptogams of the period. Williamson and Scott have recently described synthetic forms intermediate be- tween Ferns and Cycads. Another group of Carboniferous Cryptogams is believed to 360 CALAMITES [cu. xxi. have been related to the insignificant EquisetaceaB of our ponds and ditches, These Calamites, however, exhibit the exogenous mode of growth and attained to vast dimensions. Often only Fig. 508. Fig. 509. Pecopteris elliptica, Bunb., nat. size. Caulopteris primceva, Liudl., \. Fig. 611. Fig. 512. Calamites Stickowii, Brong., natural size. Common in coal throughout Europe. Stem of fig. 510, as restored by Sir W. Dawson. Eadical termination of a Calamite. Nova Scotia. the cast of the interior is preserved in a fossil state, though sometimes the woody matter remains reduced to a thin shell of coal. OH. XXI.] SPHENOPHYLLITES 361 Masses of whorled leaves referred to the genera Annularia (fig. 513), Sphenophyllum (fig 5\4:},Si 1 ndAsterophyllites (fig. 515), are believed by some botanists to have belonged to plants with Calamite-like stems, but the reasons given for the identification are not in all cases satisfactory. The greater number of the gigantic Cryptogams of the, Carboniferous period appear to have been related to the existing Fig. 513. Fig. 514. Annularia sphenophylloides, Zenk. Sphenophyllum erosum, LindL and Hutt. Fig. 515. Asterophyllites foliosus, Lindl. and Hutt. Coal-measures, Newcastle. Lycopods and Selaginellas. But while these modern plants usually creep along the ground and the erect forms are only a few inches in height, the Lepidodendra of the Carboniferous formed great trunks rising to a height of forty or fifty feet and ex- hibiting an exogenous structure. The ancient forms exhibit a dichotomous mode of branching and the peculiar scars on their stems which mark the position of the leaflike appendages. But what is of more importance to the botanist is the circum- stance that we find the fruits or cones (Lepidostrobus) some- 362 LEPIDODENDKON [CH. XXI, times attached to the branches of the plant and exhibiting all the peculiarities of the heterosporous Selaginella, both macro- spores and microspores being present in them. Fig. 516. a. Lycopodium densum, Labill. Living species. New Zealand. 6. Branch ; natural size. c. Part of same, magnified. Fig. 517. Fig. 518. Lepidodendron Sternbergii, Brong. Coal-measures, near Newcastle. Fig. 517. Branching trunk, 49 feet long, supposed to have belonged to L. Sternbergii. (Foss. Flo. 203.) Fig. 518. Branching stem with bark and leaflets of L. Sternbergii, . (Foss. Flo. 4.) Fig. 519. Portion of same nearer the root. Natural size. (Ibid.) Other closely related forms, presenting a different pattern of leaf- scars, -have been called Sigillaria; but it is very doubtful if the form and arrangement of the leaf- scars constitute a safe basis of classification in these ancient forms of vegetation. The stems of both Lepidodendra and Sigillariae are some- OH. XXI.] STIGMAKIA 363 Fig. 520. Fig. 521. a. Lepidostrobus ornatus, Brong. (Strobilus or cone), Shropshire ; nat. size. b. Portion of a section showing the large spo- rangia in their natural position. c. Microspores occurring in these sporangia, highly magnified. Sigillaria l&vigata, Brong. times found attached to large dichotomously branching root- stocks, known as Stigmaria. These forms are illustrated in the figures below- (Note S, p. 606). Fig. 522. Stiginariae attached to a trunk of Sigillaria. Pig. 523. Stigmaria flcoides, Brong. I natural size. (Foss. Flo. 32.) 364 CARBONIFEROUS LAND-SHELLS Fig. 524. In considering the terrestrial flora of the Carboniferous system, it must be remembered that the plants preserved in a fossil state are nearly all such as would grow in marshy flats like those found in the deltas of great rivers. Of the plants that may have flourished in the higher ground at the same period we have compara- tively little knowledge ; but in some Surface of^nTther indiyi^Tof of the beds of Carboniferous Sand- same species, showing form of stone great trunks of Coniferous trees fifty or sixty feet in length have been found, occupying an inclined position ; these were probably trees that had been carried down by rivers and formed ' snags ' like those seen in the Mississippi at the present day. Among plants found in a fossil state, purely herbaceous forms would have little chance of being included. That even the lowly cellular plants, represented by delicate filaments, existed in the Carboniferous period is proved by the circumstance pointed out by the late Dr. P. M. Duncan and other authors that the calcareous organisms of the Carboniferous (Corals, Echinoderms, and Mollusca) are often found to be penetrated by fine cavities produced by these parasitic algae. The terrestrial fauna asso- ciated with this remarkable flora of the Carboniferous is of great interest, though as yet very im- Fig. 526 Fig. 525. Pupa vetusta, Daw. a. Natural size. Zonites (Conulus) prisons, Carpenter. b. Magnified. perfectly known. In Nova Scotia Sir J. W. Dawson has dis- covered in the hollows of old tree-trunks representatives of the oldest known land-shells (pulmoniferous gastropoda) which have been referred to Pupa (fig. 525) and Zonites (fig. 526). And associated with these are remains of Myriapoda (fig. 527) and Arachnida (Spiders and Scorpions) ; while many forms of insects abounded in the Carboniferous. Some of the insects were closely CH. XXI.] CARBONIFEKOUS INSECTS 365 related to the Coleoptera (Beetles), the Orthoptera (Cockroaches, Crickets, &c.) (fig. 527), the Neuroptera (Ants, &c.), and other living orders ; but others seem to have been curious synthetic types for which naturalists have been compelled to establish new orders. Fig. 527. 6 XyloUus Sigillarice, Dawson. Coal, Nova Scotia and Great Britain. a. Natural size. b. Anterior part, magnified, c. Caudal extremity, magnified. Air-breathing vertebrates were probably represented in the Carboniferous by Stegocephali (Labyrinthodonts) of very simple structure, like the Dendrerpeton of North America, and the Archegosaurus (fig. 529) and other European forms. Of some of the latter not only the skeletons, but impressions of the skin have been described by H. von Meyer (fig. 530). Fig. 528. Wing of a Grasshopper, Gryllacris lithanthraca, Goldenb., nat. size. Coal, Saarbriick, near Treves. In North America footprints of animals of considerable size are sometimes found on the surfaces of sandstone slabs that are also traversed by sun-cracks. The footprints were probably those of air-breathing animals, possibly Amphibians, like those of which the bones have been found. It is interesting to notice also that the Carboniferous rocks present examples of rain-prints and worm-tracks, exactly similar to those which can be seen on muddy shores at the present day (figs. 531-532). British representatives of the Carboniferous Strata. The rapid changes in the thickness and characters of the British Carboniferous have been already referred to. In Devonshire we find 366 CULM-MEASUKES [CH. xxi. wha.t is known as the ' Culm facies ' of the Lower Carboniferous, consisting of alternations of hard shales, impure limestones and grey - wackes, with much carbonaceous matter, but no useful coal-beds. In the south-western group of coal-fields (South Wales, Bristol, and Fig. 529. Archegosaiinis minor, Goldf. Fossil Amphibian from the Coal-measures, Saarbriick. Fig. 530. Imbricated covering of skin of Archegosaurus medius, Goldf. Magnified. Somerset, Forest of Dean, and Mendip Hills) the Lower Carboniferous strata are of moderate thickness, while the upper beds of the system (Coal-measures) attain an enormous thickness. The strata, which are often much bent and folded, include many coals poor in hydrocarbons CH. XXI.] OF DEVONSHIRE 367 (steam-coals) and abo anthracites. These coal-h'elds present a close analogy with those of Northern France, Belgium, and Westphalia. In the Midland coal-fields (Staffordshire, Warwickshire, Leicestershire, Coalbrook Dale, &c.) the Lower Carboniferous is generally absent, and Fig. 531, Scale oue-sixth the original. Slab of sandstone from the Coal-measures of Pennsylvania, with footprints of air-breathing amphibian and casts of cracks. the Coal-measures rest directly upon the older rocks. In the York- shire and the Lancashire and Cheshire coal-fields, the Lower Carbo- niferous attains a great thickness, consisting of purely marine strata (Carboniferous limestone and shale) in the southern part of the area, which alternate with more and more freshwater and terrestrial beds 368 COAL-MEASUKES AND MILLSTONE GRIT [CH. xxi, as we pass northwards. In Scotland estuarine beds with coal are found from the top to the bottom of the series, and even in the arenaceous division (Calciferous sandstone) which in the northern part of Great Britain forms the lowest member of the Carboniferous Fig. 532. Fig. 533. l?!g. 532. Carboniferous rain-prints with worm-tracks (a, 5) on green shale, from Cape Breton, Nova Scotia. Natural size. Fig. 533. Casts of rain-prints on a portion of the same slab (fig. 532), seen to project on the under side of an incumbent layer of arenaceous shale. Natural size. The arrow represents the supposed direction of the shower. General notice of the divisions. The Coal-measures of the North of England differ, to a certain extent, from those of the south-west ; but a typical series would include the following strata, beginning at the top. 1. Red and grey sandstones, clays, and some- times breccias, with occasional coal- seams and streaks of coal and Spirorbis-limestone with Leper- ditia infiata, Murch. sp. 2. Middle coals, yellow sandstones, clays, and shales, with numerous workable coal-seams resting on fire-clays : fos- sils, Anthracosia, Anthracomya, Beyrichia, Estheria, Spirorbis. 8. Lower beds, gannister beds, flag- stones, shales, and thin coals, with haid siliceous layers beneath the coal-seams. Flagstones inter- calated. Fossils, Aviculopecten, Lingula, Goniatites, Orthoceras. Bone-bed, with fish and Laby- rinthodonts. In Scotland the equivalents of the uppermost beds above men- tioned are probably a red sand- stone group without coals, over- lying workable (flat) coals, and in the North-west of England these beds are barren here and there, as at Wigan ; but at Manchester they are important and coal-bearing. At Burnley, on the other hand, the beds are absent. The Millstone grit, well seen in South Wales, is grandly developed beneath some Coal-measures, and feebly beneath others, or it may be wanting. For along a line drawn from Shropshire through South Staffordshire and Leicestershire, to the Wash, a ridge of Palaeozoic rocks existed in Carboniferous times, on which little or usually no marine accumulation took place. Hence the Coal-measures at Coal- brookdale, South Staffordshire, rest upon Silurian rock with a very little or no gannister grit intervening. This ridge of old rocks, or ' central barrier,' was a Car- boniferous land-surface, and the GH. XXI.] CARBONIFEROUS LIMESTONE 369 grits collected on either flank, in- creasing in thickness far away to the north and west, and attaining a thickness of 9,000 feet in North Staffordshire, 12,130 feet in South Lancashire, and 18,700 feet in North Lancashire. The thickness of the grits at the edge of the Staffordshire coal-field is only 200 feet, and it is 3,000 feet in Western Yorkshire. The grits vary greatly in their lithology. Some are very rough and massive, others are fine-bedded micaceous sandstones and flags, whilst the bulk are jointed or are strata of varying thicknesses, and with the grains distinctly visible. All the sandstones are felspathic, and the grains are often united by a felspathic matrix. The area whence the grits came, carried by marine currents, was in the north- west. Thin coal-seams and coal- plants are found in some places in the grits, and sometimes a marine fauna exists, including fossils of the same species as those found in the lower strata called Carboniferous limestone. The grits are divided into the Rough Rock, or first grit, which underlies the lower Coal- measures ; the Flag Rock or Haslingdon Flags, or second grit, with shales and thin coal; the third grit of gritstone, flagstone, shale, and thin coals, with marine fossils; the Kinderscout grit, or fourth grit : this last forms the Peak in Derbyshire. In Scotland the Moor-rock, with thin seams of coal, is the equivalent of the English grits, and its very moderate thickness diminishes in Ayrshire, where it consists of a few beds of sand- stone at the base of the Coal- measures. Carboniferous Xiimestone series. In Yorkshire there is a downward continuation of sand- stones and shales, resembling those of the Millstone grits with inter- calated limestones, some of which are thickly crowded with en- crinites. Phillips called these the Yoredale series, and they attain the thickness of from 800 to 1,000 feet in Yoredale. The genera of marine fossils which are found in these strata are Nautilus, Orthoceras, Phragmoceras, Goniatites, Euom- phalus, Bellerophon, Productus, Spirifera, Phillipsia, Zaphrentis, &c., and these are common in the underlying carboniferous limestone. Beds of thin coals occur in the lower Yoredale strata. These strata are not found in the Centre and South of England, where the true Mountain or Carboniferous limestone exists. This important limestone, well seen in Derbyshire, South Wales, and Somerset, is massive, well bedded, and light-bluish, grey, reddish, or black m colour, and it may be either compact or crystal- line. The limestones are thickest where the grits above are thinnest, and have suffered much denuda- tion where they are at the surface. The fossils contained in them are very numerous, and in some places encrinites compose much of the rock, whilst Foraminifera are equally abundant elsewhere. The base of this important set of strata varies locally. In South Wales and Somersetshire the lower part merges into a shale Lower Lime- stone shale and this into bottom beds of yellow and green sand- stones and marls with plant- remains, and a bone -bed with Placoid fish-remains. This rests on Old Red Sandstone. In some parts of Yorkshire there are alter- nations of sands and clays at the base with plant-remains, and in the west of the county con- glomerates form the base, and rest upon Silurian rocks the Old Red Sandstone of the south-west not being present. Elsewhere, either the base of the limestone has not been seen, or it rests on very old rocks without the intervention of any beds of shaie. In Central England, where the other sedimentary beds are reduced to about 3,000 feet, the Carbonife- rous Limestone attains an enormous thickness, and, according to Mr. Hull's estimate, as much as 4,000 feet at Ashbourne, near Derby. To. a certain extent, therefore, we may consider the calcareous member of the formation as having originated simultaneously with the accumu- BB 370 CARBONIFEROUS LIMESTONE [CH. XXI. lation of the materials of grit, sandstone, and shale, with seams of coal; just as strata composed of mud, sand, and pebbles, several thousand feet thick, with layers of vegetable matter, are now in process of formation in the cypress swamps and delta of the Mississippi, while coral reefs are simultaneously forming on the coast of Florida, arid in the sea of the Bermuda Islands. For we may safely con- clude that in the ancient Carboni- ferous ocean those marine animals which secreted calcium carbonate were never freely developed in areas where the rivers poured in fresh water charged with sand or clay ; % and the limestone could only be- come several thousand feet thick over parts of the ocean bed which were being slowly depressed, the water remaining perfectly clear for ages.(Note?T, p. 607). The Carboniferous Limestone, with its associated Yoredale series, diminishes in thickness northwards, and undergoes remarkable changes in its lithology and fossils. In Northumberland, beds of coal are found right down to the bottom of the representatives of Lower Lime- stone and Shale, constituting what -is known as the ' Tuedian Series.' In Scotland Sir A. Geikie notices that the massive limestones dwindle down and are replaced by thick courses of yellow and white sand- stone, dark shale, and seams of coal and ironstone. Limestone beds are met with in thin sheets only. The whole formation is divided into the Carboniferous limestone and the underlying Calciferous sandstones. These last- mentioned strata consist of red and yellow sandstones with many- coloured marls, which pass insen- sibly into the Upper Old Red Sand- stone beneath. They are very unfossiliferous, but Sphenopteris affinis, Lindl., is common. Above the red sandstones is the Cement- stone group, of different coloured sandstones, shales, oil shales, and argillaceous limestones. In the West of Scotland these beds are poor in fossils. In the area of the Firth of Forth the Cement-stone group con- tains ironstones, seams of coal, oil shales, and sandstones ; and these last contribute to the building materials of Edinburgh. The oil shales yield petroleum on distilla- tion. Amongst the limestones of the group are the Burdie House lime- stones, composed of the tests of an Ostracod Crustacean, Leperditia Okeni, Munst., and containing fish, of which Megalichthys is a promi- nent form. Seams of coal occur, and one called the Houston coal is worked in Linlithgowshire. Sphenopteris, Lepidostrobus, A.raucarioxylon, and Lepidodendron are found in them. The Carboniferous limestone group of Scotland probably repre- sents the upper part of the English limestone in age, and consists of a few seams of encrinital limestone, shales, fire-clays, and seams of coal. The thickest of the limestones, the Hurlet, is in places 100 feet thick ; it overlies a seam of coal and pyri- tous shales, and above it are other important coal-seams and iron- stones. These last contain marine fossils, and the coals have plants and fish -remains and those of Labyrinthodontia. Some of the limestone -seams are very per- sistent over wide areas. In Ireland the Carboniferous rocks of the North have their lower series like the Scottish Calciferous sandstones. But in the southern districts there is a deep group of black and dark-grey shales, impure limestones, and grey and green grits with slates, which overlie the Old Eed Sandstone, and are beneath the base of the Carboni- ferous limestone. This group is the Carboniferous slate. Its age is not quite certain ; it may be either the equivalent of the Lower-lime- stone shale of the South-west of England or be part of the Devonian formation. The Carbonif erouslime- stone covers a large part of Ireland, and it alternates with sandstones towards the north. Deposition of tne Carbo- niferous formation. It has been mentioned (p. 368) that Coal- measures rest upon Silurian and old rocks in some parts of the CH. xxi.] OEIGIN OF THE COAL-MEASUKES 371 central barrier. This was the land of the age in the first instance. The sea flowed in upon ths Old-Red- Sandstone terrestrial area, north of the Bristol Channel, and to the north of the central barrier also, and the shales and sandstones of the lowest marine deposits accumu- lated. Further north there was a land vegetation, at times, during this age. Sinking of the greater part of the area continued, probably along lines of fault, and a considerable depth of limestone was formed, and, as time elapsed, this became an arenaceous deposit in Yorkshire and northwards. Here and there were land-surfaces, and coal-plants accumulated and formed coal. In Scotland the depression persisted, but silting up of the sea-floor, and volcanic disturbances and ejections, enabled the terrestrial surfaces to be formed over and over again. Then came a long period of wear and tear of land, mostly situated in the north- west, and the age of the Millstone Grit set in. Even during its time there were a few land-surfaces which produced coal. Subsequently the depression still continued, and the deep Coal-measures accumu- lated. The amount of volcanic energy displayed was great at certain epochs of the Carboniferous age, and will be noticed further on. Lastly, enormous curving and dislocation of the Carboniferous rocks, and great denudation of their exposed surfaces, took place Thou- sands of feet of Coal-measures were worn off before the deposition of the Permian rocks, and subsequently. It would appear that after the depo- sition of the Coal-measures, a thrust acted from north to south and south to north, forming great curvatures of the strata, the long axes being east and west. Denudation occurred, and the Permian deposits accumu- lated. Then curving occurred in the opposite direction, the axes of the curved strata being north and south. Hence more or less basin- shaped areas were produced ; and denudation wore off and displayed the edges of the underlying grits and limestones on the edge of the several basins. The term Coal-field is applied to an area where coal is visible at the surface at its edges or outcrops, or where it is not too deeply seated to be worked. There are about twenty principal coal-fields in Great Britain, and several smaller ones. Some of these form complete basins, entirely circumscribed by the lower members of the formation, others have one part of the basin visible, the rest being covered up by Permian or other strata, and the rest are bounded by faults. Coal formed on land. In South Wales, where, as already pointed out, the Coal-measures at- tain a great but variable thickness, the sandstones and shales appear to have been formed in water of moderate depth, during a slow, but perhaps intermittent, depres- sion of the surface, in a region to which rivers were bringing a never-failing supply of muddy sedi- ment and sand. The same area was alternately covered with vast forests, such as we see in the deltas of great rivers in warm climates, which are liable to be submerged beneatl\. fresh or salt water, should the land sink vertically a few feet. In one section, near Swansea in South Wales, where the total thickness of the Coal-measures is 3,246 feet, we learn from Sir H. de la Beche that there are ten princi- pal masses of sandstone. One of these is 500 feet thick, and the whole of them make together a thickness of 2,125 feet. They are separated by masses of shale, vary- ing in thickness from 10 to 50 feet. The intercalated coal-beds, sixteen in number, are generally from 1 to 5 feet thick, one of them, which has two or three layers of clay inter- posed, attaining 9 feet. At other points in the same coal-field the shales predominate over the sand- stones. Great as is the diversity in the horizontal extent of indivi- dual coal-seams, they all present one characteristic feature, in having, each of them, what is called its underclay. These underclays. co- extensive with every layer of coal, consist of arenaceous shale, some times called fire-clay, because it can be made into bricks which B B 2 ,372 UNDERCLAYS [en. xxi stand a furnace-heat. They vary in thickness from 6 inches to more than 10 feet, and, as Sir William Logan pointed out, are character- ised by enclosing the peculiar fossil plant called Stigmaria. It was also observed that, while in the over- lying shales or ' roof ' of the coal, ferns and trunks of trees abound, without any Stigmaria, and are flattened and compressed, those singular plants of the underclay most commonly retain their natural forms, unflattened and branching freely, sending out their slender rootlets in all directions. A num- ber of species of Stigmaria were known to botanists, and described by them, before their position under each seam of coal was pointed out, and before their true nature as the roots of trees (some having been actually found attached to the base of Sigillaria stumps) was recog- nised. Now that all agree that these underclays are ancient soils, it follows that where we find them they attest the terrestrial nature of the plants which formed the overlying coal, which consists of the trunks, branches, and leaves and spores of the plants which had their roots in the clay. The trunks have generally fallen prostrate in the coal, but some of them still remain at right angles to the ancient soils (see fig. 64, p. 60). Professor Gbppert, after ex- amining the fossil plants of the coal-fields of Germany, has detec- ted, in beds of pure coal, remains of every family of plants of which representatives occur fossil in the Carboniferous rocks. Many seams, he remarks, are rich in Sigillaria, Lepidodendron, and Stigmaria, the latter in such abundance as to appear to form the bulk of the coal. In some places, almost all the plants were Catamites, in others ferns. Between the years 1837 and 1840, six fossil trees were dis- covered in the coal-field of Lan- cashire, where it is intersected by the Bolton Railway. They were all at right angles to the plane of the bed, which dips about 15 to the south. The distance between the first and the last was more than 100 feet, and the roots of all were embedded in a soft argilla- ceous shale. In the same plane with the roots is a bed of coal, 8 or 10 inches thick, which was found to extend across the railway, to the distance of at least ten yards. Just above he covering of the roots, yet beneath the coal- seam, so large a quantity of the Lepi- dostrobusvariabilis, Lindl., was dis- covered, enclosed in nodules of hard clay, that more than a bushel was collected from the small openings around the base of some of the trees. The exterior trunk of each was marked by a coating of friable coal, varying from one quarter to three-quarters of an inch in thick- ness; but it crumbled away on removing the matrix. The dimen- sions of one of the trees are 15^ ft. in circumference at the base, 7 feet at the top, its height being 11 feet. All the trees have large spreading roots, solid and strong, sometimes branching, and traced to a distance of several feet, and presumed to extend much further. In a colliery near Newcastle a great number of specimens of Sigil- laria occur in the rock, retaining the position in which they grew. Not less than thirty, some of them 4 or 5 feet in diameter, were visible within an area of 50 yards square, the interior being sandstone, and the bark having been converted into coal. (See fig. 65, p. 60.) It has been remarked that if, instead of working in the dark, the miner were accustomed to remove the upper covering of rock from each seam of coal, and to expose to the day the soils on which ancient forests grew, the evidence of their former method of growth would be obvious. Where coal occurs on Gannister a gritty sandstone there is no underclay, and usually marine re- mains are found above the seam. In this instance, the vegetation did not grow where it became mine- ralised, but was carried by water- power from some other locality and deposited. The numerous coal-seams oc- curring one over the other, in a CH. XXI.] CANNEL COAL 373 series of often 10,000 feet of verti- cal measurement, indicate that the plants grew on a rapidly subsiding area, into which the sea occasionally penetrated. There are also coal-seams com- posed of the variety known as ' Cannel coal ' (called also ' parrot- coal ' in Scotland, from the noise it gives out in burning), which ap- pears not to have been formed directly from growing plants, but from the black peaty mud derived from their decay and partial de- composition. If the black muds formed by the bursting of peat mosses were to collect in hollows and undergo induration and chemi- cal change, a material would pro- bably be produced not very dissimi- lar to Cannel coal. The Cannel coals often contain a very large proportion of ash, and thus pass insensibly into the highly bitu- minous shales known as oil-shales, from the fact that when heated in retorts they yield various petro- leum oils. Many of these, like the rock of Torbane Hill (Torbanite), are of considerable economic value. In some parts of the earth's crust the destructive distillation of carbonaceous rocks has resulted from natural processes, and accu- mulations of liquid hydrocarbons (natural oils) and of gaseous hy- drocarbons (natural gas) have taken place. Deep borings sometimes tap these accumulations, and then jets of oil or emanations of gas issue at the surface and can be collected and utilised for purposes of illumination and heating. Some of these natural oils and gases are connected with the rocks of Car- boniferous age, but others are found in association with strata of very different age. Clay-ironstone occurs as bands and nodules or in thin layers in the Coal-measures, and they are formed, says Sir H. de la Beche, of ferrous carbonate mingled me- chanically with earthy matter, like that constituting the shales. The nodules have generally formed around some organic object, and in some instances, like the Mussel - band ironstone, the valves of a shell of a mollusc have been converted into ferrous carbonate. Eobert Hunt found that decom- posing vegetable matter, such as would be distributed through all coal strata, prevented the further oxidation of the ferrous salts, and converted the peroxide into pro- toxide by taking a portion of its oxygen to form carbon dioxide. Such carbon dioxide meeting with the protoxide of iron in solution, would unite with it and form a ferrous carbonate ; and this min- gling with fine mud, when the excess of carbon dioxide was removed, might form beds or nodules of argillaceous ironstone. Marine beds intercalated in Coal-measures. In the coal- fields, both of Europe and America, the association of freshwater, brackish- water, and marine strata with coal-seams of terrestrial origin is frequently recognised. Thus the upper member of the Coal-mea- sures noticed on p. 368 was formed under brackish-water and marine conditions. The characteristic fos- sils are a small bivalve, having the form of a Cyclas or Cyrena, also a small Ostracod, Leperdi- tia inflata, Murch., and the shell of a minute tubercular an- nelid of an extinct genus called Microconchus (fig. 496) allied to Spirorbis. In many coal-fields there are freshwater strata, some of which contain shells termed Anthracosia and Anthracomya, now referred to freshwater groups like the Unio- nidce of the present day; but in the midst of the coal-series of Yorkshire and other districts we find thin and sometimes widely distributed seams abounding in the remains of fishes and marine shells like Orthoceras, Goniatites Lis- teri, Sow., and Aviculopecten pa- pyraceus, Groldf. These facts show that in the estuaries in which the coal-seams were probably formed the sea occasionally broke in and sometimes occupied the area for a greater or less length of time, 374 DEVONIAN AND OLD BED SANDSTONE [CH. xxn. A very full account of the Memoirs of the Geological Survey : several coal-fields of the British 'The Yorkshire Coal-field,' by Islands has been given by Professor Prof. A. H. Green; 'The Leicester- Hull in his 'Coal-fields of Great shire Coal-field,' by E. Hull; 'The Britain.' Fuller details on many Geology of Edinburgh,' by H. H. questions connected with the Car- Howell and A. Geikie, and in boniferous strata will be found in the various English treatises on the Reports of the Coal Commis- Geology, sions, and also in the following CHAPTER XXII THE DEVONIAN SYSTEM Relations of the Devonian Devonian Corals, Brachiopoda, Cephalopoda, and Trilobites The Devonian Fish and their Relationships to Living Forms The Devonian Flora and its Relation to that of the Car- boniferous Devonian Strata of Devon and Cornwall Upper, Middle, and Lower Devonian Old Red Sandstone Relations to Devonian Proof of Freshwater Origin Old Red Sandstone of Scotland, Lower, Middle, and Upper Old Red Sandstone of England and Wales Old Red Sandstone of Ireland. Nomenclature and classification of tbe Devonian strata. The name of Devonian was first proposed by Lonsdale for the series of strata underlying the Culm-measures in Devonshire, the fossils of which, as he showed, present many analogies with those of the Carboniferous on the one hand, and with those of the Silurian on the other hand, but are clearly distinct from those of both these systems- The fossils of the British Devon- shire strata are, however, not very numerous, and are generally badly preserved ; but in Central Germany, and especially in the Eifel district, rocks of the same age are found crowded with the most beautiful and exquisitely preserved fossils. Hence some authors have preferred to call this system of strata by the name of ' Eifelian,' but the older term Devonian is now almost universally employed by geologists. ^ In most parts of the British Islands, however, we find between the Carboniferous and Silurian strata a series of red sandstones, with conglomerates, argillaceous beds, and im- pure concretionary limestones, which contain no marine fossils but yield the remains of fish, crustaceans, land-plants, and, more rarely, of freshwater mollusca. From their relations we may infer that these strata which, from their position below the coal-bearing rocks, are known as the Old Red Sandstone are, speaking generally, contemporaneous (homotaxial) with the Devonian marine strata. This conclusion is confirmed by the fact that certain fish and crustaceans are common to the two sets of strata. We thus find side by side beds of marine and fresh- CH. XXII.] DEVONIAN COKALS 375 water origin deposited during the same geological period the former constituting the Devonian and the latter the Old Bed Sandstone. Characteristics of the Devonian Flora and Fauna. In the Devonian strata we find not only those obscure impressions which may possibly represent seaweeds, but well-preserved por- tions of gigantic Laminarians, to which the name of Nemato- phycus has been given. Both in Devonshire and the Eifel, Corals are particularly Fig. 534 Fig. 535. Favosites (Packgpora) ccrvicornis, Blairiv., nat. size. S. Devon, from a poli shed specimen. A Tabulate Coral. a. Portion of the same magnified, to show the tabulae and pores. Fig. 536. Heliophyllum Halli, E. & H. A Rugose Coral. Middle Devonian. After Nicholson. Heliolites porosa, (-roldf. sp., nat. size. a. One of the corallites magnified. Middle Devonian, Torquay. Ply- moutli, Eifel. abundant, and they nearly all belong to the group of the Tetra- coralla (Rugosa). Among the common forms in the Devonian may be mentioned Favosites (fig. 584), various forms of Cyatho- phyllids (like HeliopJiyttum^g. 535), Heliolites (fig. 536), and the curious and highly characteristic operculate corals Calceola (fig. 537), which were formerly mistaken for Brachiopods. With the true Corals are found many other coral-like structures, like the Monticuliporida, which are probably allied to the Bryozoa, and the Stromatoporoidea, usually grouped with the Hydrozoa, 876 DEVONIAN BRACHIOPODA [CH. xxn. The Graptolites, which are so abundant in the Older Palseo- zoic rocks, are only represented by a few doubtful forms in the Devonian. Fig. 537. Calceola sandalina, L., f . Eifel ; also South Devon. a. Corallum. b. Operculum. The Crinoids of the Devonian period are rare in Devonshire but very abundant in the Eifel ; they are distinct from, though closely related to, those of the Carboniferous. In the Devonian, too, we find, side by side, forms of the Silurian Cystoidea and the Carboniferous Blastoidea. Among the Brachiopoda we find many forms of Spirifera (figs. 538, 539), Productus, Orthis,Athyris, Atrypa, Chonetes,&c., with certain genera peculiar to the Devonian system, such as String ocephalus (fig. 540), Uncites (fig. 541), Rensselceria, Megan- teris, &c. Fig. 538. Spirifera disjunct a, Sow., |. Syn. Sp. Verneuilii, Murch. Upper Devonian, Boulogne. Fig. 539. Spirifera mucronata, Hall, nat. size. Devonian of Pennsylvania. The Lamellibranchiata are represented by a number of genera, some of which are peculiar to the system. The genus Megalodon (fig. 542) is an abundairt and characteristic one. Gastropods of Mesozoic affinities, like Pleurotomaria, are XXH.J DEVONIAN MOLLUSCA 377 found mingled with forms like Murchisonia, which are abun- dant in the Older Palaeozoic ; while the Pteropoda, which are so abundant during the last-mentioned epoch, are represented in the Devonian by Conularia (fig. 543), Tentaculites, and other genera. Fig. 54(X Fig. 541. a Stringocephalus Burtini, Def., a. Valves united. 6. Interior of ventral or large valve, showing thick partition and portion of a large process which projects from the dorsal valve across the .shell. Fig. 542. Uncites gryphus, Def., . Middle Devonian. S. Devon and the Continent. Fig. 543. Megalodon cucullatus, Sow. Eifel ; also Bradley, S. Devon. a. The valves united. 6. Interior of valve, showing the large cardinal tooth. Fig. 544. Conularia ornata, D'Arch. and De Vern., . Kefrath, near Cologne, Fig. 545. Entomis serratostriata, Sandb. sp., Weilburg, &c. ; Cornwall ; Nassau ; Saxony; Belgium. a. Nat. size. Clymcnia linenris, Miinst. Petherwyn, Cornwall ; Elbersreuth, Bavaria, 378 DEVONIAN TBILOBITES [CH. XXII. Among the Cephalopoda of the Devonian we find, side by side with the Older Palaeozoic genera like OrtJioceras. Phrag- moceras, Cyrtoceras, &c., the oldest Ammonoidea in the Gonia- tites, and the remarkable and very characteristic genus Cly~ menia (fig. 544), which is confined to the Upper Devonian. The Arthropoda of the Devonian include the Ostracod Entomis (fig. 545), the bivalve shells of which are found cover- ing the surfaces of many of the shales. With the Eurypterids (usually found in the freshwater deposits of the period) we find a number of Trilobites, though these are no longer present in such numbers and variety as in the Older Palaeozoic formations. Fig. 546. Fig. 548. Phacopslatifrons,'Br'6Tm,T&,t. Bronteus flalellifer, Goldf. Homaionotusarmatus^iir- size. Characteristic of the nat. meister, f . Lower De- Devonian in Europe, Asia, Mid. Devon ; S. Devon ; vonian ; Daun, in the and N. and S. America. and the Eifel. Eifel ; and S. Devon. Species of Phacops (fig. 546), Bronteus (fig. 547), and Homalonotus (fig. 548), often distinguished by an abundance of spines, tubercles, or other external ornaments, are particularly characteristic of the Devonian fauna. As already remarked, a few of the fish-remains so abundant in the freshwater deposits of this age are also found associated with the marine fossils of the Devonian. The freshwater fauna of this period is a very interesting one, as it is the oldest known. It includes a representative of the Unionidae (Anodonta, fig. 549), and a number of Crustaceans, including the Eurypterids Pterygotus (figs. 550, 551),Eurypteru.s, Slimonia, &c. The curious bodies known as Parka decipiens, Flem. (figs. 552-554), are believed to be egg- cases of some of these large Crustaceans. OH xxii.] OLD RED-SANDSTONE CRUSTACEANS Fig. 549. Fig. 550. 379 Anodonta Jukesii, Forbes, $. Lpper Devonian, Kiltorkan, Ireland Pterygotus anglicus, Ag. For- mrshire. Ventral aspect. Re- stored by Dr. H. Woodward, F.R.S. . Carapace, showing the large sessile eyes at the anterior angles. b. The metastoma or post-oral plate ("serving the office of a lower lipX c. c. Chelate appendages (anten- nules). d. First pair of simple palpi (a- tennce). f. Second pair of simple palpi (mandibles'). f. Third pair of simple palpi (first maxillae). (/. Pair of swimming feet with their broad basal joints, whose serrated edges serve the office of maxillce. h. Thoracic plate covering the first two thoracic segments, which are indicated by the figures 1, 2, and a dotted line. 1-6. Thoracic segments. 7-12. Abdominal segments. 13. Telson, or tail-plate. Fig. 552. Pterygotus anglicus, Ag. Middle portion of the back of the head, called the ' Seraphim ' Fig. 551 13 Fig. 553. Parka decipiens, Flem. In sandstone of lower beds of Old Red, Ley's Mill, Forfarshire. Parka decipiens, Flem., nat. size. In shale of Lower Old Red, Park Hill, Fife. 880 FISHES OF THE [CH. XXII. Pig. 554. Old Bed Sandstone Shale of _Forfarshire. With impres- sion of plants and ova of Crustaceans. Nat. size. . Two pairs of ova (?) resem- bling those of large Sala- manders or Tritons on the same leaf. b, b. Detached ova. Most interesting of all are the remains of fishes found in these freshwater strata of Devonian age. In addition to a few representatives of the Bays, we find very remarkable forms of heterocercal ganoids, in such forms as Cephalaspis (figs. 555, 556), Pteraspis, &c. (see Note U, p. 607). Fig.5iw. Cephalaspis Lysllii, Ag. Length 6| inches. Prom a specimen found at Glamis. in Forfarshire. (See other figures, Agassiz, vol. ii. table 1 a and 1 6.) a. One of the peculiar scales with which the head is covered when perfect. These scales are generally removed, as in the specimen above figured. J, c. Scales from different parts of the body and tail. Fig. 556. Cephalaspis Lyellii, Ag. Eestoration. (After Page.) CH. XXII.] OLD KED SANDSTONE 381 By far the greater number of the Old Eed Sandstone fishes belong to the suborder of Ganoids, called Crossopterygidce or fringe-finned, by Huxley in 1861, in consideration of the peculiar manner in which the fin-rays of the paired fins are Polypterus. Living in the Nile and other African rivers. a. One of the fringed pectoral fins. c. Anal fin. d. Dorsal fin, or row of finlets. Fig. 558. &. One of the ventral fins. Iloloptychius. As restored by Professor Huxley. a. The fringed pectoral fins. c. Anal fin. 6. The fringed ventral fins. d, e. Dorsal fins. Fig. 559. Restoration of Osteolepis. Pander. Old Red Sandstone, or Devonian. a. One of the fringed pectoral fins. b. One of the ventral fins. c. Anal fin. d, e. Dorsal fins. arranged so as to form a fringe round a central lobe, as in the recent Polypterus (see a, fig. 557), a genus of which there are several species now inhabiting the Nile and other African rivers. The reader will at once recognise in Osteolepis (fig. 559), one of the common fishes of the Old Ked Sandstone, many points of 382 FISHES OF THE [CH. xxn. Fig. 560. Scale of Holoptychius nobilissimus, Ag. Clashbermie, \ nat. size. analogy withPolypterus. They not only agree in the structure of the fin, as first pointed out by Huxley, but also in the posi- tion of the pectoral, ventral, and anal fins, and in having an elongated body and rhomboidal scales. On the other hand, the tail is more symmetrical in the recent fish, which has also an apparatus of dorsal finlets of a very abnormal character, both as to number and structure. As to the dorsals of Osteolepis, they are two in number, which is unusual in living fish. Among the ' fringe-finned ' Ganoids we find some with rhomboidal scales, such as Osteolepis, figured above ; others with cycloidal scales, as Holoptychius (figs. 558, 560). In the genera Dipterus and Diplo- pterus, as Hugh Miller pointed out, and in several others of the fringe-finned genera, as in Gyroptychius and Glyptolepis, the two dorsals are placed far backwards, or directly over the ventral and anal fins. The Asterolepis (one of the Placodermata) was a ganoid fish of large dimensions. A. Asmusii, Eichwald, a species character- istic of the Old Bed Sandstone (Devonian) of Russia, as well as of the same rocks in Scotland, attained, ac- cording to Hugh Miller, the length of between twenty and thirty feet. They were partly clothed with strong bony armour, embossed with starlike tubercles. Asterolepis occurs also in the Devo- nian rocks of North America. Amongst the interest- ing points which have been recorded about the ganoid fish, Professor Huxley has observed that, while a few of the Palaeozoic and the majo- rity of the Secondary Ganoids resemble the living Bony Pike (Lepidosteus), or the Amia, genera now found in North- and Central-American rivers, the Crossopterygidae of the Old Red are closely related to the African Polypterus of the Nile and the rivers of Senegal. In 1870, a species of another genus of the . 561. Pterichthys, Agassiz ; upper side, showing slime-canal ; as restored bv H. Miller. CH. xxii.] OLD RED SANDSTONE 383 Dipnoid fish, Ceratodus Forsteri, Krefft, was found living in the rivers of Queensland, Australia. If many circumstances favour the theory of the freshwater origin of the Old Red Sandstone, this view of its nature is not a little confirmed by our finding that it is in Lake Superior and the other inland Canadian freshwater seas, and in the Missis- sippi and African rivers, that we at present find those fish which have the nearest affinity to the fossil forms of this ancient forma- tion. The peculiar family of Crossopterygidse of which we have a living example in the Polypterus of, the Nile had many Fig. 562. Fig. 563. Palceopteris hibernica, Schimp. (Cy- Bifurcating branch of Lepidodendron clopteris hibernica, Ed. Forbes.) Griffithsii, Brong. Upper Devo- (Adiantites, Gopp.) Upper Devo- nian, Kilkenny, nian, Kilkenny. representatives in Devonian times, including such representative genera as Holoptychius (fig. 558), Osteolepis (fig. 559), Glypto- lepis, &c. Among the anomalous forms of Old Bed fishes not referable to Huxley's Crossopterygidse, and which are even doubtful Ganoids, having many structures which relate them to modern Siluroids amongst the Teleosteans, are the genera Pterichthys, Cephafatpi*, Pteraspis, and Coccosteus. With regard to Pterich- thys, some writers have compared its shelly covering to that of Crustaceans, with which, however, it has no real affinity. The wing-like appendages, whence the genus is named, were first supposed by Hugh Miller to be paddles, like those of the turtle ; and there can now be no doubt that they do really correspond with the pectoral fins (fig. 561) (Note U, p. 607). 384 OLD RED-SANDSTONE FISH [CH. XXII. The genus Cephalaspis, or ' buckler-headed,' from the extra- ordinary shield which covers the head (figs. 555, 556), has the orbits close together, nearly in the centre of the shield, which has either side backwards. on a horn carried Pteraspis, of the same family, has also been found by the Rev. Hugh Mitchell in Old Red beds, Perthshire ; and it is interesting to note that this genus came in during late Silurian times. Mr. Powrie enumerated no less than five genera of the suborder Acanthodidae, the spines, scales, and other remains of which have been detected in the grey flaggy sand- stones, the chief genera being Acanthodes, Di- placanthus, and Cheira- canthus. Fig. 564. Fig. 565. Cone and branch of Lepido- dendron corrugalum. Lower Carboniferous, New Brunswick. Psilophyton princeps, Dawson. Species charac- teristic of the whole Devonian series in North America. a. Fruit, natural size. b. Stem, natural size. c. Scalariform tissue of the axis, highly magnified. In the Old Red Sandstone of Caithness Dr. R. H. Traquair has discovered the remains of a minute fish of very rudimentary CH. xxii.] OLD-RED SANDSTONE PLANTS 385 organisation, which appears to have curious affinities with the Lamprey and Hag, and to be referable to the group of the Marsipobranchia. He has called this curious, ancient and rudi- mentary type of fish Palczospondylus Gunni. While the Dipnoi are represented by Dipterus, the fore- runner of the genus Ceratodus, which lived on from the earliest Mesozoic to the present day, vertebrates of higher organisation than fishes have not as yet been met with in Devonian strata (see Note U, p. 607). The terrestrial flora of Devonian times does not appear to have differed in its general characters from that of the Car- boniferous period. Gigantic ferns likePalceopteris (fig. 562), with true Lepidodendrids (figs. 563, 564), are found mingled with some peculiar types like the Psilophyton of Sir J. W. Dawson, the affinities of which are somewhat doubtful (fig. 565). The form is interesting on account of its great antiquity. The flora of the Old Eed Sandstone is poor, but extremely interesting from its foreshadowing the later grand Carboniferous flora (Note S, p. 606). In the Upper Old Red there are only twelve species of plants, and the following genera are represented : Adiantites, Cala- mites, Filicites, Sagenaria, Sphenopteris, Trichomanites, and Knorria. The Lower Division contains Lepidodendron, also a Coniferous plant, and Psilophyton. The earliest known insects were brought to light in 1865 in the Devonian strata of St. John's, New Brunswick, and are referred by Mr. Scudder to the group Pseudo-neuroptera. One of them, a Platephemera, measured five inches in expanse of wing. It was an ancient May Fly with some peculiar structures not found in living representatives of the group. The genus Xenoneura has a remarkable union of characters which are found in different genera, at the present day. It is a lace-winged form of the May-fly group, furnished with a stridu- lating or musical organ like a Grasshopper. Such a genus is said to constitute a synthetic type. British representatives of the Devonian system. Marine strata of Devonian age are only found in the British Isles in Devon- shire and Cornwall. The rocks are much folded, faulted and altered : and, except in certain limestone beds, fossils are few and badly pre- served in them. By a comparison of the fossils of the Devonshire strata with those of the richly fossilif erous beds of the Eifel, Mr. Ussher has been able to make out the following succession in South Devon, which may be placed in general parallelism with the divisions recog- nised in North Devon. c c 386 UPPER AND MIDDLE DEVONIAN [CH. xxn. Upper Devonian Middle Devonian Lower Devonian South Devon Cypridina (Entomis) shales Goniatite limestone and shales Middle Devonian lime- stones, Stringocephalus limestone (Ashprington Volcanic series) Eifelian shales and shaly limestones, with Calccola sandalina, Lam. Grits and sandstones, with Homalonotus, Pleuro- dictyum, &c. North Devon Pickwell Down sand- stones (without fossils) Morte slates (with obscure fossils). Ilfracombe beds (with Stringocephalus lime- stone) Hangman Grits and Fore- land sandstones. Lyn- ton slates In North Devonshire the unfossiliferous Pickwell Down sand- stones are overlain by the Baggy, Marwood and Pilton beds, but these are now generally regarded either as Carboniferous in age or as con- stituting a transition series between the Devonian and Carboniferous. Similar strata intermediate in age between the Devonian and Car- boniferous are found in Ireland, and are known as the Carboniferous slate and the Kiltorcan beds. Upper Devonian Rocks. The slates and sandstones of Barn- staple contain the Brachiopod Spirifera disjuncta, Sow. (fig. 538), which has a very wide range in Europe, Asia Minor, and even China ; also Strophalosia caperata, Sow., together with the large Trilo- bite, Phacops latifrons, Bronn (fig. 546), which is all but world- wide in its distribution. The fossils are numerous, and comprise about 150 species of mollusca, a fifth of which pass up into the overlying Carbo- niferous rocks. To this Upper Devonian belong a series of lime- stones and slates well developed at Petherwyn, in Cornwall, where they have yielded 75 species of fossils. The genus of Cephalopoda called Clymenia (fig. 544) is represented by no less than 11 species, and strata occupying the same position in Germany are called Clymenien- Kalk, or sometimes Cypridinen- Schiefer, on account of the number of minute bivalve shells of the Crus- tacea called Entomis (Cypridina) serratostriata, Sandb (fig. 545), which is found in these beds in the Ehenish provinces, the Harz, Saxony, and Silesia, as well as in Cornwall and Belgium. Middle Devonian Rocks. We come next to the most typical portion of the Devonian system, including the great limestones of Plymouth and Torquay, as well as the slates and impure limestones of Ilfracombe, all replete with shells, trilobites, and corals. Of the co- rals 52 species are enumerated by Mr. Etheridge, none of which pass into the Carboniferous formation above or came from the Silurian strata below, although many genera are common to the three systems. Among the genera are Favosites, Heliolites, Smithia, Heliophyllum, and Cyathophyllum. The Helio- phyllum Halli, E. and H., a Kugose Coral (fig. 535), and Heliolites porosa, Goldf., an Alcyonarian (fig. 536), are species peculiar to this formation. Stromatopora occurs, and a few Bryozoa. With the above are found no less than 10 genera of Echinodermata, 6 of which are stone-lilies or Crinoids ; some of them, such as Cupressocrinus, are distinct from any Carboniferous forms. The mollusca also are less characteristic; of 26 genera of Brachiopoda, 19 are common to the Carboniferous series. The Stringo- CH. XXII.] LOWER DEVONIAN 387 cephalns Burtini, Defr. (fig. 540), and Uncites gryphus, Defr. (fig. 541), may be mentioned as ex- clusively Middle-Devonian genera, and extremely characteristic of the same division in Belgium. The Striiigocephalus is also so abund- ant in the Middle Devonian of the banks of the Rhine as to have suggested the name of Stringo- cephalus-Limestone. The only two species of Brachiopoda common to the Silurian and Devonian for mations are Atrijpa reticularis, L., which seerns to have been a cos- mopolitan species, and Stroplio- mena rhumboidalis, Wile. Among the Lamellibranchiate bivalves common to the Plymouth limestone of Devonshire and the Continent, we find the Megalodon (fig. 542). There are also 13 genera of Gastropoda, which have yielded 45 species, 5 of which pass to the Carboniferous group, namely, Acro- culia vetusta, Loxonema ru- gifera, Phil., L. tumicla, Phil., Murckisonia angulata, Phil., and M. spinosa, Phil. The Pteropod Tentaculites occurs in England, and on the Continent is found the genus Conularia (fig. 548). The Cephalopods have species of Cyrtoceras, Goniatites, Orlho- ceras, Nautilus, and nearly all of them are distinct from those in the Upper Devonian Limestone, or Clymenien-Kalk of the Germans, already mentioned. Although but 6 species of Trilobites occur, the characteristic Bronteus flabelliftr, Goldf. (fig. 547), is far from rare, and all collectors are familiar with its fanlike tail. In this same group, called, as before stated, the Stringo- cephalus or Eifel Limestone in Germany, several fish - remains have been detected, and among others the remarkable Old Red genus Coccosteus, covered with its tuberculated bony armour ; and these ichthyolites serve, as Sir R. Murchison pointed out, to identify this middle marine Devonian with the Old Red Sandstone of Britain and Russia. Beneath the Eifel Limestone (the great central and typical member of the Devonian on the Continent) lie certain schists called by German writers ' Calceola- Schiefer,' containing in abundance Calceola sandalina, L. (fig. 537), which was once considered a Brachiopod, but which has been shown to be an operculate coral. This is by no means a rare fossil in the slaty limestone of South Devon, and, as in the Eifel, is confined to the middle division of the system. Ziower Devonian Hocks. A great series of sandstones and glossy slates, with Crinoidea, Brachiopoda, and some corals and Bryozoa, occurring on the coast at Lynmouth and the neighbourhood, and called the Lynton Group, form the lowest member of the Devonian in North Devon. Traces of fish-remains occur, andPteraspis, a genus of Silurian fish, has been detected. Among the 18 species of all classes enumerated by Mr. Etheridge, two-thirds are common to the Middle Devonian ; but only one, the ubiquitous Brachiopod Atrypa reticularis, L., can be identified with Silurian species. Among the characteristic forms are Alvaolites suborbicularis, Lam., also common to this formation on the Rhine, and Orthis arcuata, Phil., very widely spread in the North Devon localities. But we may expect a large addition to the number of fossils whenever these strata shall have been caremlly searched. The Spirifer- sarid stone of Sandberger, as exhibited in the rocks bordering the Rhine between Coblentz and Caub, belong to this lower division, and the same broad-winged Spi- rifexr distinguish the Devonian strata of North America. Among the Trilobites of this era is the genus Phacops (fig. 546), and several large species of Homa- lonotus (fig. 548) are conspicuous. The genus is still better known as a Silurian form, but the spinose species appear to belong ex- clusively to the ' Lower Devonian,' arid are found in Britain, Europe, and the Cape of Good Hope. c c 2 388 THE OLD RED SANDSTONE [CJT. 21 Old Red Sandstone. Over the greater part of the British Islands we find developed the freshwater facies of the Devonian, which is known as the Old Bed Sandstone. In South Wales and Hereford we find a great thickness (10,000 feet) of red and green shales, flagstones, sandstones, and conglomerates, with some impure concretionary limestones ; these pass downwards conformably into the Silurian and upwards into the Carboniferous. In the transition beds a few marine fossils are found mingled with freshwater forms ; but in the great mass of the strata of this age only fishes and a few traces of land plants have been found. In Scotland, the Old Bed Sandstone can be separated into three subdivisions, each of which contains a characteristic fish-fauna. The Upper Old Bed Sandstone, which is found both in Fife and the Orkney Islands, and consists of yellow and red sandstone, contains many forms of Holoptychius, Pterichthys, Glyptopomus, Glyptolasmus, &c., and appears to graduate upwards into the Carboniferous. In Caithness a great series of flag- stones, alternating with variegated sandstones, contains a very rich fauna including C heir acanthus, Cheiroleins, Dipterus, Diplacanthus, &c., with many remarkable examples of the small phyllopod Estheria minuta, Goldf ., and some plant remains : these are regarded by many geologists as constituting a distinct subdivision, the Middle Old Bed Sandstone. The Lower Old Bed Sandstone, which contains many forms of Cephalaspid fish and Eurypterids and appears to graduate downward into the Silurian, is well developed in Perthshire and Forfar- shire. The Scottish strata of Old Bed Sandstone age are of enormous thickness, and include many masses of very coarse conglomerate, which by some authors have been thought to be of glacial origin. That the Old Bed Sandstone was of freshwater origin there can be little doubt, and some geologists have even attempted to define the limits of the great freshwater lakes in which its beds were laid down. The Old Red Sandstone of Scotland. Murchison divided the Old Bed Sandstone into three groups, which he supposed were more or less contemporaneous with the three divisions of the Marine Devonian. But Sir A. Geikie regards the Old Bed Sandstone as constituting only two divisions. He considers the Old Bed Sand- stone to have been deposited in separate basins or lakes, which were five in number. 1. Lake Orcadie, north of the Grampian range, and including the Orkneys. 2. Lake Caledonia, occupying the central valley of Scotland between the Highlands to the north and the Silurian uplands to the south. It probably was prolonged across the Firth of Clyde into the north of Ireland. 8. Lake Cheviot, in the south-east of Scotland and north of England. 4. The Welsh Lake, bounded by the Silurian hills to the north and west. 5. Lake Lome, a district in the north of Argyllshire, on the flanks of the South-west Highlands. The two- fold division of the Old Bed is seen, according to this author, typically in Lake Caledonia. The Upper Old Bed, as he shows, merges gradually into the Lower Carboniferous strata above, and the Lower Old Bed passes conformably into the Silurian formation below ; but there is complete unconformity between the two series. He further notices the occurrence in Lanarkshire of Silurian fossils a Graptolite, Spir- orbis Lewisii, Sow., and Orthoce- ras dimidiatum, Sow. about 5,000 feet above the base of the Old Bed. He states: 'This interesting fact serves to indicate that though geo- graphical changes had elevated the Upper Silurian sea-floor, partly into land and partly into inland water- basins, the sea outside still contained CH. XXII.] OF SCOTLAND 389 an Upper Silurian fauna, which was ready on any favourable oppor- tunity to re-enter the tracts from which it had been excluded. ' The Middle and Lower Old Bed Sandstones attain a depth of deposits in the central district of Scotland of 20,000 feet, and the strata present, everywhere, evi- dences of shallow-water conditions. There are proofs that local eleva- tion occurred during the ages of general subsidence, which enabled the deposits to accumulate. In Lanarkshire the strata rest on Silurian rocks conformably, but on others unconformably. The strata, which are red, brown, chocolate- coloured, grey, and yellow, include sandstones, shales, flags, coarse con- glomerates, and occasional corn- stones and limestones. The grey flags and thin grey and olive shales and ' calm- stones ' are almost con- fined to Forfarshire, and in the north-east part of the basin are known as Arbroath flags. One of the most marked features is the occurrence of prodigious masses of interbedded volcanic rocks having a thickness of more than 6,000 feet in this central basin. As a rule, the deposits of this area are singularly unfossiliferous, though the Ar- broath flags have been proved to be rich in the remains of fish and Crustacea. In Forfarshire and Perthshire plant-remains are found. The Old Red Sandstone of the northern area contains the dark grey, bituminous schists and flag- stones whose fossil fish were so well described by Hugh Miller, and the calcareous flagstones of Caithness, resting on red sand- stones and conglomerates. These last repose upon the up-tilted Silu- rian rocks. Upper Old Red Sandstone. The highest beds of the series in Scotland, lying immediately below the Carboniferous formation, con- sist of yellow and red sandstones and conglomerates, well seen at Dura Den, near Cupar, in Fife, where, although the strata contain no mollusca, fish have been found abundantly, and have been referred to Holoptychius nobilissimus, Ag., -H. Andersoni, Ag., Pterichthys major, Ag., and to species of Glyp- topomus and other genera. The number of individuals of species at Dura Den, crowded pro- fusely through the pale sandstone, indicates, according to Sir A. Geikie, that the fish were killed suddenly and covered with sediment rapidly. Sir R. Murchison groups with this upper division of the Old Red of Scotland certain light-red and yellow sandstones and grits which occur in the northernmost part of the mainland and extend also into the Orkney and Shetland Islands. They contain Catamites and other plants which agree, generically, with Carboniferous forms, and overlie the Caithness flags uncon- formably. The Fish fauna of the Upper Old Red Sandstone numbers 25 species belonging to 15 genera. Sir A. Geikie notices that a band of marine limestone of Devonian age, lying in the heart of the Old Red in Arran, is crowded with or- dinary Carboniferous Limestone shells, such as Productus giganteus, Mart, sp., P. semireticulatus, Mart, sp. ; but none occur in the great series of sandstones overlying the limestone. These species do not reappear until we reach the lime- stones of the Carboniferous age, yet all these organisms must have been living before the deposition of the Arran limestone, and, of course, long prior to the formation of the Carboniferous limestone. Across the border districts, the sandstones and conglomerates of the Upper Old Red rest uncon- formably on Silurian rocks; and Old Red Sandstone with breccias and conglomerates appears under the Carboniferous formation along the flanks of the Cumberland and Westmoreland Hills, and in corre- sponding succession as far south as Flintshire and Anglesea. The Fish-remains, which have made the Old Red Sandstone so in- teresting, belong mainly, but not entirely, to the middle and lower divisions. While the Upper Old Red has 25 species, the Middle and Lower Old Red contain 85 species distributed among 36 genera. In this portion of the series there are 390 OLD RED SANDSTONE [CH. XXH. 12 species of Placoid fish, and all the fish, and the Ray, no skeletons are rest belong to the Ganoids. In ex- preserved; but fin-spines, called planation of this statement, it may Ichthyodorulites, and teeth occur, be said that Agassiz divided the On such remains the genera Devonian fish into two great orders, Onchus, Homacanthus, Ctena- namely, the Placoids and Ganoids. canthus, and Cosmacanthus, with Of the first of these, which at the many others occurring in the Old present time comprises the cartila- Red Sandstone, have been esta- ginous fish, like the Shark, the Dog- blished. The Old Red Sandstone of Southern Britain The grandest exhibitions, says Sir E. Murchison, of the Old Red Sandstone in England and Wales appear in the escarpments of the Black Mountains and in the Vans of Brecon and Caermar- then, the one 2,862, and the other 2,590 feet above the sea. The mass of red and brown sandstone in these mountains is estimated at not less than 10,000 feet, clearly intercalated between the Carboniferous and Silurian strata. No shells or corals have ever been found in the whole series, not even where the beds are calcareous, forming irregular courses of concre- tionary lumps called ' cornstones,' which may be described as mottled, red and green, earthy limestones. The fishes of this lowest English Old Red are Ceplialaspis and Pteraspis, speci- fically different from representatives of the same genera which occur in the uppermost Ludlow (Silurian) tilestones. Crusta- ceans also of the genus Eurypterus are met with. Besides the bodies called Parka decipiens, Flem. (figs. 532- 534, p. 379), there are found the spore-cases or floats of a lowly organised plant called Pachytheca. The Old Red Sandstone of Ireland. In Ireland, as in Scotland, the upper division of the Old Red Sandstone lies unconformably upon the lower, and in South Wales the upper beds overlap the lower strata, ' indicating,' wrote Sir A. Ramsay, ' great disturbance and denudation,' but not presenting any insuperable difficulty as to the freshwater origin of the strata. A dearth of calcareous matter over wide areas is character- istic of the Old Red Sandstone. This is, no doubt, in great part due to the absence of marine deposits and the scarcity of freshwater animals with calcareous shells. In the county of Cork, in Ireland, a similar yellow sand- stone occurs containing fish of genera characteristic of the Scotch Old Red Sandstone, as, for example, Coccosteus (a form represented by many species in the Old Red Sandstone and by one only in the Carboniferous group) and Glyptole})is, which is exclusively confined to the * Old Red.' In the same Irish sand- stone at Kiltorcan has been found an Anodonta or freshwater mussel, the only shell hitherto discovered in the Old Red Sand- CH. xxn.] OF ENGLAND AND IEELAND 391 stone of the British Isles (see fig. 549). In the same beds are found the Fern (fig. 562) and the Lepidodendron (fig. 563), and twelve other species of plants, some of which agree specifically with species from the Lower Carboniferous beds. This fact lends some support to the opinion, long ago advocated by Sir Kichard Griffith, that the yellow sandstone, in spite of its fish- remains, should be classed as Lower Carboniferous an opinion which is not generally adopted by geologists. Between the Mountain Limestone and the yellow sandstone in the South-west of Ireland, there intervenes a formation no less than 5,000 feet thick, called the ' Carboniferous slate ; ' and at the base of this, in some places, are local deposits, such as the Coomhola Grits, which appear to be beds of passage between the Carboniferous and Old Red Sandstone groups. The most trustworthy account ing descriptions of the Old Red of the Devonian strata of Devon- Sandstone of Scotland and its fossils shire and Cornwall is contained in are to be found in the writings of the papers of Mr. Ussher, of the the late Hugh Miller, and also in Geological Survey. Very interest- the works of Sir A. Geikie. CHAPTEE XXIII FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE NEWER, PALEOZOIC STRATA OF THE BRITISH ISLES The Devonian rocks of the Eifel of the Ardennes and Brittany of the Carinthian Alps, the Ibeyian peninsula, and Russia Carboniferous strata of France, Germany, and Russia Permian strata of Central Germany, the Alps, and the Ural Mountains Devonian strata of the United States, Canada, and the Arctic Regions Carboniferous strata of the United States Permian strata of Texas and Nebraska De- vonian, Carboniferous, and Permian of India and Australia. NEWER PALEOZOIC BOCKS OF EUROPE Devonian strata of the Eifel district. The Oldest of the Newer Palaeozoic strata, the Devonian or Eifelian, find their fullest representation in the district of Rhenish Prussia, where limestones and other strata abounding with beautiful, well-preserved fossils occur. The general classification adopted for these strata is as follows : Upper < Clymenia Limestone and Cypridina (Entornis) Shales. Goniatite Limestone. 392 NEWEK PALEOZOIC BOCKS [CH. XXIIIo / Stringocephalus beds. jCalceolabeds. lian ( Zone of Spirifera cuUrijugata, Kom. Coblenz Slates and Quartzite (Spirifer Sandstones). Hunsriick Slates. Sericitic Slates of the Taunus. Lower Eifelian The Upper and Middle Eifelian are composed of limestones with beds of shale, the strata yielding a great number of fossils. The Lower Eifelian consist of rocks, in places much altered, which attain a thickness of 10,000 feet ; these rocks being chiefly quartzites, felspathic sandstones (greywackes), and phyllites that sometimes assume almost a gneissic aspect. of limestone they contain shells similar to those of Devonshire, thus confirming, as Sir Roderick "has pointed out, the contemporaneous origin which had been previously assigned to formations exhibiting two very distinct mineral types in different parts of Britain. The calcareous and the arenaceous rocks of Russia, above alluded to, alternate in such a manner as to leave no doubt of their having been deposited in different parts of the same great period. "While in North- Western and Central Russia we find these alter- nations of the marine (Devonian) and of the freshwater (Old Red Sandstone) types,in the Ural district there is a completely marine series similar to that of the Eifel, but ex- hibiting many interesting diffe- rences in the order of succession of the beds and in the species of or- ganisms present in them. Carboniferous strata of Europe. The divisions of the Carboniferous rocks of France and Germany can be generally paral- leled with those of this country. In Germany, as in the South- West of England (Devonshire), we some- times find the richly coal-bearing beds replaced by masses of barren measures (the ' Culm facies ' of the Carboniferous rocks). When we pass to Russia, however, we find the marine facies (Fusulina lime- stones, &c.) forming the upper member of the series, and the pro- ductive Coal-measures below them, while in this country, as we have seen, the opposite is the case. Devonian of other parts of Western Europe. In the Ardennes to the west, and in Thuringia, the Harz, and Bohemia to the east, the Devonian strata are exhibited with divisions that can be approximately paralleled with those of the Eifel. The Devonian strata also appear in Brittany. The French geologists usually classify the Devonian in the following groups : Upper ( Famenian Devonian I Frasnian Middle ( Givetian Devonian 1 Eifelian / Coblenzian Lower Devonian I Taunusian I Gedinnian In the Carinthian Alps, strata of Lower, Middle, and Upper De- vonian age lie conformably upon the Upper Silurian rocks, and in Southern France, and in Spain and Portugal, slates, limestones, and sandstones of this age have been long known, and the formation as displayed in Asturias has now been fully described by M. Barrois. Devonian of Russia. The Devonian strata of Russia extend, according to Sir R. Murchison, over a region more spacious than the British Isles ; and it is re- markable that, where they consist of sandstone like the ' Old Red ' of Scotland and Central England, they are tenanted by fossil fishes often of the same species and still oftener of the same genera as the British, whereas when they consist CH. XXIII.] OF EUROPE AND AMERICA 393 The general succession of the Carboniferous strata in Russia is as follows : Limestones with Upper Carboniferous Lower Carboniferous Fusulina. Stage of Spirifera mosquensis, Fisch., at base. Limestones with Productive gi- ganteus, L. Productive coal- bearing strata. Stage of Produc- tus mesolobus, Phil., at base. Permian Strata of Europe. The main features of the British Permian are reproduced in Central Germany. There the upper mem- ber (the Zechstein) attains a con- siderable thickness, and in Thurin- gia it includes the Kupferschiefer, a bed containing fishes and other fossils mineralised by copper py- rites. This stratum was formerly largely worked as a copper ore. The Zechstein rests unconformably on the Rothliegende, and has a much more restricted development than the latter formation. In France, the Permian is only repre- sented by its lower member. In the Alpine district and in Sicily, however, we find the marine type of the Permian well exhibited in the Bellerophon and Fusulina, limestones. The same fauna is found in beds on the western slopes of the Ural Mountains (Artinsk stage of Karpinsky), and stretching through Asia Minor into Northern India. NEWER PALAEOZOIC STRATA OF AMERICA In the United States strata of Newer Palaeozoic age attain a grand development, but it is by no means easy to correlate the various divisions of this great mass of strata with the European Permian, Carboniferous, and Devonian systems respectively. A number of very distinct life- provinces are now recognised in this area the Acadian (including New England and the Eastern part of British America), the Appala- chian, the Mississippian, and the Michigan. In these several life- provinces while a general parallelism can be detected between the fossils of the successive divisions and those of the great divisions of the European Carboniferous there are a large number of species peculiar to the American continent, not a few which are restricted to one or other of these particular areas. In the western territories of North America, the Carboniferous strata resemble those of Russia and Eastern Asia, rather than those of Western Europe. estuary of the St. Lawrence, a mass of sandstone, conglomerate, and shale referable to this period occurs, rich in vegetable remains, together with some fish- spines. Far down in the sandstones of Gaspe Dr. Dawson found in 1869 an entire spe- cimen of the genus Cephalaspis, a form very characteristic, as we have already seen, of the Scotch Lower Old Red Sandstone. Some of the sandstones are ripple-marked; and towards the upper part of the whole series a thin seam of coal has been observed, measuring, together with some associated carbonaceous shale, about three inches in thickness. It rests 011 an underclay in which are Devonian strata in the United States and Canada. Between the Carboniferous and the Silurian strata in the United States and Canada, there inter- venes a great series of formations referable to the Devonian group, comprising some marine strata abounding in shells and corals, and others of shallow- water and littoral origin, in which terrestrial plants abound. The fossils, both of the deep and shallow-water strata, are very analogous to those of Europe, the species being in some cases the same. In Eastern Canada Sir W. Logan has pointed out that in the peninsula of Gaspe, south of the 394 DEVONIAN FLOEAS [CH. XXIII. the roots of Psilophyton (see fig. 565). At many other levels root- lets of this same plant have been shown, by Principal Dawson, to penetrate the clays, and to play the same part as the rootlets of Stig- maria in the coal formation. We had already learnt from the works of Goppert, Unger, andBronn, that the European plants of the De- vonian epoch resemble generically, with few exceptions, those already known as Carboniferous ; and Dr. Dawson, in 1859, enumerated 82 genera and 69 species which he had then obtained from the State of New York and Canada. A perusal of his catalogue, comprising Coni- ferce, Sigittarice, Calamites, Aste- rophyllites, Lepidodendra, and from beneath the Carboniferous on the borders of Pennsylvania and New York, where both formations are of great thickness. The number of American De- vonian plants has now been raised by Dr. Dawson and others to 160, to which we may add about 80 from the European flora of the same age, so that already the vege- tation of this period is beginning to be nearly half as rich as that of the Coal-measures which have been studied for so much longer a time and over so much wider an area. The Psilophyton^ above alluded to, is very widely distributed in Canada. Its remains have been traced through all the members of the De- vonian series in America, and Dr Fig. 668. SJoggins ShoidieR. Cotequid Upper Silurian- Diagram showing the curvature and supposed denudation of the Carboniferous strata in Nova Scotia. A. Anticlinal axis of Minudie. B. Synclinal of Shoulie River. 1. Coal-nieasures. 2. Lower Carboniferous. ferns of the genera Cyclopteris, Neuropteris, Sphenopteris, and others, together with fruits, such as Cardiocarpum and Trigonocar- pum, might dispose geologists to believe that they were presented with a list of Carboniferous fossils, the difference of the species from those of the Coal-measures, and even a slight admixture of genera unknown in Europe, being natu- rally ascribed to geographical dis- tribution and the distance of the New from the Old World. But fortunately the Coal formation is fully developed on the other side of the Atlantic, and is singularly like that of Europe, both litholo- gically and in the species of its fossil plants. There is also the most unequivocal evidence of rela- tive age afforded by superposition, for the Devonian strata in the United States are seen to crop out Dawson has lately recognised it in specimens of Old Red Sandstone from the North of Scotland. It is a remarkable result of the recent examination of the fossil flora of Bear Island, L.t. 74 30 N., that Professor Heer has described as occurring in that part of the Arctic region (neai'ly twenty-six degrees to the north of the Irish locality) a flora agreeing in several of its species with that of the yellow sandstones of Ireland. This Bear Island flora is believed by Professor Heer to comprise species of plants some of which ascend even to the higher stages of the European Carboniferous formation, or as high as the Mountain Lime- stone and Millstone Grit. Palaeon- tologists have long maintained that the same species which have a wide range in space are also the most persistent in time, which may OH. xxm.] TABLE OF NEWER PALJEOZOIC STRATA 395 I Limestones and Kansas with g Cephalopod schis Sandstones and Nebraska Fusulina beds Upper barren Co I 1 S i 1 11 y S | o g 2g = o "g !| 'i I ll IS 6 3l s,lsi Jil . 11 i | S JO t-1 S I !~ja | 3*11 I ril | | g I l|l 1 111 r O >-i 03 U2 PnCO II ! II "^ (D ' >~o so s OQC3 02000" sa, J. Hook. Some of the fish remains are of the placoid order, and may be re- ferred to the genus Onchus, to which the spine (fig. 590) belongs. The minute scales (fig. 591) may. also belong to a placoid fish. The fragments of another predaceous genus, Plectrodus mirabilis, Ag. (fig. 592), have also been detected, together with some specimens of Pteraspis ludensis, Salt. As is 1. Xiudlow Formation. This has been subdivided into two Eiirts the Upper Ludlow and the ower Ludlow. Each of these may be distinguished near the town of Ludlow, and at other places in Shropshire and Herefordshire, by peculiar organic remains; but out of 392 species found in the Ludlow formation as a whole, not more than 5 per cent, are common to the overlying Devonian, and nearly all of those are fish and Crustacea. On the other hand, 129 of these species occur in the underlying Wenlock deposits. a. Upper Ludlow, Downton Sandstone. At the top of this sub- division there occur beds of fine- grained yellowish sandstone and hard reddish grits which were for- merly referred by Sir R. Murchison to the Old Red Sandstone, under the name of ' Tilestones.' In mine-* ral character this group forms a transition from the Silurian to the Old Red Sandstone ; but it is now ascertained that the fossils agree in great part specifically, and in gene- ral character entirely, with those of the underlying Upper Ludlow rocks, many passing upwards. Among these are Orthoceras bullatum, Sow., Platyschisma helicites, Sow. sp., Bellerophon trilobatus, Sow., Chonetes latus, Sow., &c. Crustacea of the genera Pterygotus, Eury- pterus,&nd.Stylonurusa l remetwit\\, and Fish Cephalaspis, Pteraspis, Svapliaspis, Aucfyenaspis, and 406 LOWER LUDLOW [CH. XXIV. usual in bone-beds, the teeth and bones are, for the most part, frag- mentary and rolled. Associated with these fish defences or Ichthyodoru- lites, and closely resembling them, are numerous prongs or tail spines of large phyllopod crustaceans which have been, and still are fre- quently, mistaken for the dorsal spines of fish. Grey Sandstone and Mudstone, (f-c. The next subdivision of the Upper Ludlow consists of grey, cal- careous sandstone, or very com- monly a micaceous rock, decompos- ing into soft mud, and contains, besides the shells mentioned at p. 404, Lingula cornea, Sow., Orthis orbicularis, Sow., a round variety of 0. elegantula, Dalm. (fig. 576), Modiolopsis platyphylla, Salt., Grammy sia cingulata, His. sp., all characteristic of the Upper Lud - low. The lowest or mudstone beds contain Rhynchonella navicula, Sow. (fig. 580), which is common to this bed and the Lower Ludlow. Usually, in Palaeozoic strata older than the Coal, the Brachiopoda greatly outnumber the Lamelli- branchiata. But it is remarkable that in these Upper Ludlow rocks the Lamellibranchiata outnumber the Brachiopoda, there being 56 species of the first and only 27 of the last group. Amongst the genera represented are Avicula and Pteri- nea, Cardiola, Ctenodonta (sub- genus of Nucula), Orthonota, Mo- diolopsis, and Palcearca. Some of the Upper Ludlow sandstones are ripple-marked, thus affording evidence of gradual depo- sition; and the same may be said of the accompanying fine argilla- ceous shales, which are of great thickness, and have been provin- cially named ' mudstones.' In some of these shales, stems of Crinoidea are found in an erect position, having evidently become fossil on the spots where they grew at the bottom of the sea. The facility with which these rocks weather into mud, proves that, notwith- standing their antiquity, they have not been subjected to any great chemical changes, but are nearly in the state in which they were originally deposited. b. Lower Ludlow beds. The chief mass of this formation consists of a dark grey argillaceous shale with calcareous concretions, having a maximum thickness of 1,000 feet. In some places, and especially at Aymestry in Hereford- shire, a subcrystalline and argilla- ceous limestone, sometimes 50 feet thick, overlies the shale, and ap- pears rising above the denuded Lower Ludlow shales. It is not very continuous, so that the shales of the Lower and the strata of the Upper Ludlow come together around it. Sir R. Murchison classed this Aymestry limestone as holding an intermediate position between the Upper and Lower Ludlow. It is distinguished by the abundance of Pentamerus Knightii, Sow. (fig. 573), also found in the Wenlock limestone and shale. This genus of Brachiopoda is exclusively Palaeo- zoic. The name was derived from TreWe, pente, five, and /xe'pos, meros, a part, because both valves are divided by a central septum, making four chambers, and in one valve the septum itself contains a small chamber, making five. The size of these septa is enormous compared with those of any other Brachiopod shell; and they must nearly have divided the animal into two equal halves ; but they are, nevertheless, of the same nature as the septa or plates which are found in the inte- rior of Spirifera, Uncites, and many other shells of this order. Murchison and De Verneuil dis- covered this species dispersed in myriads through a white limestone of Silurian age on the banks of the Is, on the eastern flank of the Urals in Russia, and a similar species is frequent in Sweden. Three common shells in the Ay- mestry limestone are Lingula Lewisii, Sow. (fig. 581); RhyncJio- nella Wilsoni, Sow. (fig. 579), which is also common to the Lower Ludlow and Wenlock limestones; Atrypa reticularis, L. (fig, 578), which has a very wide range, being found in every part of the Silurian system, and even passes up into the Middle Devonian series. The Aymestry Limestone con- tains many shells, especially bra- CH. XXIV.] WENLOCK BEDS 407 chiopoda, corals, trilobites, and other fossils, amounting in the whole to 84 species, all except three or four being common to the beds either above or below. The Lower Ludlow Shale con- tains many large Cephalopoda not known in newer rocks, such as Phragmoceras and Lituites. (See figs. 584, 585.) The latter is partly straight and partly convoluted in a very flat spire. Orthoceras ludense, J. Sow. (fig. 583) also occurs. A species of Graptolite, Mono- graptus priodon, Gein. (fig. 567, p. 398), occurs plentifully in the Lower Ludlow. The Graptolites will be noticed further on, but they became extinct during the Ludlow age. Star-fish, as Sir R. Murchison pointed out, are by no means rare in the Lower Ludlow rock. These fossils, of which 6 extinct genera are now known in the Ludlow series, represented by 13 species, remind us of various living forms of the orders Asteroidea and Ophiu- roidea now found in our British seas, but their anatomical details differ greatly. The two great orders of the class Crustacea in the Ludlow rocks are the Merostomata and the Phyllo- poda, and they predominate over the Trilobita, which were waning as a great group, and were destined to be -jme gradually extinct during tlyj Devonian, Carboniferous, and Permian periods. Of all the genera of Trilobita so common in the Silurian and Cam- brian formations, only two, Homa- lonotus and Phacops, survived the changes which introduced the Devonian formation. Six of the species of Merostomata pass up into the OJd Red Sandstone. Oldest known fossil fish. Until 1859 there was no fossil fish known older than the bone-bed of the Upper Ludlow ; but Cyathaspis (Pteraspis) ludensis, Salt., has been found in Lower Ludlow shale at Church Hill, near Leintwardine, in Shropshire, by the late J. E. Lee, of Caerleon. These fish were long regarded as the oldest representatives of the vertebrate series, but Dr. Lindstrom has recently found Cyathaspis in the Gothland Limestone of Sweden, which is of Wenlock age, while, in America, Walcott has described fish-remains as occurring in strata that are believed by him to be of Ordovician age. 2. Wenloclc Formation. We next come to the Wenlock for- mation, which has been divided into a, Wenlock limestone and Wenlock shale ; and b, Woolhope limestone, Tarannon shale, and Denbighshire grits. a. Wenlock Limestone. This limestone, otherwise well known to collectors by the name of the Dudley Limestone, forms a continuous ridge in Shropshire, ranging for about 20 miles from S.W. to N.E., about a mile distant from the nearly parallel escarpment of the Aymes- try limestone. This ridgy pro- minence is due to the solidity of the rock, and to the softness of the shales above and below it. Near Wenlock it consists of thick masses of grey subcrystalline limestone, replete with corals, Encrinites, and Trilobites. It is essentially of a concretionary nature ; and the con- cretions, termed ' ball-stones ' in Shropshire, are often very large, even 80 feet in diameter. They are composed chiefly of carbonate of lime, the surrounding rock being more or less argillaceous. Some- times this limestone is oolitic. All the limestones of the Upper Silurian form great lenticular masses, and thin out so as to have their space occupied by the shaly strata of the lower and upper divisions of the same great age. Among the corals in which this formation is so rich, 7G species being known, the ' Chain-coral,' Haly- sites catenulatus, L. sp. (fig. 569), may be pointed out as one very easily recognised, and widely spread in Europe, ranging through all parts of the Silurian and Ordo- vician, from the Ludlow to near the bottom of the Llandeilo rocks. Another coral, the Favosites goth- landica, Lam. (fig. 570), is met with in profusion, in large hemi- spherical masses, which break up into columnar and prismatic frag- ments. Another common form in 408 WENLOCK AND WOOLHOPE LIMESTONES [OK. xxiv. the Wenlock limestone is the Omphyma subturbinata (fig. 568), which, like many of its modern companions, reminds us of some cup- corals ; but all the Silurian genera belong to the Palaeozoic type before mentioned (p. 351). Among the numerous Crinoidea, several peculiar species of Cyatho- crinus, Crotalocrinus, &c., con- tributed their dismembered cal- careous stems, arms, and cups towards the composition of the Wenlock limestone. Of Cystoidea there are a very few remarkable forms, most of them peculiar to the Silurian system ; as, for exam- ple, the Pseudocrinites, which was furnished with pinnated fixed arms, as represented in the figure (fig. 571, p. 399). The Brachiopoda preponderated over most of the other groups, no less than 22 genera and 101 species being found. Atrypa Barrandei, David., Or this (equivalvis, David., Siphonotreta a/nglica, Mor., are special forms; about 11 species pass up into the Aymestry limestone. Examples are Atrypa reticularis, L. sp., and Orthis elegantula, Dalm. The Crustacea are represented by Eurypterida, which appear for the first time, including the ge- nera Pterygotus and Eurypterus, and by Ostracoda and Trilobites. The Trilobite Calymene Blumen- bachii, Brong.' (fig. 586), is common, and it ranges from the Llandeilo group to near the top of the Silurian. It is often found coiled up like the common wood-louse, and this is so usual a circumstance among certain genera of Trilobites as to lead us to conclude that they must have habitually resorted to this mode of protecting themselves when alarmed. Sphcerexochus mines, Beyr. (fig. 588), is almost globular when rolled up, the fore- head or glabella of this species being extremely inflated. The other common species are Encrinurus punctatus, Emmr. sp., and Phacops caudattts, Brong. (fig. 587), which is conspicuous for its large size and flattened form. In the genus Homalonotus the tripartite division of the dorsal crust is almost lost (see fig. 589) ; it is characteristic of this division of the Silurian series. Wenlock shale. Fine grey and black shales, with most of the fossils common to the overlying limestone. In the Malvern district it is a mass of finely levigated argillaceous matter, attaining, according to Pro- fessor Phillips, a thickness of 640 feet ; but it is sometimes more than 1,OCO feet thick in Wales, and is worked for flagstones and slates. The prevailing fossils, besides corals and Trilobites, and some Crinoidea, are several small species of Orthis, Atrypa, and Rhynchonella, and numerous thin-shelled species of Orthoceras. About six species of Graptolites occur in this shale, Graptolithus Flemingii, Salt., being peculiar, whilst Monograptus priodon, Gein. (fig. 567), ranges through into the Ludlow group. b. Woolhope limestone under- lies the Wenlock shale, and consists of grey shales, with nodular lime- stone. It is well seen in the valley of Woolhope, and at Malvern there is much shale beneath. The fos- sils of the Woolhope limestone are principally Crustacea, all of the Trilobite group, and Brachiopoda. Examples of the fossils are Pha- cops caudatus, Brong. (fig. 587), Homalonotus delphinoccphalus, Green sp. (fig. 589), Strophomena imbrex,Pa,nd.., Rhynchonella Wil- soni, Sow. (fig. 579), and Euca- lyptocrinits polydactylu.s, M'Coy. This limestone is in large lenticular masses, and is overlapped at its edges by the underlying shales which then join continuously with the Wenlocks above. There is a very persistent set of beds of fine light grey or blue shales, (termed 'paste-rock '), which lie on the Upper Llan.dovery strata over a considerable tract of country from the Conway into Caermarthenshire, just as the Woolhope limestone covers these last-mentioned strata in Shropshire and Herefordshire. These Tarannon shales are 1,000 to 1,500 feet thick in places, and con- tain numerous species of Grapto- lites, corals of the genera Favosites and Cyathophyllum: one of the Crinoidea, Actinocrimts pulcher, CH. xxiv.] LLANDOVERY OR MAY-HILL BEDS 409 Salt., which passes up into the Lower Ludlow, and the Brachiopod Lingula Symondsii, Salt. The Tarannon shales are covered, in Denbighshire, by grits and sand- stones at least 8,000 feet thick, which pass into hard shales of pro- bably Wenlock age. These Den- bighshire grits form mountain ranges in North and South Wales, and produce a very sterile soil. They were formed probably during the time of the accumulation of the Wenlock deposits. It is inte- resting to note that these grits do not pass up into the base of the Old Red Sandstone, but lie unconform- ably below it, indicating great ter- restrial movements before its depo- sition. This is very different from the state of things sixty miles off, where the Old Eed rests conform- ably on the underlying Silurian. Dr. Hicks found vegetable re- mains in the Denbighshire grits, such as the PacJnjtheca already noticed (p. 405), and a remarkable marine Alga NematopTiycus, which probably resembled the great branching Lessonia of the present day in its habit. Many of these great marine Algae of the existing ocean measure 30 feet in length and a foot in diameter. The marine fossils include sponges (Cliona prisca, M'Coy) ; corals (Favosites aspera, D'Orb., Syringopora serpens, L. sp.) ; there are 19 species of Brachiopoda, all common to the Wenlock limestone; and Cephalopoda, of the genera Orthoceras, Phragmoceras, and Cyrtoceras. 3. Xilandovery group Upper Iilandovery rocks. The succession of these strata has been noticed, and it must be re- membered that the Wenlock group rests conformably on the Upper Llandovery beds, which in their turn cover the worn and denuded surfaces of the disturbed, curved, and faulted underlying rocks to which they are unconformable. Upper Llandovery rocks, named May-Hill Sandstones by Sedgwick after the locality in Gloucestershire, where they are so well displayed, appear on the coast of Pembroke at Marloes Bay. They range across South Wales until they are over- lapped by the Old Red Sandstone and emerge again in Caermarthen- shire, and can be traced north-east- wards as a narrow strip at the base of the Silurian series, from a few feet to 1,000 feet or more in thick- ness, as far as the Longmynd. where, as a conglomerate, they wrap round that ancient pre- Cambrian ridge and disappear. In the course of this long tract they pass successively and unconform- ably over Lower Llandovery, Cara- doc, Llandeilo, and pre-Cambrian rocks. They consist of brownish and yellowish sandstones with cal- careous nodules, having sometimes a conglomerate at the base derived from the waste of older rocks. The fauna of the Upper Llan- dovery rocks consists of 240 species, and there are only 91 of these which do not pass up into the Wenlock group, so that the physical uncon- formity of the two groups is ac- companied by no paleeontological break of importance. The Lamelli- branchiata become of importance in this fauna, as do the Gastropoda of the genera Holopella, Acroculia, Bhaphistoma, and Turbo. The Brachiopoda number in species more than double those of any other class, there being 65 species, including Pentamerus oblongus, Sow. (fig. 574), Stricklandinia lirata, Sow. sp. (fig. 575), S. lens, Sow. sp. (fig. 574), Orthis calli- gramma, Dalm., 0. elegantula, Dalm., Strophomena compressa, Sow. Among the corals Favosites and Heliolites are found. The first Echinoid occurs, Palcechinus Phillipsice, Forbes, and its plates abut one against the other and do not overlap. Tentaculites is found (fig. 582), and also Cornulites. Pentamerus oblongus, Sow., accompanied by Stricklandinia lirata, Sow. sp., have a wide geo- graphical range, being also met with in the same part of the Silurian series in Russia and the United States. The Trilobites are of the genera Illcenus, Oalymene, Encri- nurus and Phacops, Xiower Iilandovery rocks. The Upper Llandovery strata rest unconformably OH the Lower, and 410 LOWER LLANDOVERY [CH. XXV. there is a clear physical break; but the palseontological break is not of corresponding importance to it. The hard slaty rocks and con- glomerates, from 600 to 1,000 feet in thickness, of the Lower Llan- dovery group contain a fauna of 68 genera and 204 species. More than one-half (104) of the species pass up into the Upper Llandovery strata. Etheridge explains that the lapse of time which occurred between the disturbance of the Lower Llando- very rocks and the deposition of the Upper, was not of sufficient duration to cause the extinction or migration of the older fauna or the introduction of a perfectly new one. The physical change in all probability was not very widely felt. The Brachiopoda are numerous in the Lower Llandovery rocks, and the genera Pentamerus and Stricklandinia appear for the first time ; the species with the most numerous individuals being Strick- landinia lens, Sow. sp., S. lirata, Sow. sp., and Pentamerus oblon- gus, Sow. sp., P. undatus, Sow. Sir Eoderick Murchison's ' Siluria ' may still be referred to by students as containing the fullest account of the strata of this age in the typical area; see also ' The Jreology of South Shropshire,' by Prof. Lapworth and Mr. W. W. Watts, Proc. Geol. Ass. 1894. The works of M. Barrande contain descriptions of the numerous and sp., especially the first named . The genus Murchisonia occurs among the Gastropoda, and Bellerophon with Conularia (a Pteropod) is also represented. The Trilobites are remarkable, because no less than 18 species pass into this group of strata from lower rocks, and 10 pass upwards into the Upper Llandovery group ; and this passage of forms is noticed also in the Actinozoa, but in a greater degree. The Graptolites are very rare in the English Lower Llan- dovery strata. It appears that the Lower Llan- dovery strata, having a fauna, the half of which lived on, in the Upper Llandovery rocks, and 105 species of which are also found fossil in the underlying Caradoc or Bala strata, are occasionally unconformable to these last. The importance of this palseontological continuity, asso- ciated with unconformity, may be estimated by the fact that the underlying Caradoc formation con- tains 614 species, and thus one- sixth of that fauna passes upwards into the Silurian. well-preserved fossils of the period which are found in Bohemia. Lindstrb'm and other authors have described the Scandinavian strata and fossils, while a good summary of our knowledge of the Silurian fossils of North America will be found in Dana's ' Manual of Geo- logy,' and in the Correlation papers of the U.S. Geological Survey. CHAPTER XXV THE ORDOVICIAN SYSTEM Classification of Ordovician strata Characteristics of the marine Fauna Foraminifera Graptolites Echinodermata Brachiopoda Gastro- poda Cephalopoda Worms Trilobita and their organisation Bala or Caradoc strata Llandeilo beds Arenig beds or Stiper-stones Ordovician strata of the Lake District Ordovician strata of Scotland. Nomenclature and Classification of the Ordovician strata. Mnch confusion in nomenclature with respect to the system of strata containing Barrande's second fauna has arisen from the unfortunate misunderstanding between Sedgwick and CH. xxv.J ORDOVICIAN NOMENCLATURE 411 Murchison. The followers of Mnrchison, with the powerful support of the Geological Survey, have insisted on calling the system ' the Lower Silurian,' while the supporters of Sedgwick, comprising many of his pupils at Cambridge, have named it ' Upper Cambrian.' Attempts at a compromise, like the proposal to call the period Cambro- Silurian or Siluro- Cambrian, have not met with much success ; and hence those geologists who think that general convenience should be the main considera- tion in framing a classification and nomenclature, have gladly welcomed the suggestion of Professor Lapworth to call the dis- puted strata ' Ordovician.' This name is derived from the ancient British tribe (Ordovices) which inhabited the district where the strata are best developed ; and hence the term may be regarded as strictly parallel in its derivation with the names Cambrian and Silurian. D'Orbigny called this system of strata Silurian, distinguish- ing the true Silurian as Murchisonian, while de Lapparent applied to it the name of Armorican, subsequently withdrawing the name in favour of Lap worth's suggestion. The dispute concerning names between Sedgwick and Murchison was not confined to the three great systems them- selves, but extended to the nomenclature of the smaller divisions of the strata containing the second fauna. Most geologists now recognise a threefold division of the Ordovician, and the names applied to them by Sedgwick and Murchison respectively were as follows : Sedgwick. Murchison. Bala. Caradoc. Llandeilo. Upper Llandeilo. Arenig. . Lower Llandeilo. Characteristics of the Ordovician Fauna and Flora. Although many of the forms regarded as seaweeds in the Ordo- vician and underlying Cambrian are now recognised as being the trails and markings of animals, there are some examples of true algae. Among these must probably be classed the curious Calcareous algae (Girvanella, &c.) which sometimes make up a great portion of the limestone beds, and the other obscure organisms concerned in producing rocks of oolitic and pisolitic structure. Among the Protozoa, we have many Foraminifera, like those which have been instrumental in the formation of the Glauconite sand and Glauconite limestone of Russia. In recent years thick and important siliceous deposits, made up of Radiolarians, have been found in Scotland and other countries. As a rule, however. 412 OEDOVICIAN GEAPTOLITES [CH. XXV. neither the Foraminifera nor the Radiolaria are so well pre- served as to enable us to make accurate comparisons between the early forms of these lowly organisms and their descendants of later periods. Siliceous sponges, like Astylospongia and Fig. 593. Fig. 594. Didymograptus Murchi.wni ,' Beck, \. Llandeilo flags, Wales. Fig. 595. Didymograptus geminus, Hisinger, sp. Sweden. Fig. 596, Diplograptus pristis, His., nat. size. Llandeilo beds, Waterford. Fig. 597. Diplograptus folium, His. ( Plvylloyrapt us.) Dumfriesshire ; Sweden. Llaudeilo flags. Raslrites peregrinus, Barrande, nat. size. Scotland ; Bohemia ; Saxony.; L'.andeilo flags. Echinosphcerites balticus, Eich., nat. size. (Of the family Cystoidea.) a. Mouth. I. Point of attachment of stem. Lower Silurian, S. and N. Wales. Aulocopium, are not rare, and perhaps belong to synthetic forms combining characters which in later times distinguished Hexactinellid and Lithistid forms. The Graptolites of the Ordovician are very numerous and interesting, and include great numbers of branched and double forms like Didymograptus (figs. 593, 594), Diplograptus (fig. CH. XXV.] ORDOVICIAN BRACHIOPODA 413 Fig. 539. 595), Phyllograptus (fig. 596), Tetragraptus, Dichograptus, &c. ; with a few peculiar simple forms like Rastrites (fig. 597), and the complex, netlike genus Retiolites (Note V, p. 607). Stromatoporoidea make their appearance in the Ordovician, though they attain their maximum de- velopment in the Silurian and Devonian. True Corals are abun- dant in some of the lime- stone beds, but do not occur in the same profu- sion as in the Silurian. The coral-like structures are Tetracoralla (Eugosa), Tabulate forms (Monticuliporidae), and Hydrocorallinae. Palwaster asperrimus, Salt. Caradoc. Welshpool. Fig. 600. Fig. 601. OrtMs tricenaria, Conrad. New York ; Canada. \ nat. size. Fig. 603. Orthis vespertilio, Sow. Shropshire ; N. and S. Wales. \ nat. size. Slrophomena grandis. Sow., nat. size. Caradoc Beds, Horderley, Shrop- shire; and Coniston, Lancashire. Fig. 604. Siphonotreta unguiculata, Eichwald, nat. size. From the lowest Silurian Sandstone, ' Obolus grits,' of Petersburg. a. Outside of perforated valve. b. Interior of same, showing the ter- mination of the foramen within. (Davidson.) Obolus Avollinis, Eichwald, " nat. size. From the same locality. a. Interior of the large or ventral valve. b. Exterior of the upper (dorsal) valve. (Davidson, ' Palason- tograph. Monog.') Echinoderms are principally represented by Cystoidca, like Echinosphcerites (fig. 598), a group which attained its maximum development at this period. The Crinoids, however, are rare in the Ordovician, while some starfish, like Palceaster, are found. 414 OKDOVICIAN MOLLUSCA CH. XXV. Fig. 605. It is by the abundance and variety of its Brachiopod fauna that the Ordovician system is especially distinguished. Many species of Ortliis (figs. 600, 601), Leptcena, and StropTiomena (fig. 602), occur in it, with certain genera not found in any other system, like Porambonites, Orthisina, and Platystrophia ; and others found only in the Ordovician of Kussia and North America, like Obolus (fig. 604), Siphonotreta (fig. 603), Trimerella, &c. In striking contrast with this abundant and remarkable Brachiopod fauna, the Lamellibranchiata of the Ordovician are A fossil characteristic of 3 -, the Trenton Limestone. rare and comparatively unimportant, and no sinupalliate forms are found in it. The Gastropoda include Murchisonia (fig. 605), and the remarkable form Maclurea. Pteropods like Theca, Tenta- culites, and Conularia are abundant. Fig. 606. Afurchisonia gracilis, Hall. Nat. size. Fig. 607. Urthoceras (Endow rs) duplex, Wahlenh. Russia aud Sweden. (From Murchison's ' Siluria.') a. Lateral siplmncle laid bare by the removal of a portion of the chambered shell. 6. Continuation of the same seen in a transverse section of the shell. The Cephalopods include many forms of Nautilus, Orthoceras, and other genera of the Nautiloidea like those of the Silurian. The abundance and variety of the Vermes of this period are testified to by the numerous tracks and burrows formed by these organisms (see fig. 607). The Trilobita of the Ordo- vician are only inferior in numbers and variety to the Bra- chiopoda ; among the most abun- dant genera are Asaphus (fig. 608), Ogygia (fig. 609), Trinu- cleus (figs. 610, 611), Lichas, Acidaspis, &c. Ostracods and Phyllopods are the only Arthropods besides the Trilobites which are present in any abundance in the Ordovician. Arenicoliles lincaris, Hall. Arenig beds, Stiper-stones. . Parting between the beds, o* planes of bedding. en. xxv.] ORDOVICIAN TRILOBITES 415 Traces of Ganoid fish are said to have been found in the Ordovician of Colorado, but it is doubtful if the beds have really the age which has been assigned to them. Fig. 608. Fig. 609. Asaphus tyrannus, Murch., J. Llandeilo ; Bishop's Castle, &c. Offiigia flnchii, Burra., Syn. Asaphus Buchii, Brong. Builth, Radnorshire ; Llandeilo, Caermarthenshire. Fig. 611. Young individuals of Triiiudsuscon- centricus, Eaton. a. Youngest state. Natural size and magnified ; the body rings not at all developed. b. A little older. One thorax joint. c. Still more advancpd. Three tho- rax joints. The fourth, fifth, and sixth segment are succes- sively produced, probably each time tire animal moulted its Trinuclcns concent ricus, Eaton, nat. size. crust. Syn. T. Caractaci, Murch. Ireland ; Wales ; Shropshire ; N. America Bohemia. British representatives of the Ordovician System. The Ordovician consists of three members which following the nomen- clature of Sedgwick are called Bala, Llandeilo, and Arenig ; while the names adopted by Murchison and the Geological Survey for the same divisions were Caradoc, Upper Llandeilo flags, and Lower Llandeilo flags. 1. The Bala and Caradoc group. The Caradoc Sandstone was so named by Murchison, from Caer Caradoc in Shropshire. It consists of shelly sandstones of great thickness, sometimes contain- 416 BALA OR CAKADOC [cif. XXV., ing much calcareous matter. In the Bala district there is an upper and a lower limestone, with an intermediate series of sandy and slaty strata; the lower limestone has a great extension in North Wales. These very fossiliferous strata contain a large number of Brachio- poda, Strophomena grandis, Sow. sp. (fig. 002), 8. ddtoidea, Cour. sp., Chonetes plicata, Sow. sp., and Rhynchonella nasuta, M'Coy, being amongst the characteristic species, whilst Orthis vespertilio, Salt. (fig. 601), and Orthis tricena- ria, Conrad (fig. 600), are common to this group and the Llandovery. There are no less than 109 species of Brachiopoda in the group, and they outnumber the Lamellibran- chiata with 76 species, as is almost always the case in the Ordovician rocks of every country. Their pro- portional numbers can by no means be explained by supposing them to have inhabited seas of great depth, for the contrast between the Palaeo- zoic and the present state of things has not been essentially altered by the late discoveries made in our deep-sea dredgings. We find the living Brachiopoda so rare as to form a very small fraction of the whole bivalve fauna ; whereas, in the Ordovician rocks, where the Brachiopoda reach their maximum, they are greatly in excess of the Lamellibranchiata. There may, indeed, be said to be a continued decrease of the proportional number of Brachio- poda, as compared with that of the Lamellibranchiata, in proceeding from older to newer rocks. The Gastropoda are very nu- merous, and amongst the 53 species only two are known in lower rocks, and 10 pass upwards, so that a cha- racteristic series of 41 species exists. They indicate, as do the Lamelli- branchiata, shallow-water condi- tions, some of the genera being Mur- chisonia, Holopella, Rhaphistoma, and Turbo. Pteropoda and Hete- ropoda are found, and there are 47 species of Cephalopoda, of which 39 are peculiar to the groups of strata. Lituites, Orthoceras, and Cyrto- ceras are common genera The Crustacea are the most conspicuous fossils in this group of strata, and no less than 123 specif :s of Trilobita have been discovered. Only 15 pass upwards, and amongst them are Calymene Blumenba- chii, Brong., Encrinurus puncta- tus, Emmr. sp., Lichas laxatus, M'Coy, and Phacops caudatus, Briinn. Some of the most charac- teristic genera are Harpes, Sal- teria, and Cyclopyge, Certain of the Trilobita found in the Caradoc group lived on into the Silurian, and Trinucleus concentricus, Ea- ton (fig. 611), Calymene Blumen- bachii, Brong., and Ampyx rostra- tus, Sars., are examples of this. The Trilobite order of Crustacea has a great number of genera which are found in the Palaeozoic rocks, from the Permian downwards. It was at its maximum of numbers in the Caradoc age, and diminished rapidly in the Devonian, becoming rare and extinct in the Carboniferous and Permian formations. The tri- lobed body, with a cephalic or head- shield bearing a pair of eyes, with body rings and a hinder shield or pygidium, are well seen in most of the order. Some have the angles of the cephalic shield prolonged into long spines, like Trinucleus (fig. 611). Burmeister was of opinion that the Trilobita underwent a series of transformations analogous to those of living Crustaceans. Barrande has obtained complete proofs of these metamorphoses, and he has been able in more than twenty species to trace the different stages of growth from the young state, just after its escape from the egg, to the adult form. He has followed some of them from a point in which they show no eyes, no joints, or body rings, and no distinct tail, up to the complete form with the full number of segments. This change is brought about before the animal has at- tained a tenth part of its full dimensions, and hence such minute and delicate specimens are rarely met with. Some of his figures of the metamorphoses of the common Trinucleus are copied (figs. 610, 611, a-c, p. 415), and of Sao (fig. 616, p. 423). Until recent years it was omy CH. XXV.] LLANDEILO 417 the upper surface of the Trilobites that was known to naturalists. But the studies of Walcott and Becher have shown that the Trilo- bita were furnished with numerous delicate, jointed, swimming-appen- dages on their under side, with antennae attached to their head- shields. The Rugose Corals, the Alcyo- naria and Hydrocorallina, were well represented in the Bala beds by Cyathophyllum, Heliolites, and Halysites. The Echinodermata were represented by the Cystoidea (fig. 598), and 23 species are cha- racteristic ; one also (EcliinospTie- rites arachnoidea, Forbes) passes up into the Llandovery group. The Cystoidea were stalked as a rule, and became extinct during the Devonian age. It is interest- ing to note that the Blastoidea, of which Pentremites is a well- known Carboniferous genus, began to nourish when the Cystoidea began to seriously diminish in numbers, and gained their maxi- mum development during the Car- boniferous age. All were Palaeo- zoic. Crinoidea existed in the Caradoc age, and also Asteroidea of the genera Palceaster (fig. 599) and Stenaster. Graptolites occur in conside- rable abundance, but they are chiefly found in peculiar localities where black mud abounded. The formation, when traced into South Wales and Ireland, assumes a greatly altered mineral aspect, but still retains its characteristic fossils. It is worthy of remark that when these strata occur under the form of ' trappean tuff ' (volcanic ashes of De la Beche), as in the crest of Snow- don, the peculiar species which dis- tinguish it from the Llandeilo beds are still observable. The formation generally appears to be of shallow- water origin, and in that respect is contrasted with the group next to be described. Sir A. Ramsay esti- mates the thickness of the Bala beds in North Wales, including the contemporaneous volcanic rocks, stratified and unstratified, as being from 10,000 to 12,000 feet. 2. Zilandeilo group. These strata at Llandeilo, a town in Caer- marthenshire, consist of dark- coloured argillaceous and micaceous flags, frequently calcareous, with a great thickness of shales below them, generallyof a black colour. They are also seen at AbereiddyBay in Pem- brokeshire, and at Builth in Rad- norshire, where they are interstrati- fied with volcanic matter. They are conformable with the overlying Caradoc group. A still lower part of the Llan- deilo rocks consists of a black car- bonaceous slate of great thickness, frequently containing sulphide of iron, and sometimes, as in Dum- friesshire, beds of anthracite are present. It has been conjectured that this carbonaceous matter may be due in great measure to large quantities of embedded animal matter, for the number of grapto- lites included in these slates is cer- tainly very great. In North and South Wales 25 species of Grapto- lites occur in the Llandeilo flags. The double Graptolites, or those with two rows of cells, such as Diplograptus (fig. 595), Climaco- graptus, and Dicranograptus, are conspicuous. Didymograptus (figs. 598, 594) is a branching form with one series of cells. The leaflike Phyllograpti (fig. 596) and the remarkable curved form Rastrites (fig. 597) are also found in the Ordovician. The Brachiopoda number 34 species, 23 of which pass up into the Caradoc Sandstone, while five genera Acrotreta t Crania, Bhyn- chonella, Strophomena, and Lep- tcena appear for the first time. The Lamellibranchiata are Car- diola interrupta, Brod., Modiolop- sis expansa, Portl., Ctenodonta va- ricosa, Salt., and Palcearca amyg- dalus, Salt. Some pass to the Cara- doc, and the genus Car diola appears for the first tinle. Murchisonia and Ophileta are common Gastropoda, and the Pteropods belong to the genus Theca. Cephalopoda are not abundant in the British Llan- deilo formation ; but Orthoceras, Endoceras, and Piloceras are com- mon genera. On the Continent of Europe the Orthoceratidse are very common (fig. 606). The genera Asaphus (fig. 608), EK 418 AEKNIG [CH. XXV. Ogygia (fig. 609), and Trinucleus (fig. 611) form a marked feature of the Trilobite fauna of this age, which comprehends 18 genera and 45 species. 3. Arenig 1 or Stiper-stones group. Next in descending or- der, and forming the base of the Lower Silurian, are the shales and sandstones in which the quartzose rocks called Stiper-stones in Shrop- shire occur. For a long time the only organic remains in these Stiper-stones were the tubular burrows of Annelids (see fig. 607, Arenicolites linearis, Hall), which are remarkably common in the Lowest Silurian in Shropshire, the North-west Highlands of Scotland, and in the State of New York in America. Similar burrows are now made, on the retiring of the tides, in the sands by lobworms, which are dug out by fishermen and used as bait. Sedgwick recognised this group, which he called the Arenig or Skiddaw, and separated it from the overlying and conformable Llan- deilo. Salter, however, distin- guished the break between the Are- nig and the underlying Tremadoc group, and Dr. Hicks has defined the succession in South Wales, and described a great fauna in the Arenigs of the St. David's district. It must be remembered that there is no stratigraphical unconformity be- tween the Arenig and the groups above and below it ; though the pa- laeontological break is considerable, for out of the 150 species of fossils of the Arenig strata only 25 have been found either in beds above or below them. Only 8 genera, com- prising 9 species, pass from the Arenigs to the Llandeilo above, and 11 genera, comprising 46 species, which had lived in Tremadoc times, reappear in the Arenig. No less than 40 genera appear for the first time in the Arenig group, and this in itself gives a definite impor- tance to it. Of these there ore 16 genera of the Hydrozoa cf the Graptolite group. Didymoarap- tua (figs. 593, 594), Calk-grap- tus, Diplograptus, and Tetra- graptus are examples. Four genera of Annelida, Eelmin- tholithes, Stellascolithes, Nereites, and Palceochorda, appeared ; with the genera of the Trilobita, &glina, Trinucleus, Barrandia, Calymene, Phacops, Placoparia, IllcBnus, and Homalonotus. Ribeiria and Re- doina were new Lamellibranchs, and Ophileta, Pleurotomaria, Rhaphistom a, new genera of Gas- tropoda. Orthoceras occurs, and there are four species of the genus in the Welsh and Shropshire area. The Corals, Bryozoa, and Echino- dermata are n ot represented. Phyl- lopoda of the genus Caryocaris are peculiar to the group, and there are only 18 species of Brachiopoda the special forms being Dinobo- lus Hicksii, Salt., Siphonotreia micula, M'Coy, Discina sp., Or- tliis striatula, Emmons, O.remota, Salt., and 0. alata, Sow. This Arenig group may there- fore be conveniently regarded as the base of the Ordovician system ; some authors, however, prefer to include in the Ordovician the un- derlying Tremadoc slates. Sedgwick noticed that the Are- nig dark slates, shales, flags, and bands of sandstone were associated with masses of igneous rock, and it is evident that while the sedi- mentary strata were accumulat- ing, volcanic action was going on. Hence great thicknesses of felsitic or rhyolitic lavas and tuffs were erupted and spread over the land and the sea-floor, and were inter- stratified with the fossiliferous sediments. Some of the most im- portant Welsh mour tains consist mainly of these volcanic materials such as Cader Idris, the Arans, Arenig Mountains, and others. Ordovician Strata of the lake District. In this area the Bala and Caradoc are probably re- presented by the Asgill shales, the Coniston limestone, and the under- ly m g great volcanic series, the Bor- rovvdale. The Arenigs find their equivalents in the Skiddaw slates. Ordovician Strata of Scotland. The Ordovician of the Borderland consists of greatly contorted strata, which have been worn and denuded into hills of moderate height, and deep valleys. In the Girvan district there are. CH. XXVI.] ORDOVICIAN OF SCOTLAND 419 besides conglomerates and meta- morphosed rocks, calcareous beds, which represent the Llandovery, Bala, and Llandeilo groups. In the Moffat district there are gritty or coarse-grained greywacke, fis- sile flagstones, conglomerates, and bands of black carbonaceous shales, with cream-coloured clay and iron- stone nodules. The black bands occur in lenticular masses in the greywacke, and, although they cover a vast area, are contorted, often reversed, and highly fossili- ferous. Lapworth has shown that there are three horizons at which these black bands occur, and at each is found a profusion of Grap- tolites. The highest, or Lower Besides the works cited at the end of the last chapter, reference should be made to numerous valu- able papers on the Ordovician of Wales by Dr. Hicks, published in the ' Quart. Journ. Geol. Soc.,' voh. Llandovery the Birkhill shales contain zones of Bastrites, Mono- graptus, and D^plograptus, and there is a decided palseontological break between this horizon and the next below, or the Hartfell shales, which are of Bala age, and contain Dicellograptus, Pleurograptus, and Climacograptus. The Glen- kiln shales are upper Llandeilo in age, and contain Didymograptus. Still older beds, of Arenig age, are found in the Scottish Borderland containing thick beds of chert, in which the researches of Professor Nicholson and Dr. G. J. Hinde have revealed the presence of great numbers of Radiolarians. xxxi. and xxxii. &c., and to those of Professor Nicholson and Mr. Marr on the rocks of the Lake District, ' Quart. Journ. Geol. Soc.,' vols. xxxiii., xliv. &c. CHAPTER XXVI THE CAMBRIAN SYSTEM Divisions of the Cambrian System Cambrian Flora and Fauna Sponges Graptolites Echinodermata Brachiopoda Mollusca Annelida Trilobita The oldest known fossils of the Lower Cambrian Period Upper Cambrian, Tremadoc slates and Lingula Flags Middle Cambrian, Meneviaii beds, Harlech grits, and Llanberis slates Lower Cambrian, Comley Sandstone Cambrian of Scotland, Durness Limestone, Gir^an Limestone 'Fucoid'- and Olenellus-beds. Nomenclature and Classification of the Cambrian strata. This system of strata, being the oldest in which a marine fauna has been detected, is of the highest interest to the geologist. There is fortunately, now, little difference of opinion as to the name which should be applied to it. Sedgwick proposed the name ' Cambrian ' as early as 1835, and it has now come into almost universal use. Barrande, it is true, suggested that the strata should be called 'Primordial,' or 'Primordial Silurian ; ' but a name which suggests that no earlier fossiliferous rocks will ever be found is clearly objectionable. Marcou and other geologists in America have tried to revive the name Taconic, which was proposed by Emmons in 1842. But there are serious doubts as to how far the strata indicated by that name are identical with the true Cambrian. EE 2 420 CAMBKIAN NOMENCLATUKE [CH. xxvi. During the last few years very important strata have been detected in many different districts which are regarded as con- stituting the base of the Cambrian ; so that the system is now considered to consist of three members, which are named as follows : Upper Cambrian : Olenus beds. Middle Cambrian : Paradoxides beds. Lower Cambrian : Olenellus beds. Professor Lapworth has proposed to employ the name of Taconic for the lowest division of the Cambrian, but the sugges- tion does not appear to have met with any general acceptance. In America the Lower Cambrian has been called 'Georgian,' the Middle Cambrian ' Acadian,' and the Upper Cambrian ' Potsdam ian.' It must be borne in mind that the older writers, following Murchison, ascribed to the Cambrian all the strata containing unsatisfactory fossils or none at all, and it is only in recent years that the true characteristics of the three great Cambrian faunas have come to be recognised. Many of the strata formerly confounded with the Cambrian, like the Torridonian and the Longmyndian, are now proved to be of pre-Cambrian age. Characteristics of the Cambrian Fauna and Flora. This oldest known fauna, although poor in the number of species represented in it, is remarkable for the variety and high organi- sation of many of the forms of life which it contains. In the British Islands, the Cambrian fossils are usually rare and badly preserved ; in most cases the rocks have undergone such an amount of change as to obliterate the traces of organisms, many of the argillaceous beds exhibiting slaty cleavage. The number of species known from the Cambrian of Br'tain does not probably exceed 200; and in all the European localities taken together the number has been estimated as not exceeding double that amount. The Cambrian fossils are somewhat more numerous and better preserved in North America, and a con- siderable number of forms have been described ; but the known Cambrian species from all parts of the world probably fall below 1,000, which number is small in comparison with that of the fossils found in the Ordovician and the Silurian. Nevertheless, in spite of this paucity of species in the Cam- brian fauna, it is to be noted that in it all the great divisions of the Animal Kingdom are represented except the Vertebrata. The Trilobites, moreover, are of great size, and of by no means lowly organisation. Hence we have no ground whatever for believing that this oldest known fauna is in any proper sense of this term CH. xxvi.J CAMBRIAN FAUNA 421 ' primordial,' and represents the beginning of life on the globe. If, as all palaeontological study indicates, the newer faunas are derived by descent with modification from older ones, then the period preceding the Cambrian Period, during which life flourished on the globe, must probably have been at least as great as that which has elapsed between the Cambrian Period and the present day. There is another feature of the Cambrian fauna which it is desirable to bear in mind. Even at that early period we find clear indication of a geographical distribution of life-forms, The species of Trilobites, Brachiopoda, &c., found in Britain, Scandinavia, Bohemia, and North America respectively, present general analogies with one another, but are not identical. The great majority of the forms of life found in the Cambrian seem, to have been inhabitants of the deep sea, and up to the present time the shallow-water marine forms of the period remain unknown, as do also the freshwater and terrestrial fauna and the terrestrial flora. The Cambrian fauna is mainly made up of Brachiopoda and Trilobites, and the representatives of other groups are as a rule by no means abundant. Traces of marine algae have been found, and a few Forami- nifera and doubtful Kadiolarians have been described. Sponges are represented by Protospongia (of which only the spicules of the dermal layer are known) and Archceocyathus, Graptolites are comparatively rare in the Cambrian, and are represented by few species ; but the curious Dictyon&ma may possibly have been closely related to the group of Khabdophora (Graptolita). Fig. 612. Certain markings on the surface of Cambrian rocks have been thought by Nathorst and others to indicate the exist- ence of Medusae at that early period. The Echinodermata are represented only by Crinoidea (Dendrocrinus), Cys- toidea (Protocystites), and Starfish (Palce- asterina) . Lingulella Davisii, The Brachiopoda nearly all belong to the division of the Inarticulata, and include &.' D/Sorte^'by cleavage, the persistent types Lingula (Lingulella, fig. 612) and Discina, with the peculiar genera Obolus, Obolella, Acrotreta, AcrotJiele, Kutorgina. Only a few forms of Lamellibranchiata have been found in the Cambrian, such as Palcearca and Ctenodonta. A few Gastropods (Pleurotomaria, &c.) occur in the Cambrian. 422 CAMBRIAN MOLLUSCA [CH. XXVI. Many forms of Pteropods, however, are found, including species of Theca, or Hyolithes (fig. 613). Cephalopoda are first found in the youngest of the Cambrian strata the Tremadoc, which are now placed by some authors at the base of the Ordovician. They are represented by Orthoceras and Cyrtoceras (fig. 614). Vermes are represented in the Cambrian by many tracks and burrows, like that known as Histioderma (fig. 615). Fig. 613. Fig. 614. Theca (Cleidotfiecd) operculata, Salt., nat. size. Lower Tremadoc beds, Tremadoc. Cyrtoceras precox, Salt., mag. Tremadoc rocks, N. Wales. A. Dorsal edge, place of siphuncle. B. Aperture. C. Ventral edge. Fig. 615. 1 2 fltstioderma hibernica, Kin. Oldhamia beds. Bray Head, Ireland. 1. Showing opening of burrow, and tube with wrinklings or crossing ridges. probably produced by a tentacled sea worm or annelid. 2. Lower and curved extremity of tube with fine transverse lines. The Trilobita of the Cambrian are of very great interest. Most of the species appear to have been deep-water forms, and were destitute of eyes. Some of the Paradoxides were of great size (more than a foot in length). As was shown by Barrande in the case of Sao, the metamorphoses of these early forms can be followed by the study of the fossils. Among the principal genera are Olenus (fig. 617), Conoceplialus (fig. 618), Micro- CH. XXVI.] CAMBRIAN TEILOBITES 423 discus, Ellipsocephalus, Sao (fig. 616), Dikelocephalus (fig. 621), Agnostus (figs. 619, 620), Paradoxides (figs. 622, 623), and Ole- nellus. Fig. 616. I * The small lines beneath indicate the true size. In the youngest state, , no segments are visible; as the metamor- phosis progresses, b, c, the body seg- ments, begin to be developed ; in the stage d the eyes appear, but the facial sutures are not completed ; at e the full-grown animal, half its true size, is shown. Sao hirusta, Barr.,in its various stages of growth. Fig. 617. Mg. 618. Fig. 619. Fig. 620. Olenus micrurus, Baiter, natural size. Fig. 621. Conocoryphe striata (Syn. Agnostus integer, Agnostus Rex, Conocepkalus striatus, Emmr.), natural size. G-inetz and Skrey. Beyr., nat. size and magnified. Barr., nat. size. Skrey. Fig. 622. 623. Dikelocephctlits minnesott nsis, Dale Owen, J diameter. A large Trilobite of the 01 e- noid group. Potsdam Sand- stone. Falls of St. Croix, on the Upper Mississippi. Paradoxides bohemicus, Barr., about \ natural size. Paradoxides David-is, Salt., ^ nat. size. Menevian beds, St. David's and Dolgelly. Besides the numerous Trilobites, we find other representa- tives of the Arthropods in several forms of Phyllopods like Hymenocaris (fig. 624), and of Ostracods like Leperditia. 424 OLDEST KNOWN FOSSILS fen. xxvi. Fig. 624. The Oldest known Fossiliferous Rocks. The lowest Cambrian strata (Taconic of Lapworth, ' Georgian ' of American authors) are of great interest to geologists as containing the oldest known marine fauna. This fauna, though small, con- tains representatives of a considerable number of forms of invertebrate life. Plants (algae) are doubtfully represented by a number of somewhat obscure impres- sions. We find the remains of sponges (Arcliceocyathus), possibly of graptolites (Diplograptus and Climacograptus ?), and an obscure Cystidean (?). Many Bra- chiopoda (fig. 625) (Lingulella, Kutorgina, Obolella, &c.), a Lamellibranch (Fordilla), some Gastropods (Platyceras, Scenella, &c.), and many Pfceropods (Hyolithcs, &c.). Fig. 625. Hymenocans vermicauda, Salter. A Phyllopod Crustacean. \ natural size. a. Obolella crassa, Hall sp., inat. Fig. 626. b. Lingulella ella, H. & W., Jnat. Fig. 627. c. Salterella pulchclla, Bill., nat. Fig. 628. Olentllus Callavei, Lapw., J nat. Olenellus Lapworthi, Olenellus armatus, From the Comley (Holly- Peach, Peach, bush) Sandstone of nat. size. x 2 nat. Shropshire. From the Lowest Cambrian Strata of N.W. Scotland. CH. XXVI.] UPPER CAMBRIAN 425 Remains of Annelids (Salterella, fig. 625, c, &c.) abound in many of the beds, but the most important fossils are the repre- sentatives of the Arthropoda. We find an Ostracod (Leperditia), a Phyllopod (Protocaris), and a mimber of Trilobita, including Olenellus and allied genera (figs. 626-628), and Agnostus. British Representatives of the Cambrian System. The Cambrian, like the overlying Ordovician, is usually regarded as con- sisting of three members the Upper Cambrian, including the Tremadoc slates and Lingula flags ; the Middle Cambrian, con- sisting of the Menevian beds, the Llanberis slates, and the Harlech grits ; and the Lower Cambrian, represented by the Hollybush sand- stone of Central England and the 'Fucoid beds ' of Scotland. The Upper, Middle, and Lower Cambrian are sometimes known respectively as the Olenus beds, the Paradoxides beds, and the Olenellus beds, from the genera of Trilobites which characterise those several divisions. Upper Cambrian, or strata characterised by Olenus. These consist of a great thickness of strata, among which the following divisions have been established by geologists. Tremadoc slates. The Trema- doc slates of Sedgwick are more than 1,000 feet in thickness, and consist of dark earthy grey slates occurring near the little town of Tremadoc, situated on the north side of Cardigan Bay in Caernar- vonshire. They were traced subsequently to Dolgelly, and of late years strata of the same age have been dis- covered and carefully examined by Dr. Hicks, at St. David's promon- tory and Ramsey Island, South Wales, where there are dark earthy flags and sandstones 1,000 feet thick, with many fossils. They rest conformably upon thick Lin- gula flags. Subsequently Mr. Cal- laway has shown that the Shineton shale of Shropshire is of Lower Tremadoc age. The fauna is very remarkable, and differs consider- ably in North arid South Wales ; it contains at least 84 species, and many great groups of the inverte- brata appear in the rocks for the first time. The Crinoidea, Aste- roidea, and Cephalopoda are re- presented therein, for the first time in the world's history. There are many new genera of Trilobita, such as Nesuretus, Psilocephalus, Niobe, Angelina, Asaphus, and Cheirurus, besides some which existed in the lower rocks, such as Agnostus, Conocoryphe, and Olenus. The Crinoid Dendrocrinus and Asteroid Palceasterina, the Cephalopoda Orthoceras sericeum, Salt., and Cyrtoceras prcscox, Salt. (fig. 614), of the Upper Tremadocs, are the first known. The Lamellibranchs Ctenodonta, Palcearca, Glyptarca, Davidia, and Modiolopsis make their appearance. The Brachiopoda belong to the genera which existed in the underlying strata, and the species Lingulella Davisii, M'Coy, and Orthis Carausii, Salt., and the genera Obolella and Lingula, are common to both groups. The North Wales Tremadocs contain 9 species of Pfceropoda, principally of the genus Tlieca\ and JBellerophon is found amongst the Heteropoda. Bhabdophora or graptolites were discovered in Tremadoc rocks by Callaway, and belong to the genus Bryograptus. Phyllopod Crusta- cea exist in the Upper Tremadocs, and the characteristic Trilobita are Angelina Sedgwickii, Salt., Asaphus affinis, McCoy, sp., and an Olenus. Dictyonema sociale, Salt., and Bryozoa occur, and in the strata below also. By Mr. Marr and some other authors the Tremadoc beds are regarded as forming the base of the Ordovician system, and not the top of the Cambrian. Lingula flags. Next below the Tremadoc slates in North Wales, lie micaceous flagstones, bluish and black slates and flags, with bands 426 MIDDLE CAMBRIAN [CH. xxvr. of grey flags and sandstones, in which in 1846 Mr. E. Davis discovered the Lingulella (fig. 612) named after him, from which was derived the name of Lingula flags. These beds are more than 5,000 feet thick, and have been studied chiefly in the neighbour- hood of Dolgelly, Ffestiniog, and Portmadoc in North Wales, and also at St. David's in South Wales. They have yielded 26 genera and 69 species of fossils, of which 9 only are common to the overlying Trema- doc rocks. They include Dictyone- masociale,Sn,lt.,Agnostiisprinceps, Salt., Ampyx prcenuntius, Salt., Conocoryphedepressa, Salt., Olenus impar, Salt., Lingulella Davisii, M'Coy, L. lejns, Salt., Obolella, and Orthis. In the Lingula flags Olenus (fig. 618), Agnostus, Anopolenus, Microdiscus, Paradoxidcs, and Co- nocoryphe are prominent forms of Trilobita, and Hymenocaris vermi- cauda, Salt. (fig. 624), is a common species of the Phyllopod Crustacea. The Lingula flags may be divided into two zones, an upper and lower, the middle zone of older authors being of less value. Amongst the fossils of the upper zone is Dictyonema sociale, Salt., which occurs at Keys End Hill, Malvern, and in North Wales. No less than 30 species of Crus- tacea belonging to the genera of Trilobita just noticed occur, and only 4 pass up into the Tremadocs. The Brachiopoda are of 8 species, 6 of which pass upwards, and the genera are Lingula, Lingulella, Obolella, Kutorgina, and Orthis, the two characteristic species being Lingula pygm&a, Salt., and Obo- lella Salteri, Hicks. In the Lower Lingula flags, which rest conform- ably on the Menevian strata, Cru- ziana, a supposed Annelid, occurs, and Scolioderma and Helminthites are characteristic worms. Nine genera and 25 species of Crustacea are found. Agnostus limbatus, Salt., and A. nodosus, Salt., Olenus mi- crurus, Salt., 0. gibbosus, Salt., are peculiar to these lower flags, and so is the Phyllopod Hymenocaris ver- micauda, Salt. The three genera of Brachiopoda represented are Lingulella, Orthis, and Obolella. Finally, two species of Theca occur. In Merionethshire, according to Sir A. Kamsay, the Lingula flags attain their greatest development ; in Caernarvonshire, they thin out so as to have lost two-thirds of their thickness in eleven miles ; while in Anglesea and on the Menai Straits, both they and the Tremadoc beds are entirely absent, and the Lower Silurian rocks rest directly on pre-Cambrian strata. Middle Cambrian, or strata characterised by Paradoxides, are also of great thickness. They have been studied in South Wales by Dr. Hicks, who has established the following subdivisions in the series. Menevian beds. Immediately beneath the Lingula flags there occurs a series of dark grey and black flags and slates, alternating at the upper part with some beds of sandstone, the whole reaching a thickness of from 500 to 600 feet. These beds were formerly classed, on purely lithological grounds, as the base of the Lingula flags ; but Messrs. Hicks and Salter, to whose exertions we owe almost all our knowledge of them, have pointed out that the most characteristic genera found in them are unknown in the Lingula flags, while they possess many forms from the under- lying groups of strata. They there- fore proposed to place these beds at the top of the Middle Cambrian, under the term ' Menevian,' Mene- via being the Latin name of St. David's. The beds rre well ex- hibited in the neighbourhood of St. David's in South Wales, and near Dolgelly and Maentwrog in North Wales. They are the equivalents of Etage C of Barrande's Primor- dial Zone. Fifty-two species have been found in the Menevians, which are very rich in fossils for so early a period. Nineteen species are common to the overlying Lin- gula flags, but none pass up to the Tremadoc rocks. Twelve genera and 32 species of Trilobita occur, and some forms are of large size ; Paradoxides Davidis, Salt, (see fig. 623), the largest Trilobite known in Great Britain, nearly 2 feet long, is peculiar to the Menevian. The CH. XXVI.] PAKADOXIDES BEDS 427 other genera are Agnostus, Ano- polenus, Conocoryphe, Holocepha- lina ', and the special genera of Tri- lobita are Arionellus, Erinnys, Microdiscus, and Carausia. The Trilobite with the largest number of rings, Erinnys venulosa, Salt., occurs here in conjunction with Agnostus and Microdiscus, the two genera with the smallest num- ber. Blind Trilobites are also found, as well as those which have the largest eyes, such as Microdis- cus on the one hand, and Anopo- lenus on the other. Olenus did not then exist. The Ostracod Leperditia occurs, and the genera Orthis, Discina, and Obolella amongst the Brachio- poda. Several Pteropoda have been found, with the Cystoidean Protocystites. Several species of Protospongia of the Spongida, and Arenicolttes, and Serpulites amongst the Annelida conclude the fauna. The discovery and description of this remarkable as- semblage of early forms, we owe to the careful labour of Dr. Hicks. Harlech grits and Llanberis slates. The sandstones of Harlech attain a thickness of no less than 6,000 feet without any interposition of volcanic matter; and in some places in Merionethshire they are still thicker. Until recently these rocks were supposed to contain no fossils. Now, however, through the labours of Dr. Hicks, they have yielded at St. David's a rich fauna of Trilobita, Brachiopoda, Phyllo- poda, and Pteropoda, showing, to- gether with other fossils, the exist- ence of a series of by no means low organisms at this very early period. Already the fauna amounts to 29 species, referred to 16 genera ; of these, 8 genera and 12 species are common to the Menevian group above. Although more recent dis- coveries of fossils and changes of opinion concerning the nomencla- ture and limits of species may have somewhat modified the numerical relations of faunas in this and simi- lar cases, yet the general relations as worked out by the late Mr. Etheridge for Sir Charles Lyell remain substantially unaffected. A new Trilobite, called Plutonia Sedgwickii by Dr. Hicks, has been met with in the Harlech grits of St. David's. It is comparable in size to the large Paradoxides Da- vidis, Salt., before mentioned, has well-developed eyes, and is covered all over with rough tubercles. In the same strata occur other genera of Trilobites, namely, Conocoryphe, Paradoxides, Microdiscus, Ag- nostus, pni the Pteropod Theca (fig. 613), all represented by species peculiar to the Harlech grits of that area. The sandstones of this formation are often rippled, and were evidently left dry at low tides, so that the surface was dried by the sun and made to shrink and present sun-cracks. There are also distinct impressions of raindrops on the surfaces of many strata. Fossils occur yet earlier in the Harlech group of St. David's in the lower red shales that immediately overlie the conglomerate at the base of the Cambrian formation. The only forms yet found are Lingulella ferruginea, Salt., L pri?ncBva, Hicks, Leperditia Cam- brensis, Hicks, and Discina Caer- faiensis, Hicks. The slates of Llanberis and Penrhyn in Caernarvonshire, with their associated sandy strata, attain a great thickness, sometimes about 3,000 feet. They are probably of the same age as the Harlech and Barmouth beds last mentioned, for they may represent the deposits of fine mud thrown down in the same sea, on the borders of which the sands above mentioned were ac- cumulating. The Middle Cambrian age of at least a portion of theee strata has been determined by the finding in them of a Conocoryphe (C. Viola, Woodw.). In South Wales the beds of St. David's with Lingulella pass down to a con- glomerate, and a similar indication of a physical break is found in North Wales, according to Dr. Hicks and Professor Hughes. Below are rocks the age of which is disputed (p. 434). Iiower Cambrian, or strata characterised by Olenellus. On the flanks of Caer Caradoc in Shropshire, Professor Lapworth 428 LOWEK CAMBRIAN [CH. XXVI. has discovered, in beds known as the Comley sandstone, remains of the Trilobite genus Olenellus (fig. 626), with several Brachiopods, which like it are characteristic of the Oldest Cambrian. The higher portions of this sandstone, however, contain Paradoxides, and must be referred to the Middle Cambrian. Worth-west Scotland suc- cession. In the North-western Highlands of Scotland are found limestones of Upper Cambrian age. The relation of these strata to those above and below them is now happily settled. Beneath the limestone are sandy shales and quartzites with Annelid tubes, of Salterella Maccullochi, Salt, sp., resting on and overlapping red sandstone, grit and conglomerate (Torridon sand- stone), to which a Cambrian age was formerly ascribed, but which are now known to be pre-Cambrian. Eecent discoveries by the Geo- logical Survey have shown that the whole, or nearly the whole, of the strata in the north of Scotland (which were formerly ascribed to the Ordovician, and are of con- siderable thickness) are really of Cambrian age. The Durness lime- stone, at the top of the series, con- tains a remarkable assemblage of fossils, including Orthoceras and Maclurea, which, as pointed out by Salter, have a very close re- semblance to the fauna of the Calciferous sandstone of North America. These strata are probably of the same age as the Tremadoc that is, they form the top of the Cambrian series, according to the classification now generally accep- ted by geologists, and they are said in places to attain a thickness of 1,500 feet Below the lime- stones there occur sandy beds with worm-burrows (Serpulite grit or Salterella grit), and argillaceous beds in which are markings that have been taken to represent sea- weeds, and these have been called the fucoid beds. In these 'fucoid beds ' the officers of the Geological Survey have detected remains of several species of Olenellus (figs. 627, 628), and other fossils of the Lowest Cambrian zone. Beneath the fucoid beds occur thick masses of quartzite perforated by in- numerable worm- burrows. In the Lowlands of Scotland, beneath the series of Ordovician strata, we find at Girvan in Ayr- shire beds of limestone containing a somewhat similar fauna to that of the Durness limestone, underlain by a remarkable series of rocks of volcanic origin. Dr. Hicks' s papers on the Cam- brian and Ordovician strata have already been referred to, and Dr. Callaway's and Professor Lap- worth's accounts of the Cambrian rocks of Shropshire will be found in ' Quart. Journ. Geol. Soc.,' vol. xxxiv., and in the ' Geol. Mag.,' 1888 and 1891. The discovery of the Olenellus fauna in the ' fucoid ' beds of the North-wt st of Scotland is described by Messrs. Peach and Home, ' Quart. Journ. Geol. Soo-' vols. xlviii. and xlix. CH. XXV1I.J BOHEMIA AND SCANDINAVIA 429 CHAPTER XXVII FOREIGN DEPOSITS WHICH ARE HOMOTAXIAL WITH THE OLDER PALAEOZOIC STRATA OF THE BRITISH ISLES Older Paleozoic strata less altered than the British Basin of Bohemia ' Primordial ' strata of Barrande Stages D, E, F Older Palaeozoic strata of Scandinavia Olenellus beds Alum Shales Limestones and Schists Russia and other parts of EuropeNorth America- Geographical distribution of life forms in Cambrian times Table showing equivalence of strata in different areas. THE British representatives of the Older Palaeozoic rocks, though they are of such great thickness, are usually much altered, especially in their lower portions, and, slaty cleavage having been developed in their fine-grained beds, the fossils are often rendered obscure or altogether obliterated over wide areas. It is a fortunate circumstance that, in other parts of the world, we find strata occupying the same position in the geological series as our Ordovician and Cambrian, but with fossils in a much better state of preservation thanjn this country. OLDER PALAEOZOIC STRATA OF EUROPE Bohemia and Central Europe. One of the most inte- resting districts for the study of the Older Paleozoic rocks is Bohemia, where the strata lie in a basin upon the Archaean rocks, and are but little altered. The beds are crowded with fossils in the most admirable state of preserva- tion, and the able French geologist Barrande devoted his life to the collection and description of these ancient and remarkable relics of a number of extinct faunas. Resting upon crystalline and metamorphic rocks with unfos- siliferous schist (the stage A of Bar- rande) we find a series of Grey- wackes (stage B of Barrande) which contain traces of Annelids and Brachiopoda, and are believed to represent the Lowest Cambrian (Olenellus zone). Upon these last lie a series of greenish slaty rocks 300 feet in thickness (stage C of Barrande) which have yielded a very rich fauna, that of the Middle Cambrian. These beds were re- garded by Barrande as containing the oldest known fauna, and were styled by him ' Primordial beds.' These 'primordial' strata are covered by other argillaceous, sandy, and calcareous beds (the stages D and E of Barrande), D representing the highest Cambrian and Ordo- vician strata and E the Silurian. The Older Palaeozoic rocks of Bo- hemia, though so rich in fossils, present many local peculiarities, and it is not possible to correlate with absolute precision the minor divisions of the Bohemian rocks with those of our own country. Barrande entertained the belief that certain assemblages of fossils belonging to the older series of strata may sometimes be found living on in strata of younger age ; and for such assemblages of fossils he proposed the name of ' colonies.' Professor Lapworth, Mr. Marr, and other geologists have shown that Barrande's so-called ' colonies ' are portions of older strata faulted in among newer beds. Scandinavia. The Older Palaeozoic rocks of Sweden cover a considerable area, and, though of insignificant thickness in compari- 430 NORTH AMERICA [CH. XX VII. son with our British strata of the same age, they agree much more closely with our own series, both in the sequence of beds and the types of fossils represented, than do the strata of the same age in Bohemia. The base of the series is formed by thick masses of fel- spathic sandstone and conglomerate, containing very few fossils except obscure and doubtful impressions of plants. These strata, which are called ' Fucoid Sandstones,' ' Eo- phyton Sandstones,' and ' Sparag- mites,' probably represent, in their lower part at least, the Torridon sandstone, but at their summit we find sandy beds containing the Olenellus or Lower Cambrian fauna. The thin argillaceous strata known as Alum shales, which overlie these sands, contain in their lower portion the Para- doxides fauna, and in their upper portion the Olenus fauna ; the highest zon.e of the Cambrian is represented by beds with Dictyo- graptus. The Ordovician is repre- sented in Scandinavia by lime- stones containing Orthoceras and Cystoidea, covered by shales with graptolites and Trinucleus, with OLDER PALAEOZOIC STRATA OF NORTH AMERICA many characteristic Trilobites of the Arenig, Llandovery, and Bala groups. The Silurian of Scandi- navia is represented by beds of limestone containing Pentamerus, and many May-Hill and Wenlock types, covered by a conglomerate and nodular limestone with Ludlow fossils. In the Baltic Provinces and North Germany occur a series of strata which can be fairly well paral- leled with those of Scandinavia. At the top of the Cambrian, we find in Russia a bed of glauconite sand which, though only a few feet thick, contains a very interesting as- semblage of fossils, including the ' conodonts,' formerly thought to be teeth of fishes, but now re- garded as belonging to annelids, and the internal casts of many species of foraminifera. Other Parts of Europe. In the Ardennes, in Normandy and Brittany, in the Pyrenees and the Iberian peninsula, and in the island of Sardinia, we also find many very interesting developments of the Cambrian, Ordovician, and Silurian rocks. Worth America. In the North American continent the Lowest Cambrian is represented by shales, quartzites, and lime- stones containing the Olenellus fauna (the Georgia beds of the United States geologists). The Middle Cambrian or Paradoxides beds consist of slaty beds 2,000 feet thick, well exposed in New Brunswick, Newfoundland, and various localities in the United States, while the vast masses of 'Potsdam sandstone,' 6,000 feet thick, above these, belong to the highest Cambrian or Olenus beds. Above the Potsdam sandstone the Calciferous sandstone, which in part at least may represent the Upper Cambrian, has a fauna strikingly like that of our Dur- ness and Girvan limestones. The divisions, known as the Chazy limestones, Trenton limestones, Utica shales, and the Hudson River and Cincinnati groups, are on the same general horizon as our Arenig, Llandeilo, and Bala groups, while the Clinton and Medina sand- stones represent our May-Hill beds, the Niagara limestone con- tains similar fossils to our Wenlock, and the Waterlime a^d Onondaga salt groups agree generally with our Ludlow beds. It is found, however, that in different parts of the United States these several divisions present remarkable dif- ferences in mineral characters and fossils. If we compare the fossils of the Older Palaeozoic strata in the British Islands with those of Scan- dinavia and Bohemia on the one hand, and with those of the North American continent on the other, we shall find that, while the same or similar genera are repre- sented in the several divisions, the actual species in these several areas are often distinct. It is evi- dent that at this the earliest period CH. xxvii.] TABULATION OF OLDER PALAEOZOIC 431 forms similar to that which pre- vails at the present day must have already existed. of the earth's history of which we have records of the marine life, a geographical distribution of life- The general parallelism of the deposits in the four typical areas in which the Older Palaeozoic rocks have been best studied is illustrated in the following table : Mi o 3 i fc ' 2 ' " a .5',! * -Is 8 lull w o> c3 w fill "c3'~ ?%* ive to zy l Cinci Riv Trent Chaz 2. .p n s a e o> s BI-M- ^ ^ I ss ^ . ^ g - iS, -g ' Ci ! . l -g 3 ^ S 5 S.ZS JSEVIOIA -oaao JSEVIHSMVO A very accurate account of the foreign strata of Older Palaeozoic age is given in De Kayser and Lake's ' Text-book of Comparative Geology.' The whole of the foreign ier>resentatives of the lowest Cam- brian or Olenellus beds have been described, and their fossils figured, by the Director of the United States' Geological Survey, Mr. C. "Walcott. 10th Ann. Eep. U.S. Geol. Survey, (1890). 432 OLDEST KNOWN STKATA [CH. xxvm. CHAPTEK XXVIII SEDIMENTAEY KOCKS OF PRE- CAMBRIAN AGE Existence of stratified and other Kock-masses underlying the Older Palaeozoic Deposits Rocks of both Igneous and Aqueous Origin Obscure Traces of Fossils Thickness and Extent of pre-Cambrian Bocks pre-Cambrian strata of the British Isles Pebidian Arvonian Dimetian Fundamental Gneiss, or Lewisian Caledonian, or Dai- radian Mai vernian Mon ian Uriconian Longmyndian The Tor- ridon Sandstone, or Torridonian Pre-Cambrian of Europe and North America Huronian Laurentian, Upper and Lower Algonkian. and Archaean Pre-Cambrian of India Traces of Fossils in pre-Cambrian Rocks. WITH the disappearance of well-marked and clearly recognis- able assemblages of fossils, the work of making out a chrono- logical sequence of sedimentary formations comes to an end. Nevertheless the geologist is acquainted with the fact that beneath the rocks containing the oldest Cambrian fauna there lie many others some evidently of aqueous origin, others no less clearly of igneous origin nearly all showing traces of having undergone great? alterations. From the circumstance that these rocks contain only imperfect, doubtful, or fragmentary remains of fossils, or none at all, it has not been found practicable to arrange them in a definite chronological sequence. Although the Cambrian fauna is the oldest assemblage of marine forms of life which has been discovered by geologists in the earth's crust, yet there are enormous thicknesses of rocks under- lying the beds containing the Cambrian fossils that a^e certainly of sedimentary origin and of greater age, and in these we some day hope to find definite traces of earlier forms of life. The pre-Cambrian strata are often many thousands of feet in thickness, and include varieties of arenaceous, argillaceous, calcareous, ferruginous, and other deposits similar to those which make up the fossiliferous rocks of the earth's crust, these being more or less intimately associated with igneous and metamorphic rock masses of a highly crystalline character. If we examine the map of the world compiled by Marcou to illustrate the dis- tribution of the various geological formations in the land areas of the globe, it will be found that the areas occupied by the pre-Cambrian strata are nearly equal to those covered by all the fossiliferous rocks of Palaeozoic, Mesozoic, and Cainozoic age. That the Cambrian fauna does not represent the beginning CH. xxvni.] STRATA OLDER THAN CAMBRIAN 433 of life upon the globe all biologists and geologists must agree. That fauna contains representatives of all the great divisions of the animal kingdom except the Vertebrates ; and two of the higher groups of the Invertebrata the Crustacea and Cepha- lopoda are represented in the Cambrian fauna by forms of complex organisation. If there has been a gradual evolution and progression of life-forms in the past, it has been argued by many paleontologists that the periods during which life existed upon the globe before the dawn of the Cambrian period must at least equal that which has elapsed between the beginning of the Cambrian and the present day. The fact that the pre-Cambrian strata contain beds of lime- stone and graphite has often been adduced as an argument in favour of the view that plants and animals must have existed in those earlier periods of the earth's history, which have as yet yielded no distinct relics of these ancient forms of life in the shape of fossils. It is perfectly true that nearly all the calcareous and carbonaceous rocks found associated with the fossiliferous strata owe their origin to animal- and plant-life ; but it must not be forgotten that both calcium carbonate and carbon in the form of graphite may sometimes be produced by the operation of purely chemical and inorganic agencies. Nothing more strikingly illustrates the great value of palaeonto- logical evidence to the geologist than the fact that it has been found impossible in the absence of the evidence afforded by fossils to bring into any kind of correlation the various deposits underlying Cambrian strata. Hence all designations applied to such pre-Cambrian deposits have a purely local value. Various names have been proposed for those rocks which clearly underlie the Cambrian, and are, therefore, older than that system of strata. The older writers spoke of them as Primary or Azoic rocks, but when the discovery of Eozoon sug- gested the existence of life-forms during these earlier periods they were called Eozoic. In more recent times the names pre- Cambrian and Archaean have been generally applied to them, though the latter term, as we shall see, is now often used in a more restricted sense. We will first consider the terms applied to these pre-Cambrian strata in the British Islands, and then proceed to discuss the nomenclature of homotaxial rocks in other parts of the globe. The Cambrian rocks of the neous rocks plutonic and volcanic. British Isles are underlain not only It is usual to term all rocks be- by great thicknesses of sedimen- neath the Cambrian by the names tary rocks without recognisable fos- of pre-Cambrian and Archaean, sils, but by metamorphic and ig- The discrepancy of opinion re- F F 434 PRE-CAMBKIAN STRATA [CH. XXV II garding the geological structure of the North-west Highlands, men- tioned on p. 428, will prepare the student for similar diversities of opinion regarding the age of the rocks which underlie the fossilife- rous Lower Cambrian of Wales and the equivalent formations else- where. The Geological Survey con- sider that there is no break present, and that the volcanic and meta- morphic rocks underlying the f ossili- ferous strata are really part of one great Cambrian series. On the other hand, many other geologists consider that they have sufficient evidence to state that there is a great break at the base of the fossiliferous series, a con- glomerate existing there which contains the products of the de- nudation of two, if not three, more ancient groups of rocks. One of these lower groups was volcanic, and the other and older was meta- morphic. They (with a third, ac- cording to Dr. Hicks) are included under the term pre-Cambrian, but their relations to the North Ameri- can pre-Cambrian rocks and simi- lar formations in other areas are not determinable. Dr. Hicks's researches in the St. David's area tend to prove that there is a vast thickness of unfossili- forous rocks beneath the Cambrian conglomerate, which he groups as follows : 1. The Pebidian. A volcanic series, made up of ejectamenta, more or less stratified, alternating with schistose, metamorphosed clays, and sandstones. Spherulitic felstone, greenish and purplish felspathic breccias, silvery-white schists, purple shales, light- green clay slates, greenish, red- dish, and purplish indurated ashes, often conglomeratic, are found, and also contemporaneous rhyo- lytic lavas in the form of felstone. The upper beds are red and purple ashy schists. The Pebidian series rests unconformably on the next group, and has a different structure from the overlying Cambrian, to the basal conglomerate of which it con- tributes pebbles. The upper rocks are mostly basic in character. 2. The Arvonian consists of breccias, hiillefiintas, quartz fel- sites, and of rhyolites. Dr. Hicks states that this series rests uncon- formably on the underlying Dime- tian. Some authors, who accept the Dimetian and Pebidian forma- tions of Dr. Hicks, find themselves unable to recognise his Arvonian as a distinct formation. 3. The Dimetian. These low- est rocks, the base of which has not been seen, form an anticlinal axis, flanked by the Pebidian, and partly by unaltered Cambrian strata. The Dimetian rocks are quartz porphyries, often with doubly pyramidal and sub-angular phend- crysts of quartz, and crystals of fel- spar, in a matrix of grey or green felspathic material. Fine-grained quartz - felsites, ashy, shale-like rocks, with more or less distinct lines of lamination, occur, and com- pact granitoid rocks, without mica, and with quartz in excess over the orthoclase felspar. Granitoid gneiss, with quartziferous breccias and schists, and quartzites are present. These rocks contribute to the con- glomerate at the base of the Cam- brian, and were metamorphosed before their denudation occurred. Professor Bonney has shown that a quartz-felsite, or ancient igneous flow, closely resembling more modern rhyolites, underlies the Cambrian conglomerate in North-west Caernarvonshire. And both he and Dr. Hicks have proved the occurrence of a pre-Cambrian series in Anglesea greatly resem- bling that of South Wi.ies. The oldest rock in Scotland is that called by Sir B. Murchison ' the fundamental gneiss,' which is found in the north-west of Ross- shire and in Sutherlandshire, and forms nearly the whole of the adjoining island of the Lewis, in the Hebrides. It has a strike from north-west to south- east, nearly at right angles to the metamorphic strata of the Gram- pians. On this fundamental, He- bridean or Lewisian gneiss, in parts of the Western Highlands, rocks of doubtful age rest unconform- ably. These rocks have been called Caledonian by Dr. Callaway, and Dalradian by Sir Archibald Geikie. CH. XXVIII.] OF THE BRITISH ISLES 435 The central axis of the Mal- vern chain consists of hornblendic gneisses and contorted schists, on which rest, uncoiiformably, sand- stones of Cambrian age. Many years since, Dr. Holl noted these rocks as being of pre-Cambrian age, and later authors have called them Malvernian. Professor Blake considers that these older rocks of Anglesea, with some fossiliferous strata referred by other geologists to the Older Palaeozoic, constitute a great pre- Cambrian system which he calls the Monian. Dr. Callaway thinks that the patches of ancient rhyo- litic lavas and other igneous masses about the Wrekiii and adjoining districts in Shropshire are of pre- Cambrian age and constitute a system which he proposes to call the Uriconian. The great mass of slates and slaty flagstones, con- stituting the mountainous tract of the Longmynd in Shropshire, which was formerly considered as lying at the base of the Cambrian, is now recognised as being of pre- Cambrian age, and has been called by Dr. Callaway the Longmyndian. In the absence of fossils in these various formations the task of cor- relating the Lewisian, Dimetian, Caledonian, or Dah'adian, the Ar- vonian, Pebidian, Monian, Uri- conian, and Longmyndian series, is a hopeless one. It may even be doubted if among these systems there are not some metamorphosed rocks of Palaeozoic age. The Torridon Sandstone or Torridonian. This great system of strata, occurring in the North- West of Scotland and some of the islands of the Hebrides, was first described by Dr. Macculloch in 1819 ; he showed, in opposition to the views of Murchison and Sedgwick, that it is distinct from the Old Eed Sandstone of Scotland, which it somewhat resembles, and that it underlies the strata contain- ing what are now known to be Cam- brian fossils. The Torridon strata now occupy a comparatively small area, but the patches which have es- caped denudation are evidently por- tions of what was once a great and widely spread sj stem of strata. The rocks show but little signs of altera- tion, and consist of strata of white, pink, and purplish-red sandstones, often containing pebbles and pass- ing into conglomerates. Both in the upper and lower portions of the series, bands of dark grey argillaceous rocks are found which have yielded what appear to be tracks and burrows of Annelids, but, up to the present time, no more definite traces of organisms. The thickness of this series of unaltered strata is estimated by the Geological Survey to be no less than 8.000 to 10,000 feet, and that they are older than the Cambrian is shown by the fact that strata containing the characteristic oldest Cambrian fauna, with a number of species of Olenelhis, are found uncon- formably overlying them. That an enormous period of time must have elapsed between the deposition of the great mass of unaltered Torri- donian strata and the overlying Cambrian is shown by the great unconformity and overlap existing between them, the Cambrian strata being found lying on every portion of the Torridonian series, and pass- ing transgressively from it to the Archaean or fundamental gneiss. The general relations to one another of these oldest rocks of the British Islands is shown in the sec- tions on the next page (figs. 629-31). European ire-Cambrian Formations. On the Continent of Europe it was the custom as in this country until comparatively recent years, to group all strata of clearly sedimentary origin with the Cambrian, and to restrict the names pre-Cambrian and Archaean to highly crystalline rocks. But it is now clearly recognised that in such deposits as the f elspathic sandstones and conglomerates of Brittany, the Obermittweida conglomerates and similar deposits of Central Germany, and analogous strata in Scandi- navia, we have masses of sedi- mentary rocks, often of great thick- ness, containing only very obscure traces of organisms and underlying the whole series of Palaeozoic rocks ; yet these strata lie on, and are evidently younger than, the great masses of granite, gneisses, and FF2 486 PRE-CAMBRIAN FORMATIONS OF [CH. xxvm. schists pebbles and fragments of which are frequently included in the stratified masses. American pre- Cambrian Formations. In the North American Continent the first at- tempts to classify the pre-Cambrian deposits were made by Sir William Logan and his colleagues on the Canadian Geological Survey. These Fig. 629. 3' include fragments of a great series of crystalline rocks, granites, gneisses, and schists. Huronian series. The strata called Huronian by Sir W. Logan consist chiefly of a quartzite with great masses of greenish chloritic slate. Limestones are rare in this series, but one band of 300 feet in thickness has been traced for WNW Queenaig (2,673 feet) ESE 2 23 Section near Inchnadamff, Sutherland (after Murchison). 1. Fundamental or Lewisian gneiss of Murchison (Archaean). 2. Torridon Sandstone resting unconformably on 1. 3'. Quartzite of Cambrian age resting unconformably upon 2. 3. Metamorphosed rocks (Caledonian of Callaway, Dalradian of Geikie), thrown by the action of great reversed faults over 1, 2, and 3'. Fig. 630. Suilvein Assynt Beinn More J 2 83* Section across Suilvein and Beinn More, Sutherland. 1. Fundamental or Lewisian gneiss of Murchison (Archaean). 2. Torridon sandstones resting unconformably on 1. 3. Quartzites and (3) limestones full of annelid burrows, and containing a rich Cambrian fauna. 36. Metamorphosed rocks, gneisses and schists (the Caledonian of Callaway and Dalradian of Geikie) faulted over the Cambrian, Torr Ionian, and Archaean rocks- Fig. 631. The same sections as interpreted by Nicol and Lapworth. a. Fundamental Gneiss. b. Torridonian. c. Cambrian. d. Upper Gneiss forced over the older rocks by great reversed faults (thrusts) o, o, 0, o. e. Old Red Sandstone. observers recognised the fact that beneath the Palaeozoic rocks of Canada there occur series of com- paratively unaltered sedimentary deposits without fossils, to which they gave the name of Huronian, and that these rest upon and considerable distances to the north of Lake Huron. No organic re- mains have yet been found in any of the beds, which are about 18,000 feet thick, and rest unconformably on the Laurentian rocks. Laurentian group. Under- CH. xxvui.J EUROPE AND NORTH AMERICA 437 lying the Huronian, northward of the river St. Lawrence, there is a vast series of crystalline rocks of gneiss, mica- schist, quartzite, and limestone, more than 80,000 feet in thickness, which have been called Laurentian, and which are already known to occupy an area of about 200,000 square miles. They had undergone great disturbing move- ments before the Potsdam sand- stone and the other ' primordial ' or Cambrian rocks were formed. The newer portion of the Laurentian series is unconformable to the older. Upper Laurentian, Norian, or Labrador series. The Upper Group, more than 10.000 feet thick, consists of metamorphic crystalline rocks in which no organic remains have yet been found. They consist of gneisses and granitoid rocks with Labradorite and Anorthite felspars. There are also crystalline limestones and quartzites. These felspathic rocks sometimes form mountain masses almost without any admix- ture of other minerals ; but at other times they include augite, horn- blende, and hypersthene. The iri- descent felspar, Labradorite, is found in Labrador. These rocks cover a great area in the Adiron- dack Mountains. Lower Laurentian. This for- mation, about 20,000 feet in thick- ness, is, as before stated, uncon- formable to that last mentioned ; it consists in great part of massive gneiss of a reddish tint with ortho- clase felspar. Beds of nearly pure quartzite, from 400 to 600 feet thick, occur in some places. Hornblendic and micaceous schists are often interstratified, and beds of lime- stone usually crystalline. Beds of graphite (plumbago) also occur, and it has naturally been conjectured that this pure carbon may have been of organic origin before it underwent metamorphism. There are several of these lime- stones which have been traced to great distances, and one of them is from 700 to 1,500 feet thick. In the most massive of them Sir W. Logan observed in 1859 what he considered to be an organic body. It had been obtained the year before by Mr. J. McCulloch at the Grand Calumet on the river Ottawa. This supposed fossil, to which the name of Eozoon canadense, Daws., was given, has now, however, been shown to be of inorganic origin (see p. 74). The geologists of the United States Geological Survey, especially the late Professor R. D. Irving and Mr. C. R. van Hise, have by their studies thrown much new light on the nature and classification of the pre- Cambrian rocks. They recog- nise that under the Cambrian strata of the North American continent two great series of rocks, each many thousands of feet in thickness and covering vast areas, may be traced. The older series consist of highly crystalline rocks, granites and norites, gneisses and schists, with occasional crystalline limestones and beds of graphite, and it is to these highly crystalline rocks that the American geologists propose to restrict the term Archaean. Between the Cambrian and the Archaean there exist vast thicknesses of strata, sometimes but little altered, at other times dis- playing clear evidence of con- siderable metamorphic action, and only exhibiting few and almost in- determinable traces of organisms. These strata the American geolo- gists propose to call Algonkian, and as alternative names they have proposed ' Eparchian ' (lying on the Archaean), ' Agnotozoic ' (containing unknown forms of life), and ' Pro- terozoic' (containing the earliest forms of life). It should be noted, however, that the term Proterozoic has been already applied by Pro- fessor Lapworth to the faunas which we have called the Older Palaeozoic. Among the Algonkian groups of strata, the United States geologists include the Huronian of Logan and a series of strata which, in the Lake Superior region, appear to lie unconformably upon the Huronian, and have been called the Keweena- wan; other groups which ap- parently underlie the Huronian have received the not very eu- phonious names of Animike, Kee- watin, and Coutchiking; and simi- lar terms have been applied to locally developed pre- Cambrian de- 438 FOSSILS IN PRE-CAMBRIAN ROCKS [CH. xxvm. posits in other parts of the North American continent. The relics of living beings in the pre-Cambrian stratified rocks (Algonkian or Agnotozoic of Ameri- can authors) are of a very frag- mentary and often doubtful charac- ter. Besides the obscure annelid markings found in our own Torri- donian fragments, fossils doubtfully referred by Walcott to Lingula, Discina, Hyolithes, and Stromato- pora, with traces of Trilobites, have been found in pre-Cambrian strata of the Grand Canon of the Colo- rado. In Minnesota a Lingula- like shell has been found in beds of similar age, and tracks of organic origin, with other obscure indica- tions of living beings, have been found in pre-Cambrian strata near Lake Superior and in Newfoundland. In India, geologists have given the names of the Gwalior system, the Dha"rwar system, the Bijawar system, the AraValli system, the Cuddapah system, and the older Vindhyan system to masses of more or less altered beds, containing scarcely any traces of organisms, which in some cases can be shown to underlie the Palaeozoic rocks. Noetling has recently described The pre-Cambrian strata of Great Britain will be found de- scribed in detail, and their relatioi s discussed, in Papers in the ' Quart. Journ. Geol. Soc.' by Dr. Hicks, Dr. Callaway, and Professor Blake. The relations of the Cambrian and pre-Cambrian strata of the North- west of Scotland to one another were long the subject of controversy, and papers on the subject will be found in the same journal by Sir R. Murchison, Prof. Nicol, Sir A. a series of strata as underlying beds containing Olenellus in North- West India. He confirms the con- clusions of Waagen that this series of strata, containing fossils, named by the latter as Neobolua Warthi, N. Wynnei and Hyolithen Wynnei with Stenotheca, and various re- mains of Annelida, is really of older age than the Lowest Cambrian with Olenellus. If these conclusions be substantiated, we have probably in- dications in this district of a new system of fossiliferous strata of greater antiquity than the Cambrian. In Brittany, Barrois and Cayeux have described the pre-Cambrian rocks of that county as containing great numbers of shells of Radio- larians, Foraminifera, and Sponges. If the organisms they have de- scribed as belonging to these groups are rightly referred to those three divisions of the animal king- dom, it is remarkable that the oldest Protozoa and Sponges were of much smaller dimensions than those of the Palaeozoic and overlying rocks. Future discoveries may, it is hoped, lift the veil of mystery which still envelopes the life-history of the oldest known sedimentary rocks of the earth's crust. Geikie, Sir A. Ramsay, Prof. Hark- ness, Dr. Hicks, Dr. Callaway, Prof. Bonney, Messrs. Peach and Home. For a summary of the various opinions put forward by different authors the student is referred to a paper by Mr. Hudler ton, ' Proc. Geol. Assoc.,' 1878, and to the Ad- dress to the Geological Section of the British Association at Aberdeen, 1885. See also Prof. Lapworth's ' Secret of the Highlands,' ' Geol Ma.,' 1883. OH. xxix.] THE GEOLOGICAL EECOKD 489 CHAPTEK XXIX GENERAL REVIEW OF THE SUCCESSION AND CHARACTERS OF THE SEDIMENTARY ROCKS Fossils not found uniformly distributed in Sedimentary Formations Imperfection of our Knowledge of Freshwater and Terrestrial Con- ditions during past Geological Times Existence of Organisms before Cambrian Times Illustrations of the great Imperfection of the Geological Record ' Time-ratios ' of the Geological Eras Date of Appearance of different Forms of Life as modified by new Discoveries of Fossils General Order in which Life-forms have appeared upon the Earth Groups of Animals and Plants which have predominated in successive Periods Synthetic Types Specialised Types Per- sistent Types Summary of Palaeontological History Table of Fos- siliferous Sedimentary Formations. Variations in the Number of Fossils found in difl'erent Formations. It will be seen from the foregoing chapters that as we go backwards in time the records of the changes which have taken place, both in the earth's crust and in the animals and plants which have inhabited it, become more and more fragmentary and obscure. In this respect the history of the earth resembles that of the human race. Marine Strata more frequently preserved tban Fresh- water or Terrestrial Deposits. The Cainozoic strata include deposits of marine, freshwater, and terrestrial origin, and the forms of vegetable and animal life which existed while these strata were being deposited are almost as well known to us as those of the present day. In the case of the plants and Invertebrata, most of the Cainozoic fossils can be referred to existing genera. The Vertebrata of the Cainozoic, however, differ greatly from existing forms, and the farther we go back in the Tertiary series, the more remarkable and anomalous are the forms of mam- malian and reptilian life which are found in the strata, while the actual proportion of invertebrate forms still living steadily diminishes. But while it is true that the younger strata are, as a general rule, much more highly fossiliferous than the older ones, there are many exceptions to this rule. Formations con- taining beds of limestone, like the Carboniferous and Jurassic, may yield many more fossils than those in which calcareous beds are wanting. It can be shown, in innumerable cases, that strata which must once have been crowded with fossils now exhibit only few and obscure traces of organisms. Between the Tertiary and Cretaceous strata we have evidence of a very great break ; for in the Cretaceous system almost 440 IMPERFECTION OF THE BECOED [CH. xxix. every one of the Tertiary species is seen to be absent, and in the place of the familiar types of plants and animals we find wonderful assemblages of strange and curious forms. In the Jurassic and the Triassic almost all the forms that inhabited the seas are different from those of the Cretaceous and from one another. Although the Mesozoic systems contain some inter- calated strata of freshwater origin, yet only few and imperfect traces of the terrestrial life of those vast periods remain for our study. The Newer-Palaeozoic rocks still exhibit alternations of marine and freshwater strata ; but the marine faunas and floras are far better known than the freshwater one. In the Coal- measures we are presented with the earliest important record of a terrestrial flora. When we reach the Older-Palaeozoic rocks, all relics of freshwater and terrestrial life are wanting, though marine forms of life are well represented. In passing from the Silurian to the Ordovician, and from the latter to the Cambrian, the number and variety of the marine types rapidly diminish, though in the earliest of the Cambrian faunas all the great groups of invertebrate life are still represented. The terrestrial flora of Older Palaeozoic times is practically unknown to us. Existence of living 1 Beings before the Cambrian Period. The pre-Cambrian stratified rocks include masses of sedi- ments, which may not improbably rival in thickness the whole of the fossiliferous formations. Yet the only forms of life as yet detected in them are a number of very minute Protozoa (Radiolarians and Foraminifera), with some Brachiopoda and Pteropoda and obscure tracks and markings, indicative of the existence of other forms of life, but not sufficiently definite to reveal the real nature and character of the organisms. Judging from the nature and degree of development of the oldest known Cambrian fossils, periods of time must have elapsed between the first appearance of life on the globe and the commencement of the Cambrian Epoch, at least as vast as those which separate the Cambrian fauna from that of the present day. Imperfection of the Geological Record. We have seen that the series of stratified rocks must not be looked upon as containing a complete and unbroken series of records of the earth's past history. In many cases, indeed, the gaps in the succession of strata must represent periods of time of vaster duration than those represented by the thickest masses of strata themselves. Dr. R. D. Eoberts has endeavoured to give some idea of the fact that the geological record in our islands is a most imperfect and fragmentary one. In the adjoining woodcut, taken from DIAGRAM SHOWING THICKNESSES OP STRATIFIED DEPOSITS IN EUROPE WITH BREAKS IN THE SERIES. Scale 1 inch = 16,000 feet. Break {Lower Old red Ledbury Shalea.&c Ludlow Ireah j Invertebrates. 1 1 I 3 jiil 1 E < " 'Ji -j ! t nca < JO QCQ W Pkfe 00 TERTIARY ! CRETACEOUS JURASSIC 1 1 TRIAS 1 PERMIAN CARBONIFEROUS ! DEVONIAN | SILURIAN ORDOVICIAN CAMBRIAN Al Protozoa A 2 Coalenterata A3 Echinodermata A 4 Arthropoda A 5 Mollusca (with Molluscoida) B 1 Insects B 2 Terrestrial and Fresh- water Mollusca C 1 Fish C 2 Teleosteans D 1 Amphibians D 2 Reptiles D 3 Dinosaurs E Birds F 1 Non-placental Mammals F 2 Placental Mammals G 1 Plants G 2 Dicotyledons and Palms 446 PREDOMINANT TYPES [CH. xxix. abounded in earlier periods of the earth's history to those forms in which these characters are found separated from one another in distinct species or genera (' specialised types '). While many of the forms of life show such remarkable and constant changes, others (like Nautilus and Lingula] lived on through the geological periods with but little change, and these we speak of as ' persistent types.' Although new discoveries may modify our views concerning the exact period at which certain groups of the Vertebrata made their appearance on the earth's surface, as shown in the table, it is not likely that any new facts which may be learnt by future research will seriously modify our conclusions concerning the order of those appearances. There is clear evidence that the general rate of change among the Vertebrates was more rapid than in the case of the Invertebrates ; and in the higher Verte- brates (Mammalia and Aves) it was more rapid than in the lower ones (Reptilia, Amphibia, and Pisces). Among the Vertebrates, as among the Invertebrates, we find remarkable synthetic or generalised types constituting the earlier represen- tatives of each group, and these ave followed by more specialised form?, gradually approximating in structure to those which are now living. A few Vertebrates, like Ceratodus among the fishes and the Rhyncooephalians among the reptiles, may be considered to be persistent types. Predominance of certain Types of Animal and Vege- table litte at particular Periods of the Earth's History. Although doubt must always exist as to the exact time of the appearance on the earth of particular forms of life, nothing can be more certain than the fact that during successive periods of the earth's history different groups of animals and plants attained a wonderful development, and characterised the epoch by their numbers and variety of forms. It is equally clear that the dominant types of each succeeding period belong to groups of higher and higher organisation. The Older Palaeozoic rocks yield few forms of life, except those of the Invertebrata, and among these the Graptolites, the Brachiopoda (especially curious inarticulate forms) which altogether outnumber the rare Lamellibranchiata and the remarkable Trilobita are especially conspicuous. In the Newer Palaeozoic period we find the Corals, Echinodermata, the articulate Brachiopoda, with the anomalous Stromatoporoidea and Monticuliporida, existing in great num- bers. The Graptolites have disappeared, and the declin- ing Trilobita are replaced by forms of the Eurypterida, the Xiphosura, and Crustaceans. The Cystoidea are replaced by the Blastoidea, while the Crinoidea and other groups of the CH. xxix.] SYNTHETIC TYPES 447 Echinodermata attain a very striking development. What is a most remarkable fact, however, about the life of the Newer Palaeozoic era, is the abundance and variety of the forms of fishes of that early period, while the closely related Amphibians also make their first appearance. The Mesozoic era is distin- guished by the appearance of many Sponges, Corals, and Echinodermata much more closely related in their structure to those of living forms than are those of Palaeozoic times. The Brachiopoda lose their overwhelming predominance, and many living genera of Lamellibranchiata and Gastropoda make their appearance in great numbers. The most noteworthy peculiarity of the Mesozoic era, however, is the profusion and variety of the forms of life known as Ammonites and Belemnites, and the replace- ment of the Palaeozoic Arthropoda (Trilobita and Eurypterida) by forms not very dissimilar to those which now exist. Among the Vertebrata, Fish and Amphibians lose their predominance, and the Eeptilia acquire a wonderful development. Instead of the four or five orders of the present day, we find the Reptilia represented by nearly twenty orders (see Appendix C.), and the reptiles of the period are remarkable alike for their singularity and variety of form, and for the enormous dimensions which they attained. Among the Reptilia were singular bird-like forms (Dinosauria) and equally remarkable mammal-like types (Theriodontia) ; but true birds and mammals all apparently belonging to lowly and synthetic types made their appear- ance during Mesozoic times. The Mesozoic was the ' Age of Reptiles ; ' the Cainozoic ' the Age of Mammals.' As the Mesozoic reptiles of aberrant forms disappeared, the mammalia in great numbers and often of vast size came into exist- ence. The earliest forms were synthetic types, but, as we trace them through succeeding periods, the specialised types (like camels, horses, and elephants) appear, and gradually acquire their distinctive and peculiar characters. The Inverte- brata of the Mesozoic era differ far less from those of the present day than do the Vertebrata. Ammonites and Belemnites disappear, the Brachiopoda decline in numbers and become subordinate to the Lamellibranchiata, and the existing genera and species appear in ever-increasing numbers, as we follow the succession of the Tertiary strata. What is true of the animal life of past ages is equally true of the plant life. Of the marine algae excepting those rare forms which have a calcareous skeleton our knowledge is neces- sarily limited. The oldest terrestrial flora known is that of the Newer Palaeozoic rocks. Making every allowance for the fact that the remains of plants found are usually those growing in 448 PALJEONTOLOGICAL HISTORY [CH. xxix. marshy situations, and that hence they may not fairly represent the entire plant-life of the period, the Carboniferous flora is a very remarkable one. The abundance and enormous size of the Cryptogams are very striking phenomena ; and still more won- derful is the fact that at this early period these Cryptogams, whether allied to the recent Filices, Lycopodiacese, or Equiseta- ceae, all exhibit the exogenous mode of growth now found almost alone in the Phanerogamous plants. The Cycads and Conifers, and curious forms possibly intermediate between them and the Cryptogams, which existed in considerable numbers in Newer Palaeozoic times, became still more abundant, and constituted the dominant forms of vegetation in the Mesozoic era. But during the Cretaceous we witness the incoming of the existing flora, Cryptogams and Gymnosperms declining in numbers and size, and being replaced by the Angiosperms, both Monocotyledonous and Dicotyledonous. It is interesting to notice that the epochs which mark great changes in the terrestrial flora do not coincide with those which witnessed the great changes in the marine fauna. The number of groups of animal and plant life which have become extinct during past geological times, and their propor- tions to those now living on the earth, are illustrated in Appen- dices B and C. Summary of Palaeontologlcal History. A review of the facts which have been ascertained concerning the appearance and disappearance of the forms of life during past geological periods leads us to the following conclusions. 1. The species of animals and plants die out or disappear, one by one, in consequence of the conditions for their existence becoming unfavourable, or from their failure to maintain a competition with other forms. Many examples of species that have certainly become extinct in historical times ar? known such as the Great Auk, the Dodo, and Steller's Sea-cow. Great numbers of individuals may be destroyed by 'catastrophes,' such as earthquakes, volcanic eruptions, or floods, but no proof has ever been obtained of a species having thus become extinct. 2. The new forms of life which have been constantly coming into existence upon the earth during past geological times have appeared one by one. Great changes in the fauna and flora of a district can always be correlated with the lapse of long periods of time. When we have a continuous series of deposits, however, the new forms of life make their appearance ' as single spies, and not in whole battalions.' 3. The new forms of life that thus make their advent seem in all cases to be related and generally very closely related to forms that have preceded them. The supposed cases of the CH. xxi::.] AND ITS LESSONS 449 sudden appearance of types without any precursors break down upon rigid examination of the evidence. 4. Animals or plants of more complex organisation die out and are replaced by new forms more rapidly than those of simpler structure Vertebrates change more rapidly than Mol- lusca, and Mollusca more rapidly than Foraminifera. 5. During the later geological periods, ' life-provinces ' were identical with those of the present day ; but as we go backwards in time the limits of these provinces become less clearly defined ; and in all the earlier periods of the earth's history (Mesozoic and Palaeozoic), though there were life-provinces, these had no re- lation whatever to those of the existing flora and fauna. 6. As a general rule, the most highly specialised forms of life have made their appearance on the earth later than the less specialised. Many of the older forms are what naturalists call ' synthetic types,' and exhibit, in combination, characters now displayed only in different species, genera, families, or orders. 7. There are certain cases like those of the horses (see p. 178), the camels, the elephants, and other highly specialised groups in which ancestral forms have been discovered in suf- ficient numbers to enable us to trace out with tolerable accuracy the general line of their descent, and the successive modifica- tions by which these remarkable types have assumed their peculiar characters. 8. On the other hand, there are undoubtedly many remark- able groups of animals and plants, both living and extinct, con- cerning which there is at present no palseontological evidence available which would enable us to trace their probable descent from pre-existing types. This, however, is no more than we might expect if we bear in mind the necessarily imperfect cha- racter of the geological record (Note W, p. 608). 9. Hence it must be conceded that, with respect to a large proportion of the known forms of animal and plant life, it is impossible to construct ' genealogical trees ' on the basis of palaeontological evidence. 10. But in spite of the fact that the chance of finding ancestors in the direct line of descent for living species is often a remote one, yet the evidence afforded of the existence of forms collaterally related to them is sometimes of very great value if it be rightly interpreted. 11. Although types which serve to bridge over gaps in our series of existing life-forms seem sometimes to arise, without any forerunners that can be regarded as linking them with pre- existing groups, yet instances of this kind often disappear and G G 450 TABULAR VIEW [CH. XXIX. become more and more easily explicable as the result of further research and as new discoveries are made. 12. Much of the difficulty of tracing the descent of forms of life, from the study of palaeontological evidence, arises from the imperfect preservation of fossil types, and the consequent impossibility of making complete comparisons with living types. Of the actual relations of the soft parts of the Graptolithida, Stromatoporida, Monticuliporida,&c., with those of living groups, the evidence is unfortunately altogether wanting. Such being the facts of the palaeontological history, it re- mains for the zoologist and botanist to find their explanation, and to say with what theory or theories of the origin of species that history is most consistent. TABULAR VIEW OF THE FOSSILIFEBOUS STRATA. SHOWING THE ORDER OF SUPERPOSITION OR CHRONOLOGICAL SUCCESSION OF THE PRINCIPAL GROUPS, WITH REFERENCE TO THE PAGES WHERE THEY ARE DESCRIBED IN THIS WORK. SYSTEMS BRITISH DEPOSITS I Clyde marine strata, with canoes Danish I (p. 164) Lake-dw River gravels of the South of (p. 240 W England (p. 161) Dordogii (p. 239 pq Ho Cavern deposits of Kent's Hole, Brixham, &c. (p. 160) Champla Arneri Older v earths OH Glacial drift of Scotland and the Loess of OQpH North of England (p. 165) Deposits E EH Erratics of Chici.ester, &c. (p. 159 Australi. (p. 168) bones (p. 241 ^ Glacial drifts with marine shells of Glacial < Moel Tryfaen, &c. (p. 168) (.p. 237 Glacial d Glacial formations of East Anglia (p. 166) (p. 245 FOREIGN DEPOSITS Danish Kitchen-middens (p. 159) Lake-dwellings of L vvitzerland (p. 240) Dordogne caves reindeer period (p. 239) Cham plain period of North America (p. 245) Older valley gravels and brick- niens (p. 161) e (p. 162) Deposits in caverns of Liege, &c. (p. 159) Australian cave-breccias with bones of extinct marsupials (p. 241) Glacial drift of Northern Europe (p. 237) Glacial drift of the Alps (p. 238) North America CH. xxix.] OF THE FOSSILIFEROUS STRATA 451 SYSTEMS BRITISH DEPOSITS FOREIGN DEPOSITS H Forest-bed of Norfolk cliffs Marine beds at base of Etna (p. 183) (p. 233) ll Chillesford sands and clays (p. 184) Norwich crag (p. 185) Sicilian strata (p. 233) Lacustrine strata of the Val d'Ar- ^ H O no (p. 234) [227, 229) rt H H Red crag (p. 185) German and French pliocene (pp. "^ O ^ White crag (p. 187) Diestien and Antwerp crags ^ rn ^ (p. 228) [232) g* Stone-bed at base of crags (p. 189) St. Erth beds (p. 190) Lenham beds (p. 189) Subapennine marls and sands (p. Pliocene of North America (p. 244) Beds of Pikernii, Greece (p. 236) EH o Strata of Sivaliks, India (p. 236) QJ E] Faluns of Touraine (p. 226) gS Bordeaux (p. 227) Swiss beds of Oeningen (p. 231) 1? 51 H Wanting Marine molasse of Switzerland w 2 fes (p. 231) [235) S M H Congeria beds of Vienna basin (p. R Q Strata of Mayence basin (p. 229) B O Beds of Superga, Turin (p. 232) ~ H Marine Miocene of India, &c. (p. ^ 236) Wanting Upper Oligoceue of Germany (p. 228) Calcaire de la Beauce (p. 223) H Henipstead (Hamstead) beds Gypseous series of Montmartre hj (p. 203) (p. 222) [(p. 223) H Berabridge series (p. 204) Strata of the Lima^ne, Auvergne Q Brockenhurst series (p. 207) Aquitanian and lower molasse of o Switzerland (p. 230) Headon series (p. 206) Rupelian and Tougrian of Belgium H (p. 225) [(p. 228) ^v |^] Brown coal of the Lower Rhine go Septaria clays and marine beds of H Egeln (p. 228) PI Deposits of Vienna basin (p. 235) d5 Croatian brown coal (p. 234) O Nari series of India (p. 236) PI Marine gypseous series (p. 222) ^ Barton sands and Barton clay Calcaire de St. Ouen (p. 222) . (p. 207) Gres de Beauchamp (p. 222) Jx Wemmelian beds of Belgium PH (p. 222) [(p. 242) <1 Uinta group of North America 1 Bracklesham beds (p. 208) Bournemouth beds (p. 210) Calcaire grossier of France (p. 221) Laekenian and Bruxellian of Bel- H gium (p. 222) R* Bagshot beds (p. 213) Arctic leaf-beds (p. 201) P Bovey Tracey beds (p. 212) Nummulitic limestone of Europe, 05 Asia, and Africa (p. 229) 1 London clay (p. 214) Bridger group of North America (p. 242) OH Oldhaven beds (p. 216) Woolwich and Reading series Sables de Cuise (p. 221) Paniselian, Ypresian, Landenian, \J R (p. 217) and Heersian beds of Belgium M (p. 221) Thanet sands (p. 218) Argile plastique (p. 221) Zeuglodon beds of North America (p. 242) P Wahsatch beds of North America (p. 242) Montian beds of Belgium and France (p. 220) TABULAR VIEW [CH. xxix. SYSTEMS BRITISH DEPOSITS Wanting Upper chalk (p. 254) Middle chalk (p. 261) Lower cha'k with chalk marl and Upper Greensand (p. 262) Blackdown beds (p. 261) Gault (p. 264) Sands of Folkestone, Sandgate, and Hythe (p. 266) Atherfield clay (p. 267) Tealby series of Lincolnshire (p. 268) Speeton clay of Yorkshire (p. 268) Punfield beds (p. 274) Wealden clays and Hastings sands (p. 271) FORKIGX DEPOSITS Laramie beds of North America (p. 337) Maestricht and Faxoe beds (p. 327) Pisolitic limestone (Danian of France) (p. 327) Senonian of France (p. 249) Turonian of France (p. 249) Cenomanian of France (p. 249) White chalk of Sweden and Russia (p. 254) Sands of Aix-la-Chapelle (p. 326) Quader sandstein and Planer-Kalk of North Germany (p. 326) Hippurite limestone of South France (p. 330) Sands and clays of New Jersey, U.S. (p. 334) Series of Western United States (p. 335) Aptian, Urgonian, and Neocomian of Europe (p. 331) Wealden of Hanover (p. 326) Upper, middle, and lower Purbecks (p. 286) Portland stone and sand (p. 292) Kimeridge clay (p. 292) Coral rag and calcareous grit (p. 294) Oxford clay (p. 295) Kellaways rock (p. 296) Great or Bath oolite (p. 296) Stonesfield slate (p. 299) Fuller's earth (p. 301) Inferior oolite and sand (p. 301) Upper lias sands and clay (p. 304) Marlstone (p. 304) Middle lias clay (p. 305) Lower lias clays and limestone (p. 305) Zone of Avicula contorta (' Penarth beds ') (p. 308) Marls with Exogyra virgula of Argonne (p. 325) Lithographic slate of Solenhofen (p. 283) Nerinean limestone of the Jura (p. 295) White Jura (p. 326) Brown Jura (p. 326) Lias or black Jura (p. 326) Rhastic (Kossen beds) (p. Keuper or Upper New Red sand- stone, &c. (p. 317) Red shales of Cheshire and Lanca- shire, with rock-salt (p. 321) Dolomitic conglomerate of Bristol (p. 319) Wanting Banter or Lower New Red sand- stones of Lancashire, Cheshire, &c. (p. 320) Keuper beds of Germany, &c. (p. 324) St. Cassian and Hallstadt beds, with rich marine fauna (p. 328) Coalfields of Richmond (Virginia) and Chatham (N.C.) United States (p. 333) Muschelkalk of Germany (p. 324) Bunter-Sandstein of Germany, &c. (p. 324) CH. xxix.] OF THE FOSSILIFEROUS STRATA 453 SYSTEMS BRITISH DEPOSITS Upper Permian of Cumberland (p. 343) Middle Permian, magnesian lime- stone, and marl slate of Durham and Yorkshire (p. 343) Lower Permian sandstones and breccias of Penrith, Dumfries, shire, &c. (p. 347) FOREIGN* DEPOSITS Dark-coloured shales of Thuringia, &c. (p. 393) Zechstein or dolomitic limestone (p. 393) Mergelschiefer and Kupfer- schiefer (p. 393) Roth-todt-liegendes of Thuringia (P. 393) Magnesian limestones, &c., Russia (p. 393) Sandstones of Artinsk (p. 393) of Coal-measures of South Wales, &c. (p. 349) Coal-measures of Midlands and North of England (p. 367) Flat coals of Scotland (p. 368) Millstone grit (p. 368) Yoredale series of Yorkshire (p. 369) Tuedian coal-measures of North- umberland (p. 349) Carboniferous limestone and shale of England (p. 369) Carboniferous limestone and carboniferous slate of Ireland (p. 370) Edge coals of Scotland (p. 370) Calciferous sandstone series of Scotland (p. 370) Coalfields of France, Belgium, and Germany (p. 392) Fusulina limestones (p. 393) Upper and lower Coal -measures of the United States (p. 394) Productus limestones of Russia (p. 393) oal- Coal-measures of Russia (p. 393) Conglomerates, limestones, and shales of Appalachians (p. 394) Pilton group of North Devon (p. 386) Petherwyn beds of Cornwall (p. 386) Yellow sandstones of Dura Den (p. 389) Kiltorcan beds of Ireland (p. 390) Ilfraccinbe beds a id limestones of Torquay and Plymouth (p. 386) Flagstones of Forfarshire (p. 389) Bituminous schists of Gamrie, Caithness, &c. (p. 389) Sandstones and slates of the Fore- land and Lynton (p. 386) Sandstones and cornstonesof Here- fordshire (p. 390) Clymenien-Kalk and Cypridinen- Schiefer of the Eifel (p. 391) Goniatite beds of the Eifel (p. 391) Limestones of Eifel with under- lying Calceola schists (p. 392) Catskill, Chemung, and Portage beds of North America (p. 393) Devonian strata of Russia (p. 392) Hamilton, Helderberg, and Oris- kany strata of North America (p. 393) Sandstones of Gaspe (p. 393) 454 TABLE OF STRATA [dr. XXIX SYSTK.MS BRITISH DEPOSITS FOREIGN* DEI-OMTS SILT7BIAN Upper Ludlow formation, Tile- stones, Downton sandstone and bone-beds (p. 405) Lower Ludlow formation, Aymes- try limestone (p. 406) Wenlock limestone and shale (p. 407) Woolhope limestone and grits (p. 408) Tarannon shales and Denbighshire grit (p. 408) Upper Llandovery or May Hill sandstones (p. 409) Lower Llandovery beds (p. 409) Gothland limestone of Scandi- navia (p. 430) Onandaga salt group of North America (p. 430) Etage E of Bohemia (p. 429) Niagara Limestone of North America (p. 430) Pentamerus limestones and shales of Scandinavia and Kussia (p. 430) Clinton and Medina sandstones of North America (p. 430) OBDOVICIAN Bala limestone and Caradoc beds (p. 415) L'.andeilo beds (p. 417) Arenig or Stiper-stones group (Lower Llandeilo of Murchison) (p. 418) Trinucleus shales and Cystidean limestone of Scandinavia (p. 430) Graptolitic shales of Scandinavia. Etage D (d a , d,, d 4 , d,) of Bohemia (p. 429) Orthoceras limestone of Scandi- navia (p. 430) Hudson river, Cincinnati, Trenton, and Chazy beds of North America (p. 431) CAMBBIAN Durness and Girvan limestones of Scotland (p. 428) Tremadoc slates (p. 425) Lingula flags (p. 426) Menevian beds of Wales (p. 426) Harlech grits and Llanberis slates (p. 427) Comley sandstones and Olenellus beds of Scotland (p. 427) Calciferous sandstones of North America (p. 430) Alum shales of Scandinavia (p. 430) 'Primordial' beds (Etage C) of Bohemia (p. 429) Potsdam sandstones of North America (p. 430) Olenellus beds of Scandinavia (p. 430) Olenellus slates of Georgia, &c., North America (p. 430) Pebidian beds of Wales (p. 434) Arvonian (?) beds of Wales (p. 434) Dimetian beds of Wales (p. 434) Uriconian beds of the Midland (p. 435) Longmyndian strata of the Mid- lands (p. 435) Torridonian strata of Scotland (p. 435) Lewisian strata of Scotland (p. 434) Huronian strata of Noi ch America (p. 436) Keeweenawan strata of North America (p. 437) Animike, Keewaten, and Cout- chiking beds of North America (p. 437) G walior, Dhwanvar, Bigawar, Aravalli, and Cuddapah beds of India (p. 438) Upper Laurent! an and Lower Laurentian of North America (P- 437) 455 PAET III VOLCANIC ROCKS CHAPTEE XXX VOLCANIC ROCKS, THEIR NATURE AND COMPOSITION Relation of volcanic Rocks to the sedimentary liypogene Rocks Nature of Action taking place at Volcanic vents Lavas and their Varieties Fragmental materials ejected from Volcanoes Scoriae, lapilli, dust, pumice, bombs Formation of volcanic Tuffs Alteration of volcanic Rocks by solfataric and atmospheric agencies Chemical composition of lavas Acid, intermediate and basic lavas Rhyolites and Soda- rhyolites Andesites, Trachytes, Phonolites and Tephrites Alteration of Andesites Propylites and Porphyrites Basalt and Melaphyres Tachylytes and Variolites Basaltic and Palagonite Tuffs. Relations of Volcanic Rocks to those of other classes. The aqueous or fossiliferous rocks having now been described, we have next to examine those which may be called volcanic in the most extended sense of that term. Suppose a a in the a. Hypogene formations, plutonic and metamorphic. b. Aqueous formations. c. Volcanic rocks. annexed diagram to represent the crystalline formations, such as the granitic and metamorphic ; b b the fossiliferous strata ; and c c the volcanic rocks. These last are sometimes found, as was explained in the first chapter, breaking through a and 6, sometimes overlying both, and occasionally alternating with the strata b b. Nature of Volcanoes and of Volcanic Action. Volcanoes are apertures in the earth's crust, through which various materials, usually in a highly heated condition, find their way to the surface. The substances thrown out of volcanic vents are 456 VOLCANIC ACTION [CH. xxx. sometimes in a gaseous condition, sometimes liquid, and at other times solid. The gases given off by volcanoes are chiefly water-gas or steam, sulphurous acid, hydrochloric acid, and (during the later stages of the history of a volcano) carbon dioxide ; but many other substances, such as boric acid, hydrofluoric acid, ammonium chloride, and various metals and metallic sulphides, in a vaporised condition, also escape from volcanic orifices. The chief liquid thrown from volcanic vents is water, when the temperature is not so high as to convert it into steam. Many hot and mineral springs are clearly connected with volcanic activity within the earth's crust ; and, as shown by the late Mr. Robert Mallet, ' geysers,' in all their essential characters, are identical with explosive volcanoes, though hot water instead of molten Lava is thrown out from them. The water of geysers and hot springs often contains silica, calcium carbonate, and other materials in solution, and these substances are deposited around them. In most of the ordinary volcanoes, however, various kinds of rock, either in a molten or a solid state, are ejected and accumulate round them to form conical volcanic mountains, the vent remaining as an aperture or cup-shaped hollow ('crater') at the summit or en the side of the volcanic cone. Nature of lavas. When liquid, this ' lava ' (as the molten rock is called) looks like a red- or white-hot slag, but it usually gives off great quantities of steam and other gases, water being evidently imprisoned in the midst of the molten mass, and escaping into the atmosphere when the pressure is relieved by the lava reaching the surface. Some lavas are so liquid that they Fisr. 633. Ropy Surface ' of lava scream. flow like rivers over the surface of the earth ; and such lavas generally exhibit remarkably rough and cindery surfaces, due to the escape of steam and gases from them as they flow along. Other lavas are remarkably viscous, sometimes moving along like glaciers at the rate of a few inches a day; lavas of this type usually exhibit a smooth surface, which is often wrinkled and twisted so as to resemble coils of rope, * ropy surfaces ' (see fig. 633 and fig. 653, p. 471). Ejected Fragments. The solid materials thrown from volcanic vents consist of blocks of lava, sometimes compact, but CH. xxx.] VOLCANIC PRODUCTS 457 more frequently distended by gas so as to resemble a cinder (scoria). When the scoriae are small (about the size of a nut) they are called by the Italian name of lapilli, and when reduced to a granular or sandy condition they form ' puzzolana,' while when perfectly comminuted they are known as volcanic dust or volcanic ash. Volcanic scoriae are sometimes spoken of as ' cinders,' which in outward aspect they greatly resemble. Fine volcanic dust in the same way is often called ' ash ; ' but it must be remembered that these terms only indicate the general appearance, and not the origin of the substances. There is no real analogy between the pieces of half-burnt coal known as ' cinders ' and the masses of mixed silicates, which have been distended by gases, while they were in a fused condition, that we call scoriae ; equally little is there in common between the ' ash,' or incombustible residue left by the burning of coal, &c,, and the fine dust produced by the trituration of scoriae and pumice. Volcanic scoria and dust are so like cinders and ash in outward appearance, that it is almost impossible to avoid using these names for them ; it must always be remembered, however, that volcanic materials are not, like cinder and ash, products of combustion. There is, indeed, little or no burning taking place at a volcanic vent, nor does the action of a volcano depend on combustion. The red glow above a volcanic vent is due to reflection from the clouds of steam and dust above the crater of the surfaces of glowing lava within it. The loud rumbling sounds, the trembling of the ground, the intense darkness, the lightning flashes, and the heavy falls of rain which accompany and follow violent volcanic eruptions are all consequences of the escape of great masses of watery vapour from the midst of masses of molten rock in which it has been occluded, and the ejection of fragments of lava by the agency of this escaping steam. A few inflammable gases, it is true, escape and take fire on reaching the outer air, but these are not highly luminous, and ' flames ' are never conspicuous in a volcanic outburst. Round or fusiform masses of lava, partially distended by gas, which have assumed a more or less regular form by rotation during their flight through the atmosphere, are known as volcanic bombs. These must not, however, be confounded with the fragments of scoria coated with lava, over which they have rolled, these being known to geologists as pseudo-bombs. "When a lava is glassy and becomes distended by gas, it forms the well-known material called ' pumice.' Sometimes the vol- canic glass is drawn out into delicate threads like the ' Pele's hair ' of Hawaii, or it may give rise to the beautiful material 458 SCOKIACEOUS LAVAS [CH. XXX. Fig. 634. described by the late Professor Dana as occurring in the same district and known as ' thread -lace scoria.' Scoriae which have been buried in the earth's crust often have their cavities or steam-holes filled with various minerals, these having been formed by the solvent action of water permeat- ing the substance of the lava. Such rocks are said to exhibit an amygdaloidal structure, from the Greek word amygdalon, an almond (see fig. 634). Rocks of this kind, indeed, some- times very closely resemble the well-known sweetmeat known as ' almond-hardbake.' The substances which fill up the cavities in these amygda- loidal rocks are usually opal, quartz, calcite, or the various crystallised hydrous silicates known as ' Zeolites.' Besides lava in various forms, volcanoes frequently Scoriaceons lava in part converted into an amygdaloid by infilling of its cavities. Montague de la Veille, Department of Puy- de-D6me, France. discharge fragments of rock torn from the sides of their vents, often at great depth from the surface. Such ejected blocks may be of aqueous origin and contain fossils ; but they are often much altered and sometimes have become completely crystalline in consequence of their contact with the masses of molten lava. Volcanic Tuffs. The loose materials ejected from volcanoes often become cemented to form a more or less hard and solid stone. Of this class is the ' peperino ' of the Italian geologists, a light spongy rock often used as a building material, and made up of lapilli cohering to form masses that can be easily quarried. Scoriae, lapilli, puzzolana, pumice, and volcanic dust, when acted upon by atmospheric waters containing carbon dioxide, undergo chemical changes, the calcium silicate being converted into carbonate which acts as a cement to the whole mass, while in other cases secondary silicates are formed, which play the same part. The general name applied to rocks formed of coherent fragmented volcanic material is ' volcanic tufa ' or ' tuff,' which must of course be clearly distinguished from the calcareous tufa deposited by mineral springs. The ' trass ' of the Eifel district is a rock composed of lapilli and dust which when quarried and exposed to the air sets to form a very solid r. xxx.] PORPHYRITIC LAVAS 459 and useful building-stone. The fine dust of volcanoes, when mixed with water, often sets in the same way into a hard mud. The various tuffs and volcanic muds not unfrequently contain remains of plants and land- shells ; when deposited in the sea, they may enclose shells and other marine organisms. Lavas and fragmental materials about dormant and extinct volcanoes (solfataras) are often found greatly affected by such volcanic emanations as sulphurous acid, hydrochloric acid, carbon dioxide, &c., and in consequence of this solfataric action many of the minerals of which volcanic rocks are composed are found to be much altered or even converted into ' pseudomorphs,' while new substances such as quartz, chalcedony, opal, &c.,are found filling their cavities and fissures. Chemical Composition of lavas. Lavas when analysed are found to consist of mixtures of various silicates among the chief of which are the silicates of aluminium, calcium, magnesium, iron, sodium, and potassium ; but water and other compounds of hydrogen are almost invariably present also. In some cases, the proportion of silica in lavas is very high, from 66 to 80 per cent. and the rock is said to be an acid lava. In other cases the proportion of silica is low 55 to 40 per cent. and such a lava, in which the bases prepon- derate over the acid silica, is called a basic lava. Lavas in which the proportion of silica varies from about 55 to 66 are called by English geologists intermediate lavas (' laves neutres ' of French authors). Structure of Lavas. Lavas differ from one another greatly in structure or texture. Some lavas are almost destitute of crystalline matter and form glassy, vitreous, or hyaline masses. At other times the lava-rocks, while perfectly com- pact, display, instead of the 'vitreous lustre ' of glass, the ' resinous lustre ' of pitch ; such rocks are known as ' pitch stones.' Many other lavas exhibit a finely granular or stony appearance. Lavas often contain included crystals of felspar or some other mineral scattered through them (' phenocrysts ' of American geologists), and such rocks are said to have a porphyritic structure. Glassy and stony lavas may alike exhibit this porphyritic structure. The original porphyry of the an- cients (porfido rosso ant'ico) is an old andesitic lava in which the base or ground-mass has acquired by alteration a rich purple tint, while white felspar crystals are seen scattered through it (see fig. 635). The term ' porphyritic ' is now, however, applied to any rock containing large scattered crystals, without reference to its colour. When the base or ground-mass of a lava is studied in thin sections Fig. 635. Porphyry. White crystals (phenocrysts) of felspar in a dark purplish base. 460 STKUCTURES IN GLASSY LAVAS [CH. xxx. under the microscope, we find more or less glassy matter through which are scattered embryo and minute crystals (' crystallites ' and ' microlites '). Sometimes these microlites are found grouping them- selves to form skeleton crystals, the spaces around being left more or less clear, and forming ' courts of crystallisation ; ' at other times the Fig. 636. Crystallites, microlites, and skeleton crystals in glassy rocks. a. Globulites, &. margarites, c. trichites, d. belonites, e. f. g. fern-like aggregates of microlites from the pitchstone of Corriegills, Arran, h. skele- ton crystal of magnetite from a tachylyte or basalt-glass. microlites group themselves into spherical aggregates (simple or compound), usually minute but occasionally several inches or even feet in diameter which, are called ' spherulites ' (see fig. 638). When empty cavities, probably formed by contraction during crystallisation, exist in these spherulites. they are known as lithophyses' (hollow spherulites). The disposition of the porphyritic crystals, the micro- lites and the crystallites, in a lava often indicates, by their parallel arrangement, that the molten mass has been in motion after the development of these structures within it. The rock is then said to exhibit a fluidal or banded structure (see fig. 637). Movement is also indicated by the drawing out of gas-bubbles, which exist in Fig. 637. Fig. 638. Glassy rock exhibiting banded or fluidal, Glassy rock, partially devitrified. and and perlitic structure. showing spherulitic structure. almost all lavas, and have been formed by escaping steam ; such lavas, when glassy, are said to be ' pumiceous ' in structure ; they sometimes exhibit a beautifully satiny sheen (' Schiller '), due to the presence of these drawn-out cavities. Some glassy lavas are traversed by numerous fine cracks, straight and curved, and in con- sequence of these they often produce interference colours and some- times break up into small round particles. Such lavas are said to CH. XXX.] &HYOLITES, ETC, 461 have a 'perlitic' structure, and they are then spoken of as per- lites (see fig. 637). The ' axiolitic ' structure of Zirkel appears to arise when spherulites tend to form along the sides of perlitic and other cracks in glassy rocks. When minute ' lath-shaped ' microlites of felspar are entangled in a mass of glass the ground-mass of the lava is said to be a ' microlitic felt ' (see figs. 641, 642) ; when somewhat larger crystals of felspar are enclosed in augite or some other mineral, the lava is said to have an ' ophitic ' (diabasic) structure (see fig. 688, p. 518) ; the fracture along cleavage-planes of a mineral enclosing others in this way gives rise to the appearance known as ' lustre mottling ' in rocks. Acid Lavas, Rhyolites, &c. The lavas of acid composition are now generally spoken of by geologists as Rhyolites ; occasionally, however, the terms Liparites and Quartz-trachytes are applied to them. Forms of these lavas which have undergone partial recrystal- lisation (secondary devitrification, resulting from hydration, kaolini- sation, and other chemical changes) are spoken of as ' quartz-felsites, Pig. 039. Fig. 640. Obsidian from Iceland, showing flow structure in the parallel arrangement of the microlites and incipient spheru- lites formed of aggregates of crystal- lites. Rhyolite from Hungary, with a devitrified spherulitic base. The clear rounded crystals are quartz, containing glass and stone cavities. The rectangular crystals are felspar (orthoclase). ' porcellanites,' ' halleflintas,' 'pyromerides,' ' porphyroides,' &c. The rhyolites are usually pale in colour (white, grey, or pink), and they have a low specific gravity (varying from 2*6 in the stony varieties to 2-3 in the glassy forms). They much more usually exhibit glassy varieties than do any other classes of lavas. When crystallised they always contain quartz, in distinct crystals or in corroded and rounded grains, with porphyritic crystals of orthoclase ; more rarely plagioclase, biotite, hornblende, or augite ; magnetite and apatite also occur, and, as accessories, garnet, cordierite, zircon and other minerals. Olivine, however, is very rarely, if ever, found in the rhyolites. The most stony rhyolites, or those in which the glass is not con- spicuous, are known as ' lithoidites ' by the American petrographers (see fig. 640). A form in which the porphyritic crystals are so large and numerous that the rock resembles at first sight a granite, is called ' Nevadite.' Many of the stony rhyolites exhibit beautifully porphyritic and spherulitic structures. When the rock is fine-grained and com- 462 ANDESITES, TRACHYTES [CH. xxx. pact it is often called ' hornstone.' Varieties exhibiting the lustre of pitch or resin are known as pitchstone (' retinite ' of French authors), while those in which the ground-mass is perfectly vitreous or glassy are called Obsidian (see fig. 639). Pitchstones and Obsidians exhibit the banded, fluidal, spherulitic, perlitic and pumiceous structures in great perfection. There is often a complete passage from the most stony or highly crystalline forms of rhyolite, through compact and resinous types to the glassy obsidian, and its frothy form, pumice. Fragments of these materials accumulate to form rhyolite- and pumice4uffs, which often contain opal and tridymite. Most rhyolites contain a large proportion of potash as compared with soda, but in some cases the proportion of soda is higher than the potash. Such lavas are called soda-rhyolites or ' quartz-pantelle- rites, from their occurrence in the Island of Pantelleria. They often contain, in addition to the quartz-crystals, ' phenocrysts ' of a soda- orthoclase (anorthoclase) or soda-microcline, and of a soda-horn- blende (.Enigmatite or Cossyrite). As the rhyolites more readily assume the glassy condition than any other class of lavas owing to the large proportion of alkalies which they contain the special characteristics of glassy rocks are peculiarly well exhibited by them. It is among the rhyolites that we find the most striking examples of spherulitic and perlitic structure, while it is the same class of rocks that furnish the finest and most perfect varieties of ' pumice.' When the alumino-alkaline silicates of the rhyolites are acted upon by sulphurous acid, various hydrous sulphates are formed ('alunite,' &c.), and the altered rock is known]as ' alumstone ; ' this rock, when roasted and washed, yields crystals of alum. In Hungary and other districts, extensive deposits of alumstone are found which have evidently been produced . by solf ataric action on ordinary rhyolites. By the action of atmospheric agents (carbon dioxide and water) on rhyolites, various stages of alteration and decomposition are brought about, and different names have been applied to these altered forms. In this country many of the altered rhyolite lavas are known as felsites and quartz-felsites ; rocks of this class with a spherulitic structure are called by the French geologists ' pyromerides ; ' when incipient foliation has been produced in them they have been called ' porphyroides,' and when this action has gone further uiey may be converted into ' quartz-schists.' Intermediate Lavas. The lavas of intermediate composition contain from 55 to 66 per cent, of silica ; they have as a rule a darker colour than the rhyolites and a density varying from 2-8 in the stony varieties to 2-5 in the glassy forms. They more rarely assume this glassy form, however, than do the rhyolites ; but many forms of pitchstone and obsidian, with spherulitic and perlitic varieties, and some pumices, are of intermediate composition. The lavas of inter, mediate composition usually contain quartz only as an accessory constituent, while olivine, though^occasionally present in them, cannot be regarded as one of their essential minerals. Andesites. By far the largest and most important group of the intermediate lavas is that known as the ' Andesites.' They are rocks composed essentially of crystals of plagioclase (soda-lime-felspar), with a pyroxene (augite or enstatite), hornblende, orbiotite and more or less glassy base. Usually the ground-mass of the rock exhibits the CH. XXX.] PHONOLITES AND TEPHKITES 463 aggregation of felspar microlites in a glassy base known as a ' micro- litic felt,' so that this structure is often spoken of as being typically andesitic. The Andesites f all naturally into two great groups, the hornblende- and biotite-andesites (see fig. 642), which incline towards the acid lavas, and the pyroxene-andesites (augite- and enstatite-andesites) inclining towards the basic lavas (see fig. 641). There are, however, curious links between these types rocks which contain at the same time biotite and enstatite, or hornblende and augite. Some andesites con- tain a very high percentage of silica, and when this crystallises out as quartz the rock is known as quartz -andesite or ' dacite.' These Pig. 641. Fig. 642. Augite-andesite, from Hungary. The white corroded crystals are felspar (plagioclase). In the upper part is seen a crystal of augite with its cha- racteristic cleavage. The grouud mass is a microlitic felt, and the rock contains much magnetite. Hornblende-mica andesite from Hungary. The colourless zoned crystals are felspar (plagioclase). In the upper part are seen two crystals of hornblende with characteristic cleavage. In the centre is a crystal of mica (biotite). The ground-mass is a microlitic felt. rocks may be regarded as belonging to the acid rather than the inter- mediate series. On the other hand, some of the augite- and enstatite- andesites are very dark-coloured, heavy rocks, and only differ from the basalts in not containng olivine. Besides the large and widely distributed group of the andesites. the intermediate class of lavas contains several other groups of rock of considerable interest. Trachytes. The name trachyte was originally given to all light- coloured lavas in opposition to the black basalts ; the term is now applied by petrographers to rocks consisting of crystals of orthoclase felspar (Sanidine) with subordinate plagioclase and hornblende, or augite set in a more or less glassy base (see fig. 648). Trachytes of ten contain as accessory constituents sodalite, melilite, olivine, and other minerals. The trachytes are far less common than the andesites, but are found not unfrequently not only among the products of recent volcanoes, but among those of earlier periods of the earth's history. Phonolites and Tephrites. When a rock of the trachy tic type contains considerable quantities of the felspathoids, nepheline, leucite, hauyn, or nosean, it is called a phonolite (' clinkstone ' of the old authors). The phonolites, though abundant in certain districts, like Bohemia and Central France, are not very widely distributed, In 464 INTERMEDIATE AND BASIC LAVAS [OH. xxx. this country, we find a good example of a phonolite only in the Wolf's Kock off the Land's End in Cornwall (see fig. 644), while some of the trachytes of Haddingtonshire contain small quantities of nepheline and approximate to phonolites in their structure and mineral con- stitution. Bocks of andesitic type, that is, with a plagioclastic felspar pre- dominating, which contain the felspathoids, nepheline, leucite, hauyn, &c., are known as tephrites. They form a class which is even less widely distributed than the phonolites. Fig. 643. Fig. 644. Trachyte from Ischia. The large colour- less crystals are orthoclase(sanidine) ; the dark crystals are hornblende. The base is formed of microlites of felspar with glass. Angite also is present. Phonolite from the Wolf's Roct, Cornwall. The long cracked crystals are sanidine, the clear square and hexagonal ones nepheline, and the dark-zoned ones nosean. Lavas of intermediate composition undergo changes much more easily than the acid lavas, owing to their smaller proportion of silica and their higher percentage of iron oxides. Andesites which have been altered by solfataric action are called propylites\ among trachytes which have been similarly altered we find the domites of Central France. When altered by the various atmospheric agents, the andesites are converted into the rocks known as porphyrites, though some of the rocks called by this name are plutonic rather than volcanic rocks. The phonolites, owing to the instability of their felspathoid constituents, are especially liable to alteration and disin- tegration. Basic lavas. The lavas of basic composition contain less than 55 per cent, of silica ; they are always dark and frequently even black in colour, and have a density varying from 2-9 in the stony varieties to 2'7 in their glassy forms. When quartz is present in these lavas, which is a rare occurrence, there is reason to believe that it has been caught up in the flowing mass, and is not an original constituent of the rock. The felspars present are lime-soda varieties (labradorite and anorthite) ; the ferro-magnesian silicate is usually augite ; a con- siderable amount of magnetite or titanoferrite generally occurs in them. In addition to the minerals already mentioned, we nearly always find some olivine present in the basic lavas ; and this mineral not unfrequently forms conspicuous granules, or even nodules of considerable size. Some authors, indeed, regard olivine as an essen- tial constituent of the basic lavas. CH. xxx.] BASALTS 465 Basalts, &c. The general name applied to these basic lavas is basalt, the term dolerite being reserved for the coarser-grained Varieties, which, as we shall see hereafter, occur more frequently as plutonic rocks. It was shown by Cordier that, in spite of its com- pact appearance, when viewed by the naked eye, ordinary basalt is really an aggregate of minerals felspar, augite, olivine, and magne- tite. By grinding the rock to powder and carefully levigating it, Cordier was able to separate the several minerals according to their different densities ; the same result can be much more easily attained at the present day by putting powdered basalt into liquids of different specific gravity ; in a still more simple manner, we may examine not only the several minerals in the rock, but also their relations to one another, by making thin transparent sections of basalt and studying them under the microscope (see fig. 645). All the basalts contain more or less of a glassy material between their crystals ; and basalts with an exceptional quantity of such glassy material are called magma-basalts. Basalts sometimes contain, in addition to their essential minerals, enstatite, hornblende, or biotite. Locally distributed, we find basalts containing one or other of the felspathoids leucite, nepheline, hauyn, or melilite and these are known as leucite-basalts, nepheline-basalts, Fig. 645. Fig. 646. Isle of Mull. The lath-shaped Leucitite from near Rome. The large crystals are felspar (plagioclase), the polygonal crystals with symmetrical granular ones augite, the black opaque inclusions are' leucite. Aug'ite, biotite, grains magnetite, and the large ciys- and magnetite are also present with a tals in high relief olivine. glassy base. hauyn-basalts, and melilite-basalts. Basalt-like rocks with leucite and nepheline, but without olivine, are called by continental authors leucitite (see fig. 046) and nephelinite. The basalts show much less tendency to pass into a glassy state than do rocks of more acid composition. Occasionally, however, as in the surfaces of lava-streams and the margins of dykes, where rapid cooling has taken, place, we find basalt-glass or tacliylytc formed. This tachylyte may exhibit the banded, spherulitic, and per- litic structures so characteristic of vitreous rocks. Pele's hair and the thread-lace scoria of Hawaii are beautiful pumiceous forms of a basalt-glass. Owing to their smaller proportion of silica and the amount ex? H H 466 PHASES OF [CH. xxxi. iron-oxides which they contain, the basaltic lavas undergo easy de- composition, often losing their black colour and assuming reddish and brownish tints. Melaphyres are forms of these altered lavas. When the basalt was scoriaceous, the cavities become filled up with various secondary minerals calcite, quartz, zeolites, chlorites, &c. and an amygdaloidal rock is produced (see fig. 634, p. 458). When a basalt-glass with spherulitic structure undergoes alteration, a rock is produced, which, from its fancied resemblance to the skin of a small-pox patient, has long been known as variolite. The fragmental materials derived from the basaltic lavas are known as basalt-tuffs ; the variety known as palagonite-tuff, which is common in Sicily and Iceland, contains the hydrous glass of secon- dary origin called by mineralogists ' palagonite.' There are a very few lavas in which the proportion of silica is so low that they must be classified with the ultra-basic rocks, but these are of rare and exceptional occurrence. Fuller details concerning the in the ' British Petrography ' of Mr. nature and structure of lavas will Teall, and in the treatises on the be found in the treatises on petro- microscopic characters of rocks by graphy by Dr. Hatch, Mr. Harker, Fouque and Michel Levy, Zirkel and and Mr. Rutley already referred to, Rosenbusch. CHAPTER XXXI ORIGIN AND STRUCTURE OF VOLCANIC ROCK-MASSES Explosive and effusive action of Volcanoes Origin of Volcanic Cones Internal structure of Volcanic Cones Origin of Volcanic Craters Formation of Volcanic Dykes Varieties of Volcanic Dykes Alteration of Rocks on the sides of Volcanic Dykes Contact Metamorphism Alteration of Sandstone, Shale, Limestone, and Coal Interbedded and contemporaneous Volcanic Rocks contrasted with intrusive or subsequent masses Columnar and globular structures in Lavas. Different Kinds of Volcanic Action. Volcanic activity is of a twofold nature explosive and effusive. Sometimes great volumes of steam escape from the vent with terrible violence, carrying up considerable rock-masses, with bombs, scoriae, lapilli, and dust, to the height of many miles into the atmosphere and in such quantities as to completely darken the whole district around for hours, days, or even weeks. The larger fragments, when they fall back to the vent, are re-ejected, and this takes place again and again, till all are gradually reduced to an impalpable powder. This volcanic dust mingling with the rain, produced by the condensa- tion of steam, sometimes flows down in rivers of mud, which consolidate to form heds of volcanic tuff. At each explosion of steam from the midst of the molten rock in a volcanic vent, a fresh surface of the glowing lava is exposed, and it is the ruddy CH. xxxi.j VOLCANIC ACTIVITY 467 reflection of this upon the clouds of vapour and dust which is so frequently mistaken especially at night for flames, and has led to volcanoes being termed incorrectly ' burning mountains.' The friction between the ascending column of vapour, the ejected fragments, and the sides of the vent gives rise to the generation of electricity and the wonderful displays of lightning so common during volcanic outbursts. At other times, lava issues quietly from a volcano, with but comparatively little escape of steam ; and the mass of molten rock flows as ' lava streams,' which are often of enormous volume. This effusive action may, like the explosive action, go on continuously for long periods for days, weeks, months, or even years. In many volcanoes, there occur alternations of explosive and effusive action. At the beginning of each eruption steam at high tension escapes from the vent, and explosions, following one another in rapid succession, discharge into the atmosphere vast quantities of fragmental materials. As the violence of the paroxysm gradually dies out, however, the explosive is succeeded by effusive action, and streams of lava flow out in the place of the violent discharges of scorias and dust. But in some volcanoes the action is almost always explosive ; thus the great volcanoes of Java appear to be wholly built up of loose materials projected from their vents, there being few if any examples of lava-streams. In other volcanoes, like those of the Hawaiian Islands, the action appears to have been almost entirely effusive ; and the volcanic cones are built up of thick sheets of lava, piled one on the top of another, with hardly any layers of scoria or dust between them. A striking example of effusive volcanic action on the grandest scale was afforded to geologists in 1788, when, at Skaptar Jokul, in Iceland, a great fissure opened, on which only some small scoria-cones were thrown up, but the two streams of lava issuing from this vent were in bulk equal to the mass of Mont BJ^,nc ! A century later, in 1883, the most violent explosive eruption on record occurred at Krakatoa in the Sun da Straits. There was no outflow of lava, but pumice and dust were thrown to the height of sixteen miles into the air, the pulsations of the atmo- sphere travelled two and a half times round the globe, violent waves were produced in the ocean, which were registered on the tide-gauges all over the world, and ejected materials were scattered over a circle with a radius of 1,000 miles ! External form, structure, and origin of volcanic Moun- tains. In the case of Monte Nuovo in Southern Italy (see fig. 647), we have a volcanic cone, more than four hundred feet in height, with a large and deep crater at its summit, which was H H 2 468 VOLCANIC CONES [CH. XXXI. formed by a series of explosive outbursts, lasting for three days, in the year 1538. The origin of the great volcanic cones with crater-shaped summits has been explained in the ' Principles of Geology ' (chaps, xxiii. to xxvii.), where Vesuvius (see fig. 648), Fig. 647. Monte Nuovo, near Naples. Fig. 64a Vesuvius, with the old encircling crater of Monte Somma. Fig. 649. Barren Island in the Bay of Bengal, with au active cone rising in the midst of a large ancient crater. Etna, Santorin, and Barren Island (see fig. 649), are described. The more ancient portions of those mountains or islands, formed long before the historic period, exhibit the same external features and internal structure as those of the extinct volcanoes of still higher antiquity. All these volcanoes were produced bv the CH. XXXI.] AND THEIR STKUCTUEES 469 same agencies which cause modern volcanic eruptions, and their materials belong to the same groups of rocks and only differ slightly in physical characters and in chemical constitution. Ancient and modern Cones and Craters. In regions where the eruption of volcanic matter took place in the open air, and where the surface has never since been subjected to great Fig. 650. Group of volcanic cones in Auvergne, with their sides breached by the outflow of lava-streams. aqueous denudation, cones and craters abound. Many hun- dreds of such cones still remain in Central France, in the ancient provinces of the Auvergne, Velay, and the Vivarais, where they form chains of hills. Although probably none of the eruptions have happened within the historical era, the ancient streams of lava may still be distinctly traced, descending from many of the craters, and following the lowest levels of the existing valleys (see fig. 650). Fig. 651. Ideal section of a scoria- or tuff-cone, showing the arrangement of the materials of which it is built up. The origin of the cone and crater of the modern volcano is now well understood, the growth of many having been watched during volcanic eruptions. A chasm or fissure first opens in the earth, from which great volumes of steam are evolved. The explosions are so violent as to splinter the rocks in which the volcanic vent is opened, and hurl up into the air fragments of 470 SCORIA-CONES [CH. xxxi, broken stone, parts of which are shivered into minute portions. This stone is, in part, the rock which is penetrated by the up- rushing steam, gases, and hot water, but mainly the volcanic rock which had been gradually forced up in a molten state. The showering down of the various ejected materials around the orifice of eruption gives rise to a conical mound, in which the successive envelopes of ash and scoriae form layers, dipping on all sides from a central axis (see fig. 651). In the meantime a hollow, called a crater, has been kept open in the middle of the mound by the continued passage upwards of steam and other gaseous fluids. After a while, molten rock, quite liquefied (lava), usually ascends through the vent by which the gases make their escape. Although extremely heavy, this lava is forced up by the expansive power of entangled gaseous Fig. 652. Small scoria-cones thrown up on the surface of lava-streams, Vesuvius. (After Schmidt.) fluids and steam. Quantities of the lava are also shot up into the air, and burst into minute fragments called ash. Blocks of solid lava are ejected also, being more or less scoriaceous. The lava sometimes flows over the edge of the crater, and thus thickens and strengthens the sides of the cone ; but sometimes it breaks down the cone on one side (see fig. 650), and often it flows out from a fissure at the base of the hill, or at some dis- tance from its base. The lava in cooling assumes a clinkery or scoriaceous appearance. Small cones made up of scoriae thrown out in a pasty condi- tion may accumulate into steep-sided, bottle-shaoed, or chimney- like, piles (see fig. 652) ; ordinary * cinder ' or scoria cones have a slope of about 35; 'tuff cones' are formed of lapilli, puz- .zolana or dust, which, when mingled with water, flow freely AND LAVA-CONES 471 and accumulate to form hills with a slope of about 17. Lava- cones vary in form according to the liquidity or viscosity of the material; we have steep-sided, massive 'mamelons,' like the phonolite-volcanoes of Bohemia (see figs. 653-654), or the domitic ' puys ' of Auvergne on the one hand ; or greatly flat- tened domes with a slope of only a few degrees, as in the Hawaiian volcanoes on the other hand. Fig. 653. Mamelon ' composed of a ropy lava, Isle of Bourbon. (After Bory de St. Vincent.) Fig. 654. Diagram showing tlie probable internal structure of a lava-cone like fig. 653. Internal Structure of Volcanic Cones The mode of origin of volcanic cones, as above described, is admirably exhibited when such cones have been partially swept away by the action of the waves of the sea or of rivers. We then find that the Scoria- or ' Cinder '- cones, like those of Auvergne, are composed of materials often ex- hibiting a most perfect stratification ; the thin layers of which the cones are made up slope outwards and inwards, as shown in the dia- gram (fig. 651, p. 469). The degree of slope of the materials varies with their nature, rough cindery masses lying in steeper slopes than the finely comminuted matter, which, mingled with water, often flows as mud to consolidate as tuffs. Lava-cones are formed by the successive 472 CONES AND [CH. XXXI. outwelling of the liquid materials (see fig. 654). If this be viscous we get steep-sided domes like those of Bourbon, Bohemia, &c. ; if the lava be very liquid, exceedingly depressed or flat domes are produced like the volcanoes of Hawaii, which have a slope of only 4. The majority of volcanic cones are made up of alternations of lava and f ragmen tal materials, and these are known as compound cones. By continual ejections from a vent, volcanoes may gradually grow up into conical mountains many thousands of feet high, like Cotopaxi (fig. 655). The volcanoes of Hawaii rise to a height of 30,000 feet from the ocean floor on which they stand. When ' lateral ' or ' parasitical ' eruptions occur on the sides of a volcano, it may lose its regular conical form and assume characters like those exhibited by Etna, the flanks of which are covered by parasitical cones. Fig. 655. Snow- line Line of forest vegeta- tion Cotopaxi (19,600 ft.), the highest active volcano in the world, seen from a distance of 90 miles. (After Humboldt.) Origin of Volcanic Craters. There is no doubt that the craters of volcanic cones are formed by violently explosive or paroxysmal outbursts. In fig. 656 we give a copy of Mr. Scrope's drawing of the crater formed at the summit of Vesuvius by the great eruption of 1822; it was more than 1,000 feet in diameter, and at least 1,000 feet deep. At an earlier date, as shown by the drawings of Sir William Hamilton, the cone rose to a much greater height, and within the crater small cones were formed, by gentle and long- continued eruption (see fig. 657). Since the great paroxysm of 1822 the vast crater has been filled up and the cone re-formed, though constant changes have taken place in the size and form of the sum- mit-crater, and in the number of small cones within it. The tendency of the violent eruptions is to produce large craters truncating the summit of a volcano ; but gentle and long-continued eruptions build up a cone within the crater, the sides of which may in the end become confluent with those of the great mountain itself, the height of which thus becomes greatly increased. This arrangement of cone within crater is very characteristic of volcanic mountains. Sometimes craters are formed pf enormous dimensions ; when com- CH. XXXI.] CEATEKS Pig. 656. 473 Great crater formed at Vesuvius during the eruption of 1822, 1,000 ft. in diameter and 1,000 ft. deep. (From a drawing by Scrope.) Fig. 657. Summit of Vesuvius in 1767. (After Sir William Hamilton.) Fig. 658. Crater-lake of GustavjJa, Mexico. 474 SUBMAKINE VOLCANOES [CH. XXXI. posed of tuffs or other materials impermeable to water they give rise to circular crater-lakes like fig. 658. There are crater-lakes in Italy, (Bracciano and Bolsena), which are respectively ten and twelve miles in diameter. It was held by Von Buch and Elie de Beaumont that craters were formed in volcanoes owing to the mountain being pushed up like bubbles and bursting at the top. But this ' theory of elevation-craters,' as it was called, has now been completely aban- doned by geologists. At the same time it must be remembered that Fig. 659. Graham's Isle, a submarine volcano thrown up in the Mediterranean in 1831. in some craters, like that of Kilauea in Hawaii, the lake of lava at the bottom may undermine the sides, and thus tend to enlarge the area of the crater. Volcanoes are occasionally submarine, like the volcano thrown up in the Mediterranean and known as Graham's Island (see fig. 659), Fig. 660. Part of the chain of extinct volcanoes called the Monts Dome, Auvergne. (Scrope.) and build their way up to the surface, being exposed to the action of waves and tides. Some volcanic eruptions, both at the present day and in the remote past, took place along lines of fracture of the earth's crust, and lava welled out, and sheets and flows were produced on a grand scale with the formation of only small cones. Ancient vol- canoes were as large as the modern, and as active. They show by their linear arrangement that they were formed on lines of fissure (see fig. 660), and followed the law of occurring on areas which are undergoing elevation. Denudation has, in many instances, worn the CH. XXXI.] VOLCANIC DYKES 475 old volcanoes nearly to the surface of the earth, yet some of the remains of the central vent and of the sloping layers around it enable the original dimensions to be estimated. All through the earth's history, internal heat, and the presence of water in deeply seated rocks, have given rise to volcanic action. Volcanic matter, in the form of lava, bursts forth under certain circumstances through the body of the volcanic cone, or if it does not reach the outside, it solidifies within, and is called a dyke. Kg. 661 Section across the Binn of Bnrntisland, Fife. (After Geikie.) 1. Sandstones ; 2. Limestone ; 3. Shales. 6 6. Interbedded basalts ; / t, Bedded tuffs. T. Tuff filling the old volcanic orifice of the Binn ; E. Basalt dykes. Similar outbursts occurring beneath the surface of the earth cause masses of volcanic rock to be injected through and between the sedi- mentary strata. By denudation we have exposed to our view great masses of material which have formed the centres and lower portions of volcanic cones. Such rudiments of volcanoes were called by Darwin the ' basal wrecks ' of volcanoes. Among the lava-masses injected into volcanic cones and the strata underlying and surrounding them, we recognise dykes, intrusive sheets (or ' sills '), laccolites (or lenticular intrusions), and the still larger bosses out of which whole mountains may have been carved by denudation. Volcanic Dykes. The leading varieties of the volcanic rocks, basalt, andesite, and rhyolite, for example, are sometimes found in Basaltic dyke in the Val del Bove, Etna, from which a lava-stream is seeu to proceed. (After Abich.) dykes penetrating stratified and unstratified formations, and these are examples of intrusive or subsequent volcanic ejections. Fissures have already been spoken of as occurring in all kinds of rocks, some a few feet, others many yards in width. If such a parallel-sided fissure be filled with molten rock, or lava, the material on consolida- tion forms a wall-like mass known as a dyke. In volcanic cones it is sometimes possible to trace the actual connection between a dyke filling a fissure in the side of the volcano and a stream of lava which aas flowed out at the surface (see fig. 062). It is not uncommon to 476 PECULIAEITIES EXHIBITED BY [CH. xxxi. find such dykes passing through strata of soft materials, such as tuff, scoriae, or shale, which, being more easily removed by denudation than the volcanic rock, are often washed away by the sea, rivers, or rain, in which case the dyke stands out prominently on the face of precipices, or on the level surface of a country, as may be seen in Madeira (see fig. 663) and in many parts of Scotland. Fig. 663. Dyke in valley, near Brazen Head, Madeira. (From a drawing by Basil Hall.) In the islands of Arran and Skye, and in other parts of Scotland, where sandstone, conglomerate, and other hard rocks are traversed by lava-dykes, the converse of the above phenomenon is also seen. The dyke, having decomposed more rapidly than the containing rock, has once more left open the original fissure, often for a distance of many yards inland from the sea-coast. There is yet another case, by no means uncommon in Arran and other parts of Scotland, where the strata in contact with the dyke, and for a certain distance from it, have been hardened, so as to resist the action of the weather more Fig. 664. Ground plan of dolerite dykes traversing sandstone, Arran. than the dyke itself or the surrounding rocks. When this happens, two parallel walls of indurated strata are seen protruding above the general level of the country and following the course of the dyke. In fig. 664 a ground-plan is given of a ramifying dyke of dolerite, cutting through sandstone on the beach near Kildonan Castle, in Arran. The larger branch varies from 5 to 7 feet in width, which will afford a scale of measurement for the whole. CH. xxxi.] VOLCANIC DYKES 477 Some volcanic dykes may be followed for leagues, uninterruptedly, in nearly a straight direction (like the Cleveland dyke, which runs from the Yorkshire coast right through the south of Scotland), showing that the fissures which they fill must have been of extraordinary length. The materials of the dykes or flows which have been injected through and between strata were hot, pasty, and full of water and gases under pressure, and they acted upon and locally metamorphosed, more or less, the strata on either side and above and below them. The volcanic matter, moreover, became more or less crystalline on cooling. Usually, the sides and surfaces of such intrusive masses have a finer crystalline texture than the middle part, and occasionally the surfaces in contact with the strata are actually glassy. Columnar structure (the columns being at right angles to the walls of the dyke), spheroidal structure, and other forms of jointing occur in dykes. Some dykes are of composite character, different kinds of rocks entering into their composition. In certain cases a segregative action appears to have gone on within the molten material filling a dyke, and the sides and centre thus come to be formed of rocks of different chemical composition. In other cases a dyke has been reopened, and the fissure or fissures formed in it may be injected with materials of a different composition from that of the original dyke. Rocks altered by volcanic dykes Contact Metamorphism. After these remarks on the form and composition of dykes them- selves, it may be well to describe the alterations which they some- times produce in the rocks in contact with them. The changes are usually such as the heat of melted matter and of the entangled steam and gases might be expected to cause. In some instances, however, little or no change happened in the surrounding rocks- Plas-Newydd : Dyke cutting through shale. A striking example of contact metamorphism, near Plas-Newydd, in Anglesea, has been described by Henslow. The dyke is 134 feet wide, and consists of dolerite. Strata of shale and argillaceous limestone, through which it cuts perpendicularly, are altered to a distance of 30, or even, in some places, of 35 feet from the edge of the dyke. The shale, as it approaches the igneous rock, becomes gradually more compact, and is most indurated where nearest the junction. Here it loses part of its laminated structure, but the bedded character is still discernible. In several places the shale is converted into hard porcellanous jasper. In the most hardened part of the mass, the fossil shells, principally Producti, are nearly obliterated ; yet even here their im- pressions may frequently be traced. The argillaceous limestone under- goes analogous changes, losing its original texture as it approaches the dyke, and becoming granular and crystalline. But the most extra- ordinary phenomenon is the appearance in the shale of numerous crystals of analcime and garnet, which are seen to be confined to those portions of the rock affected by the dyke. Some of the garnets contain as much as 20 per cent, of lime, which they may have derived from the decomposition of the fossil shells. The same mineral has been observed, under very analogous circumstances, in High Teesdale, by Professor Sedgwick, where it also occurs in shale and limestone, altered by basalt. Antrim : Dyke cutting through chalk. In several parts of the county of Antrim, in the north of Ireland, Chalk with flints is 478 VOLCANIC DYKES [CH. xxxi. traversed by basaltic dykes. The chalk is there converted into granular marble near the basalt, the change sometimes extending 8 or 10 feet from the wall of the dyke, being greatest near the point of contact, and thence gradually decreasing till it becomes evanescent. 'The extreme effect,' says Dr. Berger, 'presents a dark brown crystalline limestone, the crystals running in flakes as large as those of coarse primitive (metamorphic) limestone ; the next state is saccharine, then fine-grained and arenaceous ; a compact variety, having a porcellanous aspect and a bluish-grey colour, succeeds : this, towards the outer edge, becomes yellowish white, and insensibly graduates into the unaltered chalk. The flints in the altered chalk usually assume a grey yellowish colour.' All traces of organic remains are effaced in that part of the limestone which is most crystalline. The annexed drawing (fig. 665) represents three basaltic dykes traversing the chalk, all within the distance of 90 feet. The chalk contiguous to the two outer dykes is converted into a finely granular marble, m m, as are the whole of the masses between the outer dykes and the central one. In some cases the change undergone by the chalk is of a chemical nature, and the rock, besides being indurated Fig. 665. Basaltic dykes in chalk in Island of Rathlin, Antrim. Ground plan as seen on the beach. (Conybeare and Buckland.) and crystallised, is also dolomitised. The complete contrast in the composition and colour of the intrusive and invaded rocks in these cases renders the phenomena peculiarly clear and interesting. Another of the dykes of the north-east of Ireland has converted a mass of red sandstone into hornstone. By another, the shale of the Coal-measures has been indurated, assuming the character of flinty slate ; and at Portrush the shaly clay of the Lias has been changed into flinty slate, which still retains numerous impressions of Ammonites. In the infancy of geological science the aqueous origin of basalt was maintained by Werner and his disciples. They mistook the altered Lias clay of Antrim for basalt, and referred to the occurrence of Ammonites in the rock as a proof that this rock could not be of igneous origin. It might have been anticipated that beds of coal would, from their combustible nature, be affected in an extraordinary degree by the contact of melted rock. This is seen to be the case in one of the doleritic dykes of Antrim, which, passing through a bed of coal, reduces it to a cinder for the space of 9 feet on each side. At Cockfield Fell, in the North of England, a similar change is observed. Specimens taken at the distance of about 30 yards from the dyke CH. xxxi. J AND INTRUSIVE SHEETS 479 are not distinguishable from ordinary pit-coal ; those nearer the dyke are like cinders, and have all the character of coke, while those close to it are converted into a substance resembling soot. It is by no means uncommon, however, to meet with similar rocks almost wholly unchanged in the proximity of volcanic dykes. This great inequality in the effects of the igneous rocks may often arise from an original difference in their temperature, and in the nature of the entangled gases such as is ascertained to prevail in different lavas, or in the same lava near its source and at a distance from it. Sometimes the extreme alteration produced near a volcanic dyke may be ascribed to the circumstance that the fissure now filled with solid rock may have constituted a channel through which enormous quan- tities of molten material have flowed up to the surface during vast periods of time. Xnterbedded or Contemporaneous Plows. Lava-streams consisting of volcanic rock which has flowed over the surface and altered the underlying rocks are scoriaceous in their upper part, and usually at their lower surface also. Sedimentary deposits accumulated upon the flows are of course not found to be altered physically or chemically by the contact, for before they were deposited the volcanic flow had cooled. Such flows are said to be interbedded. They may have occurred during the progress of the deposition of strata all around, during any particular geological period, and the fossils of the bed below and above the volcanic flow may be of the same species. Hence the flow thus interbedded is said to be contem- poraneous. Interbedded or contemporaneous flows occur as compact sheets or as fragmented masses, and they conform to the plane of the underlying stratum. They are not found to have broken into or altered the overlying strata in any way. Both of their surfaces are scoriaceous or vesicular, and this peculiarity may extend through the whole sheet. Beds of tuff and be Fig. 666. other fallen materials may interstratified with the flows. The fragmentary volcanic rocks of the present day, such as ashes and blocks, fall on the surface and do not influence the underlying strata. In past geological periods the tuffs and ash-beds and the breccias similarly covered other rocks in vast de- posits, more or less stratified, and no metamorphism resulted. In the illustration (fig. 666), from the Lower Carboniferous rocks of Linlithgowshire, a black shale (1) is at the bottom, and has the remains of terrestrial plants ; and there are other shales, num- bered 3, 5, 7, ( .). Between them are bands of pale yellowish volcanic tuff with lapilli or ejected pieces of an older lava (Nos. 2, 4, 6, 8). A coarse agglomerate tuff lies on the top of all (No. 10). The distinction between volcanic materials which have accumulated Interstratified volcanic tuff and" shale. (After Geikie.) 480 COLUMNAR STRUCTURE [CH. XXXI. Fig. 667. on land and on the sea-floor respectively is often not very easy. Tuffs or volcanic ashes collect on the floor of the Mediterranean, and are dredged up with the living mollusca, and lava-currents have, during historical times, flowed into the Bay of Naples, and have become columnar in their structure. Intrusive volcanic sheets are distinguished from contemporaneous or interbedded lava flows by not exhibiting scoriaceous upper and under surfaces ; by their affecting by contact metamorphism the strata above as well as below them ; by their usually more crystalline and less scoriaceous character ; and by the fact that they often are seen to cut across and to send offshoots into sur- rounding beds. Columnar and globular struc- ture. One of the characteristic forms assumed by volcanic rocks is the colum- nar, a structure often displayed in a very striking manner by basaltic lavas. The columns are sometimes straight, at other times curiously curved and twisted. In section they are polygonal (with a tendency towards hexagonal forms), and they are often divided longitudinally by equi- distant joints, which sometimes exhibit curved surfaces of articulation ; in certain cases the angles of one division of a column are found to project and to form processes which fit into sockets in the adjoining divi- sions (see fig. 667). Columns of different varieties often occur in the Basaltic column, divided by curved cross joints and with ' ball-and-socket '-articulations. Fig. 668. Lava-stream cut through in the valley of the Ardeche, with thick vertical columns in its lower part, and thinner columns, irregularly disposed, in its upper part. (After Scrope.) same lava stream, the thick straight articulated columns being found in the lower, and the smaller curved forms in the upper portion ; and the line of junction between the two kinds is in many cases very distinctly marked (see fig. 668). It is this peculiar combination of columns of different kinds which gives rise to the beautiful and CH. XXXI.] ITS NATURE AND ORIGIN 481 well-known features of the Isle of Staifa ; it is also seen in many iavas of more recent date. It being assumed that columnar rock has consolidated from a fluid state, the prisms are found to be always at right angles to the cool- ing surfaces. If these surfaces, therefore, instead of being either perpendicular or horizontal, are curved, the columns ought to be inclined at every angle to the horizon ; and there is a beautiful exemplication of this phenomenon in one of the valleys of the Lava of La Coupe d'Ayzac, near Antraigue, in the Department of Ardeche. Fig. 670. Vivarais, a mountainous district in the South of France, where, in the midst of a region of gneiss, a geologist encounters unexpectedly several volcanic cones of loose sand and scoria. From the crater of one of these cones, called La Coupe d'Ayzac, a stream of lava has descended and occupied the bottom of a narrow valley, except at those points where the river Volant, or the torrents which join it, have cut away portions of the solid lava. The accompanying sketch (fig. 669) represents the remnant of the lava at one of these points. It is clear that the lava once filled up the whole valley to the dotted line d a ; but the river has gradually swept away all below that line, while the tribu- tary torrent has laid open a trans- verse section ; by which we perceive, in the first place, that the lava is composed, as is usual in that district, of three parts : the uppermost, at a, being scoriaceous ; the second, b, presenting irregular prisms of small diameter ; and the third, c, with regular columns of great thickness, which are vertical on the banks of the Volant, where they rest on Columnar basalt in the Vicentin. (Fortis.) horizontal base of gneiss, but which are inclined at an angle of 45 at g, and are nearly horizontal at /, their position having been everywhere determined, according to the law before mentioned, by the form of the original valley. I I 482 VARIETIES OF COLUMNAR [CH. XXXI. In fig. 670, on the preceding page, a view is given of some of the in- clined and curved columns which present themselves on the sides of the valleys in the hilly region north of Vicenza, in Italy, and at the foot of the higher Alps. Unlike those of the Vivarais, last mentioned, the basalt of this country was evidently submarine, and the present valleys have since been hollowed out by denudation. In vertical dykes, as has been already remarked, the columns are horizontal ; they start from the outer walls of the dyke, and meet in an irregular line towards its centre. The columnar structure is by no means peculiar to the volcanic rocks of the basaltic type ; it is also observed in trachyte, and other more acid rocks, although in these it is rarely exhibited in such regular polygonal forms, and never with the ball-and-socket joints, which form so conspicuous a feature in many basaltic columns. It has been already stated that basaltic columns are often divided by cross-joints. Sometimes each segment, instead of an angular Fig. 671. Basaltic pillars of the Kasegrotte, Bertrich-Baden, halfway between Treves and Cobleuz. Height of grotto, from 7 to 8 feet. assumes a spheroidal form, usually produced by weathering, so that a pillar is made up of a pile of balls, usually flattened, as in the Cheese-grotto at Bertrich-Baden, in the Eifel, near the Moselle (fig. 671). The basalt there is part of a small stream of lava, from 30 to 40 feet thick, which has proceeded from one of several volcanic craters, still extant, on the neighbouring heights. In some masses of decomposing basalt, dolerite, and other volcanic rocks, the globular structure is so conspicuous that the rock has the appearance of a heap of large cannon-balls. According to M. Delesse, the middle of each spheroid has been a centre of con- traction produced by cooling ; Professor Bonney has assigned the globular, ' curvitabular,' and other structures exhibited in volcanic rocks to the same cause. To similar contraction we may attribute some cases of columnar structure in sedimentary strata, such as volcanic ash, shale and sandstone, which have been affected by the proximity of volcanic dykes or the overflow of lav a- streams. OH. xxxi.] AND GLOBULAK STRUCTUKE 483 Fig. 672. Scrope gives as an illustration of this structure a glassy rhyo- lite or ' pitchstone-porphyry ' in one of the Ponza islands, which rise from the Mediterranean, off the coast of Terracina and Gaeta. The globes vary from a few inches to three feet in diameter, and are of an ellipsoidal form (see fig. 672). The whole rock is in a state of decomposition, ' and when the balls have been exposed a short time to the weather, they scale off at a touch into numerous concentric coats, like those of a bulbous root, inclosing a compact nucleus. The laminfe of this nucleus have not been so much loosened by decomposition ; but the application of a ruder blow will produce a still further ex- foliation.' This spheroidal struc- ture may be also seen in volcanic ash at Burntisland and in many other rocks ; the perlitic struc- ture, before referred to, is an ex- ample of the same kind of action exhibited on a very minute scale in rocks of vitreous texture. For further information on the nature of volcanic action, Scrope's ' Volcanoes ' may be consulted, and the volume on the same subject in Globiform pitchstone. Chiaja di Luna, the ' International Scientific Series.' Isle of Ponza. (Scrope.) CHAPTEE XXXII PRINCIPLES ON WHICH THE CHRONOLOGICAL CLASSIFICATION OF VOLCANIC ROCKS IS BASED Variations of mineral character in the volcanic rocks of different periods Not essential, but due to alteration Age of lava flows and intrusions Tests to be applied to determine relative age of volcanic masses Sources of error in drawing inferences Great value of fossils when found Test by included fragments Order in which volcanic roclis have been erupted Views of Bunsen, Durocher, Bichthofen, and later authors. Age of volcanic phenomena. Having in the former part of this work referred the sedimentary strata to a long succession of geological periods, we have now to consider how far the volcanic formations can be classed in a similar chronological order. The tests of relative age in this class of rocks are four : 1st, superposition or intrusion, with or without alteration of the 484 DETEKMINATION OF AGE [CH. xxxn. rocks in contact ; 2nd, organic remains ; 3rd, mineral characters; 4th, included fragments of older rocks. Besides these four tests it may be said, in a general way, that volcanic rocks of pre-Palseozoic and Palaeozoic age differ, in a certain degree, from those of the Secondary or Mesozoic age, and these again from the Tertiary and Recent. Not, perhaps, that they differed originally in a much greater degree than the modern volcanic rocks of one region, such as that of the Andes, differ from those of another, such as Italy, but because all rocks permeated by water, especially if its temperature be high, are liable to undergo a slow metamorphosis. Although subaerial and submarine denudation removes, in the course of ages, large portions of the cones and of the upper or more superficial products of volcanoes, yet these are some- times preserved by the occurrence of subsidence, and the burying of the volcanic rocks under sedimentary deposits. In this way the volcanic structures may be protected for ages ; but even in this case they will not remain unaltered, because they are percolated by water, often at high temperatures, and charged with silica, iron-salts, and other substances in solution, whereby gradual changes in the constitution of the rocks are induced. Every geologist is aware how often silicified trees occur in volcanic tuffs, the perfect preservation of their internal structure showing that they had not decayed before the petrifying material was supplied. The porous and vesicular nature of a large part, both of the basaltic and trachytic lavas, affords cavities in which silica, calcite, and the zeolites are readily deposited. The minerals of the zeolite family, which are so commonly found in such amygdaloidal cavities, are closely related in composition to the felspars, but contain water. Daubree and others have shown that the zeolites are formed by the action of percolating water upon the felspathic ingredients of racks. From these con- siderations it follows that perfect identity of appearance and character in very ancient and very modern volcanic formations is scarcely to be expected. Age of Intrusive dykes or sheets. After the differences between intrusive and contemporaneous volcanic flows have been considered (see p. 479), there should be no difficulty in understanding the relation of the age of strata and of the flows which pass through or amongst or above them, especially when the strata are capable of being classified in definite geological groups or formations. The nearly vertical dykes and the more or less horizontal sheets must be younger than the strata they penetrated and CH. xxxn.] OF VOLCANIC INTRUSIONS 485 influenced physically and chemically. How much younger does not always appear, because, for instance, a dyke which was formed in Tertiary times may come to the surface of the earth through Carboniferous strata which now form the surface rock there, denudation having worn away the overlying strata before or since the intrusion took place. A careful survey of the general distribution of the strata around the volcanic area, however, will often enable the question of age to be settled. Age of interbedded or contemporaneous lava flows. The flow must be younger than the stratum it overlies, and has metamorphosed more or less, and older than the stratum which rests on its scoriaceous upper surface. The fossils in the two sets of strata determine their age and the relative antiquity of the flow. The presence of similar species in the two sets of strata proves the flow to have been truly contemporaneous, for the volcanic outburst was evidently only an episode in the history of the period. The finding of very different species of fossils in the two sets of strata leads to a less definite conclusion regarding the age of the flow, which may be of any age between the ages of the under- lying and overlying beds. In the annexed figure (fig. 673) a flow, 6, is placed under D, between the strata a and c of the Carboniferous formation. It appears to be interbedded and contemporaneous. But both the Fig. 673. J) strata a and c have been more or less altered by it, so that it was injected subsequently to the deposition of both of them. Under E, the same flow covers #, having pierced it and baked it on the top. Hence this flow is certainly younger than a. We may, however, be easily deceived in supposing the vol- canic rock to be intrusive, when in reality it is contempora- neous ; for a sheet of lava, as it spreads over the bottom of the sea, cannot rest everywhere upon the same stratum, either because these have been denuded, or because, if newly thrown down, they thin out in certain places, thus allowing the lava to cross their edges. Besides, the heavy igneous mass while fluid will often, as it moves along, cut a channel into beds of soft mud and sand. Suppose that the submarine lava F (fig. 674) has come in contact in this manner with the strata a, b, c, and that 486 USE OF FOSSILS [CH. xxxii. after its consolidation the strata d, e, are thrown down in a nearly horizontal position, yet so as to lie unconformably to F, the appearance of subsequent intrusion will here be complete, although the lava is in fact con- Flg ' 674> temporaneous. We must not, therefore, hastily infer that the rock F is intrusive, unless we find the overlying strata d, e, to have been altered at their junc- tion, as if by heat. The test of age by superposi- tion is strictly applicable to all stratified volcanic tuffs, according to the rules already explained in the case of sedimentary deposits (see pp. 128, 129). If masses of volcanic rock intercalated among strata have evidently participated in the movements that have produced the curvatures and fractures of the strata, the volcanic masses, even if intrusive in origin, must clearly be of more ancient date than the period when the movements that have affected the whole series of deposits took place. Test of agre by organic remains. We have seen how, in the vicinity of active volcanoes, scoriae, pumice, fine dust, and fragments of rock are thrown up into the air, and then showered down upon the land, or into neighbouring lakes or seas. In the tuffs so formed, shells, corals, or any other durable organic bodies which may happen to be strewn over the bottom of a lake or sea, will be embedded, and thus continue as per- manent memorials of the geological period when the volcanic eruption occurred. Tufaceous strata thus formed in the neigh- bourhood of Vesuvius, Etna, and Stromboli, and other volcanoes now in islands or near the sea, may give information of the relative age of these tuffs at some remote future period when the activity of these mountains is extinguished. By evidence of this kind we can establish a coincidence in age between volcanic rocks and the different Palaeozoic, Secondary, and Tertiary fossiliferous strata. The tuffs alluded to may not always be marine, but may include, in some places, freshwater shells ; in others, the bones of terrestrial quadrupeds. The diversity of organic remains in formations of this nature is perfectly intelligible, if we reflect on the wide dispersion of ejected matter during eruptions, such as that of the volcano of Coseguina, in the province of Nicaragua, January 19, 1835. Hot cinders and fine scoriae were then thrown up to a vast height, and covered the ground as they fell to the depth of more than 10 feet for a distance of CH. xxxii.] INCLUDED FRAGMENTS, ETC, 487 8 leagues from the crater in a southerly direction. Birds, cattle, and wild animals were scorched to death in great nunv bers, and buried in ashes. Some volcanic dust fell at Chiapa, upwards of 1,200 miles, not to leeward of the volcano as might have been anticipated, but to windward, a striking proof of a counter- current in the upper region of the atmosphere ; and some on Jamaica, about 700 miles distant to the north-east. In the sea, also, at the distance of 1,100 miles from the point of eruption, Captain Eden of the ' Conway ' sailed 40 miles through floating pumice, among which were some pieces of con- siderable size. The importance of studying the fossils contained in the strata beneath and above contemporaneous flows has been already pointed out. Test of age by mineral composition. As sediment of homogeneous composition, when discharged from the mouth of a large river, is often deposited simultaneously over a wide area, so a particular kind of lava flowing from a crater during one eruption may spread over an extensive district ; thus in Ice-> land, in 1783, the melted matter, pouring from Skaptar Jokul, flowed in streams in opposite directions, and formed a con* tinuous mass, the extreme points of which were 90 miles distant from each other. This enormous current of lava varied in thickness from 100 feet to 600 feet, and in breadth from that of a narrow river-gorge to 15 miles. Now, if such a mass should afterwards be divided into separate fragments by denu- dation, we might still perhaps identify the detached portions by their similarity in mineral composition. Nevertheless, this test will not always avail the geologist ; for, although there is usually a prevailing character in lava emitted during the same eruption, and even in the successive currents flowing from the same volcano, still, in many cases, the different parts even of one lava- stream, or, as before stated, of one continuous mass of lava, vary much in mineral composition and texture. Test by included fragments. Where the evidence of superposition alone would be insufficient, we may sometimes discover the relative age of two sets of volcanic rocks, or of an aqueous deposit and the volcanic rock on which it rests, by find- ing fragments of one included in the other. It is also not uncommon to find a conglomerate almost exclusively composed of rolled pebbles of lava, associated with some fossiliferous stratified formation in the neighbourhood of igneous rock-masses. If the pebbles agree, generally, in mineral character with the latter, we are then enabled to determine its relative age by knowing that of the fossiliferous strata associated with the con- glomerate. The origin of such conglomerates is explained by 488 SUCCESSION OF VOLCANIC ROCKS [en. xxxn. observing the shingle beaches composed of pebbles of igneous rock in modern volcanoes, as at the base of Etna. Order of appearance of different volcanic rocks. In Auvergne, the Eifel, and other countries where acid and basic lavas are both present, the acid rocks are for the most part older than the basaltic. These rocks do, indeed, sometimes alternate to some extent, as in the volcano of Mont Dore, in Auvergne ; and in Madeira, acid rocks overlie an older basic series ; but the acid rock occupies mou3 generally an inferior position, and is cut through and overflowed by basalt. It seems that in each region where a long series of eruptions have occurred, the lavas con- taining quartz and acid felspar have been first emitted, and the escape of the more basic kinds has followed. Von Kichthofen, who has given the subject of the succession of volcanic materials in North America and Europe great attention, believed that the volcanic rocks might be arranged in five groups, and that their appearance has been in the same order all over the world : 1, Propylite ; 2, Andesite ; 3, Trachyte ; 4. Rhyo- lite ; 5, Basalt. Basalt, he affirmed, is always the last of a series, although some of the other groups may not have been present. The explanation of the eruption of acid volcanic materials in the first instance, and of basic rocks subsequently, may be that as suggested by Bunsen, Durocher, and later authors a differentiation of molten materials within the earth's crust may take place, and the upper and lighter materials may be ejected before the lower and denser ones (Note X, p. 608). CHAPTER XXXIII VOLCANIC ROCKS OF CAINOZOIC AGE The latest exhibitions of volcanic energy in the British Islands The thermal springs of Bath, &c. Tertiary volcanoes of the west of Scot- land and the north of Ireland First period of eruption Second period of eruption Third period of eruption Tertiary volcanic rocks of other parts of Europe Vesuvius Auvergne Newer Pliocene vol- canoes of Italy Older Pliocene volcanoes of Italy and the Eifel Oligocene and Miocene volcanoes of Auvergne and the Eifel Eocene volcanic rocks of Monte Bolca Tertiary volcanic rocks of the Atlantic Islands Of India The United States and Australia. Latest exhibitions of volcanic activity in the British Islands. We shall now select examples of contemporaneous volcanic rocks of successive geological periods, to show that igneous causes have been in activity in all past ages of the world. CH. xxxm.] LATEST VOLCANIC ACTION IN BRITAIN 489 They have been perpetually shifting the places where they have broken out at the earth's surface, and we can sometimes prove that those areas which are now the great theatres of volcanic activity were in a state of perfect tranquillity at remote geological epochs ; while, on the other hand, in places where at former periods the most violent eruptions took place at the surface and continued for a great length of time, there has been an entire suspension of igneous action in historical times. The most recent volcanic rocks in the British Islands are those occurring in the Hebrides and the North of Ireland, and they rest upon strata containing Upper-Chalk fossils. They are products of three more or less distinct periods of eruption, which all clearly belong to the Tertiary era ; but in the absence of intercalated strata containing marine fossils it is not possible to determine their exact position in the geological series. The lavas of the second period of eruption are found alternating with clays con- taining the leaves and other remains of terrestrial plants. There seems to be no reason to doubt that these belong to the Older-Tertiary period, and some palaeophytologists are inclined to assign them to a very early part -of that period. Considering, however, the difficulty of exactly correlating geological deposits by the evidence of land-plants only especially when that evidence consists almost entirely of detached leaves it is per- haps not wise to do more than assign the first two periods of eruption to the Older Tertiary. In the interval between the second and third periods of eruption great denudation took place, and there was probably a considerable lapse of time so there is little doubt that the third period of eruption must have fallen within the Newer-Tertiary period. The fact that all traces of cones and craters have been removed, and that only very bulky lava-streams, or the centres of volcanic cones of considerable size, have been preserved, points to the conclusion that a long period of time must have elapsed since these latest British volcanoes became extinct. Of volcanic action we find no trace in the British Islands at the present day beyond certain hot springs, like that of Bath. This spring has a constant temperature of from 117 to 120 F., it contains about 144 grains per gallon of mineral matter in solution, and it has certainly flowed since Koman times. As illustrating how much may be effected by slow continuous action, as contrasted with violent and spasmodic activity, it may be interesting to refer to calculations which have been made by Lyell and Ramsay concerning the effects pro- duced by the Bath spring in historical times. It is probable that this comparatively small thermal spring has relieved the earth's crust, in 2,000 years, of as much heat as was dissipated in the 490 TEBTIABY VOLCANOES [CH. xxxm. eruption forming Monte Nuovo during three days in 1538. Nor was the actual amount of matter brought from the earth's in- terior and deposited on its surface by the Bath spring, less than that resulting from the outburst that produced the Monte Nuovo ; but while, in the latter case, the materials were piled up round the volcanic orifice, and remain to our view, in the former they were carried into the Avon, from the Avon into the Severn, and by the latter delivered to the ocean. The Tertiary Volcanoes of the British Islands. The volcanic area stretching through the North of Ireland and the inner Hebrides belongs to a ' petrographic province,' which is prolonged northwards and includes both the Faroe Islands and Iceland. While volcanic activity has died out in the southern part of this province, it is still rife in the extreme north, within the island of Iceland. It is possible that certain other rocks to the southward, like the granite of Lundy Island and the phonolite of the Wolf's Eock, mark a still fur- ther extension of this area of volcanic activity in Tertiary times. All through this ' petrographical province ' the different kinds of rocks erupted exhibit remarkable analogies in character, and in the order in which they made their appearance. To use the expressive term suggested by Professor Iddings, there is a marked ' consan- guinity ' among the products emptied in different parts of the area. First Period of Volcanic Activity. The rocks ejected dur- ing the earliest period were of two kinds. First, great masses of ande- sitic lavas forming bulky lava-streams and belonging to the class of more basic pyroxene-andesites (augite-andesites, and enstatite- andesites) and the acid mica-andesites and hornblende-andesites ; these andesites varied too, not only in their mineralogical consti- tution, but in their structure, and we find every gradation from stony to glassy types. As is so common in Auvergne and other districts, we sometimes find lavas of more basic types true basalts- alternating with the andesitic flows. There were five great centres within the district of the Inner Hebrides from which these andesitic lavas were erupted and accumulated to form the basal portions of volcanoes of vast dimension. These centres of volcan ; 3 action are now found constituting the following districts St. Kilda, the centre of the Isle of Skye, the Small Isles (Bum, Eigg, &c.), the peninsula of Ardnamurchan, and the Isle of Mull. At each of these centres there is evidence of great solfataric action having taken place, after the eruption of the andesitic lavas, and by this action the rocks in question were converted into the altered forms known as ' propylites.' It is interesting to note that, as pointed out by Von Bichthofen, volcanic activity in the districts of Hungary and the Western Territories of North America commenced with the eruption of andesitic lavas and their conversion by solfataric action into propylites. In the North of Ireland, according to the researches of the geological surveyors of that country, the rocks erupted during this earliest period consisted mainly of basalts, which are found lying upon the eroded surface of the hard chalk (white limestone) of Antrim. Towards the close of this first period, materials of highly acid character were intruded into these andesitic lavas and the underlying rocks, and these consolidated CH. xxxiii.] OF THE BKITISH ISLANDS 491 to form the great masses of granite, micropegmatite, and quartz- felsite, found at each of these five districts, and also in the island of Arran. These acid rocks are seen to send numerous veins into the andesites and to include fragments of them. The rocks erupted during the first period appear generally to have suffered so much denudation before the ejection of the materials of the second period, that there is some lack of evidence concerning the amount of mate- rial which actually reached the surface as lavas. Some masses of acid lavas (rhyolites), however, were certainly ejected in the Inner Hebrides at this time ; and in Antrim there was formed the beautiful volcanic mass of Tardree and Sandy Braes, consisting of rhyolites varying from very coarse-grained stony types (Nevadites) through many spherulitic and banded varieties to glassy forms (pitchstone- porphyry). Second Period of Volcanic Activity. The second period of volcanic eruption within the British Islands during the Tertiary period was marked by the outflow of immense sheets of basaltic lava, which accumulated to a great thickness in places exceeding 2,000 feet. These basaltic lavas strikingly resemble the rocks of the same composition in the Faroe Islands and Iceland, and like them often exhibit the ophitic structure (see fig. 688, p. 518) ; they only rarely alternate with other lavas of more acid composition (andesites). That these lavas were poured out from the same great vents as gave rise to the older andesitic lavas (now converted into propylites) there is no room for doubting. The basaltic lavas, it is true, were of a moro liquid character and spread farther from the centres of eruption than did the andesitic streams ; but in this respect they strikingly resemble the modern basalts of Iceland and the Sandwich Islands. There is clear evidence, however, that, as in Etna, the basalts were not always ejected from the central vents of the great volcanoes of the period, but from parasitical cones on the flanks of the mountains ; and a little to the north of Tobermory,in Mull, we find evidence of one such parasitical cone of exceptional dimensions, the plug of lava consoli- dated in the vent of the volcano being clearly traceable. There is no evidence that any of the volcanic rocks ejected during the first period of activity were thrown out beneath the waters of the ocean all traces of tuffs with marine remains being wanting. That the rocks of ths second period were of subaerial origin there can be no doubt, for the basaltic lavas alternate with beds of tuff, river gravels, beds of lignite and old soils (burnt to a red colour by the heat of the lava), with other evidences of lacustrine, fluviatile, and terres- trial conditions. Some of the lakes in Ireland and Scotland which existed during this period contain deposits of a pisolitic ironstone, evidently formed by the action of organisms like those which at the present day form the lake-ores of Sweden (seep. 49). The plant-remains found in the strata intercalated with the basalts of this second period consist of the living fern Onoclea sensi- bilis, L., with forms of Thuja, Sequoia, Gingko, Platanus, Corylus, &c., similar to those found in beds alternating with the basalts of Greenland, and having decidedly American affinities. The late Pro- fessor Heer regarded this flora as a Miocene one, while Mr. Starkie Gardner is disposed to refer it to the Montian or very oldest Eocene. In the present state of our knowledge it is probably not safe to attempt any more exact correlation of these strata than is 492 BRITISH TERTIARY VOLCANOES [CH. xxxm. involved in placing them in the Older Tertiary. At the end of this second period of activity, the five great volcanoes of the Hebrides probably rivalled Etna in their dimensions. Third Period of Volcanic Activity. The third period of eruption was characterised by a great variety of volcanic products. Very acid rocks (rhyolites), with many different types of andesite and basalt, appeared in numerous sporadic outbursts usually thrown up to form lines of ' puys ' like those of Auvergne, radiating from the five great centres which had become extinct. Most of the rhyolites appeared to have belonged to the class of soda-rhyolites or quartz-pantellerites, and these exhibit many interesting and glassy varieties. The andesites also are sometimes represented by beautiful glassy (vitrophyric) forms. As a rule, it is found that, owing to ex- tensive denudation, the surface ejections from the youngest British volcanoes, have all disappeared, and we can study only the fissures along which these eruptions took place these, being filled with con- solidated rock, constitute dykes of rhyolite, andesite, and basalt. Such dykes of rhyolite, both stony and glassy, are found traversing the granites, and all the other rocks in the Isle of Arran, and some of these belong to the class of ' composite dykes ' (see p. 477). Certain of the andesite dykes, like those of Eskdale, and some of the basalt dykes, like that of Cleveland, can be traced cutting across rocks of all ages for more than one hundred miles, and these bear witness to the great length of the chains of ' puys ' which were formed on lines radiating from the great volcanoes of the earlier periods of eruption. In a few exceptional cases, where the volcanoes were of larger size, more considerable relics have escaped denudation. Thus at Beinn Shiant, in Ardnamurchan, we find numerous lava-streams, of both stony and glassy augite-andesite, with great beds of volcanic tuff preserved under these sheets of hard lava, the whole forming the basal relic of a by no means insignificant volcano. In the Island of Eigg the remains of two streams of glassy lava which have flowed in succession down the same valley, cut in the basalts of the second period of eruption, and covered beds of gravel derived from it, have been described by Hugh Miller and Sir Archibald Geikie. Unfortunately the only organic remains pre- served in this gravel are fragments of coniferous \ood, which have been named Pinites eiggensis by Witham, so that the exact geological age of these latest volcanic rocks of the British Iblands remains doubtful (see Note Y, p. 609). TERTIARY VOLCANIC ROCKS OF THE EUROPEAN CONTINENT latest Exhibitions of Volcanic Activity in Europe. - Besides the volcanoes of Iceland, there are at present five active volcanoes on the shores or islands of the European continent namely Etna, Vesuvius, Stromboli, Vulcano, and Santorin. But several submarine eruptions have occurred in recent years in the Mediterranean ; and in various parts of Italy, Hungary, Germany, and France numerous thermal springs bear testimony to the fact that the igneous forces, though dormant, are not extinct in the area. CH. xxxiii.] RECENT VOLCANOES OF EUROPE 493 One portion of the lavas, tuffs, and trap-dykes of Etna, Vesuvius, the island of Ischia, and the Lipari Islands has been produced within the historical era ; another and a far more considerable part origi- nated at times immediately ante- cedent, when the waters of the Mediterranean were already in- habited by the existing species of mollusca, but when certain species of Elephant, Rhinoceros, and other quadrupeds, now extinct, inhabited Europe. Vesuvius. In the ' Principles of Geology ' the history of the changes which the volcanic region of Campania is known to have undergone during the last 2,000 years has been traced. The aggre- gate effect of igneous operations during that period is far from insig- nificant, comprising as it does the formation and the repeated recon- struction of the modern cone of Vesuvius since the year 79, and the production of several minor cones in Ischia, together with that of Monte Nuovo in the year 1538. Lava-currents have also flowed upon the land and along the bottom of the sea ; volcanic sand, pumice, and scoriae have been showered down so abundantly that whole cities were buried ; tracts of the sea have been filled up or con- verted into shoals; and tufaceous sediment has been transported by rivers and land-floods to the sea. There are also proofs, during the same recent period, of a permanent alteration of the relative levels of the land and sea in several places, and of the same tract having, near Pozziioli, been alternately upheaved and depressed to the amount of more than 20 feet. In connection with these convulsions, there are found, on the shores of the Bay of Baise, recent tufaceous strata, filled with articles fabricated by the hands of man, and mingled with marine shells. It has also been stated, that when we examine this same region, it is found to consist largely of tufaceous strata, of a date anterior to human history or tradition, which are of such thickness as to consti- tute hills from 500 to more than 2,000 feet in height. Some of these strata contain marine shells which are exclusively of living species, others contain a slight mixture (1 or 2 per cent.) of species not known as living. The ancient part of Vesuvius is called Somma, and consists of the remains of an older cone which was partly destroyed by the first historic explosion (see fig. 648, p. 468). Auvergrne. Although the latest eruptions in Central France seem to have long preceded the historical era, they are so modern as to have a very intimate con- nection with the present superficial outline of the country and with the existing valleys and river-courses. Among a great number of cones with perfect craters, one called the Puy de Tartaret sent forth a lava current which can be traced up to its crater and which flowed for a distance of 13 miles along the bottom of the present valley to the village of Nechers, covering the alluvium of the old valley in which were preserved the bones of an extinct species of horse, and of a lagomys and other quadrupeds all closely allied to recent animals, while the associated land- shells were of species now living, such as Cyclo- stoma elegans, Drap., Helix hor- tensis, List., H. nemoralis, L., H. lapicida, Mull., and Clausilia rugosa, Drap. That the current which has issued from the Puy de Tartaret may, nevertheless, be very ancient in reference to the events of human history, we may conclude, not only from the divergence of the mammalian fauna from that of our day, but from the fact that a Roman bridge of such form and construction as continued in use only down to the fifth century, but which may be older, is now seen at a place about a mile and a half from St. Nectaire. This ancient bridge spans the river Couze with two arches, each about 14 feet wide. These arches spring from the lava of Tartaret, on both banks, showing that a ravine precisely like that now existing had already been ex- cavated by the river through that lava thirteen or fourteen centuries ago. 494 LATER TERTIARY [CH. .XXXltl. Puy cle Gome, The Puy de Come and its lava-current, near Clermont, may be mentioned as another minor volcano of about the same age. This conical hill rises from the granitic platform, at an angle of between 30 and 40, to the height of more than 900 feet. Its summit presents two distinct craters, one of them with a vertical depth of 250 feet. A stream of lava takes its rise rt the western base of the hill, instead of issuing from either crater, and descends the granitic slope towards the present site of the town of Pont Gibaud. Thence it pours in a stated (p. 288) that Pleistocene for- mations occur in the neighbourhood of Catania, while the oldest lavas of the great volcano are Pliocene. These last are seen associated with sedimentary deposits at Trezza and other places on the southern and eastern flanks of the great cone. Cyclopean Islands. The Cy- clopean Islands, called by the Sicilians 'dei Faraglioni,' in the sea-cliffs of which these beds of clay, lava, and tuff are laid open to view, are situated in the Bay of Trezza, and may be regarded as the extremity of a promontory severed from the mainland. Here Fig. 675. View of the Isle of Cyclops in the Bay of Trezza. (Drawn by Capt. Basil Hall.) broad sheet down a steep declivity into the valley of the Sioule, filling the ancient river-channel for the distance of more than a mile. The Sioule thus dispossessed of its bed has worked out a fresh one between the lava and the granite of its western bank ; and the excavation has disclosed, in one spot, a wall of columnar basalt about 50 feet high. Newer Pliocene volcanic rocks. The more ancient portion of Vesuvius and Etna originated at the close of the Newer Pliocene period, when less than ten, some- times only one, in a hundred of the shells differed from those now living. In the case of Etna, it was before numerous proofs are seen of sub- marine eruptions, by which the argillaceous and sandy strata were invaded and cut through, and tu- faceous breccias formed. Enclosed in these breccias are many angular and hardened fragments of laminated clay in different states of alteration by heat, and intermixed with vol- canic sands. The loftiest of the Cyclopean islets, or rather rocks, is about 200 feet in height, the summit being formed of a mass of stratified clay, the laminae of which are occasion- ally subdivided by thin arenaceous layers. These strata dip to the N. W., and rest on a mass of columnar CH. XXX1II.J VOLCANOES OF EUKOPE 495 lava (see fig. 675) in which the tops of the pillars are weathered, and so rounded as to be often hemispheri- cal. In some places in the adjoin- ing and largest islet of the group, which lies to the north-eastward of that represented in the drawing (fig. 675), the overlying clay has been greatly altered and hardened by the igneous rock, and occasion- ally contorted in the most extra- ordinary manner ; yet the lamina- tion has not been obliterated, but, on the contrary, rendered much more conspicuous by the indurat- ing process. In the woodcut (fig. 676) is represented a portion of the altered rock, a few feet square, where the alternating thin laminae of sand and clay are contorted in a manner often observed in ancient meta- morphic schists. A great fissure, running from east to west, nearly divides this larger island into two parts, and lays open its internal structure. In the section thus exhibited, a dyke of lava is seen, Fig. 676. Contortions of strata in the largest of the Cyclopean Islands. first cutting through an older mass of lava, and then penetrating the superincumbent tertiary strata. In one place the lava ramifies and terminates in thin veins, from a few feet to a few inches in thick- ness. The arenaceous laminae are much hardened at the point of Fig. 677. Newer Pliocene strata invaded by lava. Isle of Cyclops (horizontal section). a. Lava. 5. Laminated clay and sand. c. The same altered. contact, and the clays are converted into siliceous schist (fig. 677). In this island the altered rocks assume a honeycomb structure on their weathered surface, singularly con- trasted with the smootti and even outline which the same beds present in their usual soft and yielding state. Dykes of Palagonia. Dykes of vesicular and amygdaloidal lava are also seen traversing marine tuff or peperino, west of Palagonia, some of the pores of the lava being empty, while others are filled with calcium carbonate. In such cases we may suppose the tuff to have resulted from showers of volcanic sand and scoriae, together with fragments of limestone, thrown out by a submarine explosion similar to that which gave rise to Graham Island in 1831. When the mass was, to a certain degree, consoli- dated, it may have been rent open, so that the lava ascended through fissures, the walls of which were perfectly even and parallel. In one case, after the melted matter that filled the rent (fig. 678) had cooled down, it must have been fractured and shifted horizontally by a lateral movement. In the second figure (fig. 679) 496 TERTIARY VOLCANOES [CH. XXXIII. the lava has more the appearance of a vein, which forced its way through the peperino. It is highly probable that similar appearances would be seen, if we could examine the floor of the sea in that part of the Mediterranean where the waves have recently washed away the new volcanic island ; for when a super- incumbent mass of ejected frag- ments has been removed by denu- dation, we may expect to see sections of dykes traversing tuff, or, in other words, sections of the channels of communication by which the subterranean lavas reached the surface. Crag of Suffolk, so well described by Mr. Searles Wood, the specific agreement between the British and Italian fossils is found to be so great, if we make due allowance for geographical distance and the difference of latitude, that we can have little hesitation in referring both to the same period, or to the Older Pliocene. It is highly probable that, between the oldest trachytes of Tuscany and the newest rocks in the neighbourhood of Naples, a series of volcanic pro- ducts might be detected of every age from the Older Pliocene to the historical epoch. Fig. 678. Fig. 679. V ' Ground-plan of dykes near Palagonia. a. Lava. &. Peperino, consisting of volcanic sand, mixed with fragments of lava and limestone. Older Pliocene Period. Italy. In Tuscany, as at Radi- cofani, Viterbo, and Aquapendente, and in the Campagna di Roma, submarine volcanic tuffs are inter- stratified with the Older Pliocene strata of the Subapennine hills in such a manner as to leave no doubt that they were the products of eruptions which occurred when the shelly marls and sands of the Sub- apennine hills were in the course of deposition. These rocks are well known to rest conformably on the Subapennine marls, even as far south as Monte Mario in the suburbs of Rome. On the exact age of the deposits of Monte Mario new light has recently been thrown by a careful study of their marine fossil shells. After the comparison of no less than 160 species of shells with the shells of the Coralline Pliocene Volcanoes of the Eifel. Some of the most perfect cones and craters in Europe may be seen on the left or west bank of the Rhine, near Bonn and And^rnach. They exhibit characters distinct from those described in other volcanic districts, owing to the large part which the escape of aqueous vapour has played in the eruptions and the small quantities of lava emitted. The fundamental rocks of the district are grey and red sandstones and shales, with some associated limestones replete with fossils of the Devonian or Old Red Sandstone group. The volcanoes broke out in the midst of these inclined strata, and when the pre- sent systems of hills and valleys had already been formed. The eruptions occurred sometimes at the bottom of deep valleys, some- CH. XXXIII.] OF EUROPE 497 times on the summit of hills, and frequently on intervening platforms. In travelling through this district we often come upon them most unexpectedly, and may find our- selves on the very edge of a crater before we had been led to suspect that we were approaching the site of any igneous outburst. Thus, for example, on arriving at the village of Gemund, immediately south of Daun, we leave the stream, which flows at the bottom of a deep valley in which strata of sandstone and shale crop out. We then climb a steep hill, on the surface of which we see the edges of the same strata dipping inwards towards the mountain. When we have ascended to a considerable height we see fragments of scoriae sparingly scattered over the surface ; until, at length, on reaching the summit, we find ourselves suddenly on the edge of a tarn, or deep circular lake basin, called the Gemunder Maar. In it we recog- nise the ordinary form of a crater, for which we have been prepared by the occurrence of scoriae scat- tered over the surface of the soil. But on examining the walls of the crater we find precipices of sand- stone and shale which exhibit no signs of the action of heat ; and we look in vain for those beds of lava and scoriae, dipping outwards on every side, which we have been accustomed to consider as charac- teristic of volcanic vents. As we proceed, however, to the opposite side of the lake, we find a consider- able quantity of scoriae and some lava, and see the whole surface of the soil sparkling with volcanic sand, and strewn with ejected fragments of half-fused shale which preserves its laminated texture in the interior, while it has a vitrified or scoriaceous coating. Other crater-lakes of circular or oval form, and hollowed out of similar ancient strata, occur in the Upper Eifel, where copious aeriform discharges have taken place, throw- ing out vast heaps of pulverised shale into the air. I know of no other extinct volcanoes where gaseous explosions of such magni- tude have been attended by the emission of so small a quantity of lava. It appears that when some of these volcanoes were in action, the river valleys had already been eroded to their present depth. Trass. The tufaceous alluvium called trass, which has covered large areas in the Eifel, and choked up some old river valleys, now partially re-excavated, is unstratified. Its base consists almost entirely of pumice, in which are included fragments of basalt and other lavas, pieces of burnt shale, slate, and sandstone, and numerous trunks and branches of trees. If, as is probable, this trass was formed during the period of volcanic eruptions, it may have originated in the manner of the fetid mud (' moya ') of the Andes. We may easily conceive that a similar mass might now be pro- duced, if a copious evolution of gases should occur in one of the lake-basins. If a breach were made in the side of the cone, the flood would sweep away great heaps of ejected fragments of shale and sandstone, which would be borne down into the adjoining valleys. Forests might be torn up by such a flood, and thus the occurrence of the numerous trunks of trees dispersed irregularly through the trass would be explained. The manner in which this trass con- forms to the shape of the present valleys implies its comparatively modern origin, probably one dating no further back than the Pliocene period. Oligoceae. Rhine- Prussia. A large portion of the volcanic rocks of the Lower Rhine are coeval with the Oligocene deposits to which most of the 'Brown Coal' of Germany belongs. The Tertiary strata of that age are seen on both sides of the Rhine, in the neighbourhood of Bonn, resting unconformably on highly inclined and vertical strata of Silurian and Devonian rocks. The Brown-Coal formation of that region consists of beds of loose sand, sand- stone, and conglomerate, clay with nodules of clay-ironstone, and occa- sionally flint. Layers of light brow n K K 498 OLDER TERTIARY [CH. XXXIII. and sometimes black lignite are interstratified with the clays and sands, and often irregulary diffused through them. They contain nu- merous impressions of leaves and stems of trees, and are extensively worked for fuel, whence the name of the formation. In several places, layers of trachytic tuff are inter- stratified, and in these tuffs are leaves of plants identical with those found in the Brown Coal, showing that during the period of the ac- cumulation of the latter, some volcanic products were ejected. The igneous rocks of the Wester- wald, and of the mountains called the Siebengebirge, consist partly of basaltic and andesitic and partly of trachytic lavas, the latter being in general the more ancient. There are many varieties of trachyte, some of which are highly crystal- line, resembling a coarse-grained granite, with large separate crystals of felspar. Trachytic tuff is also very abundant. Miocene and Oligocene volcanic rocks of Auverg-ne. The extinct volcanoes of Auvergne and Cantal in Central France seem to have commenced their eruptions in the Oligocene period, but to have been most active during the Miocene and Pliocene eras. The earliest monuments of the Tertiary period in that region are lacustrine deposits of great thick- ness, in the lowest conglomerates of which are rounded pebbles of quartz, mica-schist, granite, and other non-volcanic rocks, without the slightest intermixture of vol- canic products. To these con- glomerates succeed argillaceous and calcareous marls and limestones, containing Oligocene shells and bones of mammalia, the higher beds of which sometimes alternate with volcanic tuff of contemporaneous origin. After the filling up or drainage of the ancient lakes, huge piles of trachytic and basaltic rocks, with volcanic breccias, accumulated to a thickness of several thousand feet, and were superimposed upon granite, or the contiguous lacus- trine strata. The greater portion of these igneous rocks appears to have originated during the Miocene and Pliocene periods ; and extinct quadrupeds of those eras, belong- ing to the genera Mastodon, Rhino- ceros, and others, were buried in ashes and beds of alluvial sand and gravel, which owe their pre- servation to overspreading sheets of lava. In Auvergne, the most ancient and conspicuous of the volcanic masses is Mont Dore, which rests immediately on the granitic rocks standing apart from the freshwater strata. This great mountain rises suddenly to the height of several thousand feet above the surround- ing platform, and retains the shape of a flattened and somewhat irregu- lar cone, the slope of which is gradually lost in the high plain around. This cone is composed of layers of scoriae, pumice, and fine detritus, with interposed beds of trachyte and basalt, which descend, often in uninterrupted sheets, until they reach and spread themselves round the base of the mountain. Conglomerates, also, composed of angular and rounded fragments of igneous rocks, are observed to alternate with the above ; and the various masses are seen to dip off from the central axis, and to lie parallel to the sloping flanks of the mountain. The summit of Mont Dore terminates in seven or eight rocky peaks, where no regular crater can now be traced, but where we may easily imagine one to have existed, which may have been shattered by earthquakes, and have suffered degradation by aqueous agents. Originally, per- haps, like the highest crater of Etna, it may have formed an in- significant feature in the great pile, and, like it, may frequently have been destroyed and renovated. Respecting the age of the great mass of Mont Dore, we cannot arrive at present at any positive decision, because no organic remains have yet been found in the tuffs, except impressions of the leaves of trees of species not yet determined. It has already been stated that the earliest eruptions must have been posterior in origin to those grits and conglomerates of the freshwater formation of the CH. XXXIII.] VOLCANOES OP EUROPE 499 Limagne which contain no pebbles of volcanic rocks. But there is evidence at a few points, that some eruptions took place before the great lakes were drained, while others occurred after the desicca- tion of those lakes, and when deep valleys had already been excavated through freshwater strata. The valley in which the cone of Tartaret, before mentioned (p. 493), is situated affords an impressive monument of the very different dates at which the igneous erup- tions of Auvergne have happened ; for while the cone itself is of Pleistocene date, the valley is bounded by lofty precipices com- posed of sheets of ancient columnar trachyte and basalt, which once flowed from the summit of Mont Core in some part of the Miocene period. These Miocene lavas had accumulated to a thickness of nearly 1,000 feet before the ravine was cut down to the level of the river Couze, a river which was at length dammed up by the modern cone and the upper part of its course transformed into a lake. Eocene volcanic rocks. Monte Bolca. The fissile lime- stone of Monte Bolca, near Verona, has for many centuries been cele- brated in Italy for the number of perfect ichthyolites which it con- tains. When Lyell visited Monte Bolca, in company with Murchison, in 1828, it was ascertained that the fish-bearing beds were of Eocene date, containing well-known species of Nummulites, and that a long series of submarine volcanic erup- tions, evidently contemporaneous, had produced beds of tuff, which are cut through by dykes of basalt. There is evidence here of a long series of submarine volcanic erup- tions of Eocene date, and during some of them, as Sir R Murchison has suggested, shoals of fish were probably destroyed by the evolution of heat, by noxious gases, and tu- faceous mud, just as happened when Graham's Island was thrown up between Sicily and Africa in 1831, at which time the waters of the Mediterranean were seen to be charged with red mud, and covered with dead fish over a wide area. TERTIARY VOLCANOES OF THE ATLANTIC ISLANDS Upon the great ridge that tra- verses the Atlantic from north to south stand a number of volcanoes that were in eruption during the same periods as those of the He- brides and the North of Ireland. The date of occurrence of the out- bursts of certain of these volcanoes it has been possible to determine, in some cases from the fossils con- tained in beds intercalated with the lavas. Madeira. Although the more ancient portion of the volcanic eruptions by which the island of Madeira and the neighbouring one of Porto Santo were built up oc- curred in the Miocene period, a still larger part of the island is of Pliocene date. That the latest out- breaks belonged to the Newer Plio- cene period is inferred from the close affinity to the present floi*a of Madeira of the fossil plants pre- served in a leaf -bed in the north- eastern part of the island. These fossils, associated with some lignite in the ravine of the river San Jorge, can none of them be proved to be of extinct species, but their antiquity may be inferred from the following considerations. First the leaf-bed, discovered by Lyell and Hartung in 1853, at the height of 1,000 feet above the level of the sea, crops out at the base of a cliff fonned by the erosion of a gorge, cut through alternating layers of basalt and scoriae, the product of a vast succession of eruptions of un- known date, piled up to a thick- ness of 1,000 feet, which were all poured out after the plants, of which about twenty species have been recognised, flourished in Ma- deira. These lavas are inclined at an angle of about 15 to the north, and came down from the great central region of eruption. Their accumulation implies a long period of intermittent volcanic action, subsequently to which the ravine KK2 500 VOLCANOES OF ATLANTIC ISLANDS, [CH. xxxm. of San Jorge was hollowed out. Secondly some few of the plants, though perhaps all of living genera, are supposed to be forms not now existing in the island. They have been described by Sir Charles Bunbury and Professor Heer, and the former first pointed out that many of the leaves are of the laurel type and analogous to those now nourishing in the modern forests of Madeira. He also recognised among them the leaves of Woodwardia radicans, Smith, and Davallia ca- nariensis, Smith, ferns now abun- dant in Madeira. Thirdly fossil land shells, five per cent, of which are extinct, are found in the blown sands upon the leaf-bed. Although the greater part of the volcanic eruptions of Madeira belong to the Pliocene period, the most ancient of them are of Miocene date, as is shown by the fossil shells included in the marine tuffs which have been upraised to the height of 1,300 feet above the level of the sea at San Vicente, in the northern part of the island. A similar marine and volcanic forma- tion constitutes the fundamental portion of the neighbouring island of Porto Santo, forty miles distant from Madeira, and is there elevated to an equal height, and covered, as in Madeira, with lavas of subaerial origin. The largest number of fossils have been collected from the tuffs and conglomerates and some beds of limestone in the island of Baizo, off the southern extremity of Porto Santo. The species amount, in this single locality, to more than sixty, of which about fifty are mol- lusca, but many are only casts. Some of the shells probably lived on the spot during the intervals between eruptions, and some may have been cast up into the water or air together with muddy ejections, and, falling down again, have been deposited on the bottom of the sea. The hollows in some of the frag- ments of vesicular lava, of which the breccias and conglomerates are composed, are partially filled with calcite, being thus half con- verted into amygdaloids. Among the fossil shells common to Madeira and Porto Santo, large Cones, Strombs, and Cowries are con- spicuous among the univalves, and Cardium, Spondylus, and Litho- domus among the lamellibranchiate bivalves, while among the Echino- derms the large Clypeaster altus, L., an extinct European Miocene form, is found. The largest list of fossils has been published by Karl Mayer in Hartung's ' Madeira.' Mayer iden- tifies one-third of the Madeira shells with known European Mio- cene forms. Grand Canary. In the Ca- naries, especially in the Grand Canary, the same marine Upper Miocene formation is found. Stra- tified tuffs, with intercalated con- glomerates and lavas, are there seen in nearly horizontal layers in sea-cliffs about 800 feet high near Las Palmas. Lyell and Hartung were unable to find marine shells in these tuffs at a greater elevation than 400 feet above the sea ; bufc as the deposit to which they belong reaches to the height of 1,100 feet or more in the interior, they be- lieved that an upheaval of at least that amount had taken place. The Clypeaster altus, L., Spondylus gce- deropus, L., Pectunciduspilosus,J-i. sp., Cardita calyculata, L. sp., and several other shells, serve to identify this formation with that of the Madeiras, and Ancillaria glandiformis, Lam, which is not rare, and some other fossils, re- mind us of the faluns of Touraine. These tuffs of the southern shores of the Grand Canary, con- taining the Miocene shells, appear to be of about the same age as the most ancient volcanic rocks of the island. Over the marine lavas and tuffs, trachytic and basaltic pro- ducts of suba/erial volcanic origin, between 4,000 and 5,000 feet in thickness, have been piled, the central parts of the Grand Canary reaching the height of about 6,000 feet above the level of the sea. A large portion of this mass is of Pliocene date, and some of the latest lavas have been poured out since the time when the valleys were already excavated to within a few feet of their present depth, CE, XXXIII.] NORTH AMERICA, ETC. 501 Azores. In the island of St. Mary's, one of the Azores, marine fossil shells have long been known. They are found on the north-east coast on a small projecting pro- montory called Ponta do Papagaio (or Point Parrot) chiefly in a lime- stone, about twenty feet thick, which rests upon, and is again covered by, basaltic lavas, scoriae, and conglomerates. The pebbles in the conglomerate are cemented together by calcium carbonate. One of the most characteristic and abundant of the species, Car- dium Hartungi, Bronn, not known as fossil in Europe, is very common in Porto Santo and Baixo, and serves to connect the Miocene fauna of the Azores and the Madeiras. In some of the Azores, as well as in the Canary Islands, the volcanic fires are not yet extinct, as the re- corded eruptions of Lanzerote, Teneriffe, Palma, St. Michael's, and others attest. The late sound- ings (1873) of H.M.S. ' Challenger ' have shown the Azores, Canaries, Cape de Verde Islands, &c., to be merely the highest summits of a great submerged mountain ridge, comparable with the Andes of South America both in extent and altitude, as well as in the volcanic character of many of its most ele- vated peaks (see NotejZ, p. 609). TERTIARY VOLCANOES IN OTHER PARTS OF THE WORLD All over the globe we find evi- dence that the older parts of vol- canoes still active, and many volcanoes now extinct, were in eruption during different parts of the Tertiary period. Hindostan. A vast volcanic area exists in the Western and Central parts of the Peninsula of Hindostan, called ' the Deccan and Malwa Trap.' It covers 200,000 square miles, and is found as numerous flows of earthy basalt, amounting to 8,000 feet in thick- ness. This vast deposit seems to have come from fissures upon which the volcanic cones, which were doubtless thrown up, have not been preserved, and the flows covered an old terrestrial surface of the Upper Cretaceous age. The lava flows continued during a vast period of time (for lake- beds exist amongst them with their fossils silicified), and are older than the Nummulitic period. The United States. In the Western Territories of the United States the Sierra Nevada has a great thickness of auriferous gravels of Pliocene age, covered by basalt, and this is a part of the result of a grand series of eruptions from vol- canoes, which continued probably into the Historic period. Such flows are vast in amount in other regions, and. were one of the great phases in the development of the physical features of the continent after the upheaval of the mountain systems. The great lakes to the west of the Rocky Mountains were in existence when the outflows took place in and to the west of the mountains. The basalts form im- portant features in Nevada, Oregon, Idaho, Utah, &c. Besides the basaltic rocks, great masses of andesites (altered into propylites)with a few trachytes and phonolites, and many interesting rhy elites and obsidians were erupted during the Tertiary period in this part of the earth's surface. The geyser district of the Yellowstone Park, where so many traces of volcanic activity are still to be witnessed, has become very famous, and the phenomena displayed there have been carefully studied by the United- States geologists, Messrs. Hague and Iddings. Australia. Vast volcanic flows occurred in Australia during the Tertiary ages, and those of Queensland and of Victoria are of great importance, both geologically and economically. Marine and freshwater deposits, the ages of which can be determined by their fossils, are covered by, or 1'est upon, great thicknesses of dolerites. There are two series of outflows, an upper and a lower, and the 502 VOLCANOES OF MESOZOIC [CH. XXXIV. auriferous deposits are covered by the upper and rest on the lower. Where the upper or Pliocene basalt is absent or has been denuded, the sedimentary strata at once afford the gold-seeker his clue ; for if they contain marine fossils, they are older than the age when the denudation of exposed auriferous quartz-reefs permitted the accumulation of auriferous deposits. The marine strata are of Miocene age, and the basalt covering them underlies the aurife- rous freshwater deposits, the results of the denudation of higher ground than that covered by the older and marine series. In the northern part of Queensland, north of lat. 21, the upper volcanic series consists of well-defined craters and great lava flows, which are older than the Pleistocene marsupials (p. 241), the foreshadowers of the existing fauna. In Queensland, these Plio- cene flows cap a ' desert sandstone,' and in Victoria, gravels, conglome- rates, cement-beds, and other Plio- cene auriferous deposits. The Victorian Pliocene volcanic flow is at a considerable altitude, and has been much denuded. Beneath the Pliocene volcanic rocks an older series occurs, both in Victoria and Queensland, which has been referred to the Miocene. In the case of these older lava-flows, all traces of the cones and craters from which they were emitted would seem to have disappeared, probably through denudation. CHAPTER XXXIV VOLCANIC ROCKS OF THE MESOZOIC, PALAEOZOIC, AND ARCHAEAN ERAS Absence of evidence of volcanic action in Cretaceous and Jurassic times in the British Isles and Western Europe Triassic Volcanoes of Devonshire Permian Volcanoes of Scotland Volcanoes of the Car- bonifarous Period Buried trees of Arran Volcanoes of the Old Red Sandstone and Devonian Period Volcanoes of the Silurian, Ordovician, and Cambrian Periods Pre-Cambrian Volcanoes Pre-Tertiary Vol- canoes of other parts of the globe Cretaceous and Jurassic volcanic Rocks of Greece Newer and older Palaeozoic Volcanoes of Central Europe Pre-Cambrian volcanic Rocks of Canada. IN the British Islands we have no volcanic rocks of either Cretaceous or Jurassic age, and the same is true of Western Europe generally. It is this circumstance which led continen- tal geologists to regard the Tertiary volcanic rocks as having a totally different character and origin from the igneous products of the older geological periods ; it was found impossible, in many cases, to show a complete sequence and transition from the older to the newer volcanic rock-masses, and the former were supposed to have been extruded under the sea from great fissures in the earth's crust, being called ' trap-rocks ' (from the Swedish trappa, a stair). Volcanic Rocks of the Triassic Period. The youngest pre-Tertiary volcanic rocks which occur in the British Islands CH. xxxiv.] AND PALEOZOIC PERIODS 503 are probably those found in the south-west of England, which were described by Sir Henry de la Beche. In the southern part of Devonshire volcanic rocks are associated with New Bed Sandstone, and, according to De la Beche, have not been intruded subsequently into the sandstone, but were produced by contemporaneous volcanic action. Some beds of grit, mingled with ordinary red marl, resemble ashes ejected from a crater ; and in the stratified conglomerates occurring near Tiverton are many angular fragments of por- phyrite or altered andesite, some of them one or two tons in weight, intermingled with fragments of other rocks. These angular fragments were probably thrown out from volcanic vents, and fell upon sedimentary matter then in the course of deposition. There still appears to be some doubt, however, whether the red sandstones with which these volcanic rocks are associated may not be of Permian rather than of Triassic age, and that they were so was the view maintained by Murchison. Volcanic Rocks of the Permian Period. The researches of the officers of the Geological Survey in Scotland have led to their mapping a considerable number of small and scattered volcanic masses, consisting of small intrusive masses, and some- times of lava-sheets with interbedded tuffs. The lavas are generally porphyrites and melaphyres (altered andesites and basalts), but the evidence on which a Permian age has been assigned to them cannot be regarded in most cases as altogether satisfactory. The volcanoes, which were in nearly all cases of small size, must be regarded as examples of sporadic outbursts similar to the ' puys ' of Auvergne, and like them marking the decline of volcanic activity in the districts where they occurred. Volcanic rocks which have been referred to the Permian period by the officers of the Geological Survey occur in Ayrshire, in Nithsdale, in Dumfriesshire, and away through Central Scotland into Fifeshire. Volcanic Rocks of the Carboniferous Period. Two ex- tensive developments of volcanic rocks occur in the Carboniferous basin of the Forth in Scotland. One of these is well exhibited along the shores of Fifeshire, where the igneous masses consist of basalt, sometimes with olivine, and of dolerites. These appear to have been erupted while the sedimentary strata were in a horizontal position, and to have suffered the same disloca- tions which those strata have subsequently undergone. In the associated volcanic tuffs of this age are found not only frag- ments of limestone, shale, flinty slate, and sandstone, but also pieces of coal. Other volcanic rocks connected with the Car- 504 NEWEE PALEOZOIC VOLCANOES [CH. xxxiv. boniferous formation may be traced along the south margin of Stratheden, and constitute a ridge parallel with the Ochils, ex- tending from Stirling to near St. Andrews. These consist almost exclusively of dolerite, becoming, in a few instances, earthy and amygdaloidal. They are either interbedded with, or intruded among, the sandstone, shale, limestone, and ironstone of the Lower Carboniferous. The Cement -stone group (p. 370) accumulated, writes Sir A. Geikie, in a region of shallow lagoons, islets, and coal-growths, which was dotted over with innumerable active volcanic vents. The eruptions continued into the time of the Carboniferous limestone, but ceased before the deposition of the Millstone grit. Close-grained basalts and dolerites were formed with felsites, porphyrites, and tuffs. Beneath this group there are evidences of vast volcanic flows, some sheets being 1,500 feet thick. The most persistent zone of volcanic rocks in the Scottish Carboniferous system is that which succeeds the lower part of the Calciferous Sandstones. Composed of successive sheets of porphyrites and tuffs, it sweeps in long isolated ranges of hills from Arran and Bute on the west to the mouth of the estuary of the Forth on the east, and from the Campsie Fells on the north to the heights of Ayrshire, and still further south to Berwickshire, Liddesdale, and the English border. Erect trees buried in volcanic ash at Arran. An interesting discovery was made in 1867 by Mr. E. A. Wiinsch in Carboni- ferous or Permian strata of the north-eastern part of the island of Arran. In the sea-cliff, about five miles north of Corrie, near the village of Laggan, strata of volcanic ash occur, forming a solid rock cemented by calcium carbonate and enveloping trunks of trees, determined by Mr. Binney to belong to the genera Sigillaria and Lepidodendron. The trees with their roots occur in two distinct strata of volcanic tuff, parallel to each other, and inclined at an angle of about 40, having between them beds of shale and coaly matter seven feet thick. It is evident that the trees were overwhelmed by showers of ashes from some neighbouring volcanic vent, as Pompeii was buried by matter ejected from Vesuvius. The trunks, several of them from three to five feet in circumference, remained with their Stigmarian roots spreading through the stratum below, which had served as a soil. Arthur's Seat, Edinburgh, is the relic of a volcanic cone, and commencing as a fissure in the Calciferous Sandstone age, gave forth andesitic and basaltic lavas, of which much was forced amongst the surrounding strata, to form the Salisbury CH. xxxiv.] OF THE BEITISH ISLES 505 Crags and other intrusive sheets at the base of the hill. The eruption probably took place in shallow water, and after a while elevation occurred and agglomerates collected, forming the higher part of the mass. The volcanic relics have partici- pated in the general movements of the area since the Carboniferous age, and have since suffered great denudation. Evidences of similar sporadic or * puy-like ' eruptions during the Carboniferous period are found scattered all over the Central Valley of Scotland. The rocks of which these old Carboniferous volcanoes were composed have been described by Mr. Allport and Sir A. Geikie. They consist for the most part of andesites and basalt, but some trachytes and an occasional phonolite may also be found among them. Great sheets of melaphyre, porphyrite, and tuff are found in the Carboniferous limestone of Limerick, and to the north in Ireland. In Derbyshire, sheets of contemporaneous basaltic lava called ' toadstone ' occur in the limestone, and flows of the same age have been found in the Isle of Man. Volcanic Rocks of the Old Red Sandstone Period. By referring to the section explanatory of the structure of Forfar- shire, already given (p. 80), the reader will perceive that beds of conglomerate, No. 3, occur in the middle of the Old Red Sandstone system, 1, 2, 3, 4. The pebbles in these conglome- rates are sometimes composed of gneiss and quartzose rocks, sometimes exclusively of different varieties of lava, which last, although purposely omitted in the section referred to, is often found either intruding itself in amorphous masses and dykes into the old fossiliferous tilestones, No. 4, or alternating with them in conformable beds. All the different divisions of the red sand- stone, 1, 2, 3, 4, are occasionally intersected by dykes, but they are very rare in Nos. 1 and 2, the upper members of the group consisting of red shale and red sandstone. These phenomena, which occur at the foot of the Grampians, are repeated in the Sidlaw Hills ; and it appears that in this part of Scotland vol- canic eruptions were most frequent in the earlier part of the Old Red Sandstone period. These lavas belong for the most part to the class of porphyrites, their structure is often found to be amygdaloidal, the kernels being sometimes calcareous, but often siliceous, and forming beautiful agates. In a more or less decomposed condition these felspabhic lavas are known under the name of claystones. With them occur beds of stratified tuff and conglomerate. Some of these rocks look as if they had flowed as lavas over the bottom of the sea, and enveloped quartz pebbles which were lying there, so as to form con- glomerates with a base of igneous rock, as is seen in Lumley 506 OLDER PALEOZOIC VOLCANOES [CH. xxxiv. Den in the Sidlaw Hills. On either side of the axis of this chain of hills (see section, p. 80) the beds of massive lava, and the tuffs composed of volcanic ashes, dip regularly to the south- east, or north-west, conformably with the shales and sandstones. The geological structure of the Pentland Hills, near Edin- burgh, shows that igneous rocks were there formed during the Devonian or 'Old Eed ' period. These hills rise 1,900 feet above the sea, and consist of conglomerates and sandstones of Devonian age, resting on the inclined edges of grits and slates of Upper Silurian date. The contemporaneous volcanic rocks intercalated in this Lower Old Eed Sandstone consist of felspathic lavas or felstones, with agglomerates and ashy beds. Volcanic rocks are found associated with the strata of the Lower, Middle, and Upper Old Eed Sandstone, and occur in the Cheviot Hills (where beautiful glassy enstatite-andesites are found), in the Central Valley of Scotland, and in the Orkney and Shetland Islands. Similar rocks are found associated with Old Eed Sandstone strata in the Killarney district in Ireland, and with the Marine Devonian beds of the South- West of England. The lavas of this period comprise andesites and basalts, often much altered, and more rarely the acid rhyolites. Silurian and Ordovlcian volcanic rocks. The Upper Silurian series of the West of Ireland shows successive sheets of lava and tuffs forming conspicuous bands amongst the stratified rocks. The volcanic series of the Lake-district of the North- West of England is of vast thickness, and intervened between the Skiddaw slates and the Coniston limestone and shale. It occupied much of the Bala age, and all that of the Llandeilo, and part of the Arenig epoch of Wales. The Snowdonian hills, in Caernarvonshire, consist, in great part, of volcanic tuffs, the oldest of which are interstratified with the Bala and Llandeilo beds. There are contemporaneous felsitic lavas of this era, which altered the slates on which they repose, having doubtless been poured out over them in a melted state, whereas the slates which overlie them, having been deposited after the lava had cooled and consolidated, have entirely escaped alteration. But there are ' greenstones ' associated with the same formation, which, although they are often conformable to the slates, are in reality intrusive rocks. They alter the stratified deposits both above and below them. Volcanic action occurred largely during the formation of the Arenig strata, and felsitic or rhyolitic lavas were erupted, and interstratified with fossiliferous deposits. Tuffs added to the bulk of the whole. Cader Idris, the Arans, the Arenigs, and CH. xxxiv.] OF THE BRITISH ISLES 507 other mountains are thus built up. Similar volcanic rocks of Ordovician age occur in Scotland and Ireland. Cambrian volcanic rocks. On the western flank of the Malverns in Herefordshire, some black shales belonging to the Upper Lingula Flags are interstratified with thin sheets of vesi- cular lava that were probably erupted beneath the sea contem- poraneously with the deposition of the muddy sediment. The shales lying beneath the volcanic rock are white, as if calcined by the molten lava, while those lying above have retained their normal black colour. In speaking of this ancient volcanic outburst, the late Professor Philipps said : ' One might mistake the ferruginous and cellular stone for the subaerial reliquise of a volcano in Auvergne,' a district where the erupted volcanic matter is clearly contemporaneous with the associated sedi- mentary deposits. Pro-Cambrian volcanic rocks. Beneath the lowest fos- siliferous Cambrian rocks, and the basal conglomerate of the formation in Wales and elsewhere, is a vast volcanic series, the agglomerates, tuffs, and flows of which have been altered to a certain extent by metamorphic action. These Pebidian rocks have already been noticed (p. 434). Beneath the halleflintas is the great group known as Dimetian series, in which metamorphic and granitoid masses with acid volcanic rocks are found. The Lewisian (or Fundamental) gneiss of Scotland is largely made up of rocks which are evidently metamorphosed igneous masses, some of them having apparently been extruded in the form of lavas and tuffs during the ancient periods when these rocks were formed. PRE-TERTIARY VOLCANIC ROCKS IN OTHER PARTS OF THE WORLD The absence of Mesozoic volcanic rocks, which is so marked a feature in the British Islands and throughout Western Europe, is not noticed in other parts of the globe. Even in the eastern part of our own Continent important volcanic rocks are found intercalated with Cretaceous and Jurassic strata. Cretaceous Period. M. calcareous kernels, and a base of Virlet has shown in his account of serpentine. In certain parts of the the geology of the Morea, that Morea, the age of these volcanic certain volcanic rocks in Greece are rocks is established by the following of Cretaceous date ; as those, for ex- proofs : first, the lithographic lime- ample, which alternate conformably stones of the Cretaceous era are cut with Cretaceous limestone and through by volcanic rocks, and then greensand between Kastri and Da- a conglomerate occurs, at Nauplia mala in the Morea. They consist in and other places, containing in its great part of diallage rocks and ser- calcareous cement many well-known pentine, and of an amygdaloid with fossils of the chalk and greensand 508 FOREIGN VOLCANIC ROCKS [CH. XXXTT. together with pebbles formed of rolled pieces of the same serpen- tinous rocks as appear in the dykes above alluded to. Period of Oolite and Lias. Although the green and serpenti- nous volcanic rocks of the Morea be- long chiefly to the Cretaceous era, as before mentioned, yet it seems that some eruptions of similar rocks began during the Oolitic period; and it is probable that a large part of the volcanic masses called ophiolites in the Apennines, and associated with the limestone of that chain, are of corresponding age. Important masses of volcanic rock in the Rajmahal district of Hin- dostan are of Jurassic age. Volcanic rocks of Itteso- zoic and Palaeozoic Age. In Central Europe we find vast thicknesses of lavas and tuffs of acid, intermediate, and basic compo- sition alternating with sediments of Triassic, Permian, Carboniferous, Devonian, Silurian, Ordovician, and Cambrian age, and similar rocks are found in Southern Europe belonging to various portions of the Newer and Older Palaeozoic as well as to the Trias. The periods of volcanic eruption in our own islands in the Triassic and Permian, the Devonian, Ordovician, and Cambrian were equally periods of great igneous activity all over Western Europe. The products of volcanic activity in these several periods maintain a In his addresses to the Geo- logical Society for 1891 and 1892, Sir Archibald Geikie has given an admirable summary of the work done by the Geological Survey, so far as it -has gone, in determining the age of the various masses of volcanic rock met with in associa- tion with the strata of the British Islands. remarkable uniformity of character over tolerably wide districts, and it is thus possible to define even in these areas the boundaries of great ' petrographical provinces.' Iiaurentian volcanic rocks. The Laurentian rocks in Canada, especially in Ottawa and Argenteuil, are among the oldest intrusive masses yet known. They form a set of dykes of a fine-grained dolerite, composed of felspar and pyroxene, with occasional scales of mica and grains of pyrites. Their width varies from a few feet to a hundred yards, and they have a columnar structure, the columns being truly at right angles to the sides of the dykes. Some of the dykes send off branches. These dolerites are cut through by intrusive syenite, and this syenite, in its turn, is again cut and pene- trated by porphyritic felsite. All these old volcanic rocks appear to be of Laurentian date, as the Cambrian and Huronian rocks rest uiicon- formably upon them. Whether some of the various conformable crystalline rocks of the Laurentian series, such as the coarse-grained granitoid and porphyritic varieties of gneiss, exhibiting scarcely any signs of foliation, and some of the serpentines, may not also be of volcanic origin, is a point very diffi- cult to determine in a region which has undergone such extreme meta- morphic action. This information has been since extended and embodied in his work ' The Ancient Volcanoes of Great Britain ' ( 1897), to which the reader is referred for fuller details on this subject. For the description of volcanicrocksin other countries, we are indebted to the writings of many geologists in Europe and the United States. 509 PAET IV PLUTONIC ROCKS CHAPTER XXXV PLUTONIC EOCKS, THEIR NATURE AND COMPOSITION Analogy of the Plutonic Rocks with those of Volcanic origin Proofs of the deep-seated origin of Plutonic Rocks Chemical composition of the different classes of Plutonic Rocks Changes which they undergo Liquid cavities in the crystals of Plutonic Rocks Order in which the several minerals crystallise in Plutonic Rocks Granite and its varieties Syenites, &c. Diorites, &c. Nepheline Syenites and Theralites Gabbro and its varieties Ultra-acid Rocks Ultra-basic Rocks Peridotites Pyroxenites Amphibolites Relations of the Ultra-basic Rocks to Meteorites. THE plutonic rocks may be treated of next in order, as they are most nearly allied to the volcanic class already considered. In the first chapter we have described these plutonic rocks as a division of the crystalline or hypogene formations, and have stated that they differ from the volcanic rocks, not only by their more crystalline texture, but also by the absence of tuffs and breccias, which are the products of eruptions at the earth's surface, whether thrown up into the air or beneath the sea. They differ also by the absence of pores or cellular cavities to which the expansion of the entangled gases gives rise in ordinary lavas. From these and other peculiarities, it has been inferred that the granites have been formed at considerable depths in the earth, and have cooled and crystallised slowly, under great pres- sure, where the occluded gases could not expand. The volcanic rocks, on the contrary, although they also have risen up from below, have cooled from a melted state more rapidly upon or near the surface. From this hypothesis of the great depth at which the granites originated has been derived the name of ' Plutonic rocks.' 510 STRUCTURE AND ORIGIN [CH. xxxv. The heat which in every active volcano extends downwards to indefinite depths, must produce, simultaneously, very different effects near the surface and far below it ; and we cannot sup- pose that rocks resulting from the crystallising of fused matter under a pressure of several thousand feet, much less several miles, of the earth's crust, can exactly resemble those formed at or near the surface. Hence the production at great depths of a class of rocks analogous to the volcanic, and yet differing in many particulars, might have been predicted, even had we no plutonic formations to account for. It has, however, been objected, that if the granitic and volcanic rocks were simply different parts of one great series, we ought, in mountain chains, to find volcanic dykes passing upwards into lava and downwards into granite. But we may answer that our vertical sections are usually of small extent ; and if we find in certain places a transition from solid to porous lava, and in others a passage from granitic rocks to solid lava, it is as much as we could expect from this kind of evidence. The plutonic formations agree with the volcanic in exhibiting veins or ramifications proceeding from central masses into the adjoining rocks, and causing alterations in these last, which will be presently described. They also resemble volcanic masses in containing no organic remains ; but they differ in being more uniform in texture, whole mountain masses of indefinite extent appearing to have originated under conditions almost precisely similar. The most striking analogies between the Plutonic and the Volcanic rocks are seen, however, when we study their chemi- cal composition and their mineralogical constitution. Every variety of lava acid, intermediate and basic has its exact counterpart in the series of plutonic rocks; it is only in its structure that a lava differs from its plutonic representative. While the lavas are sometimes wholly glassy in structure, and in almost all cases crypto -crystalline or micro-crystalline in their base or ground-mass, the plutonic rocks usually exhibit a more perfectly crystalline structure and often pass into masses that consist entirely of crystals of different minerals without any intervening base or ground-mass ; in such cases we speak of the rock as being ' holocrystalline.' As was suggested by Jukes, it is probable that if we could trace a mass of pumice downwards to greater and greater depths in the earth's crust, we should find the pumice losing its porous character, and becoming solid glass (or obsidian) ; the glassy obsidian by the development in it of crystallites and microlites would gradually acquire more and more stony CH. xxxv.] OF PLUTONIC KOCKS 511 characters (rhyolite and quartz -felsite) ; and finally, as the crystals increased in size and perfection of development, the rock would assume the perfectly holocrystalline character (micropegmatite and granite). Similar changes could doubtless be traced in each variety of intermediate and basic lavas as it was followed to depths where it must have consolidated at a slower rate and under greater pressure. In each series, the lavas overlap their plutonic representa- tives. The central portions of massive lava- streams are often more highly crystalline than the materials of narrow plutonic dykes or veins. Occasionally, indeed, truly plutonic rocks, in small masses, may consolidate as a glass. While every known lava has its plutonic counterpart, there are a few deep-seated rocks, as we shall see hereafter, which seem to have no representatives among those erupted at the surface. The rocks of these peculiar types, which have only been found in plutonic dykes, constitute the class of ' dyke-rocks.' In the deeper parts of volcanic masses which have been exposed to our view by denudation, we find rocks which we may with equal propriety speak of as ' plutonic ' or * volcanic.' Many of the plutonic rocks, like their volcanic analogues, are found to have undergone great chemical, mineralogical, and structural alterations, so that materials are produced differing very greatly indeed from the original rocks. As the result of such alteration, glassy materials become crystalline (secondary devitrification) ; minerals undergo metamorphoses, without alteration of chemical composition (paramorphism) , or with such change (pseudomorphism) ; and in some cases the whole mass may become completely recrystallised with the formation of entirely new minerals. The older a rock, the more likely is it to have undergone such changes, and this circumstance led the older geologists to suppose that fundamental differences existed between the rocks of the earlier geological periods and those which have been formed in Tertiary and recent times. But the more carefully the most ancient igneous rocks are studied, the more clearly does it appear that the difference between the igneous products of the older geological periods and those of the present day are not essential but accidental being the result of mechanical and chemical changes which they have undergone since their first consolidation. There is no ground whatever for believing that the rocks formed during the earlier periods of the earth's history differ either in chemical or mineralogical charac- ters from those which are being consolidated within the earth's crust at the present day. There is one respect in which the minerals of deep-seated or 512 CAVITIES IN MINERALS [CH. XXXV. plutonic rocks strikingly differ from those of the lavas formed at the earth's surface. The minerals of lavas contain cavities sometimes filled with gas (gas-enclosures) ; or with vitreous materials (glass-enclosures) ; or with the devitrified products of glass (stone-enclosures). The minerals of deep-seated or plutonic rocks, however, frequently contain in their cavities liquids (often with movable bubbles), and the liquids sometimes have floating about in them crystals showing that they are supersaturated solutions (see fig. 680). Sometimes crystals are found containing two different kinds of liquids at the same time. By determining the coefficient of expansion of the liquids and their critical points, and by submitting them to spectral or chemical analysis, it has been proved that they are sometimes liquid carbon dioxide (it may be mixed with other gases liquefied by pressure), at other times supersaturated aqueous solutions Fig. Cavities seen in the crystals of rocks. a. Gas-cavity of irregular form. b. Liquid-cavities with bubbles (these cavities are irregular in form), c. Cavity bounded by crystalline planes of the mineral (quartz) in which it is enclosed, and containing two liquids with a bubble. Cavities thus bounded by crystalline planes are called ' negative crystals.' d. Similar cavity with liquid and bubble. The liquid contains a cubic crystal, e. Glass-cavity. /. Stone-cavity (both of these are negative crystals), ft, c, d are found in minerals of plutonic rocks ; a, e,fin minerals of lavas. of the alkaline chlorides and sulphates. The presence of these ' liquid- enclosures ' in the minerals of plutonic rocks affords striking evidence of the enormous pressures under which these rocks must have consolidated. Two methods have been suggested whereby we may possibly be able to determine the actual temperature and pressure, and hence the depth in the earth's crust, at which a crystalline rock must have been formed. Sorby pointed out that, by measuring the relative size of the cavity and the gas-bubble in a liquid- enclosure, physicists may arrive at a definite conclusion con- cerning the exact conditions of crystallisation. Eenard, on the other hand, would seek for the data required, by measuring the bulk of the crystals floating in the liquid of a cavity and com- paring this with the volume of the supersaturated solution in which they are suspended. But our knowledge of the behaviour CH. xxxv.] TYPES OF PLUTONIC ROCKS 513 of liquids and solutions at excessively high pressures and tem- peratures is insufficient to make calculations based on either kind of data of much practical value to geologists. The cavities found in the crystals of plutonic rocks are sometimes so minute and numerous that many millions of them must exist in every cubic inch of the rock. In form, these cavities are very varied ; sometimes they are most irregular and exhibit fine ramifications that communicate with one another ; in other cases they present the crystal-faces of the mineral in which they are enclosed forming what mineralogists know as negative crystals (see fig. 680, c, d, e, f). The holocrystalline or granitic forms of rocks corresponding to the chief types of lavas are named as follows : Lava Khyolite Holocrystalline form Granite. Trachyte Syenite Phonolite Elseolite-syenite Andesite Diorite Tephrite Theralite ,, Basalt Gabbro The commonest plutonic rocks, Granite, Diorite, and Gabbro, are those which correspond to the most abundant lavas Rhyolite, Andesite, and Basalt. In addition to the holocrystalline forms of plutonic rocks we find hypocrystalline or hemicrystalline varieties in which a less perfectly crystalline ground-mass is present. Many of these rocks, which are intermediate in structure between the ' granitic ' and the lava-like or ' trachytic ' forms, have received distinctive names from petrologists. Some varieties of plutonic rocks are named from the presence of a conspicuous mineral either essential, accessory, or even secondary while other types again are distinguished by the nature and amount of change which the minerals of the rock have undergone since its first formation. We have seen that the basic rocks have a higher density or specific gravity than the intermediate, and the intermediate than the acid rocks. If a plutonic rock be melted and cooled rapidly, it forms a glassy mass with a much lower specific gravity than the crystalline rock from which it was produced. The lavas have always a lower specific gravity than their plutonic and more highly crystalline counterparts. Hence the determination of the specific gravity of an igneous rock with an inspection of its degree of crystallisation may enable us to draw a safe conclusion as to its chemical composition. By examining a crystalline rock, especially in thin sections L L 514 GRAPHIC STRUCTURE [OH. xxxv. under the microscope, we may determine the order in which the several minerals have crystallised out in a magma. Most rocks exhibit minerals belonging to different ' periods of consoli- dation,' to use the term employed by French petrographers. In igneous rocks generally the order in which the several minerals have separated is that of * decreasing basicity,' as it has been denned by Eosenbusch. Firstly. Accessory minerals like apatite, zircon, sphene, garnet, &c. Secondly. Oxides of iron and titanium magnetite, rutile, and titano-ferrite. Thirdly. The ferro-magnesian silicates olivine, pyroxenes, amphiboles, and biotites. Fourthly. The alumino-alkaline silicates. The felspars in the following order : anorthite, labradorite, andesine, oligoclase, albite, orthoclase and anorthoclase, and the felspathoids. Fifthly. Quartz. But in certain cases this order appears to be subject to some modification. Acid rocks sometimes show the quartz and felspar ri 6gl to have crystallised almost simultaneously, giving rise to the graphic or pegmatitic struc- ture (see fig. 681), which when exhibited on a microscopic scale is known as micro- graphic or micropegmatitic structure. In basic rocks the augite has sometimes crystal- lised after the felspars, and the basic mineral is seen to enclose lath-shaped crystals of the more acid one ; this gives rise to the structure called Graphic granite. Portsoy. The clear colourless crystals are quartz, the clouded by the French petrographers 'ophitic,' and by the Germans ' diabasic.' In some plutonic rocks the crystals form radial and globular aggregates like the spherulites of the lavas, and rocks with this peculiarity, such as the well-known corsite, are said to exhibit an orbicular structure. Acid Plutonic Rocks. Granite and its Varieties. The granites are holocrystalline aggregates of felspar (in which ortho- clastic varieties always predominate over plagioclastic) with quartz and mica the latter mineral being sometimes replaced by hornblende and, more rarely, by a pyroxene. The orthoclastic felspar is usually allotriomorphic (that is, not bounded by its proper crystalline planes), CH. xxxv.] POKPHYRITIC STKUCTUKE 515 except when it occurs as phenocfysts or porphyritic constituents. It is often red or pink in colour, and more rarely green ; it sometimes exhibits the microcline and perthite structures. The plagioclastic felspar, usually white grey or greenish in colour, appears to be oligo- clase or allied to that species. The micas are sometimes black (lepidomelane) and sometimes white, and when the two varieties occur together, the rock is spoken of as granite with two micas. The white mica in granites is sometimes muscovite (muscovite* granites), but sometimes a colourless biotite. The quartz is almost always allotriomophic ; it is usually colourless, but sometimes milky, while in rare cases it assumes a blue or smoky tint. In the drusy cavities of granites, the crystals of the constituent minerals are found assuming their proper form (or becoming idiomorphic). The typical granites often contain two micas, one colourless and the other deeply coloured (see fig. 683). Gustav Rose proposed to call the more basic granites, with a large proportion of plagioclase, by the name of granitite. Eosenbusch applies the same name to rocks in which biotite-mica is present in considerable quantities. Hornblende- or Amphibole-granite (or granitite) is also usually a somewhat basic granite. Pyroxene-granites contain a colourless or pale green augite, or a pale-coloured and, rarely, a more ferriferous enstatite (hypersthene) ; of the latter class is the interesting Charnockite or hypersthene -granite of India, de- scribed by Mr. Holland. Special accessory minerals, such as sphene, tourmaline, garnet, cordierite, pyrite, sillimanite, andalusite, r LiiidliViViVii Granite veins traversing clay slate. Table Mountain, Cape of Good Hope (Capt. Basil Hall). Granite veins traversing gneiss. Cape Wratb. (Macculloch.) granite ; and another, still newer, traverses both the second and the first. In Shetland, according to Macculloch, there are two kinds of granite. One of them, composed of hornblende, mica, felspar, and quartz, is of a dark colour, and is seen underlying gneiss. The other is a red granite, which penetrates the dark variety everywhere in veins. Fig. 697 is a sketch of a group of granite veins in Cornwall, given by Von Oeynhausen and Von Dechen. The main body of the granite is of a porphyritic structure, with large crystals of felspar ; but in the veins it is fine-grained, and without these large crystals. The general width of the veins is from 16 to 20 feet, but some are much wider. The granites, syenites, diorites, felsites, and indeed all plutonic rocks, are frequently observed to contain metallic veins at or near their junction with stratified formations. On the other hand, similar veins which traverse stratified rocks are, as a general law. 526 VEINS AND DYKES [CH. XXXVI. more metalliferous near such junctions than in other positions. Hence it has been inferred that these metals may have been diffused through the molten mass, and that the contact of another rock at a different temperature, or sometimes the existence of rents in other rocks in the vicinity, may have caused the transfer of the metallic compounds to their present situation. Fig. 697. Granite veins passing through hornblende slate. Carnsilver Cove, Cornwall. Veins of pure quartz are often found in granite, as in many stratified rocks, but they are not traceable, like veins of granite or lava, to large bodies of rock of similar composition. They appear to have been cracks, into which siliceous matter was infiltrated. Such segregation, as it is called, can sometimes clearly be shown to have taken place long subsequently to the original consolida- tion of the containing rock. Thus, for example, in the gneiss of Gneiss. Fig. 698. Greenstone dyke. Gneiss. a, b. Quartz vein passing through gneiss and greenstone. Tronstadt Strand, near Christiania. Tronstadt Strand, near Drammen, in Norway, the annexed section is seen on the beach. It appears that the alternating strata of whitish granitiform gneiss and black hornblende schist were first cut through by a greenstone dyke, about 2^ feet wide ; then the crack a, b, passed through all these rocks, and was filled up with CH. xxxvi.] OF PLUTONIC ROCKS 527 quartz. The opposite walls of the veins are in some parts in- crusted with transparent crystals of quartz, the middle of the vein being filled up with common opaque white quartz. When masses of granite approach, or are visible at, the sur- face of the earth, their relations to the strata and rocks on all sides, and above, are often very difficult to understand. The sur- rounding rocks are often greatly altered in their stratification and mineral nature. In many localities there are great extensions of granite far below the surface, which have only become known by the coming up of veins to the surface and the alterations which have oc- curred in the rocks which have not yet been denuded off. Results of Segregative Action in Plutonic Rock-masses. The tendency of the more basic minerals in rocks to crystallise before those of acid composition, and of the still fluid materials to separate from the crop of earlier-formed crystals, may give rise to a want of homogeneous character to igneous rock-masses. But in addition to this action, there can be little doubt that, in many cases, the masses of fused silicates containing water and gases tend to break up into magmas of different composition, density, and fusibility. In addition to the composite dykes formed by the injections of fissures in an older dyke with later materials of different chemical composition, there is another class of com- posite dykes (which has been specially studied by Professors Vogt and Lawson), in which segregative action has clearly operated upon the liquid materials after they have filled the dyke, and caused the rock occupying its centre to have a different chemical composition and mineralogical constitution from that forming its sides. The same kind of action, as has been shown by Mr. Harker, takes place in plutonic intrusions of much greater dimensions than dykes, and has been described by that author as occurring at Carrock Fell. Most granitic and other plutonic rocks are also found to contain inclusions or irregular patches of different che- mical composition from the general mass of the rock. These in- clusions, as shown by the late John Arthur Phillips, belong to two distinct classes. We sometimes find fragments of schist and other rocks which have clearly been caught up in the liquid mass during its intrusion, and we can detect every gradation from frag- ments in which the sedimentary origin is obvious to others which have suffered such complete fusion and recrystallisation as to betray no signs of their origin. On the other hand there are inclusions which have undoubtedly been formed by segregative action going on in the consolidating magma ; such ' segregative inclusions ' usually consist of the same minerals as form the mass of the rock, but in different proportions ; sphene and the more basic minerals, bio- tite and hornblende, are especially abundant in these segregative masses, which are sometimes found only half enveloping a large ' phenocryst ' of the rock, while they occasionally exhibit an orbi- cular structure. The older geologists also noticed the profusion of veins, often breaking up into the most minute ramifications, which traverse many plutonic masses. In some cases, like the veins of almost pure quartz, there can be little doubt that their existence must be due to the fissuring of the rock-mass and the infilling of the fissures 528 AGE OF PLUTONIC ROCKS [CH. xxxvi. with materials ' leached out ' from the general mass, probably before its complete consolidation. In many cases these veins betray very close analogies in chemical and mineralogical characters with the segregation inclusions of the same rock. Hence the old geologists spoke of these veins as ' contemporaneous ' or ' segrega- tion veins.' It must be remembered, however, as pointed out by Professor Sollas, that such veins often differ in no essential character from true intrusive veins, and that many of the so-called ' contemporaneous segregation veins ' may really be of ' subsequent intrusive origin.' The marked tendency of the volcanoes of a particular ' petro- graphical province ' to exhibit a distinct order in the materials ejected at successive periods of eruption also points to a segrega- tive action going on in the plutonic magmas which supplied the volcanoes (see p. 488). Physicists have suggested several distinct causes as leading to this differentiation in the masses of mixed sili- cates which constitute the igneous magmas (Note X, p. CHAPTER XXXVII PLUTONIC ROCKS BELONGING TO DIFFERENT GEOLOGICAL PERIODS Plutonic Hocks were formed during the whole of the geological periods Those of the most recent period seldom exposed at the surface by de- nudation Test of the geological age of Plutonic rock-masses Relative position Intrusion and Alteration Mineral composition Included fragments Tertiary Plutonic Rocks of Western Scotland North-East Ireland Elba, &c. Difficulty of determining the age of Plutonic Rock-masses in Mountain chains Plutonic Rocks of the Cretaceous the Jurassic the Carboniferous the Ordovician and Pre-Cambrian Periods. On the different ages of the Plutonic Rocks. It has been stated that the plutonic rocks were formed under greater pressure than the volcanic, and that the pressure appears to have been produced by the weight of superincumbent rocks, and by com- pression and crushing accompanyingrock-folding and fracture. It may be that granite and similar materials underlie the deepest known strata, and that, under special conditions, they have been forced upwards and have cooled and assumed the crystalline form. Although the volcanic rocks resemble the plutonic in their general mineralogical constitution, yet it must be remembered that the rhyolites, andesites, and basalts occasionally contain minerals or associations of minerals, differing slightly from those found in granite, diorite, gabbro, and other typical plutonic rocks. If granites and similar rocks can only be formed as the result of slow cooling and the pressure of many thousands of C'H. xxxvii.] TESTS OF AGE 529 feet of superincumbent material, it follows, as we have already pointed out, that only where a sufficient time has elapsed since their consolidation for the removal of these thick overlying masses by denudation, can we expect to see such highly crystalline masses exposed at the surface. Such being the case, we shall now proceed to show that inasmuch as we can never expect very important aid from fossils in determining the age of a plutonic rock, there is even greater uncertainty in arriving at just conclusions concerning the periods at which rocks of this class were formed, than in the case of rocks of volcanic origin. Test of age by relative position. Unaltered fossiliferous strata of every age are met with reposing immediately on plu- tonic rocks; as at Christiania in Norway, where the Pleis- tocene deposits, and at Heidelberg on the Neckar, and Mount Sorrel in Leicestershire, where the New Red Sandstone forma- tions rest on granite. In these, and similar instances, inferi- ority in position is connected with the superior antiquity of granite. The crystalline rock was solid before the sedimentary beds were superimposed, and the latter usually contain rounded pebbles of the subjacent granite, but the latter never gives off veins into the rocks above. Test by Intrusion and alteration. But when plutonic rocks send off veins into the sedimentary strata, and have altered them near the planes of contact, it is clear that, like intrusive volcanic rocks, they are newer than the strata which they have invaded and altered. Examples of the application of this test will be given in the sequel. Test by mineral composition. Sometimes a peculiar mineral condition distinguishes a plutonic rock, and is found prevailing throughout an extensive region ; so that, having ascertained the relative age of the rock in one place, we can recognise its identity in others, and thus determine from a single section the chronological relations of large mountain masses. Having observed, for example, that the syenite of Norway, in which zircon and other peculiar minerals abound, has altered the Silurian strata wherever it is in contact, we do not hesitate to refer other masses of the same zircon-syenite in the south of Norway to a post- Silurian date. But too much reliance should not be placed on mineral character as a test of age ; again and again have conclusions concerning the age of rocks, based on mineral characters only, proved to be untrust- worthy. Test by included fragments. This criterion can only be of value in particular cases, because the fragments included in granite are often so much altered, that they cannot be referred M M 530 PLUTONIC ROCKS [OH. xxxvn. with certainty to the rocks whence they were derived. In the White Mountains, in North America, according to Professor Hubbard, a granite vein, traversing granite, contains fragments of slate and other rocks which must have fallen into the fissure when the fused materials of the vein were injected from below, and thus the granite is shown to be newer than those slaty and other formations from which the fragments were derived. Tertiary Plutonic Rocks. At many different points in the Hebrides, as in Skye, Mull, Bum, St. Kilda, &c., great masses of granite and gabbro occur in close association with the Tertiary volcanic rocks already described. Dr. Macculloch showed that the granites of Skye intersect limestone and shale which are of the age of the Lias. Macculloch also pointed out that the granite and gabbro of the Inner Hebrides are newer than the secondary strata of these islands, and Edward Forbes afterwards showed that in Mull there are strong grounds for believing the volcanic rocks so inti- mately associated with the granites and gabbros to be of Tertiary age. Professor Zirkel has demonstrated that the great moun- tain masses of intrusive rocks, both in Mull and Skye, consist of granite and gabbro which differ in no essential respect from the granites and gabbros belonging to the older geological periods ; in Skye, these gabbros are seen in the remarkable Cuilin Hills, which are so famed for their wild and majestic scenery. And lastly, it has been shown that the great moun- tain groups in the Hebrides, composed of granites and gabbros, constitute the relics of five grand volcanoes which were in eruption during a great part of the Tertiary period, the earlier formed masses of granite being intruded into a series of andesitic and other lavas probably of Eocene age ; while the gabbros, which break through the granites, are the consolidated reservoirs and ducts that gave rise to the great streams of basaltic lava of somewhat later age, constituting the plateaux forming so large a portion of the Hebridean Archipelago. These researches, show that the Western Isles of Scotland afford a most admirable and instructive series of illustrations not only of the intimate connection between the rocks of the volcanic and the plutoiiic classes respectively but at the same time of the perfect identity, in their nature and sequence, of the phenomena of volcanic activity during former periods of the earth's history and those which are exhibited to us at the present day. There are the strongest grounds for believing that the granites of Arran and those of the Mourne Mountains in Ireland are of the same age as the granites of Skye, Mull, Rum, &c. It has been shown by Lotti that the granites and diabases CH. xxxvii.j OF CAINOZOIC AGKE 531 (or gabbros) of the Island of Elba are like those of our own Hebrides, of older Tertiary age. In a former part of this volume (p. 229) the great Nummulitic formation of the Alps and Pyrenees was referred to the Eocene period, and it follows that vast movements which have raised those fossiliferous rocks from the level of the sea to the height of more than 10,000 feet above its level have taken place since the commencement of the Tertiary epoch. Here, therefore, if anywhere, we might expect to find hypogene formations of Eocene date breaking out in the central axis or most disturbed region of the loftiest chain in Europe. It was believed by the older investigators, and is still credited by some geologists, that in the Swiss Alps even the flysch, or upper portion of the Num- rnulitic series, has been occasionally invaded by plutonic rocks, and converted into crystalline schists of the hypogene class. It is stated that even the granite or gneiss of Mont Blanc itself has been in a fused or pasty state since the fly sell was deposited at the bottom of the sea ; and the question as to its age is not so much whether it be a secondary or tertiary granite or gneiss as whether it should be assigned to the Eocene or Miocene epoch. But the student must always be on his guard against receiving statements regarding the age of granites in disturbed areas, such as those of mountain-chains. For inversions of strata in such situations are exceedingly common, and on the grandest scale. Plutonic Rocks of the Cretaceous Period. It will be shown in a following chapter that the Chalk and the Lias have been altered by granite in the eastern Pyrenees. Whether such granite be Cretaceous or Tertiary cannot easily be decided. Sup- pose 6, c, d, fig. 699, to be three members of the Cretaceous series, the lowest of which, b, has been altered by the granite A, the modifying influence not having extended so far as c, or having but slightly affected its lowest beds. Now it can rarely be possible for the geologist to decide whether the beds d existed at the time of the intrusion of A, and alteration of b and c. or whether they were subse- quently thrown down upon c. But as some Cretaceous and even Tertiary rocks have been raised to the height of more than 9,000 feet in the Pyrenees, we must not assume that plutonic formations of the same periods may not have been brought up MM2 532 PLUTONIC ROCKS OF [CH. XXXVII. and exposed by denudation, at the height of 2,000 or 3,000 feet, on the flanks of that chain. Plutonic Rocks of the Jurassic Period. In the Depart- ment of the Hautes-Alpes, in France, M. Elie de Beaumont traced a black argillaceous limestone charged with Belemnites to within a few yards of a mass of granite. Here the limestone begins to put on a granular texture, but is extremely fine- grained. When nearer the junction it becomes grey, and has a saccharoid structure. In another locality, near Champoleon, a granite composed of quartz, black mica, and rose-coloured felspar, is observed partly to overlie the secondary rocks, pro- ducing an alteration which extends for about 30 feet downwards Fig. 700. Junction of granite with Jurassic or Oolite strata in the Alps, near Champoleon.^ diminishing in the beds which lie farthest from the granite (see fig. 700). In the altered mass the argillaceous beds are hardened, the limestone is saccharoid, the grits quartzose, and in the midst of them is a thin layer of an imperfect granite. It is also an important circumstance that near the point of contact both the granite and the secondary rocks become metalliferous, and contain nests and small veins of blende, galena, and iron- and copper-pyrites. The stratified rocks become harder and more crystalline, but the granite, on the contrary, softer and less perfectly crystallised near the junction. Although the granite is incumbent in the above section (fig. 700), we cannot assume that it overflowed the strata, for the disturbances of the rocks are so great in this part of the Alps that their original position is often inverted. The age, therefore, of the granite is doubtful. CH. xxxvn.] MESOZOIC AND PALAEOZOIC AGES 533 Plutonic Rocks of the Triassic Period. The great intrusive masses consisting of ' monzonite ' (augite- syenite), tourmaline -granite, hypersthene-dolerite, and other rocks, so well exhibited at Predazzo in the Tyrol, are now known to be of Upper Triassic age. The general relations of these rock- masses are represented in fig. 701. Both the acid and basic Pig. 701. iPredagzo OL a. Botzen porphyry, of Permian age. b, c, d. Stratified rocks of the Lower, Middle, and Upper Trias, e. Monzoni syenite, traversed by veins of hyper- sthene-dolerite, &c. /. Tourmaline-granite, g. Hypersthene-dolerite, and other basic rocks. rocks show that general dip towards the centre of the mass which is so commonly seen beneath volcanoes when the under- lying rock-masses are exposed by denudation. In the lime- stones in contact with the great intrusive rock-masses, beautifully crystallised minerals are found of precisely the same species as those ejected from Vesuvius and other recent volcanic vents. Plutonic Rocks of the Carboniferous Period. The granite of Dartmoor, in Devonshire, was formerly supposed to be one of the most ancient of the plutonic rocks, but is now ascertained to be posterior in date to the Culm-measures of that county, which from their position, and as containing true coal-plants and Trilobites of the Phillipsia group, are now known to be members of the Carboniferous series. This granite has broken through the Devonian and Carboniferous stratified formations, the suc- cessive members of the Culm-measures abutting against the granite, and becoming metamorphosed as they approach it. These strata are also penetrated by granite veins, and dykes, called ' elvans.' The granite of Cornwall is probably of the same date, and therefore as modern as the Carboniferous strata, if not newer. Plutonic Rocks of the Ordovician Period. It has long been thought that a very ancient granite near Christiania, in Norway, is posterior in date to the Ordovician strata of that region, although its exact position in the Palaeozoic series cannot be defined. Von Buch first announced, in 1813, that it was pf 584 PALEOZOIC AND [CH. xxxvi r. newer origin than certain limestones containing Orthocerata and Trilobites. The proofs consist in the penetration of granite veins into the shale and limestone, and in the alteration of the strata, for considerable distances from their planes of contact with these veins and with the central mass from which they emanate. (See fig. 702.) "When the junctions of the strata and the granite are carefully examined, it is found that the plutonic rock intrudes Fig. 702. Silurian. Granite. Silurian strati itself in veins and nowhere covers the fossiliferous strata in large overlying masses, as is so commonly the case with volcanic formations. Now this granite, which is more modern than the Ordovician strata of Norway, also sends veins into an ancient formation of gneiss of the same country ; and the relations of the plutonic rock and the gneiss, at their junction, are full of interest when we duly consider the wide difference of epoch which must have separated their origin. The length of this interval of time is attested by the following facts : The fossiliferous, or Silurian, beds rest unconformably upon the truncated edges of the gneiss, the inclined masses of which had been denuded before the sedimentary beds were superimposed (see fig. 703). The signs of denudation are two- Pig. 703. Gneiss. Granite -.Gneiss. Granite sending veins into Silurian strata and gneiss. Christiania, Norway. a. Inclined gneiss. b. Silurian strata. fold : first, the surface of the gneiss is seen occasionally (on the removal of the newer beds containing organic remains) to be rounded and water-worn ; secondly, pebbles of gneiss have been found in some of these Silurian strata. Between the origin, therefore, of the gneiss and the granite there intervened, first, the period when the masses of gneiss were denuded ; secondly, the period of the deposition of the Silurian strata on the de- nuded and inclined gneiss, a. The granite produced after this CH. xxxvn.] PKE-PAL^OZOIC PLUTONIC EOCKS 535 long interval is often so intimately blended with the gneiss at the point of junction, that all distinction is arbitrary. The whole of these rocks have been since studied with great thorough- ness by Professor Brogger, who has confirmed the conclusions of his predecessors concerning their general relations. Pre -Cambrian Plutonic Rocks. Granite appears to have been intruded into the metamorphic rocks which are the lowest in the South Wales area the Dimetian of Dr. Hicks ; and it is possible that the veins of it did not pass beyond this lowest horizon. The Lewisian or Fundamental gneiss of Scotland con- tains many plutonic rocks which are certainly older, not only than the Cambrian strata, but than the Torridon Sandstone which underlies them. The investigations of the geological surveyors in Scotland lead to the conclusion, indeed, that in the northern portion of the Western Highlands, the Fundamental gneiss series consists almost wholly of plutonic rock -masses, more or less altered by the shearing movements to which they have been subjected. In addition to the hornblendic gneiss, which is the predominant rock, we find Amphibolites and Pyroxenites (Augite rocks and Hypersthene-augite rocks), Pyroxene -gneiss and granulite, and many garnet-bearing rocks. The whole of these ancient plutonic rock-masses are traversed by numerous dykes of every age, up to the Tertiary. The Laurentian rocks of Canada have numerous veins and dykes of diabase, sometimes of great width, and they are cut across by extensive masses of syenite, with veins of reddish- brown porphyritic felsite. These intrusive rocks appear not to enter the superimposed Silurians. But it is very evident that many of the eruptive rocks found in the pre-Cambrian forma- tions are of later age, and were erupted during the Devonian or Carboniferous age. The intrusion of plutonic rocks into the gneisses and mica schists of Archaean and subsequent ages is exceedingly in- teresting, especially when fragments of the schistose rocks are found included in the plutonic masses. Very frequently there is great difficulty in determining whether a rock is a true gneiss or a granite, showing parallel arrangement of its crystals, produced by pressure during consolidation. General McMahon has shown that some of the granites of the Himalayas, which give off numerous veins into the surrounding rocks, nevertheless exhibit a marked foliated or gneissic structure. On the following page we have given in tabular form a series of analyses of volcanic and plutonic rocks, which will illustrate the intimate relations which exist between the two great classes of igneous products 536 CHIEF TYPES OF IGNEOUS ROCKS [CH. xxxvn. ANALYSES OF CHIEF TYPES OF IGNEOUS ROCKS (VOLCANIC AND PLUTONIC) 4 fc s | H a ^ 02 I p g2 K " ia i gs CQ H KO < PH 5 fc a F APLITE (Wexford), sp. gr. 2'63 . 80-2 12-2 0-7 0-9 5'6 0-4 GRANITE (with two Micas) 76'1 18-4 1-8 0-2 0-3 8-1 4'9 1-1 (Saxony), sp. gr. 2'G6 Q GRANITITE (Odenwald), sp. gr. 2'68 09-0 14-8 2-3 0-9 1-1 8-8 2'5 4-5 i 5 HORNBLENDE-GRANITITE (Oden- 65'8 18-0 4'2 0-7 2'1 5-1 1-8 2-2 1-2 < wald), sp. gr. 2'74 .... RHYOLITE (Hungary), sp. gr. 2'40 . RHYOLITE (Hungary), sp. gr. 2'46 . 76-3 69-0 13-2 17-1 1-9 = 0-2 1-9 0-7 2'8 2'3 8'7 9'7 0-6 0-9 ' BANATITE ' (Quartz Diorite) (Hun- 65-7 17-1 2-8 1-8 2'6 5'2 8-9 1-0 gary), sp. gr. 2'72 .... I SYENITE (Saxony), sp. gr. 2'73 . DIORITE (Hartz), sp. gr. 2'90 . 59'8 54-7 16-9 15-7 2'0 7-0 6'3 2-6 5'9 4-4 7-8 2-4 2-9 6-fl 8-8 1-3 1 G EL^EOLITE - SYENITE (Transylva- 56-8 24-1 2-0 O'l 0-7 9'8 C-8 1'6 nia), sp. gr. 2'48 .... H DACITE (Quartz-Andesite) (Hun- 67-2 17-0 8'5 1-2 1-5 4'5 3'7 1-6 0-9 a! gary), sp. gr. 2'50 .... H TRACHYTE (Bolsena), sp. gr. 2'55 . ANDESITE (Buffalo Peak, U.S.), sp. 59'2 56-2 18-6 5-0 6'] 4'4 1-1 4-6 3-0 7-0 4-9 8-0 6-7 2'4 1-1 1-3 B gr. 2-74 M PHONOLITE (Wolf Rock), sp. gr. 2'54 56-5 22-3 2'7 ro 1-5 11-1 2'8 2-1 GABBRO (Hartz), sp. gr. 3'02 . 49'6 16'2 1-9 12-8 5'4 9'8 1-9 0-8 3-3 DOLERITE [Diabase] (Sweden), sp. 50-2 15-0 16-9 5'8 10-5 2-2 1'4 0'7 gr. 2'98 BASALT (Etna, 1865), sp. gr. 2'77 49-7 18-2 ^_ 12'5 4-0 11-4 3'4 0-7 0-2 o LEUCITB BASALT (Bohemia), sp. 40-3 11-6 21-8 7-6 11-1 1-4 4-2 0-3 55 gr. 2'94 ! pq NEPHELINE BASALT (Germany), 40-5 14-9 i-o 11-2 8-0 14-6 2-9 2' 4'6 sp. gr. 3'04 MELILITE BASALT (Germany), sp. 33'9 9'9 15'6 16-1 15-2 2'9 6-4 gr. 3-04 LlMBURGlTE (Germany), sp. gr. 2'83 42-8 8-7 18'9 10-1 12-3 2-3 0-6 4-8 -jg LHERZOLITE (Pyrenees), sp. gr. 3-23 HORNBLENDE-PlCRITE(Odenwald), 45-0 41'4 1-0 6'6 13-9 12-0 6'8 16-0 18'4 19-5 7-2 0-2 0-9 C'5 5-6 s| sp. gr. 2'82 DUNITE (Olivine rock) (New Zea- 42-8 9'4 47-4 0-6 p land), sp. gr. 3'3 METEORITE (Chasslgny, 1815) . 85-3 - - 26-7 31-8 - - 0-7 4-9 The student will find the plu- tonic rocks fully described in the treatises of Rosenbusch and Zirkel, and in the English text-books of Rutley, Hatch, and Harker, already referred to. Illustrations and de- scriptions of the most important types of plutonic rocks in this country are published in Teall's 'British Petrography.' Valuable series of rock-analyses will be found in the works of Justus Both. 537 PART V METAMORPHIC ROCKS CHAPTER XXXVIII METAMORPHIC ROCKS, THEIR NATURE AND ORIGIN Contact Metamorphism and Eegional Metamorphism Thermo-metamor- pliism and Hydrothermal action Dynamo-metamorphism Different results of Metamorphic action Researches of Daubree and others on Thermo-metamorphic and Hydrothermal action Dynamo-metamorphic action and its results Slaty cleavage Its nature and origin Investi- gations of Phillips, Sharpe, Sorby, &c. Experimental proofs of origin of slaty cleavage Foliation, its nature and origin Relations between Cleavage and Foliation Experimental researches of Daubree, Spring, and others upon the action of pressure in producing Metamorphism. Nature of IVIetamorphic Rocks. We have now considered all the classes of rocks, except the last group, which comprises those called Metamorphic, and which result from great alteration taking place in other rocks. The term Metamorphic implies that rocks have undergone changes of chemical, mineralogical, and textural kinds, and that their internal structure and outward appearance no longer resemhle those of the original rock. Such changes and alterations as are sufficient to produce a kind of metamorphism may be studied at the present day in volcanic regions, such as Iceland, or near Naples. The flowing of lava over soil, or into streams or small lakes, produces alterations in the clays and sands, which are baked by the heat and are sometimes infiltrated with siliceous solutions altering them chemically and mechanically. Similar changes occurred under analogous circumstances in past geological ages. Thus, in examining the sides of dykes and other plutonic masses, as has been already pointed out, very striking evidence is often detected of the action of heated lavas upon the clays, sandstones, or limestones with which the igneous masses have been in contact. These may be taken as examples of local, or contact, metamorphism ; but on examining the rocks in the midst of 538 NATURE AND VARIETIES [CH. xxxvin. great mountain chains slates, schists, quartzites, crystalline limestones, gneisses, &c., they are found in positions where originally horizontal rocks have been subjected to the weight of superincumbent rock-masses, to intense lateral pressure, to heat, and to the action of percolating gases, and of water holding various materials in solution. Such rocks, which are said to have undergone 'regional metamorphism,' are found over great tracts of country. The mountains of Cornwall, North Wales, and the Lake district, illustrate the phenomena of metamorphism, but examples of still more highly altered rocks are found in the Alps, the Scandinavian peninsula, and the North-Western Highlands of Scotland, where the results of the extreme action of this ' regional ' metamorphism are fully exemplified. There are thus two classes of metamorphic rocks, recognised by geologists ; those which have been locally affected by the contact of plutonic and volcanic rock-masses, and those which have been exposed to more general action the agencies of heat and pressure operating over wide areas, and probably at great depths from the surface. We speak of the metamorphic action in the first class of rocks as ' contact ' or ' local metamorphism ' and in the second as 'general ' or ' regional metamorphism.' From a study of the ultimate chemical composition of the different varieties of metamorphic rocks (see table, p. 588), it is obvious that metamorphic action has not been restricted to any one class of rocks ; but that sedimentary strata, volcanic lavas and tuffs, and the materials of plutonic intrusions must alike have undergone great changes, and are now exhibited to us under very different aspects from those which they originally presented. There is probably no class of aqueous or igneous materials which is not represented among the metamorphic rocks bv masses of material which differing little if at all from them in ultimate chemical composition have nevertheless had the whole of their constituents recombined and recrystallised. Different kinds of Me tarn or phi c action. The two great agencies concerned in the production of metamorphism are heat and pressure. The effects produced by heat alone we speak of as Thermo -metamorphism, or, recognising the great influence exerted by the presence of water and gases in these heated masses, we often refer to it as hydrothermal action. The results produced on rocks by pressure we call Dynamo- metamorphism. Though it may be convenient to speak of these two kinds of metamorphic action as distinct from each other in their nature and their effects, it must be remembered that in most cases thermo-metamorphism and dynamo- CH. xxxviii.] OF METAMORPHIC ACTION 539 metamorphism co-operate in producing the characters found in metamorphic rocks. In local or contact metamorphism, though the chief agent of change appears to have been the heat emanating from the plutonic intrusion, yet pressure must have operated in increasing the chemical action of the water and gases imprisoned in the rock undergoing alteration, or passing into it from the igneous mass with which it was in contact. In regional metamorphism, dynamo-metamorphic action usually appears to have played a much more important part than in contact metamorphism. The researches which had been made in the distribution of underground temperature (see p. 13) rendered it highly probable, if not absolutely certain, that at a depth of 10,000 feet, or two miles from the surface of the earth, the rocks of the earth's crust must have a tempera- ture of at least 212 F. (See, however, Note B, p. 601.) During the great movements to which the strata of regions now occupied by mountain chains have been subjected, sub- sidences of 10,000 feet and of even five times that amount have been common occurrences ; and similar downward movements, as we have already shown, must have accompanied the deposition of many thick masses of sedimentary rocks, such, for example, as those of the Carboniferous system. Hence it is certain that many of these rocks have been subjected to temperature varying from that of boiling water to that of red- hot iron. A very simple calculation serves to show that rocks, when buried at the depth of 10,000 feet from the surface, are subjected to a pressure of about 37 tons to the square inch, and that there is a progressive increase of pressure in descending to still greater depths. This pressure, produced by the weight of super- incumbent rock-masses, we may speak of as statical pressure ; its effects are seen in the liquefied gases which, as we have pointed out, are found imprisoned in the cavities of deep-seated plutonic rocks, and in the water and gases occluded in volcanic rocks, which are given off into the atmosphere when the lava issues from a vent and the pressure is relieved. The effects of these statical pressures are testified to by the condition of the minerals of all rocks which, at any period of their history, have been deep-seated. The chemical changes, which these rock-forming minerals have undergone, show that they must have been com- pletely permeated by liquids and gases which, under the enor- mous pressures, were forced between the molecules of the solid crystals. Of far greater intensity and effect, however, are the pressures 540 THERMO-METAMORPHISM [CH. xxxvin. produced when great rock -masses are bent, folded, crushed, and broken across, during earth-movements such as those which are concerned in making mountain-chains. Under these conditions we find that pebbles of the hardest rocks are sometimes thrust against one another with such irresistible force as to mutually bruise and indent one another (impressed pebbles') ; rock sur- faces are ground against one another so as to produce polished and striated faces (slickensides) ; and solid materials broken into angular fragments (fault-rode) or reduced to the finest powder (mylortites). The remarkable and chemical effects produced by this dynamical action we shall presently consider in greater detail. Different ways in which Rocks have been affected by IVEetamorphic action. It may be asked, then, whether all rocks which have been buried under the same thicknesses of superincumbent strata exhibit like effects of metamorphic action. A little reflection will show that there are examples of strata like the limestones, grits, and coal-measures of the Carboniferous system which must have been long buried under many thousands of feet of superincumbent rock, but in which, nevertheless, the changes produced have been remarkably small, and of others which, under like conditions, have undergone the most intense alteration accompanied with complete recrystal- lisation of their materials. Under these circumstances, therefore, it may be desirable to inquire a little more particularly how the several agencies heat and pressure really operate in modifying the characters of rock-masses. It is in rocks which have been subjected to contact- metamorphism that we can best study the direct action of heat in producing chemical change and recrystallisation of their materials. Bocks that have been subjected to regional meta- morphism, on the other hand, best exemplify the effects of pressure, acting either alone or in combination with thermal or hydrothermal agencies. Thermo metamorphisrn, or Hydrothermal action. As all rocks contain water, it must have influenced their metamorphism j under heat and pressure, and its agency would be enhanced by the presence of various substances held in solution. In local metamor- j phism, water is introduced in excess from the intruded or overflowing volcanic rock, and also various chemical compounds in solution, with gases which act upon the surrounding strata. In regional metamorphism the excess of water does not appear to have been necessary, the original amount already contained in the rocks pro- bably being sufficient. But hydrothermal action that is, the influence of hea,ted wa.ter containing dissolved solid matter, and CH. xxxvni.] EXPERIMENTAL ILLUSTBATIONS 541 also gases, like hydrochloric acid and carbon dioxide, in solution is recognised as a potent factor in metamorphism. Thus it is known that long after volcanoes have spent their force, hot springs continue to flow out at various points in the same area. In regions also subject to violent earthquakes such springs are frequently observed issuing from rents, usually along lines of fault or displacement of the rocks. These thermal waters are most commonly charged with a variety of dissolved ingredients, and they retain a remarkable uniformity of temperature from century to century. A like uniformity is also found in the nature of the solid and gaseous substances with which they are impregnated. It is well ascertained that springs, whether hot or cold, charged with carbon dioxide, and with sulphuric, hydrochloric, boric, or hydrofluoric acids, which are often present in small quantities, are powerful causes of decomposition and chemical change in rocks through which they percolate. The alterations which Daubree has shown to have been produced by the alkaline waters of Plombieres in the Vosges, are especially instructive. These waters have a temperature of 160 F., or an excess of 109 above the average temperature of ordinary springs in that district. They were conveyed by the Komans to baths through long conduits or aqueducts. The foundations of some of their works consisted of a bed of concrete made of lime, fragments of brick, and sandstone. Through this and other masonry the hot waters have been percolating for centuries, and have given rise to various ^ zeolites Apophyllite and Chabazite among others also to Calcite, Aragonite, and Fluorspar, together with siliceous minerals, such as Opal all found in the interspaces of the bricks and mortar or constituting part of their rearranged materials. The amount of heat brought into action in this instance in the course of 2,000 years has, no doubt, been enormous, but its intensity, or the temperature developed at any one moment, has always been inconsiderable. From these facts and from the experiments and observations of Senarmont, Daubree, Delesse, Scheerer, Sorby, Sterry Hunt, and others, we are led to infer that when there are large volumes of molten matter in the earth's crust, containing water and various acids, even in excessively minute quantities, heated under pressure, these subterranean fluid masses will gradually part with their heat by the escape of steam and various gases through fissures, producing hot springs ; or by the passage of the same through the substance of the overlying and injected rocks. Even the most compact rocks may be regarded, before they have been exposed to the air and dried, in the light of sponges filled with water. According to the experi- ments of Henry, water, under a hydrostatic pressure of 96 feet, will absorb three times as much carbon dioxide as it can under the ordinary pressure of the atmosphere. There are other gases, as well as the carbon dioxide, which water absorbs, and more rapidly in proportion to the amount of pressure. The water acts also by its affinity for various silicates, which are hydrated or decomposed. Quartz can be produced under the influence of heat by water hold- ing alkaline silicates in solution, as in the case of the Plombieres springs. The quantity of water required, according to Daubree, to produce great transformations in the mineral structure of rocks is very small. As to the heat required, silicates may be produced in the 542 DYNAMO-METAMORPHISM [CH. xxxviu. moist way at about incipient red heat, whereas to form the same in the dry way requires much higher temperatures. M. Fournet, in his description of the metalliferous gneiss near Clermont, in Auvergne, states that all the minute fissures of the rock are quite saturated with free carbon dioxide ; which gas rises plentifully from the soil there and in many parts of the surrounding country. The various minerals of the gneiss, with the exception of the quartz, are all softened ; and new combinations of the acid with calcium, iron, and manganese are continually in progress. The effect of subterranean gases on rocks is well illustrated in the neighbourhood of St. Calogero, in the Lipari Islands, where the hori- zontal strata of tuff forming cliffs 200 feet high have been discoloured in places by the jets of steam, often above the boiling point, called ' stufas,' issuing from the fissures ; and similar instances are recorded by Virlet of the corrosion of rocks near Corinth, and by Daubeny of the decomposition of trachytic rocks by sulphuretted hydrogen and hydrochloric-acid gases in the Solfatara, near Naples. In all these instances it is clear that the gases must have made their way through vast thicknesses of porous or fissured rocks, and their modifying influence may spread through the crust for thousands of yards in thickness. It has been urged as an argument against the metamorphic theory, that rocks have a small power of conducting heat, and it is true that when dry they differ remarkably from metals in this respect. The syenite of Norway has sometimes altered fossiliferous strata both in the direction of their dip and strike for a distance of a quarter of a mile. But in regional metamorphism the production of gneiss and mica and other schists was a slower process than local metamorphism, and the duration of the process compensated for the diminished increments of heat, pressure, and hydrothermal action. Bischoff has shown what changes may be superinduced, on black marble and other rocks, by the steam of a hot spring ; and we are becoming more and more acquainted with the prominent part which water is playing in distributing the heat of the interior through mountain-masses of incumbent strata, and of introducing various chemical compounds into them, in a fluid or gaseous state. Dynamo-metamorphic action. While statical pressures seem to have led to the induration, or to the recrystallisationof the material of rocks, with occasional slight modifications in chemical composition, dynamical action has resulted in a rearrangement of their materials, so that the rocks often split up along planes quite distinct from those of the original bedding of the mass. When the change produced is of a mechanical character only, resulting in a rearrangement of the particles of the rock, it is called cleavage ; but when this rearrange- ment is accompanied by chemical changes and recrystallisation of the rock-materials, the result is called foliation. The development of planes of cleavage and foliation at right angles to the directions in which pressure has been exerted is a question to which the attention of geologists and physicists has long been devoted. Slaty cleavage. Sedgwick, whose essay ' On the Structure of Large Mineral Masses ' first cleared the way towards a better under- standing of this difficult subject, called attention to the fact that joints are distinguishable from planes of slaty cleavage in this, that the rock intervening between two joints has no tendency to cleave in a direc- CH. XXXVIII.] SLATY CLEAVAGE 543 tion parallel to the planes of the joints, whereas a rock is capable of indefinite subdivision in the direction of its slaty cleavage. In cases where the strata are curved, the planes of cleavage are still perfectly parallel. This has been observed in the slate rocks of part of Wales (see fig. 704), which consist of a hard greenish slate. The true bed- ding is there indicated by a number of parallel stripes, some of a Fig. 704. Parallel planes of cleavage intersecting curved strata. (Sedgwick.) lighter and some of a darker colour than the general mass. Some stripes are found to be parallel to the true planes of stratification, wherever these are manifested by ripple marks, or by beds containing peculiar organic remains. Some of the contorted strata are of a coarse mechanical structure, alternating with fine-grained crystalline chloritic slates, in which case the same slaty cleavage extends through the coarser and finer beds, though it is brought out in greater perfec- tion in proportion as the materials of the rock are fine and homo- geneous. It is only when these are very coarse that the cleavage planes entirely vanish. In the Welsh hills these planes are usually inclined at a very considerable angle to the planes of the strata, the average angle being as much as from 30 to 40. Sometimes the cleavage planes dip towards the same point of the compass as those of stratifi- cation, but often to opposite points. The cleavage, as represented in fig. 704, is generally constant over the whole of any area affected Section in Lower Silurian slates of Cardiganshire, showing the cleavage planes bent along the junction of the beds. (T. McK. Hughes.) by one great set of disturbances, as if the same lateral pressure which caused the crumpling up of the rock along parallel, anticlinal and synclinal axes caused also the cleavage. Professor McKenny Hughes remarks, that where a rough cleavage cuts flagstones at a considerable angle to the planes of stratification, the rock often splits into large slabs, across which the lines of bed- ding are frequently seen, but when the cleavage planes approach 544 ITS NATURE AND ORIGIN [CH. xxxvni. within about 15 of stratification, the rock is apt to split along the lines of bedding. He has also called attention to the fact that sub- sequent movements in a cleaved rock sometimes drag and bend the cleavage planes along the junction of the beds, indicated in the Fig. 706. Stratification, joints, and cleavage. (From Murchison's ' Silurian System.') annexed section (fig. 705). The relation of cleavage planes to joints is seen in fig. 706. The joints J J are parallel. S S are the lines of stratification ; D D are lines of slaty cleavage, which intersect the rock at a considerable angle to the planes of stratification. Mechanical theory of cleavage, Professor Phillips long ago remarked that in some slaty rocks, affected by cleavage, the form of the outline of fossil shells and trilobites has been much changed by distortion, which has taken place in a longitudinal, transverse, or oblique direction. This change, he adds, seems to be the result of a ' creeping movement ' of the particles of the rock along the planes of cleavage, its direction being always uniform over the same tract of country, and its amount in space being some- times measurable, and being as much as a quarter or even half an inch. Mr. D. Sharpe, following up the same line of inquiry, came to the conclusion that the present distorted forms of the shells in certain British slate rocks may be accounted for by supposing that the rocks in which they are embedded have undergone compression in a direction perpendicular to the planes of cleavage, and a corre- sponding extension in the direction of the dip of the cleavage. It would appear that the pressure was at right angles to the original bedding, and that it was very great. Subsequently (in 1853) Mr. Sorby demonstrated that this mechanical theory is applicable to the slate rocks of North Wales and Devonshire, districts where the amount of change in dimen- sions can be tested and measured by comparing the different effects exerted by lateral pressure on alternating beds of finer and coarser materials. Thus, for example, in the accompanying figure (fig. 707) it will be seen that the sandy bed d f, which has offered greater resistance, has been sharply contorted, while the fine-grained strata, a, 6, c, have remained comparatively unbent. The points d and /in the stratum d f must have been originally four times as far apart as they are now. They have been forced so much nearer tc each other, partly by bending, and partly by becoming elongated in CH. xxxviii.] PEOOFS OF COMPRESSION 545 the direction of what may be called the longer axes of their contor- tions, and lastly, to a certain small amount, by condensation. The chief result has obviously been due to the bending ; but, in proof of elongation, it will be observed that the thickness of the bed d f is now about four times greater in those parts lying in the main Fig- 707. direction of the flexures than in a plane perpendicular to them ; and the same bed exhibits cleavage-planes in the direction of the greatest movement, al- though they are much fewer than in the slaty strata above and below. Above the sandy bed d f, the stratum c is somewhat dis- turbed, while the next bed b is much less so, and a not at all ; yet all these beds, c, b, and a, must have undergone an equal amount of compression with d, the points a, and g having ap- proximated as much towards each other as have d and /. The same phenomena are also re- peated in the beds below d, and might have been shown, had the section been extended down- wards. Hence it appears that the finer beds have been squeezed into a fourth of the space they previously occupied, partly by condensation, or the closer packing of their ultimate particles (which has given rise to the high specific gravity of such slates), and partly by elongation in the planes of the cleavage of which the general direction is perpen- dicular to that of the pressure. 'These arid numerous other cases in North Devon are analogous,' says Mr. Sorby, 'to what would occur if a strip of paper were included in a mass of some soft plastic material which would readily change its dimensions. If the whole were then compressed in the direction of the length of the strip of paper, it would be bent and puckered up into contortions ; whilst the plastic material would readily change its dimensions without undergoing such contortions ; and the difference in distance of the ends of the paper, as measured in a direct line or along it, would indicate the change in the dimensions of the plastic material.' N N (Drawn by H. C. Sorby.) Vertical section of slate rock in tLe cliffs near Ilfracombe, North Devon. Scale one inch to one foot. , 6, c, c. Fine-grained slates, the stratifi- cation being shown partly by lighter or darker colours, and partly by differ- ent degrees of fineness in the grain. d, f. A coarser-grained, light-coloured sandy slate, with less perfect cleavage. 546 EXPERIMENTAL ILLUSTRATIONS [CH. xxxvin. Experimental demonstration of the origin of slaty cleav- age. Mr. Sorby has come to the conclusion that the absolute con- densation of the slate rocks amounts, upon an average, to about one- half their original volume. Most of the scales of mica occurring in certain slates examined by Mr. Sorby lie in the plane of cleavage (see fig. 715) ; whereas in a similar rock not exhibiting cleavage they lie with their longer axes in all -directions. May not their position in the slates have been determined by the movement of elongation before alluded to ? To illustrate this theory, some scales of oxide of iron were mixed with soft pipeclay in such a manner that they inclined in all directions. The dimensions of the mass were then changed artificially to a similar extent to what has occurred in slate rocks, and the pipeclay was then dried and baked. When it was afterwards rubbed to a flat surface, perpendicular to the pressure, and in the line of elongation, or in a plane corresponding to that of the dip of cleavage, the particles were found to have become arranged in the same manner as in natural slates, and the mass admitted of easy fracture into thin flat pieces in the plane alluded to, whereas it would not yield in that perpendicular to the cleavage. Tyndall, when commenting in 1856 on Mr. Sorby's experi- ments, observed that pressure alone is sufficient to produce cleavage, and that the intervention of plates of mica or scales of oxide of iron, or any other substances having flat surfaces, is quite unnecessary. In proof of this he showed experimentally that a mass of ' pure white wax, after having been submitted to great pressure, exhibited a cleavage more clean than that of any slate-rock, splitting into laminae of surpassing tenuity.' He remarks that every mass of clay or mud is divided and subdivided by surfaces among which the cohesion is comparatively small. On being subjected to pressure, such masses yield and spread out in the direction of least resistance, small nodules become converted into laminas separated from each other by surfaces of weak cohesion, and the result is that the mass cleaves at right angles to the line in which the pressure is exerted. In reply to Tyndall, Mr. Sorby pointed out that the white wax is really a crystalline substance made up of prismatic needles, as is seen when it is examined with a microscope, and under the influence of pressure these inequiaxed particles arrange themselves with their longer axes at right angles to the direction in which the force is applied. Darwin attributed the lamination and fissile structure of volcanic rocks of the acid series, including some obsidians in Ascension, Mexico, and elsewhere, to their having moved when liquid in the direction of the laminae. The separation of the bands sometimes results from air-cells being drawn out and flattened in the direction of the moving mass. Foliation of Crystalline Schists. After studying, in 1835, the crystalline rocks of South America, Darwin proposed the term foliation for the structure that leads to the separation of gneiss, mica-schist, and other crystalline rocks into laminae or plates. ' Cleavage,' he observes, may be applied to the structure in which divisional planes render a rock fissile, although it may appear to the eye quite or nearly homogenous. ' Foliation ' may be used when the alternating layers or plates are of different mineralogical nature, like those of which gneiss and other metamorphic schists are composed. CH. xxxviii.] NATURE OF FOLIATION 547 It will be seen, then, that foliation differs from cleavage in the circumstance that the laminae into which a cleaved rock breaks up are all of the same composition ; while those of a foliated rock consist of distinct minerals like the quartz, felspar, and mica of gneiss (see fig. 708). The thin flakes making up a foliated rock, moreover, usually have a distinctly lenticular form, and may be spoken of as folia, rather than lamina) like those of slate. There is, however, the most perfect gradation from cleaved into foliated rocks. That the planes of foliation of the crystalline schists in Norway accord very generally with those of original stratification is a con- clusion long since espoused by Keilhau. Numerous observations made by the late David Forbes in the same country (the best probably in Europe for studying such phenomena on a grand scale) seemed to confirm Keilhau's opinion. In Scotland, also, Forbes pointed out what seemed to be a striking case where the foliation is identical with the lines of stratification, in rocks well seen near Grianlarich in Perthshire. There is in that locality a crystalline lime- stone, foliated by the intercalation of small plates of white mica, so Fig. 708. Fragment of gneiss, natural size ; section made at right angles to the planes of foliation. that the rock is often scarcely distinguishable in aspect from gneiss or mica-schist. The stratification is shown by the large beds and coloured bands of limestone all dipping, like the folia, at an angle of 32 degrees N.E. In stratified formations of every age we see layers of siliceous sand, with or without mica, alternating with clay, with fragments of shells or corals, or with seams of vegetable matter : and we should expect, Forbes argues, the mutual attraction of like parti- cles to favour the crystallisation of the quartz, or mica, or felspar, or calcite along the planes of original deposition, rather than in planes placed at angles of 20 or 40 degrees to those of stratifica- tion. After a general examination of the metamorphic rocks of the Highlands, Murchison and Geikie were led to the conclusion that, throughout the whole district, foliation is coincident with the strati- fication of the rocks, and not, as had been suggested by Daniel Sharpe, with their cleavage. Scrope, on the other hand, was inclined to attribute the foliation of the crystalline schists to ' the results of internal differential movements in the constituents of the subterranean mineral matter while exposed to enormous irregular NN 2 548 VAKIETIES OF FOLIATED STRUCTURE [CH. xxxviii. pressures as well as to variations of temperature, and under these influences changing at times from a solid to a fluid state, and pro- bably back again to crystalline solidity, through intervening phases of viscosity movements and changes which must of necessity have frequently arranged and rearranged the component crystalline minerals, sometimes in irregular composition like that of granite, diorite, or trachyte, sometimes in laminar or schistose bands like those of gneiss, mica-schist, and other so-called metamorphic crys- tallines.' We have seen how much the original planes of stratification rm,y be interfered with or even obliterated by concretionary action in deposits still retaining their fossils, as in the case of the Magnesian limestone of the Permian. Hence we must expect to be frequently baffled when we attempt to decide whether the foliation does or does not accord with that arrangement which gravitation, combined with current-action, imparted to a deposit from water. Moreover, when we look for stratification in crystalline rocks, we must be on our guard not to expect too much regularity. The occurrence of wedge- shaped masses (such as belong to coarse sand and pebbles), oblique lamination, ripple-marks, Fig. 709. unconf ormable stratifica- tion, the fantastic folds pro- duced by lateral pressure, faults of various widths, intrusive dykes, remains of organic bodies of diversified shapes, and other causes of irregularity in the planes of deposition, both on the small and on the large scale, will interfere with paral- lelism. If complex and enigmatical appearances did not present themselves, it would be a serious objection to the metamorphic theory. Mr. Sorby has shown that a structure which he compares to that of ripple-marked sands can be detected in certain varieties of mica-schists in Scotland. In the diagram (fig. 709) is represented the foliation of a coarse argillaceous schist in the Pyrenees (which was examined by Lyell in 1830). In part, it approaches in character to a green and blue roofing-slate, while part is extremely quartzose, the whole mass passing downwards into micaceous schist. The vertical section here exhibited is about three feet in height, and the layers are sometimes so thin that fifty may be counted in the thickness of an inch. Some of them consist of pure quartz. There is a resemblance in such cases to the diagonal lamination which we see in sedimentary rocks, even though the layers of quartz and of mica, or of felspar and other minerals, may be more distinct in alternating folia than they were originally. General coincidence between Foliation and Cleavage in Metamorphic Rock-masses. In spite of examples, like those just cited, in which the folaition of metamorphic rocks appears to follow the original lamination (or fine bedding) of a stratified mass, Foliation of an argillaceous schist, Montague de Seguinat, near Gavarnie, in the Pyrenees. en. XXXVIIT.] EXPERIMENTAL ILLUSTRATIONS 549 there can be little doubt that, in the great majority of cases, the schistose structure is an entirely superinduced one, and that foliation, like cleavage, must be referred to the action of pressure, the planes of foliation being developed, like those of cleavage, at right angles to the direction in which the pressure is exerted. What were taken by David Forbes, and by Murchison and Geikie, as cases of the interbedding of rock-masses, with foliation parallel to the stratifica- tion, have been proved by the researches of Professor Lapworth and the officers of the Geological Survey to be really examples of rock- masses brought into juxtaposition by great reversed faults (thrust- planes), see fig. 631, p. 436. Simulations of ' false-bedded ' and ' ripple- mark' structures like those referred to in the Pyrenees appear to often result from changes in the direction of pressure in a great mass undergoing folding movements which lead to the appearances known as Ausweichungs-Clivage (the ' strain-slip cleavage ' of Professor Bonney). Such being the case, we can understand the phenomena to which attention was drawn by Darwin in South America, where over vast areas cleavage and foliation everywhere maintain a marked parallelism ; the strike of the cleavage and foliation being coincident with that of the stratification, but the dip of the planes of cleavage being inclined, often at a very high angle, to those of bedding. Experimental Illustrations of Dynamo-metamorphic action. By the method of sealing up various substances in glass tubes with water and exposing them to high temperatures, so that the confined vapour of the water exercises a powerful pressure within the tube, Daubree and other French chemists and mineralo- gists have shown that many crystallised minerals may be produced. Glasses, both natural and artificial, which are amorphous mixtures of various silicates, were found to break up under these conditions, and their various constituents recombined and crystallised out as quartz, sanidine, wollastonite, diopside, and other well-known mineral species. In this way a very considerable proportion of the minerals composing the earth's crust has been artificially prepared ; the crystals, though often of microscopical dimensions, presenting all the distinguishing characters of the -natural ones. Professor W. Spring, of Liege, has carried on a series of experi- mental researches upon the effects of pressure apart from those of high temperature. In these experiments, pressures estimated to exceed 7,000 atmospheres were employed, and the precaution was taken of applying the force so slowly that any heat generated would be dissipated, and would not interfere with the result. The con- clusions at which Spring arrived were as follows : 1. Powders of metals and other solids may, by intense pressure (especially if all interstitial air-films be removed by the action of an air-pump) be converted into solid masses indistinguishable from those produced by fusion. In powders and colloid masses pressure will produce a perfectly crystalline structure. 2. Where elements have allotropic forms, or compounds are hetero- morphous, the less dense substance may be converted into the heavier by the action of pressure. Van 't Hoff and Reicher have also shown that the temperature at which all such paramorphic changes take place is modified by pressure. 3>. Powders of metals, oxides and salts may by pressure be made fa> react chemically upon one another without the intervention of 550 EXAMPLES OF [CH. xxxvm. any liquid or gas and alloys are produced, various salts formed, and double decompositions brought about under such conditions. 4. The rubbing or sliding of the particles of solid bodies over one another under intense pressure powerfully promotes chemical action between them. 5. When the particles of solid bodies have been brought into contact by intense pressure, the chemical action between them goes on, even when the pressure is removed. 6. The action of pressure on solids is variously modified by the presence of small quantities of water or of various gases. The late Dr. Guthrie and other physicists and chemists have shown, by the study of solutions under pressure, that there is a per- fect continuity between the states of solution and fusion. As was maintained by Bunsen, the various mixtures of silicates, which con- stitute igneous rocks, are really solutions at high temperatures, the solvent being sometimes a fusible silicate, often mixed with more or less water under pressure. A very good summary of our ' Etudes sur le Metamorphisme knowledge on the cleavage of rocks des Roches, ' to Daubree's ' Geolo- will be found by the student in the gie Experimentale,' and to J. essay of Mr. Marker on the sub- Lehmann's 'Die Entstehung der ject, Brit. Assoc. Rep. 1885. For altkrystallinischen Schieferge- a discussion of the action taking steine,' and for an account of place in the metamorphism of Spring's researches to 'Journ. rocks he is referred to Delesso's Chem. Soc.' 1890, p. 404. CHAPTER XXXIX CONTACT METAMORPHISM AND REGIONAL METAMORPHISM I THE VARIETIES OF ROCKS RESULTING FROM THESE TWO KINDS OF ACTION Illustrations of the action of Contact Metamorpliism Distance to which Contact Metamorphism can be traced from the intrusive nu.ss Minerals produced by Contact Metamorphism Chief types of Rocks produced by Contact Metamorphism Andalusite, Kyanite, Sillimanite, Staurolite Rocks, &c. Rocks produced by Regional Metamorphism, Quartzites, Metamorphic Limestones, and Dolomites, Slates, Phyllites, Schists, Gneisses, Granulites, Anthracites, &c. As we have seen in the last chapter, contact metamorphism is largely the result of the action of heat, while in the case of regional metamorphism the action of heat is greatly modified by, and even subordinated to, that of pressure. Hence we are not surprised to find that the rocks formed by contact and re- gional metamorphism respectively, while having many features in common, nevertheless often present dissimilar and distinc- tive characters. Fossiliferous strata rendered metamorpbic by intrusive masses. In treating of the nature of intrusive veins of volcanic CH. xxxix.] CONTACT METAMORPHISM 551 and plutonic origin (p. 477), examples were given of alterations in the affected rocks by heat and percolating water containing chemical matters. The subject was further illustrated in noticing the methods of distinguishing the age of volcanic rocks (p. 485). It is, therefore, only necessary to cite a few additional instances of local metaniorphism. In the southern part of Norway there is a large district, on the west side of the fiord of Christiania (which Lyell visited in 1837 with the late Professor Keilhau) in which hornblendic granite protrudes in mountain masses through fossiliferous strata, usually sending veins into them at the point of contact. The stratified rocks, replete with shells and corals, consist chiefly of shale, limestone, and some sandstone, and all these are invari- ably altered near the granite for a distance of from 50 to 400 yards. The shales are hardened, and have become flinty, sometimes resembling jasper. Eibboned jasper is produced by the hardening of alternate layers of green and chocolate- coloured shale, each stripe faithfully representing the original lines of stratification. Nearer the granite the altered shale often contains crystals of hornblende, which are even met with in some places, for a distance of several hundred yards from the junction ; and this black hornblende is so abundant that eminent geologists, when passing through the country, have confounded it with the ancient hornblende -schist, subordinate to the great gneiss formation of Norway. Frequently, between the granite and the hornblende -slate above mentioned, crystalline grains of mica and felspar appear in the schist, so that rocks resembling gneiss and mica-schist are produced. Fossils can rarely be de- tected in these schists, and they are more completely effaced in proportion to the more crystalline texture of the beds and their vicinity to the granite. In some places the siliceous matter of the schist becomes a granular quartzite ; and when hornblende and mica are added, the altered rock loses its stratification, and resembles granite. The limestone, which at points remote from the granite is of an earthy texture and blue colour, and often abounds in corals, becomes a white granular marble, sometimes siliceous, near the granite the granular structure extending occasionally upwards of 400 yards from the junction ; the corals are for the most part obliterated, though sometimes preserved, even in the white marble. Both the altered lime- stone and the hardened slate contain garnets in ' many places, with ores of iron, lead, and copper, and some silver. These alterations occur equally, whether the granite invades the strata in a line parallel to the general strike of the fossiliferous beds, or in a line at right angles to their strike .both of which modes 552 METAMORPHISM BOUND [CH. xxxix. of junction will be seen by the accompanying ground- plan (fig. 710). The granite of Cornwall sends forth veins into a coarse argillaceous schist, locally termed killas. This killas is con- verted into hornblende -schist near the contact with the veins. These appearances are well seen at the junction of the granite and killas in St. Michael's Mount, a small island nearly 300 feet high, situated in the bay, at a distance of about three miles from Penzance. The granite of Dartmoor, in Devonshire, ac- cording to De la Beche, has intruded itself into the Carbonife- rous slate and slaty sandstone, twisting and contorting the strata, and sending veins into them. Hence some of the slate rocks have become ' micaceous ; ' others much indurated, exhibit cha- racters of mica-slate ; while others again are converted into a hard, banded rock with much felspar resembling gneiss. Fig. 710. >< X" ^ __^ 7//?m7; ^ unaltered Ground-plan of altered slate and limestone near granite, Christiama. The arrows indicate the dip, and the oblique lines the strike of the beds. Nowhere, however, are the phenomena of local metamorphism more beautifully illustrated than in the Western Isles of Scotland. In this district, great masses of granite and gabbro have been thrust through the various Palaeozoic and Secondary strata, during the Tertiary period ; and in the vicinity of the junctions of the igneous and the sedimentary masses most instructive examples of metamorphism may be observed. Thus limestones of the same age as those of Durness (Cambrian) are found losing, as we approach the igneous rocks, all traces of their organic remains, and at last passing into a highly crystalline or saocharoid marble suitable for statuary purposes. Clays and sandstones of various ages, under like conditions, are also found to be deprived of every trace of the organic structures originally present in them and to graduate into indurated slaty rock and CH. xxxix.] GKANITIC INTEUSIONS 553 quartzite, while the felspathic sandstones of the Cambrian are altered to a highly micaceous quartzite. We learn from the investigations of M. Dufrenoy, that in the Eastern Pyrenees there are mountain masses of granite, posterior in date to the formations called Lias and Chalk of that district, and that these fossiliferous rocks are greatly altered in texture, and often charged with iron- ore, in the neighbourhood of the granite. Thus in the environs of St. Martin, near St. Paul de Fenouillet, the chalky limestone becomes more crystalline and saccharoid as it approaches the granite, and loses all trace of the fossils which it previously contained in abundance. At some points, also, it becomes dolomitic, and filled with small veins of ferrous carbonate and spots of red hematite. The local metarnorphism of carbonaceous beds, such as coal- seams, is very interesting. The most simple result of the intru- sion of a dyke of basalt, for instance, amongst Coal-measures is for the coal to become hard and brittle, to lose its more volatile matters, and to change into anthracite, or even into graphite, and this may take place 50 yards away from the basalt. Close to the dyke, the coal may be reduced to the form of cinder, occupying a much smaller space than before ; or, as in South Staffordshire, the coal may become sooty and coked. Distilla- tion arising from the heating and alteration of coal and bitu- minous shales by the action of igneous intrusions causes the gases to find their way to the surface, and the liquid products to collect in fissures and cavities. Petroleum and asphalte are thus collected in chinks of sandstones and other sedimentary rocks, and even of the igneous rocks themselves, while natural gases (hydrocarbons) find their way to the surface. Prismatic structure, resembling miniature basaltic columns, has often been produced in coal by this local metamorphic action. On the other hand, the igneous rock has often been altered by contact with the coal, becoming white, or yellow, earthy, light, and friable. This ' white trap,' as it is called, has had its crystalline structure nearly destroyed, much of its silica and lime removed, while its iron remains to form ferrous carbonate. Extent of Contact BXetamorphism. Contact metamorphism is always distinguished by its local character ; in some cases the alteration produced can only be traced for a few feet or even inches from the planes of junction with the igneous mass ; and in few instances probably can such changes be detected at distances of over two miles. Usually, the amount of alteration increases as we approach the igneous rock, but there are some very remarkable cases in which apparent exceptions to this rule are found. The different kinds of rocks undergo very variable amount of change, 554 MINERALS PRODUCED BY [CH. xxxix. according to their chemical composition, and thus it sometimes happens that in the metamorphic zones surrounding great intrusive masses we find alternations of rocks which are much altered with others showing few signs of change. Where contact metamorphism is extreme, a very marked foliation is always developed. As a rule, the extent of metamorphism, and the distance to which it can be traced, bear a marked relation to the volume of the intrusive mass around which it is exhibited. But it should be remembered that in some cases the igneous rock may have been simply injected into the surrounding rocks, while in other cases the liquid mass may have been forced for ages through the fissure which has served as a means of communication with the surface. The amount of chemical action in the latter case would, of course, be far greater than in the former. It is worthy of notice that the minerals produced in lime- stones and other rocks in contact with igneous masses are identical with those found in the fragments of much-altered materials that are thrown from volcanic vents. Not only do we find all kinds of sedimentary materials under- going alteration around plutonic rock-masses, but volcanic and older plutonic rocks are likewise changed, while metamorphic rocks are subjected to still further metamorphism. Chief varieties of minerals produced by Contact metamorphism. The principal changes produced by the ther- mal or hydrothermal action taking place around igneous intrusions consist in the development of various crystalline minerals in the mass. Most of these minerals appear to be formed from the various elements already existing in the rock, these entering into new com- binations and forming crystals often of great beauty and perfection. But, in many cases, there can be no doubt that substances con- tained in the intrusive mass react on the materials into which it is thrust, and minerals are formed which could not be produced by simple metamorphism. Of such metasomatic changes, as they are called, we have examples in the formation of tourmaline, axinite, fluorspar, &c., through the action of the boric, hydrofluoric acid, and other gases given off from great intrusive masses like the granite of Dartmoor acting on the silicates of the invaded rock, and in the impregnation of metallic sulphides so frequently found near the contact'of igneous and other rock masses. On the other hand, we not unfrequently find evidence that the igneous rock itself is affected by the contact with materials among which it is intruded. Fragments of the surrounding rock are torn off, being fused and absorbed in the liquid and highly heated mass, and new minerals are formed during its crystallisation. All the ordinary rock-forming minerals- quartz, the various felspars, and many ferro-magnesian silicates are formed by contact metamorphism ; varieties of biotite are especially abundant, and forms of hornblende are by no means rare. There are certain silicates for the most part highly aluminous ones which are par- ticularly abundant in, and generally characteristic of, the metamor- phosed argillaceous rocks. Among the most important of these are andalusite, sillimanite or fibrolite, and kyanite, with garnet, stauro- lite, cordierite, epidote, and zoisite. Many of these minerals are very unstable, and we frequently find in the metamorphosed rocks, not the minerals themselves, but the products of their alteration, CH. XXXIX.] CONTACT METAMOEPHISM 555 such as muscovite and its hydrated forms (damourite, sericite, &c.), pinites, kaolin, the chlorites, ottrelite, and other chloritoids, the ver- miculites, &c. Many of the rocks undergoing contact metamorphism are seen to exhibit spots and markings, showing that a segregative action has gone on in the mass, and between these obscure markings and fully developed crystals of biotite, hornblende, garnet, andalu- site, &c., we find every intermediate gradation (see fig. 711). The structure of these ' spotted slates' (' flecktschiefer,' ' garbenschiefer,' etc., of the Germans) is admirably displayed in thin sections under the microscope, and changes by which the passage of amorphous and fragmental materials into beautifully crystallised minerals is effected, can be distinctly traced. Some of the garnets and other minerals found in the rocks altered by contact metamorphism are of large size, perfectly clear, and free from foreign inclusions. But in other cases, much uncrystallised material is found caught up in Fig. 711. Fig. 712. Spotted slate from Saxony. The smaller dark-coloured spots are incipient crystals of biotite ; the larger and paler-coloured ones are hornblende. In both, the work of crystallisation is incomplete, and thev include much amorphous material distributed through a base with crystalline pro- perties. Chiastolite slate, from Bavaria. The white spots are sections (longitudinal and transverse) of the variety of andalusite known as chiastolite. In these, the amorphous matter, instead of being irregularly distributed through the crystals, is arranged in cross- shaped patterns within them. The material around these crystals is but little altered. the crystals during their growth, and this sometimes shows a re- markable symmetrical arrangement within the mineral, as in the form of andalusite known as ' chiastolite ' (see fig. 712). Rocks produced by Contact ittetamorphism. Purely siliceous sands and sandstones are altered by contact metamor- phism into a quartzite or quartz -rock, the grains of quartz, besides being cemented into a solid rock, sometimes having the aqueous solutions or carbon dioxide expelled from their cavities, and occasion- ally exhibiting signs of partial fusion or even of complete recrystal- lisation (see fig. 713). More felspathic or micaceous sandstones may have many secondary minerals developed in them, and the rock may become distinctly foliated (quartz -schist). Very impure sandstones, grits, and arkoses pass into siliceous schists, and even into rocks which must be classed as true gneisses. It is in the case of the argillaceous rocks that we find the greatest variety of products resulting from the action of contact 556 VAEIETIES OF [CH. xxxrx. metamorphism. When fine-grained siliceous clays are exposed to the action of heat by contact with igneous masses, they pass into the hard compact materials often called hornstones, porcellanites, ribboned jasper, Lydian stone, &c., and in some of these materials traces of the fossils contained in the original rocks may still be detected. But clays, shales, and slates are also found passing into spotted slates of different kinds, and then into mica-slates, andalu- site- and chiastolite-slates, cordierite-slates, ottrelite-slates, &c. Garnets are frequently developed in great numbers, but of micro- scopic dimensions ; and in this way are produced some of the materials most valued as whetstones, like the celebrated Water-of- Ayr stone. In other cases the micaceous minerals assume a parallel arrangement, large garnets are developed with staurolite, andalusite, cordierite, and many other minerals, and the whole mass passes into a foliated rock very similar to the product of regional metamorphism. Fig, 713. Fig. 714. Quartzite, from Saxony, made vup of grains of clear quartz with cloudy felspars. The rock has probably been produced by the metamorphism of a felspathic sandstone or grey wacke; the outlines of the grains being still dis- tinguishable. Impure crystalline limestone, from Styria. Clear crystalline grains of calcite are seen, traversed by the cha- racteristic twinning and cleavage- planes, with some quartz-grains con- taining streams of liquid-cavities and opaque particles of graphite. Pure limestones become completely recrystallised by the action of metamorphism, and aggregates of calcspar crystals each marked by the peculiar twinning and cleavage of the mineral are formed, giving rise to the well-known saccharoid or statuary marble (see fig. 714). In some cases, magnesian compounds appear to be in- troduced into a calcareous rock during contact metamorphism, and magnesian limestones, and even true dolomites, result from the action. When the limestone contains impurities, new crystalline minerals are formed, such as tremolite, actinolite, various micas, wollastonite, felspars, zoisite, lime-garnets, quartz, &c. ('calciphyres '). The rocks thus formed occasionally exhibit a distinct foliation, and calc-schists (' cipolinos ') are produced, indistinguishable from those resulting from regional metamorphism. Many of the most beautiful marbles, and rocks containing the greatest variety of crystallised minerals like those of Monzoni in the Tyrol appear to be the result of the action of contact metamorphism on limestone rocks. That igneous and metamorphic rocks are also subjected to CH. xxxix.] METAMORPHIC BOOKS 557 contact metamorphism has been already pointed out. In some cases, the minerals produced by weathering, like kaolin, chlorites, calcites, chalcedony, &c., are altered back to others similar to those which existed in the original rocks. In other cases, new minerals have been formed, like albite, hornblende, epidote, sphene, with pyrites and other sulphides the latter being due in part at least to the introduction of certain materials given off by the igneous magma which has invaded the rocks. Regional metamorphic Rocks. These rocks, when in their most characteristic development, are wholly devoid of organic remains, and contain no distinct fragments of other sedimentary rocks. Gneiss and mica-schist may be taken as typical examples. But phyllites and some schists (the former sometimes still containing fossils or their impressions) may be considered to be less altered varieties. They sometimes appear in the central parts of mountain chains, but in other cases extend over areas of vast dimensions, occupying, for example, nearly the whole of Norway and Sweden, where, as in Brazil, they appear alike in the lower and higher grounds. However crystalline these rocks may become in certain regions, they seldom, like granite, send veins into contiguous formations. In Great Britain, those members of the series which approach most nearly to granite and other plutonic rocks in their composition, as gneiss, mica- schist, and hornblende-schist, are chiefly found in the country north of the rivers Forth and Clyde, in Wales, the Malverns, and Leicestershire (Charnwood Forest). Many attempts have been made to trace a general order of succession or superposition in the members of this family ; clay- slate, for example, having been often supposed to hold in- variably a higher geological position than mica-schist, and mica- schist to overlie gneiss. But although such an order may prevail throughout limited districts, it is by no means universal. The mechanical peculiarities of these rocks are expressed by the terms ' cleavage ' and ' foliation.' We have seen that sedimentary rocks in the immediate proximity of great igneous intrusions are found to have under- gone great induration, while the development of various crys- talline minerals has frequently taken place in them. In cases where the action of contact metamorphism has been extreme, we have shown that a very distinct foliation is often developed in the altered rocks. The similarity of the rocks thus formed especially where the metamorphism has been extreme to many of the foliated or schistose rocks, characteristic of regional meta- morphisrn, suggests that the latter may have been produced from pre-existing strata by the action of analogous chemical 558 QUAKTZITE, SCHISTS, GRAPHITE [CH. xxx.x. forces operating on a more extended scale. Thus gneiss and mica-schist may be nothing more than altered felspathic and micaceous sandstones, granular quartzite may have been de- rived from pure sands and sandstone, and the most highly crystalline quartzite may be the last stage of alteration of the same materials. Similarly, clay- slate and many forms of schist may be altered shale, and granular marble may have originated in an ordinary limestone, replete with shells and corals, which have since been obliterated ; and, lastly, calcareous sands and marls may have been changed into impure crystalline lime- stones. The anthracite and graphite associated with regional-meta- morphic rocks may have been coal ; for not only is coal con- verted into anthracite in the vicinity of some trap dykes, but we have seen that a like change has taken place generally even far from the contact of igneous rocks, in the disturbed region of the Appalachians. At Worcester, in the State of Massachusetts, 45 miles due west of Boston, a bed of plumbago or impure graphite occurs, interstratified with mica-schist. It is about 2 feet in thickness, and has been made use of both as fuel and in the manufacture of ' lead-pencils.' At the distance of 30 miles from the plumbago, there occurs, on the borders of Rhode Island, an impure anthracite in slates containing impressions of coal-plants of the genera Pecopteris, Neuropteris, Calamites, &c. This anthracite is intermediate in character between that of Pennsylvania and the graphite of Worcester, in which last the gaseous or volatile matter (hydrogen, oxygen, and nitrogen) is to the carbon only in the proportion of 3 per cent. (After tra- versing the country in various directions, Lyell came to the conclusion that the Carboniferous shales or slates with an- thracite and plants, which in Rhode Island ofte:. pass into mica-slates, have at Worcester assumed a perfectly crystalline and metamorphic texture: the anthracite having been nerarly transmuted into that state of pure carbon which is called plumbago or graphite.) The alterations already described as being superinduced in rocks by volcanic dykes and granite veins prove incontestably that powers exist in nature capable of transforming clastic and fossiliferous strata into crystalline rocks. But while all the sedimentary rocks undergo more or less complete recrystallisation, with or without the development of a foliated structure, like changes may affect all kinds of volcanic, plutonic, and metamorphic rock-masses. Tuffs and lavas, as well as the several varieties of crystalline rocks, may thus be converted into schists and gneisses. It is probable that many CH. xxxix.] SLATES AND MYLONITES 559 of the gneisses and schistose rocks which we now see exposed at the earth's surface have undergone not one cycle of change but many repeated metamorphoses. All the materials of the earth's crust, indeed, tend to pass through regular cycles of change, granites and the crystalline rocks being broken up by denuding agencies at the surface to form clastic sedimentary rocks, and these, when buried at great depths, being subjected to metamorphic action whereby they pass into schists, and even into gneiss, in which last all traces of foliation may disappear, when the rock becomes a granite. Such being the case, it is of course impossible to say from what particular variety of aqueous, igneous, or metamorphic rock a given gneiss or schist has been formed. The ultimate chemical composition of such rocks as Fig. 715. Fig. 716. Clay-slate. North Wales. Section cut Contorted mylonite. Loch Maree. The transversely to the cleavage. The rock is made up of crushed and re- flattened grains are all seen with their cemented particles. The breaking-up longer axes lying in the direction of of the mass has been produced along the cleavage, or at right angles to the fault- (or 'thrust'-) planes. The direction in which the pressure has ' cataclastic ' structure is especially acted which produced that structure. well seen under polarised light, the The flattened grains consist of kaolin different orientation of the crystal and other micaceous minerals. particles being thus made visible. quartzites and limestones of course indicates that they must have been formed from arenaceous or calcareous strata, but there are many schists, granulites, and even gneisses, which, so far as their ultimate chemical composition indicates their origin, may have been equally formed from igneous or sedimentary materials. Rocks formed by regional Met amor pbism. Among the least altered of the rocks affected by dynamo-metamorphic action are the clay-slates, and the rocks called by Professor Lapworth ' mylonites.' Clay-slates retain many of the ordinary characteristics of a clay or shale, though with a slightly increased density. But sections of slate cut transversely to the cleavage, when examined under the microscope, exhibit a remarkable parallelism of their 560 PHYLLITES AND SCHISTS [CH. xxxix. particles, all the rods and flat plates in the mass having had their positions rearranged so that they lie at right angles to the direction of the pressure which has acted upon them (see fig. 715). Mylonites are composed of the fine dust or fragments of rocks which have been crushed to powder along great fault-planes (or ' thrust- planes '), these fine particles being consolidated and often partially recrystallised (see fig. 716). Many metamorphic rocks exhibit a similar 1 cataclastic ' structure. Where pebbles and other included masses have existed in the rock they are often fractured and sometimes crushed and broken, the fragments being sometimes torn apart from one another by the shearing movements within the mass. In the same way, metamorphic rocks which contain porphyritic crystals some- times exhibit this cataclastic structure in a very marked manner, the large felspar-crystals having their edges and angles rounded off, so as to form the eyes (' Augen ') of the so-called Augengneiss (see fig. 720). Fig. 717. Pig. 718. Phyllite. Saxony. Marje up of clastic Clilorite-scliist. Moravia. Made up of materials mingled with crystals of crystalline particles of chlorite and biotite and other micaceous' minerals magnetite, showing a distinct fo'.ia- of secondary origin, developed along tion. Other minerals are occasion- the planes of cleavage. ally present in the rock. In the majority of cases, however, the movements producing cleavage and shearing in a rock -mass are attended with a certain amount of recrystallisation, as well as deformation and crushing. Thus we often find the surfaces of slaty rocks covered with crystals of mica and other minerals which are evidently of secondary origin and have been produced during the movements to which the rock- masses have been subjected. Such rocks are usually called in England mica-slates, talcose-slates, chlorite-slates, &c. (after the mineral most distinctly exhibited in them), in contradistinction to the clay-slates in which no such secondary minerals are apparent in the cleavage-planes. In France, rocks of this class are usually called phyllites (see fig. 717) ; there are phyllites, as has been shown by Professor Eeusch, of Christiania, which exhibit recognisable traces of corals, trilobites, and other fossils, which remain in spite of the partial recrystallisation of the materials of the rock. When the whole, or nearly the whole, of the materials of a rock have been recrystallised, so that it is made up of thin folia of quartz and of some other minerals, the rock is called a schist. CH. XXXIX.] GNEISS AND GRANULITE 561 The term ' schist ' is, however, used in a much more general manner in Germany, being sometimes applied to phyllites and even to clay- slates and shales. The different kinds of schist are named after the most conspicuous mineral present in them, such as mica-schists Fig. 719. Fig. 720. Mica-schist -With garnets. Made up of folia of mica (distinguished by its dis- tinct basal cleavage) and quartz with large garnets (one is seen in the lower part of the section) interspersed through the mass. Hornblende- gneiss, with ' eyes ' (Augen- gneiss). The large deformed crystals consist of felspar or, more rarely, horn- blende (one example with characteris- tic cleavage is seen in the lower part of ' the section). (see fig. 719), talc-schists, chlorite-schists (see fig. 718), hornblende- schists, actinolite-schists, tremolite-schists, epidote-schists, pied- montite-schists, &c. These schists often contain additional minerals identical with those found in rocks formed by contact meta- Fig. 721. Fig. 722. Pyroxene-granulite (' trap-granulite '). Saxony. Made up of colourless grains of felspar and quartz, with augite, hypersthene, and garnet. The inter- growth of felspn.r and garnet gives rise to the 'centric' structure seen near the middle of the section. Granulite, with kyanite. Saxony. Rounded granules of quartz and orthoclase felspar make up the mass of the rock, through which are scat- tered larger grains of garnet and kyanite (the former mineral is seen on the right of the section). morphism such as garnet, cordierite, kyanite, sillimanite or fibrolite, andalusite, staurolite, &c. The metamorphic rocks which contain felspar form the two classes of the yranulites and the gneisses. These have often the O 562 ACID AND BASIC GNEISS [CH. xxxix, same ultimate chemical composition as igneous rocks, both basic and acid, though some of them are not improbably the result of the extreme metamorphism of arkoses, felspathic sandstones (grey- wackes), micaceous flagstones, and similar sedimentary rocks. (See table of analyses, p. 588.) The distinctive structure of the granulites is seen in a rock when all the minerals present more or less rounded grains which fit together, so that under the microscope the sections resemble mosaics. The basic or pyroxene-granulites (' trap granu- lites ') consist of felspars of different species with one or more forms of pyroxene (augite, hypersthene, &c.), sometimes replaced by hornblende or biotite ; in addition, garnets are almost always present, Sometimes in considerable quantities (see fig. 721). The acid or common granulites (leptynites of the French and Weissstein of the Germans) contain much felspar, orthoclase usually predominating with quartz and garnet, to which kyanite is frequently added (see fig. 722). The most common gneisses are of acid composition and agree in their mineralogical composition with the granites, granitites, Fig. 723. Fig. 724. Micaceous gneiss. Saxony. Made up of folia of quartz, felspar, and mica. The rock has the mineralogical consti- tution of an ordinary granite, with a rery distinct foliation. Pyroxene-gneiss. Ceylon. Very coarsely crystalline. The clear crystals are quartz and scapolite, the clouded ones a basic plagioclase felspar, and the dark-coloured ones a green augite. and quartz-diorites, but are distinguished from these by their more or less distinct foliated structure (see fig. 723). The so-called protogine-gneisses have much hydrous white mica (sericite), and some gneisses contain many accessory minerals, such as garnet, cordierite, andalusite, sillimanite or fibrolite, kyanite, &c In addition to these common or acid gneisses, there are others which correspond in composition with the pyroxene-granulites and contain much basic felspar (labradorite or anorthite) with quartz, and some pyroxene (sahlite, eegerine, hypersthene, &c.), these minerals being sometimes replaced by hornblende or biotite. Many accessory or secondary minerals, such as scapolite, wollastonite, fibrolite, garnet, &c., occur in these basic or ' pyroxene '-gneisses (see fig. 724). The gneisr.es are sometimes very coarsely grained rocks ; they not unfrequently contain porphyritic crystals of felspar and other minerals, these being sometimes converted by crushing movements into ' eyes ' (Augengneiss, fig. 720). The foliation of gneiss is often CH. xxxix.] GRANULITES, ETC. 563 so obscure that it can only be seen when great rock-masses are studied in the field. Plutonic rocks, like granite, diorite, and gabbro, not unfrequently present both the granulitic and the gneissic structures ; and when such structures are exhibited by these rocks it can generally be shown that they have been subjected to movements while in a plastic or semi-plastic condition, so as to have a ' flow-structure ' developed in them. The distinction between granulitic gabbros or norites of igneous origin and pyroxene-granulites, of metamorphie origin, is often very doubtful and obscure; and in the same way it is often impossible to decide if a rock should be rightly described as a gneiss-granite or a granitic gneiss. Schists, granulites, and gneisses, even when derived from sedimentary rocks by metamor- phism, cannot be expected to exhibit traces of fossils, for all their materials have been completely recrystallised. It has been shown how insensible are the gradations from various sedimentary rocks, altered by contact metamorphism into true schists and gneisses ; and on the other hand how difficult it is to dis- criminate between certain structural varieties of undoubted plutonic rocks and the granulites and gneisses. These facts point to the conclusion that the rocks now exposed in the earth's crust may all really have passed through those cycles of change which we have been describing. No good reasons have been adduced for asserting that any of the highly crystalline rocks whether foliated or not were originally part of the globe as it first consolidated from a state of igneous fusion, or that the causes which are now acting upon and within the earth's crust were ever different in kind or in order of magnitude from those which are operating at the present day. For further information on the Petrographie,' 3rd ed., 1895). An metamorphie rocks the student is excellent summary of the subject referred topetrographical text-books will be found in Harker's ' Petrology like that of Zirkel (' Lehrbuch der for Students ' (1895), pp. 254-302. CHAPTEB XL THE FORMATION OF MOUNTAIN-CHAINS Various types of Mountain-chains Majority of Mountain-chains belong to Appalachian type Structure of the Scottish Highlands and similar denuded Mountain-chains Mountain forms due proximately to denu- dation Sequence of events in Mountain-making Geo-synclinals Ge-anticlinals Mountain sculpture. Different kinds of Mountain-chains. From what has been said in the preceding chapter, it will be inferred that the pro- duction of regional metamorphism is intimately connected with the folding and faulting of rock-masses, which have played such an important part in the formation of the great mountain- chains oo2 564 MOUNTAIN-CHAINS [CH. XL. of the globe. There are, it is true, mountains and mountain- chains in which metamorphic rock-masses do not appear. Thus we have mountains of volcanic origin of the grandest dimensions ; the tops of the great lava-cones that rise above the surface of the Pacific to form the Sandwich Islands are nearly 30,000 feet above the ocean-floor on which they stand, and in the Andes such volcanic mountains unite to form a considerable chain. In the district of the Jura, and in the western territories of the United States, we find examples of mountain-chains which owe their origin to uniclinal (monoclinal) or anticlinal foldings of the strata, or to the upheaval of rocks capable of resisting denudation along great lines of fault. In all cases it must be observed that the proximate cause of the forms assumed by mountain-masses is the action of subaerial denudation, which is especially powerful in the higher regions of the atmosphere. Appalachian type of Mountain -chains. In spite of the fact that there are certain types of mountains which have originated from causes other than those connected with the lateral movements in the earth's crust that produce the folding, fracturing, and foliation of rock-masses, it is clear that the majority of mountain structures, both in past and present times, must have owed their existence to these latter agencies. This conclusion was first fairly brought home to the minds of geologists by the brothers W. B. and H. D. Rogers, in their investigation of the structures of the Appalachian Mountains. It was shown by thes"e authors that not only are the strata greatly folded and faulted as we approach the central axis of the mountain-chain, but that great reversed faults can be traced for more than eighty miles, along which one series of rocks is seen to be forced over another, sometimes for distances of twenty miles. For such great reversed faults the hade of which may some- times nearly correspond with the horizontal plane the name of * thrust ' has since been suggested. The two authors, to whom we have referred, clearly demonstrated the origin of those types of structure which are so constantly exemplified in mountain- chains ; they showed that ordinary symmetrical anticlinal and synclinal folds like those we have described in preceding pages, yielding to lateral pressure, have their axis plane pushed over farther and farther from the vertical position (see fig. 725 a-), and that the strain on the 'middle-limb ' of the fold eventually leads to elongation and fracture (see fig. 725 b) ; this, if the tangential pressure continues to acfc, results in the formation of a typical 'thrust ' (see fig. 725 c), which is nothing but a greatly exaggerated reversed fault. One of the authors named, the late Professor H. D. Rogers, subsequently visited the Alps and showed that CH. XL.] FLEXUKES AND FKACTUKES 565 the features described in the Appalachians were repeated on even a grander scale in the Alps. The studies of Heim and other Swiss geologists have confirmed in the most striking manner the conclusions of the American geologists, and have served to explain how the features known as ' fan- structure ' Fig. 725. a. Over-folded strata. b. Overfold passing into reversed fault. c. Overthrust (the plane of faulting is often called a ' thrust-plane '). Fig. 726. ' Fan structure,' resulting from lateral pressure. At a, a the opening out of the strata is directed upwards, at b downwards. Strata showing ' double isoclinal overfolding.' (see fig. 726), and multiple folding (see fig. 727), can be ex- plained by the great lateral or tangential pressures to which the rock-masses have been subjected. structure of the Scottish Highlands. Nearly thirty years ago Professor J. Nicol in seeking to explain the relations of 566 MOUNTAIN-CHAINS [CH. XL. the rock-masses of the North- West of Scotland (see figs. 629, 630, p. 436), invoked the aid of similar lateral thrusts to those described by Rogers in the Appalachians and the Alps ; and though his views were opposed so long and so strenuously by Murchison and Geikie, their correctness has been established by the later labours of Messrs. Lapworth, Peach, Home, and other observers (see fig. 631, p. 436). By these researches it has been made manifest that just as we may often learn more about the nature and effects of volcanic action by investigating the greatly denuded basal wrecks of old volcanoes, than by studying volcanoes in actual eruption so the researches carried on in districts like Central Europe, Scandinavia, and the Highlands of Scotland may throw more light on the origin of mountains and the causes of metamorphism than is to be gained by a study of mountain-chains that have undergone less denudation. Effects of denudation in Mountain-chains. In con- nection with this subject it should be pointed out that all the great mountain-chains at present existing on the globe are of very recent age, geologically speaking. All great mountain-chains must be young mountain-chains; for so rapid is the work of subaerial waste in the higher regions of the atmosphere, that the mountain-chains of the earlier geological periods are now reduced to ' basal- wrecks ;' but in these it is often possible to study the results of the action of the forces engaged in mountain - making in a way that is not possible in mountain-chains which have been less completely dissected by denudation. Origin of Mountain-chains. The systematic study of the origin of mountain-chains, begun by the brothers Kogers more than fifty years ago, has been admirably followed up by Dana and other American geologists, and it is largely owing to their efforts that we are now able to trace the succession of operations which, in the end, result in the formation of a great mountain-chain. Geo-synclinals. -Mountain-ranges, as pointed out by Suess, usually originate along lines of weakness in the earth's crust : indeed a mountain-chain may be regarded as a cicatrised wound in the earth's solid crust. The original line of weakness may or may not be indicated by volcanic outbursts taking place along it ; but in all cases the initial stage in the development of a mountain-range con- sists of a slow but prolonged subsidence in that part of the crust which is afterwards to become a mountainous mass. The slowness of this subsidence is indicated by the fact that many thousands of feet of strata, some of them of littoral or shallow-water origin, accumulate during many successive geological periods on the sub- siding ocean floor. In this way is formed what the American geolo- gists call a geo-synclinal, which in the Appalachians consisted of a thickness of 40,000 feet of strata, and in the Alps of 50,000 feet. Ge-anticlinals. The next series of operations contributing to the formation of the mountain-chain is the action of lateral or CH. XL.] MOUNTAIN SCULPTURE 567 tangential thrusts whereby the great thickened masses forming the geosynclinal are folded and fractured, and the sundered masses being forced one over the other in the way we see so strikingly exemplified, not only in mountain-chains like the Alps and Himalayas, but in districts like the Highlands of Scotland and Scandinavia where great mountains have once existed. It is a most striking and significant circumstance that the great movements which gave rise to the folding and elevation of the strata forming the Alps and Himalayas took place at the time when the sands and clays of the northern part of the Isle of Wight and the New Forest were being accumulated ! Strata of the same age as the London Clay, the Barton Clay, and the Bracklesham beds are found in the Alps and Himalayas at heights of 10,000 and 16,000 feet respectively. Indeed it is probable that the monoclinal fold which affects the strata of the Isle of Wight with the anticlinal of the Weald and the synclinal of the London Basin are but por- tions of that series of earth-movements which, in Oligocene times and subsequently, affected the whole of the rocks of Southern Europe and Asia, and gave rise to the elevation of the Alps and Himalayas. Denudation. The third great series of operations concerned in the formation of mountain-ranges consists in the sculpturing action of denudation which has gone on, to a great extent, side by side with the work of folding, crumpling, fracturing and elevation. W T hile it is true that all the actual forms of the rock-masses consti- tuting a mountain-chain are due to the sculpturing action of denuding forces, it must not be forgotten how much that action has been con- trolled and modified by the great internal movements and changes within the earth's crust. Such, very briefly sketched, seems to have been the general succession of events in the formation of typical mountain -chains though of course local conditions have often modified the sequence in particular cases. That the metamorphism of rock-masses has been effected while they have been buried at great depths, so as to have been exposed to a moderately high temperature and at the same time subjected to intense dynamical action, there is every ground for believing. But in the nature of the case, the actual processes by which particular rock-masses of this class have been formed must always be difficult to determine. It must not be supposed from what has been said that the folding and fracturing of rock-masses is always attended with metamorphism and the production of foliation. While it is certainly true that all highly altered rocks exhibit evidence of having been subjected to great movements and tangential strains, the converse is by no means true. Strata of all ages, from the Carboniferous upwards, are found in the Alps caught up in complicated folds, and themselves bent and puckered in the most remarkable manner, yet retaining their mineralogical characters, and exhibiting their fossils almost unaltered. Doubtless the depth at which a roc-k-mass lies, and the consequent temperature which it attains while it is being subjected to folding and fracture, and other surrounding circumstances, may have much to do with determining whether the process of recrystallisation shall be set up in the mass, and foliation thus produced, or not. Mr. Mallet has suggested that the mechanical work of rock-crushing may be actually converted into heat and chemical action, and, if this be the case, then the time in which the operation is effected would 568 NATURE OF ORE DEPOSITS [CH. XL. determine whether the results of the action would accumulate to produce great results or be gradually dissipated. An excellent account of modern well to consult the Memoir by Mr views concerning the origin of Bailey Willis ' On the Mechanics of mountain-chains of different types Mountain Structure as displayed by will be found in Dana's ' Manual of the Appalachian ranges,' published Geology ' (5th edition, 1895), pp. 345- in the 13th Report of the U. S. Geo- 396. The student would also do logical Survey. CHAPTER XLI ORE-DEPOSITS AND THEIR ORIGIN Ore-deposits usually formed by Hypogene action Classification of Ore- deposits Origination of Metalliferous Veins in fissures Different ages of the formation and infilling of Veins Proofs of successive opening and refilling of Veins Comb-structure in Veins Irregularities in width of Veins Chemical Deposition in Veins Supposed relative ages of different metals. Hypogrene origin of most Ore-deposits. The various deposits in which the ores of the metals employed in the arts are found are of great practical value to mankind, and the study of their mode of occurrence and origin is of the highest theo- retical interest to geologists. Some masses of the ores of iron and manganese, like the lake-ores of Sweden (see p. 48), are evidently of aqueous origin, and a few deposits of volatile compounds, like the sulphides of arsenic and mercury, are seen to be deposited around volcanic vents. But the great majority of the ore-deposits, which are of such importance to mankind, are evidently of deep-seated or hypogene origin, and ^re closely connected with plutonic and metamorphic rock-masses. Some ores, like the ironstone of Cleveland and the Kupferschiefer of Thuringia, have been produced during the consolidation and alteration of stratified deposits. The large? part of the precious and other metals used by man is obtained, however, from veins and analogous deposits, the nature and origin of which we must proceed to consider. Different kinds of mineral veins. The mineral veins with which we are most familiarly acquainted are those of quartz and calcite, which are often observed to form lenticular masses of limited extent traversing both hypogene strata and fossiliferous rocks. Such veins appear to have once been chinks or small cavities, caused by the contraction or movement of the rock-masses which they traverse. Siliceous, calcareous, and CH. XLT.] FISSUEE VEINS 569 occasionally metallic matters sometimes find their way into such empty spaces by infiltration from the surrounding rocks. Carried by water or steam, metallic compounds may have per- meated the mass until they reached those receptacles formed by shrinkage, and thus gave rise to that irregular assemblage of veins called by the Germans a ' Stockwerk,' in allusion to the different floors on which the mining operations are in such cases carried on. The late J. A. Phillips showed that in Nevada, hot springs rise to the surface and deposit silica, with metallic ores, includ- ing gold and the compounds of mercury which incrust the walls of the fissures. The more ordinary or regular veins, usually highly inclined or vertical, have evidently been fissures produced by similar mechanical actions. They traverse all kinds of rocks, both hypogene and fossiliferous, and extend downwards to indefinite or unknown depths. We may assume that they correspond with such rents as we see caused in rocks by movement and faulting. Metalliferous veins are occasionally a few inches wide, but more commonly 3 or 4 feet, and some are as much as 150 feet in width. They hold their course continuously in a certain prevailing direction for a short distance or for miles or leagues, passing through rocks varying in mineral composition. Metalliferous veins were fissures. There are proofs in almost every mining district of a succession of faults, by which the opposite walls of rents, now the receptacles of metallic substances, have suffered displacement. Thus, for example, suppose a a, fig. 728, p. 570, to be a tin-lode in Cornwall, the term lode being applied to veins containing metallic ores. This lode, running east and west, is a yard wide, and is shifted by a copper lode (6 b), of similar width. The first fissure (a a) has been filled with various materials partly of chemical origin, such as quartz, fluor-spar, tinstone, copper-glance, arsenical pyrites, native bismuth, and nickeliferous pyrites, and partly of mechanical origin, comprising clay and angular fragments or detritus of the intersected rocks. The succes- sive deposits of spars and ores are, in some places, parallel to the vertical sides or walls of the vein, being divided from each other by alternating layers of clay, or other earthy matter. Occasionally, however, the metallic ores are disseminated in detached masses among the sparry minerals or vein-stones. It is clear that, after the gradual introduction of the tinstone and other substances, the second rent (6 b) was produced by another fracture accompanied by a displacement of the rocks along the plane of b b. This new opening was then filled with minerals, some of them resembling those in a a, as fluor-spar and quartz ; others different, the copper ore being plentiful, and the tin ore wanting or very scarce. We must next suppose a third movement to occur, 570 AGE OF VEINS [CH. XLI. breaking asunder all the rocks along the line cc, fig. 729 ; the fissure, in this instance, being only six inches wide, and simply filled with clay, derived, probably, from the friction of the walls of the rent, or Fig. 728. Vertical sections of the mine of Huel Peever. Redrtith, Cornwall. partly, perhaps, washed in from above. This new movement has displaced the rock in such a manner as to interrupt the continuity of the copper vein (6 6), and, at the same time, to shift or heave CH. XLI.] SYSTEMS OF VEINS 571 laterally in the same direction a portion of the tin vein which had riot previously been broken. Again, in fig. 730, we see evidence of a fourth fissure (d d), also filled with clay, which has cut through the tin vein (a a), and has lifted it slightly upwards towards the south. The various changes here represented are not ideal, but are exhibited in a section obtained in working an old Cornish mine, long since abandoned, in the parish of Kedruth, called Huel Peever, and described both by Williams and Carne. The principal movement here referred to, or that of c c, fig. 729, extends through a space of no less than 84 feet ; but in this, as in the case of the other three, it will be seen that the outline of the country above, d, c, b, a, &c., or the geographical fea- tures of Cornwall, are not affected by any of the dislocations, a powerful denuding force having clearly been exerted subsequently to all the faults. It is commonly said in Cornwall that there are eight distinct systems of veins, which can in like manner be referred to as many successive movements or fractures ; and the German miners of the Hartz Mountains speak also of eight systems of veins, re- ferable to as many periods. Besides the proofs of mechanical action already explained, the opposite walls of veins are often beautifully polished, as if glazed, and are not unfrequently striated or scored with parallel furrows and ridges (slickensides), such as would be produced by the continued rubbing together of surfaces of unequal hardness. In some of the veins in the Mountain limestone of Derbyshire containing galena, the vein-stuff, which is nearly compact, is occa- sionally traversed by what may be called a vertical crack passing down the middle of the vein. The two faces in contact are slicken- sides, well polished and fluted, and sometimes covered by a thin coating of lead-ore. When one side of the vein-stuff is removed, the other side cracks, especially if small holes be made in it, and fragments fly off with loud explosions (owing to the relief from strain), and continue to do so for some days. The miner, availing himself of this circumstance, makes with his pick small holes about six inches apart and 4 inches deep, and on his return in a few hours finds every part ready broken to his hand. That a great many veins communicated originally with the surface of the country above, or with the bed of the sea, is proved by the occurrence of well-rounded pebbles in them, agreeing with those in superficial alluvia, as in Auvergne and Saxony. Marine fossil shells, also, have been found at great depths, having possibly been engulfed during submarine earthquakes. Thus, the late Charles Moore described lead-veins traversing the Carboniferous limestone of the Mendips in Somerset, which at the time they were rilled must have been in communication with the Liassic sea, for he found Lias fossils in them. In Cornwall, Carne described true pebbles of quartz and slate as occurring in a tin lode of the Eelistran Mine, at the depth of 600 feet below the surface. They were cemented by tin- stone and copper pyrites, and were traced over a space more than twelve feet long and as many wide. When different sets or systems of veins occur in the same country, those which are supposed to be of contemporaneous origin, and which are filled with the same kind of ores, often maintain a general parallelism of direction. Thus, for example, both the tin and copper veins in Cornwall run nearly 572 COMB-STRUCTURE [CH. XLI. east and west, while the lead-veins run north and south ; but there is no general law of direction common to different mining districts. The parallelism of the veins is another reason for regarding them as ordinary fissures, for we observe that faults and volcanic dykes, admitted by all to be masses of melted matter which have filled rents, are often parallel. Fracture, Reopening-, and Successive Formation of Veins. Assuming, then, that veins are simply fissures in which chemical and mechanical deposits have accumulated, we may next consider the proofs of their having been filled gradually and often during successive enlargements. Werner observed, in a vein near Gersdorff, in Saxony, no less than thirteen bands of different minerals, arranged with the utmost regularity on each side of the central layer. This layer was formed of two plates of calcareous spar, which had evidently lined the opposite walls of a vertical cavity. The thirteen beds followed each other in corresponding order, Fig 731. consisting of fluor-spar, heavy spar, galena, &c. In these cases the central mass has been last formed, and the two plates which coat the walls of the rent on each side are the oldest of all. If they consist of crystalline precipi- tates, they may be explained by supposing the fissure to have remained unaltered in its dimensions, while a series Copper lode, near Eedruth, enlarged at of changes occurred in the successive periods. nature of the solutions which rose up from below ; but such a mode of deposition, in the case of many successive and parallel layers, appears to be exceptional. If a veinstone consists of crystalline matter, the points of the crystals are always turned inwards, or towards the centre of the vein ; in other words, they point in the direction where there was space for the development of the crystals. Thus each new layer receives the impression of the crystals of the preceding layer, and imprints its crystals on the one which follows, until at length the whole of the vein is filled ; the two layers which meet dovetail the points of their crystals the one into the other. But in Cornwall, some lodes occur where the vertical plates, combs, as they are there called, exhibit crystals so dovetailed as to prove that the same fissure has been often enlarged. De la Beche described the following curious and instructive example (fig. 731), from a copper-mine in granite, near Redruth. Each of the plates or combs (a, 5, c, d, e,f) is doubled, having the points of their crystals turned inwards along the axis of the comb. The sides or walls (2, 3, 4, 5, and 6) are parted by a thin covering of ochreous clay, so that each comb is readily separable from another by a moderate blow of the hammer. The breadth of each represents the whole width of the fissure at six successive periods, and the outer walls of the vein, where the first narrow rent was formed, consisted of the granitic surfaces 1 and 7. CH. XLI.] INFILLING OF VEINS 57B A somewhat analogous interpretation is applicable to many other cases, where clay, sand, or angular detritus alternates with ores and veinstones. Thus, we may imagine the sides of a fissure to be incrusted with siliceous matter after the manner observed by Von Buch in Lancerote. He noticed that the walls of a volcanic crater formed in 1731 were traversed by an open rent in which hot vapours had deposited hydrous silica, the incrustation nearly extending to the middle. Such a vein may subsequently be filled with clay or sand, and afterwards reopened, the new rent dividing the argillaceous deposit, and allowing a quantity of rubbish to fall down. Various ores and spars may then be precipitated from aqueous solutions percolating among the interstices of this hetero- geneous mass. That such changes have taken place repeatedly is demonstrated by the occurrence of occasional cross-veins, implying the oblique fracture of previously formed chemical and mechanical deposits. Thus, for example, M. Fournet, in his description of some mines in Auvergne, worked under his superintendence, observes that the granite of that country was first penetrated by veins of massive granite and then dislocated, so that open rents crossed both the granite and the granitic veins. Into such openings, quartz, accompanied by iron pyrites and arsenical pyrites, was introduced. Another movement then burst open the rocks along the old line of fracture, and the first set of deposits was cracked and often shattered, so that the new rent was filled not only with angular fragments of the adjoining rocks, but with pieces of the older veinstones. Polished and striated surfaces on the sides or in the contents of the vein also attest the reality of these movements. A new period of repose then ensued, during which various sulphides were introduced, together with chalcedonic silica of the variety known as hornstone, by which angular fragments of the older quartz before mentioned were cemented into a breccia. This period was followed by other dila- tations of the same veins, and the introduction of new sets of mineral deposits, as well as of pebbles of the basaltic lavas of Auvergne, derived from superficial alluvia, probably of Miocene or even older Pliocene date. Such repeated enlargement and reopening of veins might have been anticipated, if we adopt the theory of fissures, and reflect how few of them have ever been sealed up entirely, and that a country with fissures only partially filled must naturally offer much feebler resistance along the old lines of fracture than anywhere else. Cause of alternate contraction and swelling in veins. A large proportion of metalliferous veins have their opposite walls nearly parallel, and sometimes over a wide extent of country. But many lodes in Cornwall and elsewhere are extremely variable in size, being 1 or 2 inches in one part, and then 8 or 10 feet in another, at the distance of a few fathoms, and then again narrowing as before. Such alternate swelling and contraction are so often characteristic as 10 require explanation. The walls of fissures in general, as De la Beche pointed out, are rarely perfect planes throughout their entire course, nor could we well expect them to be so, since they commonly pass through rocks of unequal hardness and different mineral com- position. If, therefore, the opposite sides of such irregular fissures slide upon each other, that is to say, if there be a fault, as in the 574 VARIATION IN VEINS [CH. XLI. case of so many mineral veins, the parallelism of the opposite walls is at once entirely destroyed, as will be readily seen by studying the annexed diagrams. Let a, 6, fig. 732, be a line of fracture traversing a rock, and let a, 6, fig. 733, represent the same line. Now, if we cut in two a piece of paper representing this line, and then move the lower portion of this cut paper sideways from a to a', taking care that the Fig. 732. Fig. 733. ^^-^ 3 Fig. 734. two pieces of paper still touch each other at the points 1, 2, 3, 4, 5, we obtain an irregular aperture at c, and isolated cavities d d d, and when we compare such figures with Nature we find that, with certain modifications, they represent the interior of faults and mineral veins. If we move the lower part of the paper towards the left about the same distance that it was previously moved to the right, we obtain considerable variation in the cavity so produced, two long irregular open spaces, / /, fig. 734, being then formed. This will serve to show to what slight circumstances considerable varia- Fig. 735. tions in the character of the openings between un- evenly fractured surfaces may be due, such surfaces being moved upon each other, so as to have numerous points of contact. Most lodes are perpendicular to the horizon, or nearly so ; but some of them have a considerable inclination or ' hade,' as it is termed, the angles of dip being very various. The course of a vein is fre- quently very straight ; but, if tortuous, it is found to be choked up with clay, stones, and pebbles, at points where it departs most widely from vertically. Hence at places, such as a, fig. 735, the miner complains that the ores are ' nipped,' or greatly reduced in quantity, the space for their free deposition having been inter- fered with in consequence of the preoccupancy of the lode by earthy materials. When lodes are many fathoms wide, they are usually filled for the most part with earthy matter and fragments of rock, through which the ores are disseminated. The metallic substances frequently coat or encircle detached pieces of rock, which our miners call ' horses ' or ' riders.' That we should find some mineral veins which split into branches is also natural, for we observe the same in regard to open fissures. Chemical deposits in veins. If we now turn from the me- chanical to the chemical agencies which have been instrumental in the production of mineral veins, it may be remarked that those CH. XLI.] OKIGIN OF VEINS 575 parts of fissures which were choked up with the ruins of fractured rocks must always have been filled with water ; and almost every vein has probably been the channel by which hot springs, so common in countries of volcanoes and earthquakes, have made their way to the surface. For we know that the rents in which ores abound extend downwards to vast depths, where the temperature of the interior of the earth is more elevated. We also know that mineral veins are most metalliferous near the contact of plutonic and strati- fied formations, especially where the former send veins into the latter, a circumstance which indicates an original proximity of veins at their inferior extremity to igneous and heated rocks. It is, moreover, acknowledged that even those mineral and thermal springs, which, in the present state of the globe, are far from volcanoes, are nevertheless observed to burst out along great lines of upheaval and dislocation of rocks. It is also ascertained that, among the substances with which hot springs are impregnated, such as are volatile also occur in the gaseous emanations of volcanoes. The whole of these are also among the constituents of the minerals most usually found in veins, such as quartz, calcite, fluor-spar, the metallic sulphides, heavy-spar, brown-spar, and the oxides of iron. We may add that, if veins have been filled with vitreous materials from masses of melted matter, slowly cooling in the subterranean regions, the contraction of such masses as they pass from a glassy to a crystalline state would, according to experiments of Deville on granite (a rock which may be taken as a type), produce a re- duction in volume amounting to 10 per cent. The slow crystallisa- tion, therefore, of such plutonic rocks supplies us with a force not only capable of rending open the incumbent rocks by causing a failure of support, but also of giving rise to fissures whenever one portion of the earth's crust subsides slowly while another contiguous to it happens to rest on a different foundation, so as to remain un- moved. Although we are led to infer, from the foregoing reasoning, that there has often been an intimate connection between metalliferous veins and hot springs holding mineral matter in solution, yet we must not on that account expect that the contents of hot springs and mineral veins would be identical. On the contrary, M. E. de Beaumont has judiciously observed that we ought to find in veins those substances which, being least soluble, are not discharged by hot springs or that class of simple and compound bodies which the thermal waters ascending from below would first precipitate on the walls of a fissure, as soon as their temperature began slightly to diminish. The higher they mount towards the surface, the more will they cool till they acquire the average temperature of springs, being in that case chiefly charged with the most soluble substances, such as salts of the alkalies, soda and potash. These are seldom met with in veins, although they enter so largely into the composition of granitic rocks. To a certain extent, therefore, the arrangement and distribution of metallic matter in veins may be referred to ordinary chemical action, or to those variations in temperature which waters holding the ores in solution must undergo as they rise upwards from great depths in the earth. But there are other phenomena which do not admit of the same simple explanation. Thus, for example, in 576 CLASSIFICATION OF OKE-DEPOSITS [CH. XLI. Derbyshire, veins containing ores of lead, zinc, and copper, but chiefly lead, traverse alternate beds of limestone and basalt. The ore is plentiful where the walls of the rent consist of limestone, but is reduced to a mere string when they are formed of basalt, or ' toad- stone,' as it is called provincially. Not that the original fissure is narrower where the basalt occurs, but because more of the space is there filled with veinstones, and the waters at such points have not parted so freely with their metallic contents. Lodes in Cornwall are very much influenced in their metallic riches by the nature of the rock which they traverse, and they often change in this respect very suddenly, in passing from one rock to another. Thus many lodes which yield abundance of ore in granite are unproductive in clay-slate, or killas, and vice versa. Theories as to the Origin of Ore-deposits. In recent years the studies carried on in the Western States of North America, in South America, South Africa, and Australia, have shown that ore- deposits are much more varied in character than was supposed by the students of mineral veins in Saxony and Cornwall. It has been found necessary, in order to account for some of these deposits, to modify and extend the theories which were thought sufficient to explain the origin of ordinary veins. Professor Clement Le Neve Foster classifies all ore-deposits under the following heads : I. Tabular or sheet-like, including j Bed^or stratified deposits. A. Necks or pipes (like the diamond rocks of South Africa). II. Masses including . -\ B. Stockworks, or ' Network- deposits.' C. Various irregular masses of doubtful origin. The origin of veins and other ore-deposits has, according to the same authority, been variously referred to the following causes : 1. Fracture and motion with mechanical filling. 2. Fracture and injection of molten matter. | A. from above. 3. Fracture and deposition from solutions, \ B. from below. I C. from the sides. 4. Fracture and sublimation, or deposition from gases. Very much still remains to be done in the study of ore-deposits, before we can hope to supply reasonable explanations of many of the remarkable occurrences of metallic ores within the earth's crust. For further information on the Pacific Slope. A very valuable subject of Ore-deposits the student work on ' The Genesis of Ore- is recommended to consult J. A. deposits,' by Prof. F. Posepny, Phillips's ' Ore-deposits,' 1884, and with discussions by many eminent the various monographs of the geologists and mining engineers, U.S. Geological Survey on the has been published by the Comstock Lode, the Leadville and American Institute of Mining the Eureka deposits, and that on Engineers (1902). the Quicksilver deposits of the xi.ii.] AGE OF METAMOEPHIC BOOKS 577 CHAPTER XLII ON THE DIFFERENT AGES OF METAMORPHIC ROCKS, MOUNTAIN- CHAINS AND ORE-DEPOSITS How the age of Meframorphic Eocks is determined Period of original formation Period of Metamorphism Disturbed condition of Meta- morphic Eocks Age of Eocks formed by Contact-metamorphism Similarity of Eocks formed by Contact-metamorphism to those pro- duced by Eegianal metamorphism Difficulty of determining age of Eocks formed by latter process Metamorphic Eocks of the Alps Supposed Tertiary age Metamorphic Eocks of Mesozoic Age Meta- morphic Eocks of Newer Palaeozoic Age Metamorphic Eocks of Older Palaeozoic Age Metamorphic Eocks of Pre- Cambrian Age Uniformity of characters in Metamorphic Eocks of all ages Supposed parallelism of Mountain-chains formed during different periods Mountain-chains of Tertiary, Mesozoic, Palaeozoic, and Archaean Ages Ages of Ore- deposits Supposed relative ages of different metals Origin and age of Gold-deposits. Tests of the age of Metamorphic Rocks. We have seen in the earlier chapters of this work that, by means of stratigraphi- cal and palaeontological evidence, a chronological sequence can be traced among the various deposits of aqueous origin forming the earth's crust. In the case of the rocks of volcanic origin, the relations which they exhibit to stratified masses, and the fossils which they occasionally contain, enable us though often with some doubt and hesitation to refer the various lavas and tuff's to portions of the same sequence. But, when we pass from the epigene to the hypogene rocks, the task of making out a chronological succession among the intrusive or plutonic masses has been shown to be beset with far more serious difficulties, and the conclusions arrived at consequently liable to much greater uncertainty. It is in the case of the metamorphic rocks, however, that the geologist experiences the greatest amount of difficulty in deter- mining their relative ages. Not only do we find, as in the case of the plutonic rocks, that the younger hypogene rocks are but rarely exposed by denudation at the surface, but the fact that ex- treme regional metamorphism is in almost all cases connected with great terrestrial movements, prepares us for encountering repeated foldings, complicated inversions, and violent displace- ments of rock-masses ; and under these circumstances all traces of fossils having necessarily been destroyed in the re- crystallised materials geologists often find it extremely difficult, if not quite impossible, to arrive atdefinite conclusions concerning their original sequence. r P 5?8 METAMOKPHIC llOCtfS [CH. XL!!. Disturbed condition of Metamorphic Rocks. Accord- ing to the theory of metamorphisra adopted in this work, the metamorphic and foliated rocks have been deposited during one geological period, and have become crystalline at another and later period. We can rarely hope to define with exactness the date of each of these periods, the fossils having been destroyed by the process of crystallisation, while mineral characters are identical in rocks of very different ages. When we come to study the metamorphic rocks in detail, moreover, we find abundant evidence that, before or during metamorphism, rocks of the most varied geological age may have become infolded with or faulted against one another ; and further, that the work of metamorphism, resulting in recrystalli- sation and foliation, has not been accomplished in a single period, but has probably been repeated again and again at different geological epochs. Hence we must not be surprised to find that, with respect to many of these greatly disturbed and much altered rock-masses, the task of unravelling their compli- cated history has proved an insuperable one, and geologists have been unable to arrive at anything like agreement concerning all the difficult problems presented to them by the metamorphic rocks. Relative ages of Rocks formed by Contact-metamor- phism. The simplest cases are undoubtedly those presented to us in the study of contact-metamorphism. Within a distance of two miles or less, we may often find a rock crowded with fossils undergoing progressive changes, as we approach the igneous intrusion, until at last as new crystals of minerals multi- ply in the mass, and all traces of organisms are finally oblite- ratedthe rock may become highly crystalline, and even perfectly foliated. In many cases the passage from the fossiliferous to the crystalline rock can be followed in such obvious gradations, that no doubt about the geological age of the material out of which the slate, schist, or gneissose rock has been formed can possibly exist. If we are able also to determine the period of the intrusion of the igneous mass around which this contact- metamorphism is developed, we have then before us all the data necessary for settling the main facts concerning the chronology of a metamorphic rock. That this metamorphic process is going on at the present day, around great igneous intrusions, we have conclusive evidence in the fragments thrown from the vents of Vesuvius and other volcanoes. These ' ejected blocks ' have evidently been torn from the rocks through which the igneous materials are forcing their way to the surface ; they sometimes contain fossils which can be distinctly recognised, but at other times exhibit the signs CH. XLII.] OF DIFFEKENT GEOLOGICAL AGE 579 of more and more complete recrystallisation, not unfrequently accompanied with the development of a distinctly foliated structure. Stratified and fossiliferous rocks belonging to every geo- logical period, from the Tertiary downwards, are found altered in this way by igneous intrusions of every date, so that in the case of the rocks produced by contact-metamorphism the continuity and uniformity of metamorphic processes, from the earliest periods of the geological history down to the present day do not admit of doubt. Analogy between Rocks formed by Contact- and Regional-metamorphism respectively. Kecent petrogra- phical researches have undoubtedly tended towards the conclu- sion that there is a much closer analogy between the rocks produced by contact-metamorphism and those which are the result of regional metamorphism than was at one time supposed. In the zones surrounding great granitic bosses like those of Galloway, so well described by Miss I. Gardiner, we find mica- schists rich in garnets and other accessory minerals, which have undoubtedly been formed by the alteration of Ordovician greywackes and flagstones ; yet these nevertheless are not distinguishable by any important characters from the mica- schists, intercalated with gneisses, and similar rocks forming portions of districts which have been subjected to regional metamorphism. On the other hand, minerals at one time supposed to be especially characteristic of the rocks formed by contact metamorphism, such as andalusite, sillimanite, kyanite, staurolite, &c., have now been found to be much more widely distributed and to form important constituents not only of the schists but also of the granitic gneisses of districts where the rocks have resulted from regional metamorphism. Relative ages of Rocks produced by Regional meta- morphism. In the face of the almost total absence of palseonto- logical evidence, and the obliteration of all structural characters by recrystallisation and foliation, the task of defining the age of the great masses of hypogene and highly altered rocks is one surrounded by great and indeed almost insuperable difficulties. Rocks affected by slaty cleavage are certainly known belonging to every geological epoch ; the slates of Glaris contain fish of Eocene age ; and other cleaved rocks might be adduced that can be referred to almost every division of the Mesozoic and Palaeozoic epochs. In many cases, too, these ' clay-slates ' are found passing insensibly into true ' phyllites,' in which a considerable amount of chemical change has taken place in addition to the mechanical effects of pressure. p p 2 580 PEKIODS OF REGIONAL ACTION [CH. XLII. In Norway, as Professor Eeusch has so well shown, there are undoubted phyllites, approaching to true schists in structure, which contain trilobites, brachiopoda, and corals of Silurian age. Scarcely less altered rocks with Devonian fossils occur in Devonshire. It would be unreasonable to expect that rocks which can only have undergone the extremes of metamorphic change by being buried to most profound depths in the earth's crust should be frequently exhibited at the earth's surface by denudation, unless they were of great geological antiquity. When so exposed, it is often not possible to assert with confidence that they are really integral parts of the great masses among which they now lie, and have not been caught up and rolled out among rocks of far greater antiquity. Some geologists of authority, indeed, have been led to affirm that the great and wide- spread masses of gneiss and schist covering vast areas of the earth's surface have so little in common with the smaller masses that can be proved to have been formed by contact metamorphism or by the meta- morphism of sedimentary and igneous materials, that we must, in all cases, infer for these very highly crystalline masses a pre-Palaeozoic age, and probably an origin different from that of any rocks formed since the commencement of the geological record. But it must be remembered that, just as it has been found impossible by petrologists to point out any fundamental distinc- tions between the rocks formed by contact-metamorphism and those resulting from more wide -spread action at greater depths upon ancient sediments, lavas and tuffs, so the difference between the characters of the granitic gneisses and true schists on the one hand, and the phyllites and similar rocks on the other, is, to say the least, often shadowy and indefinite. It is no+ unreason- able to suppose that a more deeply seated action, a higher temperature, or the more intense or more prolonged operation of dynamic agencies, may lead to changes the result of which is seen in the most perfectly recrystallised metamorphic rocks ; for these changes differ in degree rather than in kind from those the effects of which we can so clearly follow. On the other hand, it would be rash to affirm that among the widely spread and highly crystalline rocks that underlie all the rocks, the sedimentary or volcanic origin of which can be clearly demonstrated, there may not be found some of which the origin is different from that of the undoubted metamorphic rocks some relics of a primaeval condition of the earth, when rock- masses may have been formed under conditions essentially dif- ferent from those which now prevail in the earth's crust. As, CH. XLII.] TEKTIAEY METAMOKPHIC BOOKS 581 however, the metamorphic theory admitting such extension of it as is reasonable with increasing temperature and pressure seems fully adequate to the explanation of the origin of these highly crystalline rocks, the onus of proof that other conditions prevailed during their formation rests with those who make the assertion. Examples of Rocks formed by Regional metamorphism which are of different Geological ages. Owing to the great foldings and faulting to which most metamorphic rocks have been subjected, great differences of opinion have arisen concerning the relative ages of many metamorphic masses. The following state- ments concerning certain cases, therefore, must be taken as indicating probabilities rather than actually demonstrated conclusions. XVXetamorphic Rocks of the Alps possibly of Tertiary ag-e.The existence of rocks in mountain masses of Paleozoic, Secondary, and even of Eocene age, metamorphosed into crystalline schists, has been asserted over and over again in the Alps. The late Professor Favre, of Geneva, to whom we owe so much correct know- ledge regarding the Alps, traced the origin of Mont Blanc from a time when palaeozoic rocks of Carboniferous age, with their beds of coal and plant-remains, were deposited upon a partially submerged region of gneiss and crystalline schists. Many of the strata contain the denuded remains of these schists. Some disturbance occurred, and the secondary rocks were laid down during subsidence, and finally the Nummulitic series of overlying sandstones. Then came the great movement of mountain-making, and the strata and schists were curved, folded, faulted, inverted, and thus schists were forced above the reversed fossiliferous series. The products of the wear and tear of the mountain-mass collected in the form of strata of gravels and clays on its flanks, and at last the final tangential movements came, which added to the complication by inverting the last-made strata on the flanks of the Alps, so that they appear to dip under- neath the Nummulitic group. A very remarkable paper on the geology of the Alps, by Murchison, in 1848, refers to the Pass of Martinsloch, in Glarus, 8,000 feet above sea-level. In this locality, Nummulitic beds dipping S.S.E., at a high angle, are regularly overlaid by the succeeding Flysch sand- stone, resting unconformably, and in a nearly horizontal attitude, upon the edges of which are 150 feet of hard Jurassic limestone, overlain in its turn by talcose and micaceous schists, which were regarded by Escher as similar to those which underlie these lime- stones in the valley below. The mass of Flysch appears nearly to dip beneath these limestones, which in their turn are overlain by Neocomian and Cretaceous strata. The superposition of the schists may not have been original., but may have been brought about by frac- ture and displacement along an anticlinal. Similar great inversions are seen in the Valley of Chamounix, where secondary limestones dip at a high angle towards Mont Blanc, and plunge beneath its crystal- line schists. In one of the sections described by Studer in the highest of the Bernese Alps, namely, in the Boththal, a valley bordering the line of perpetual snow on the northern side of the Jungfrau, there occurs a 582 MESOZOIC METAM ORPHIC ROCKS [CH. XLTI. mass of gneiss 1,000 feet thick and 15,000 feet long, which is seen not only resting upon, but also again covered by strata containing oolitic fossils. These anomalous appearances may partly be explained by supposing great solid wedges of intrusive gneiss to have been forced in laterally between strata to which they are found to be in many sections unconformable (see fig. 727, p. 565). The superposition also of the gneiss to the oolite may be due to a reversal of the original position of the beds in a region where the contortions have been on so stupendous a scale. Most living Swiss geologists, like Heim, Baltzar, and Renevier, believe that some of the metamorphic rocks of the Alpine chain represent Newer Palaeozoic, Mesozoic, and even Eocene rocks, which have undergone great metamorphism. This conclusion is, however, disputed by Professor Bonney and some other geologists, who explain the position of younger rocks among the schistose masses of the Alpine chain by asserting that they are due in all cases to folding and faulting of the rocks. Ittetamorphic Sedimentary Rocks of IVIesozoic Age. ~ Neumayr and other geologists have described masses of chlorite- schist, mica-schist, and gneiss in Greece, as alternating with beds of more or less altered limestone which, in some cases, contain obscure but undoubted traces of fossils. The rocks from which these meta- morphic fossils were formed are supposed to be the Hippurite lime- stones and other Cretaceous strata, and the metamorphism must have occurred in post-Cretaceous, if not Tertiary times. In the coast ranges of the Pacific Slope of California, Dr. Becker and the officers of the United States Geological Survey have de- scribed granulitic rocks, glaucophane schists, phthanites (silicified limestones) and serpentinous rocks as being formed from stratified masses of undoubted Neocomian age. Other strata of Jurassic age in California are, according to Whitney, found altered into clay-slate, talcose slate, and serpentinous rocks. Northern Apennines, Carrara. The celebrated marble of Carrara, used in sculpture, was once regarded as a type of primitive limestone. The absence of fossils, its mineral texture and composi- tion, and its passage downwards into talc-schist and garnetiferous mica-schist, gave it the appearance of a rock of great age, especially as underlying gneisses, penetrated by granite veins, are believed to graduate into the schists. The variety of opinions regarding the age of this limestone which have been published by competent authorities should warn the student against geological dogmatism in this difficult question of the age of metamorphic rocks. Most geologists believe that the marble is an altered Triassic or Jurassic limestone, and that the underlying schists are altered plutonic rocks of secondary age. Examples of metamorphic Rocks of Newer Palaeozoic ag-e. Besides the cases of altered Carboniferous rocks converted into schists which have been asserted by many geologists to occur in the Alps, we have in the district of the Taunus in Central Germany, as pointed out by Lossen, Lower Devonian strata altered into various clay slates and spotted slates, with quartzites, phyllites, mica-schist, and even sericite-gneiss. Although some authors assert these altered strata to be of Older Palaeozoic age, yet no good grounds have been adduced for assigning them to an earlier geological period than that to which they were referred by Lossen, CH. xr.ii.] PALEOZOIC METAMORPHIC ROCKS 583 In the Ardennes, Dumont and Renard have shown that a series of quartzites and phyllites graduating into highly crystalline rocks containing hornblende, mica, garnet, sphene, graphite, and many other minerals, are really the altered representatives of Devonian strata. In the metamorphosed rocks Sandberger has been able to detect two undoubted forms of Devonian Brachiopoda. IVIetamorphic Rocks of Older Palaeozoic age. The most interesting examples to the British geologist of rocks of this age are those found in the Highlands of Scotland. We have already seen that, pushed over the Lewisian gneisses and the Torridonian sandstones, and the Cambrian limestones, quartzites, and shales by great reversed faults (thrusts), we find the great masses of gneiss (Caledonian of Callaway, and Dalradian of Geikie), which cover so large a portion of Scotland, north of the valleys of the Forth and Clyde (see fig. 631, p. 436). These gneisses and schists have a very different aspect from the Lewisian or fundamental gneiss, and, as long ago pointed out by Nicol, graduate when traced southwards towards the central valley of Scotland into slaty rocks, in which, however, fossils have not yet been found. The officers of the Geological Survey are led to conclude that these ' younger schists and gneisses ' of the Highlands are really the altered representatives of Torridonian, Cambrian, and Ordovician strata and of igneous intrusions in them, and that the metamorphism of these rocks was effected in Silurian times. In Scandinavia, Cambrian, Ordovician, and even Silurian strata are similarly found converted into quartzites, phyllites, and even true schists and gneisses, the clastic origin of the rocks being be. trayed, however, by the presence of pebbles, and even in some cases by traces of fossils. In the Green Mountains of New England, Dana and others have described schists and limestones, which have been formed by the metamorphism of the Ordovician strata of the district. IVIetamorphic Rocks of pre-Cambrian age. Many of the schists and gneisses of the globe are undoubtedly older than the oldest-known fossiliferous rocks, for these latter not only overlie them, but contain fragments derived from them. As we have already pointed out, these undoubtedly pre-Cambrian gneisses and schists have not been shown to exhibit any characters by which they can be clearly distinguished from rocks of the same class be- longing to later periods. Order of Succession in IVIetamorphic Rocks. It has been remarked that, as the hypogene rocks, both stratified and unstratified, crystallised originally at a certain depth beneath the surface, they must always in order to be upraised and exposed at the surface by denudation be of considerable antiquity, relatively to a large portion of the fossiliferous and volcanic rocks. Whether they were forming during all the geological periods is a debated question ; but before any of them can become visible, they must be raised above the level of the sea, and the rocks which previously concealed them must have been removed by denudation. There is no universal and in- variable order of superposition among metamorphic rocks, although a particular arrangement may prevail throughout districts of great extent. If wo investigate different mountain-chains, we find gneiss, 584 AGE OF MOUNTAIN-CHAINS [CH. XLH. mica-schist, hornblende-schist, chlorite-schist, crystalline limestone, and other rocks, succeeding each other, and alternating with each other in every possible order. But the rule is that the thicker gneisses and most foliated schists are the oldest. It is, indeed, more common to meet with some variety of clay-slate forming the upper- most member of a metamorphic series than any other rock ; but this fact by no means implies, as some have imagined, that all clay-slates were formed at the close of an imaginary period, when the deposition of the crystalline strata gave way to that of ordinary sedimentary deposits. Such clay-slates, in fact, are variable in composition, and sometimes alternate with fossiliferous strata, so that they may be said to belong almost equally to the sedimentary and metamorphic groups of rocks. It is probable that had they been subjected to more intense hypogene action, they would have been transformed according to their chemical composition into hornblende-slate, foliated chlorite- slate, scaly talcose-slate, mica-slate, or other phyllites, such as are usually associated with schist and gneiss. Uniformity of mineral character in Hypogene Rocks. It is true, as Humboldt has happily remarked, that when we pass to another hemisphere, we see new forms of animals and plants, and even new constellations in the heavens ; but in the rocks we still recognise our old acquaintances the same granite, the same gneiss, the same micaceous schist, quartz-rock, and the rest. There is certainly a great and striking general resemblance in the principal kinds of hypogene rocks, and of the regionally metamorphosed rocks in all countries, however different their ages. But when we re- member how great has been the amount of recrystallisation of the materials, this uniformity of character may cease to surprise us. The more exact study of the oldest crystalline rocks of North America, South America, the Indian peninsula, Japan, ; but of their artificial origin there is a considerable amount of doubt. When it is remembered, however, that savage races employ for various purposes flint and similar materials, with natural fractures, and gradually adapt these for special purposes by new fractures, it is not surprising that the distinction between worked and unworked flints sometimes becomes very difficult. 604 SUPPLEMENTARY NOTES NOTE K (p. 177) The Ancestry of the Elephant. Since the appearance of the first edition of this work, much new and interesting evidence has been obtained concerning the early Tertiary forms of the Proboscidians. In the Eocene and Oligocene strata of the Fayum, in the midst of the Libyan desert, the earliest forms of these curiously specialised mammals have been found. The small animals called Mceritherium and Palceomastodon show the beginnings of the wonderful modifica- tions of the molar (grinding) teeth and of the incisors (tusks) which distinguish the existing elephants, and an almost complete series can be traced from these through the forms known as Tetra- belodon, Mastodon, and Stegodon to Eleplias. NOTE L (p. 177) The Ancestry of the Horse. The palaeontologists of the United States have in recent years greatly added to our knowledge on this subject. About a dozen genera and more than a hundred species of mammals, forming an almost complete chain of links between the unspecialised five-toed forms of the oldest Eocene and one-toed equine animals of the present day, have been described. The horse tribe is thus shown to have originated in North America and to have gradually spread thence to Europe and Asia, with which there must have been land connections. But a somewhat similar, though distinct, series of one-toed forms is now known to have originated independently in South America. NOTE M (p. 178) Origin of other Mammalian Types. Recent discoveries have also thrown much new light on the origin of other types of mammals, including many of the most aberrant ones ; thus the ancestors of the rhinoceroses, and of the camels and llamas, have been shown, from remains found in Tertiary strata, to have originated in the North American continent and to have spread thence to other areas ; many of the deer and antelopes, as well as members of the pig-tribe, migrated from Europe anu Asia : the whales, like the elephants, seem to have been first differentiated in Africa ; in the same way, the sloths and armadillos appear to have acquired their peculiar characters in South America, the continent to which they are still almost confined ; and lastly the kangaroos and their allies made their appearance in, and became distinctive of, Australia. NOTE N (p. 241) Relation of Pleistocene and Recent Mammals in the same Areas. Almost simultaneously with Darwin's pregnant discovery in South America, that the remains of gigantic mammals in Pleisto- cene deposits have such remarkable analogies with the sloths and armadillos of the same region, a collection of bones from caves in Australia, which was examined by Clift, enabled Jameson to arrive at a similar conclusion. He pointed out that the marsupials of Australia were preceded by forms, now extinct, belonging to the same remarkable group of animals. Subsequent discoveries have shown that some of these Australian Pleistocene marsupials were SUPPLEMENTARY NOTES 605 of gigantic size, like the herbivorous Diprotodon (the bones of which were at first mistaken for those of an elephant), and the possibly carnivorous Thylacoleo. It is now known that in all the continents, as well as the larger islands of the globe, like Madagascar and New Zealand, the latest Tertiary mammals and birds closely resemble those of the existing fauna, but include species that attained a great size. It is probable that in all cases it was the advent of man that led to the destruction of these large and often unwieldy mammals and birds. NOTE (p. 245) Fossil Mammals of North America. Great additions to our knowledge of the Tertiary mammalian fauna have been made by the palaeontologists of the United States through the study of the remains of fossil forms, which have been found in such abundance in the " bad lands " of the Western Territories of the North American continent. These remains appear to have been buried in volcanic tuffs, or sometimes in fluviatile or lacustrine deposits belonging to every stage of the Tertiary era, and the fossils are in many cases singularly well preserved. A recent census by Professor Osborn and Dr. Matthew shows that over 1,000 species of these terrestrial mammals have already been described ; something like 300 extinct genera have had to be instituted to receive them ; and these belong to 90 families, three-fourths of which are extinct ; while no less than 6 new orders are represented by them. The great abundance of well-preserved specimens belonging to allied species, in a series of deposits separated by no great intervals, has enabled palaeontologists to trace out with great exactness the probable lines of ancestry of many of the existing forms of mammalian life. NOTE P (p. 265) Appearance of Varieties of Species at successive Geological Horizons. The careful collection and study of specimens of the Brachiopoda by the late Mr. J. F. Walker, and of the Echinoderms by Dr. A. W. Rowe, as they occur at different stages in the Creta- ceous system, have shown how all the characters which distinguish the several species are found to undergo gradual variation, and these varietal forms can be proved to be characteristic of definite horizons. The great, chalk formation, which was evidently accumulated con- tinuously but very slowly, affords especial facilities for this kind of study. It is evident that, at one time, the chalk must have covered the greater part of the British Islands, and fragments derived from higher beds than those now found in situ occur in the Boulder Clay and other deposits formed during the removal by denudation of the great chalk mantle. The great palseontological break between the chalk and the oldest Tertiary strata indicates the lapse of an enor- mous period of time, during which denudation went on ; certain strata in North America (the Laramie formation) may not impro- bably have been deposited during this great interval. NOTE Q (p. 292) The Ancestors of Flowering Plants. It is now known that many of the plants of the Jurassic strata, which were formerly 606 SUPPLEMENTARY NOTES regarded as Cycads, really belong to a great extinct order the Cycadophyta which present a combination of the characters now regarded as distinctive of ferns, cycads, and flowering plants respec- tively. Many botanists consider these curious " synthetic " forms as being the probable progenitors of the flowering plants. In Itcnnettites (the old Cycadeoidea of geologists) it has been shown that on a stem closely resembling that of the Cycads flower-like organs of reproduction are borne ; and from such types, occurring in the Older Cretaceous strata, it is not difficult to imagine that the flowering plants (Angiosperms), which begin to appear in abundance in the Younger Cretaceous rocks, may have been derived. NOTE K (p. 317) The Oldest known Mammals and their Ancestry. It is now known that there existed, in the Triassic, Jurassic, and Cretaceous periods, numerous genera and species of mammals, mostly of small size, which were widely distributed over the globe, and lived on to the earliest part of the Eocene period, when they began to be replaced in different areas by larger and more specialised forms. These primitive mammals are referred to an extinct order variously known as Multituberculata, Allotheria, or Prototheria. But in the South-African Trias we also find Reptiles, which in their dentition, the character of their limb-bones, and other points of their struc- ture very closely resemble the primitive mammals. So close are these resemblances, suggesting a line of descent of mammals from reptiles, that Tritylodon, figured on pp. 316 and 317, is now generally removed from the Mammalia, and placed among the " Theriodont " Reptilia. Many other remarkable mammal-like reptiles from the Trias of South Africa (Karoo Beds) have been described by the late Professor Seeley and others. Still more recently many remains of these Theriodont reptiles have been discovered in the Permian and Triassic rocks of India, South America, and Russia, and also in Scotland. NOTE S (p. 343) The Nature of the Newer Palaeozoic Floras. Mauy very im- portant discoveries have been made in recent years concerning the nature and relationships of the plants which characterise the Devonian Carboniferous and Permian strata the oldest known terrestrial flora. Calcareous concretions, known as "coal-balls," are sometimes found in the midst of coal-seams, and in these, when studied in thin sections under the microscope, the minute structures of the plants of the period are found to be very perfectly preserved. Among the most interesting results of these studies, which were inaugurated by the late Professor W. C. Williamson, are the following. Many of the leaves, which closely resemble, in general form, the fronds of ferns, are now shown to belong to seed-bearing plants, the organs of fructification of which appear to be analogous to those of the Cycads. This group of plants the oldest known seed- bearers has been called by botanists Cycadofilices or Pteridosperms. Another group of seed-bearing plants, known as the Cordaitales, is also found in these Newer Palaeozoic strata, and during the later portion of that era Conifers make their appearance. Before that SUPPLEMENTARY NOTES 607 era, no terrestrial flora has yet been discovered, but in the Newer Palaeozoic rocks there are proofs that the following groups had already come into existence: Among spore-bearing plants, the Lycopods (Club-mosses), Equisitales (Horsetails), and Filices (Ferns). All of these attained great dimensions and exhibited the secondary thickening of stems, by the formation of woody tissue with medullary rays, now characteristic of the exogenous phenero- ganis. With these are found the Sphenophyllales, a group of thin- stemmed, probably climbing, plants with whorled leaves. Among the seed-bearing plants there were Pteridosperms (seed ferns), the Cordaites, and towards the close of the era Conifers. NOTE T (p. 370) Zones in the Carboniferous Limestone. In 1905 Dr. A. Vaughan was able to show that the Carboniferous limestone in the neighbourhood of Bristol is capable of being divided into a number of palaeontological zones and sub-zones, based on the distribution of the corals and brachiopoda. Other writers have since shown that this subdivision of the Carboniferous limestone series, by paljeonto- logical evidence, can be extended to other areas in this country and on the Continent. It thus appears that the division of formations into zones, based on the presence in them of characteristic fossils, which has been so well established in the case of the Mesozoic rocks by the study of the Ammonites and other forms, and in the case of the older Palaeozoic rocks by that of the Graptolites and Trilobites, is equally applicable to rocks of Newer Palaeozoic age. NOTE U (p. 385) The Fish-like Animals of the Newer Palaeozoic Rocks. It is now generally recognised by zoologists that many of the curious forms of life occurring in the Silurian and Devonian strata, such as Pterasjris, CejjJtalaspis, Ptericthys, etc., with the puzzling little Palccos2)iallage is an altered augite (brown or green) with a basal parting and sub-metallic lustre. Found generally in basic plutonic rocks (gabbros). Enstatite. IV. Pol. low. Pleoc. marked in coloured varieties. The ferriferous varieties of Enstatite are called Bronzite, the highly ferriferous varieties are known as Hypersthene. The chief amphiboles are : Hornblende. V. G = 2-9-3-4, H = 5-6. Pol. strong. Pleoc. high. Very common in metamorphic and some igneous rocks. Tremolite is a colourless hornblende, .Actinolite is a green hornblende, and Arfvedsonite, Claucophane, Riebeck'te, CYC., are soda-hornblendes, which sometimes assume blue tints. Anthophyllite (IV.) is an amphibole corresponding to Enstatite in the Pyroxene series. THE OLIVINE GROUP. This group consists of basic silicates of magnesia and iron (occasionally with lime). The minerals of the group occur in basic and especially in ultrabasic rocks (Peridotites). Olivine (or Peridot). IV. G =-- 3-4, H = 6-7. Green to black in colour. Colourless in thin sections. Pol. high tint. Very easily undergoing alteration and easily changed to serpentine. D. THE OXIDES OF IRON, TITANIUM, &c. These are very widely diffused, but form a .large part of rocks only of the basic and ultrabasic groups. Magnetite. I. G = 4-9-5-2, H = 5-5-6-5. Opaque, black, and APP. A.] ACCESSORY MINERALS 615 submetallic. Often showing alteration to red, brown, and yellow (hydrated) products. Comp. FeO, Fe. 2 3 . Titanoferrite. III. G = 4-5-5, H = 5-6. Opaque, black, and submetallic, but showing white decomposition products. Comp. FeO.TiA- Xtutlle. II. G = 4-18-4-25, H = 6-6-5. Reddish brown to yellow translucent. Usually in very small crystals (twinned), often en- closed in other minerals. Comp. Ti0 2 . Anatase and Brookite, two other forms of TiO.,, are probably formed by alteration of Rutile. Hematite (III.) (hexagonal plates, blood-red by transmitted light) is formed by oxidation of magnetite. Comp. Fe 2 O 3 . Various brown and yellow oxides result from the hydration of hematite. Corundum (A1 2 3 ) ; the Spinels (Picotite, Cbromite, &c.) and Cassiterite (SnOJ occur more rarely in rocks. E. ACCESSORY MINERALS. Apatite. III. G = 2-92, H = 4-5-5. Comp. Phosphate of calcium with chloride or fluoride of calcium. Very common in small quantities in igneous and metamorphic rocks. A phosphate of yttrium and the cerium metals called Monazite is present in very minute crystals in many rocks. Zircon. II. G = 4-6-4-7, H-7'5. Comp. Si0 2 ,ZrO,. Like the above very frequent in small crystals enclosed in all the constituents of rocks. Sphene. V. Often in wedge-shaped crystals, brown or colourless. G = 3-4-3-6. Comp. Calcium silico-titanite. A primary constituent of some rocks, and the result of alteration of titanoferrite in others. F. MINERALS ESPECIALLY FOUND IN ROCKS PRODUCED BY CONTACT METAMORPHISM. Garnet. I. G = 3-15-4-3, H = 6-5-7'5, fusible. Comp. Isomorphous mixtures of compounds of aluminium, iron, chromium, and titanium sesquioxides with the protoxides of iron, magnesium, lime, &c. Found both in igneous and metamorphic rocks. Topaz. IV. G = 3-4-3-65, H = 8. Comp. Aluminium silicate with fluorine. Colourless, higher refractive index than quartz. Tourmaline. III. Hemimorphic. G = 2-98-3-2, II = 7-7*5. Colour very various. Pleoc. strong. Comp. Borosilicate of aluminium, calcium, iron, &c. Andalusite (C/iww/oZ^e)(IV.,Sillimanite (Fibrolite) (IV.), and Kyanite (VI.) are forms of aluminium silicate. Cordierite (IV.) and Staurolite (IV.) are more complicated compounds of aluminium with magnesium and iron silicates. Epidote (V.) and Zoisite (IV.) are aluminium and calcium silicates, sometimes with iron, manganese, &c. (Piedmontite). G. SECONDARY MINERALS. Almost any of the minerals already named may occur as secondary constituents of rocks, but certain minerals like the following seldom occur as primary constituents of igneous rocks, 616 SECOND AKY AND METALLIC MINERALS [AFP. A. (a) Micaceous Minerals (pseudo-hexagonal with strong basic cleavage). Chlorites. V. Hydrous aluminium silicates with silicates of magnesium and iron. Usually exhibiting green tints. Chloritoids (Brittle Micas). V. Very varied in composition, e.g. Ottrelite, Masonite, &c. Vermiculites. Other hydrated minerals which exfoliate and curl up under the blowpipe. Kaolin. V. G = 2'6, H = 2-2-5. Hydrous aluminium silicates. This and other similar compounds form the basis of most clays. Talc (Steatite). V. G = 2'7-2'8. H = l. Comp. Hydrous magne- sium silicate. (b) Non-micaceous Minerals. Zeolites. Hydrated alumino-alkaline minerals. They boil up in the blowpipe flame, hence their name. Calcite. III. G = 2-72, H = 3. Cleavage and twinned character strongly marked. Comp. CaCo.,. Aragronite IV. G = 2'95, H = 3'5 4. Another form of calcium carbonate. Dolomite. III. G - 2-8-2-9, H = 3-5-4. Comp. (CaMg)C0 3 . Gypsum. IV. G-2-3, H = 2. Comp. CaS0 4 + 2HX). The hydrous oxides have already been noticed ; they occur frequently as secondary constituents of rocks. H. MlNEBALS OF THE HEAVY METALS (' ORES '). These are found in veins or other ore deposits, or, more rarely, diffused in small quantities through igneous, aqueous, and meta- morphic rock-masses. They are usually inter-crystallised with various sparry minerals (veinstones), such as Quartz, Calcite, Fluorspar (calcium fluoride), Barytes (barium sulphate, &c.). The ores found in the upper part of veins (' gossans ') are either oxides like Magnetite, Hematite, Cuprite (copper oxide), Zincite (zinc oxide), &c., or hydrated oxides, like those of iron (Gotbite, Xiimonite, Xanthosiderite), of manganese (Manganlte, Psilo- melane), &c. With these occur Carbonates, like those of iron (Cbalybite), of zinc (Calamine), of copper (Malachite, Azurite), of lead (Cerussite), &c., with various sulphates, silicates., phosphates, and other salts. The deeper portions of ore-deposits are usually characterised by the presence of sulphides, among the commonest of which are those of iron (Pyrite, Marcasite, Pyrrhotite), of lead (Galena), and of zinc (Blende). With these occur many complex compounds of sulphides, selenides, tellurides, arsenides, and antimonides. Some of the less oxidisable metals (gold, platinum, &c.) usually occur ' native ' (uncombined), or as alloys or amalgams ; and silver, copper, and mercury are also not unfrequently found in the unoxidised condition. Iron, alloyed with nickel or platinum, similar to the iron meteor- ites (siderites), has been found in igneous masses at Ovifak in Green- land, Santa Catharina (?) and Eibiera in Brazil, Awarua in New Zealand, Josephine in Oregon, and Ekaterinberg in the Urals, APP. B.] PLANT-CLASSIFICATION 617 APPENDIX B. CLASSIFICATION OF PLANTS, LIVING AND FOSSIL. Names printed in italic capitals are those of groups which are not found preserved as fossils. Those printed in capitals have both fossil and living representatives. Those in thick type are extinct. A. CELLULAR CRYPTOGAMS (spore -bearing plants with cel- lular tissue only). MYXOM YCKTES. (Slime-fungi living on dead organisms or causing plant-diseases). DIATOMACE^E. (Unicellular with siliceous skeletons). Tertiary. SCHIZOPHYTA. (Oscillatoria, micrococcus, bacteria, &c.) ALG.E. Freshwater and marine (including calcareous forms like Lithothamnion, SiphonieaB, Chara, &c.). Camb. FUNGI. (Peronosporites, Discomycetes, Plio- saurus.) Trias. Cret. CHELONIA. (Tortoises and turtles.) Trias | Anomodontia. (Dicynodon, &c.) Trias. Therio- J Placodontia. (Placodus.) Trias, inorpha. j Pariesauria. (Pariesaurus.) Perm. Trias. ' Theriodontia. (Cynodracon.) Perm.- Trias. KHYNCHOCEPHALIA. (Hatteria, Hyperodapedon.} Perm. LACERTILIA. (Lizards.) Eocene Pythonomorpha. (Mososaurus.) Cret. OPHIDIA. (Snakes.) Eocene CROCODILIA. (Crocodiles and alligators.) Trias. jy no _ [Sauropoda. (Atlantosaurus,Cetiosaurus.} Jura. Cret. uria I Tliero P oda " (Megalosaurus.) Trias. Cret. (Ortbopoda. (Stegosaurus, Iguanodon.) Jura.-Cret. Ornitho- | Pterosauria. (Rhamphorhynchus.) Jura. Cret. sauria. 1 Pteranodontia. (Pteranodon.) Cret. 622 CLASSIFICATION OF ANIMALS [APP. c AVES (Birds). Saururae. (Archaoptoryoe.) Jura. Ratitae-Odontolcae. (Hctperernit.} Cret. BATIT;E. (Ostriches, &c.) Miocene. Carinata Odontormae. (Ichthyornis.) Cret. CAKINATA. (Common birds.) Eocene MAMMALIA (Mammals). Allotheria. (Prototheria, Multituberculata), (Micro- lestes, &c.) Trias. Tertiary. MONOTREMATA. (Echidna, Ornithorhynchus). Pleisto- cene MARSUPIALIA. (Kangaroos, Opossums, &c.) Trias. EDENTATA. (Sloths and armadilloes.) Pliocene ( Arcbaeoceti. (Zeuglodon.) Eocene. Cetacea. \ ODONTOCETI. (Delphinus,Squalodon,Zi2)hius) Miocene. ( MYSTICETI. (Balcena.) Miocene SIRENIA. j(Dugong, Halitherium.) Eocene. ^Condylarthra. (Phenacodua.) Eocene. PERISSODACTYLA. (Horse, &c.) Eocene. AETIODACTYLA. (Ox, deer, &c.) Eocene. Litopterna. (Macraucheiiia.} Oligocene. Pleistocene. Amblypoda. (Coryphodon, Uintathcriwn, &c.) Eocene. PROBOSCTDEA. (Elephants, &c.) Miocene Toxodontia. (Toxodon.) Oligocene Pleistocene. Typotheria. (Typotherium.') Oligoeene. Pleistocene. Barypoda. (Arsinothcr'ntm.) Eocene. HYEACOIDEA. (Hyrax.) Tillodontia. (Tillotherium.) Kocene. EODENTIA. (Eodents.) Eocene INSECTIVOBA. (Insectivores.) Eocene CHEIROPTERA. (Bats.) Eocene | Creodontia. (Hycsnodon.) Eocene. j FISSIPEDIA. (Cats, dogs, &c.) Eocene ( PINNIPEDIA. (Seals.) Miocene | PROSIMIJE. (Old apes, lemurs, etc.) Eocene j SIMILE. Apes and monkeys (Dryopithccus). ( BIMANA. (Homo.) Pleistocene Carni- vora. Pri- mates. Miocene The views of botanists and zoologists on classification are under- going continual change with new discoveries. Some modifications, based on the British Museum Catalogue, have been made in these tables in the present edition. In quoting the names of species of animals and plants in this work, the convenience of the student has been consulted in preference to any attempt being made to secure absolute accuracy or uniformity of procedure. In a few cases it has been found necessary to insert two names that by which the fossil ought to be designated, and that which is familiar to most geologists. This course has been adopted in the case of the Ammonites, and of some large genera, the sub- division of which is inevitable. In all cases, however, the authors of the specific 'names are given. INDEX The names printed in italics are those of fossils and other objects illustrated by figures. Names of authors quoted are printed in capitals ABBOTT ABBOTT and HUMPHREYS, 120, 125 ABICH, 475, 521 Abnormal varieties of Acer trilobatum. Ad. Brong., 180 Absolute duration of Geological periods, 592 Acadian strata, 420 Acanthocera* rothomagensis, Befr., 262 Acer trilobatum, Ad. Broug., 180 Acid Lavas, 461 Acrodut nobilis, Ag., teeth, 279 Action of torrents, 40 ^Kchmodus Leachii, Ag., scales, 278 restored, 278 .Kgoceras planorbis, Sow. sp., 306 vEolian rocks, 18 Aerial rocks, 18 AGASSIZ, L., 168, 346 Age of Bronze, 171 of Copper, 171 of Iron, 171 of Mammoth, 170 of Reindeer, 170 of Stone, 171 Agnostus integer, Beyr., 423 rex, Barr., 423 Agnotozoic strata, 437 Aix-la-Chapelle, strata and flora of, 327 Albiau series, 265 Albite, characters of, 602 Aldeby and Chillesford beds, 184 Algonkian strata, 437 ALLPORT, 505 Alluvia of different ages, 110 Alluvium, formation of, 109 Alpine blocks on Jura, 239 glaciers, fornipr extension of, 239 Alps, fan-structure in, 565 junction of granite with Oolite in, 532 overf aiding in, 565 Alternation of freshwater and marine strata, 78 Alumino- alkaline silicates, 602 Alumstone, 462 Amaltheus margaritatus, Montf. sp., 305 Ammonites of the Chalk, 262 Neocomian, 268-269 ANCYLUS Ammonites of the Middle Oolites, 295 Lower Oolites, 298, 303, 304 Lias, 204-206 Trias, 313 Permian, 340 Ammonites A. (Acanthoceras) rothomagensis, Defr., 262 A. (Amaltheus) margaritatus, Montf., 305 A. (.Kgoceras) planorbis, Sow., 306 A. (Arietites) liueklandi, Sow., 306 A. (Arcestes) multilobatus, Brown, 313 A. (Cosmoceras) Eliziibethce, Pratt, 295 A. (Cosmoceras) Jason, Reinecke, 295 A. (Cyclolobus) Oldhami, Waagen, 340 A. (Hildoceras) bifrons, Brug., 304 A. (Hildoceras) Walcotti, Sow., 304 A. (Hoplites) Deshayesii, Leym., 268 A. (Hoplites) noricus, Schloth. sp., 269 A. (Afedlicottia) Wynnei, Waagen, 340 A. (Stephanoceras) Braikenridgii, Sow., 304 A. (Stephanoceras) Hiunphriesiantts, Sow., 303 A. (Stephanoceras) macrocephalus, Schloth. 298 A. (Trachyceras) Aon, Miinst., 313 A. (Xenodiscus) plicatus, Waagen, 340 A. Ap/ uchus of, 294 Amphibians of Carboniferous, 365 Amphiboles, 604 Amphibolites, 518 Amphigestina Hauerina, 176 Amphitheriiim Broderipii, Ow., lower jaw, 300 Prevostii, Cuv. sp., lower jaw and molar, 300 Ampnllaria glanca, 56 Amyij'laloids, 458 Analyses of types of Igneous rocks, 536 of Metamorphic rocks, 588 Ananchytes ovatus, Leske, 257 with Crania attached, 46 Anchitherium of Miocene, 178 Ancillaria sitbulata.Sov?., 57 Ancyloceras Duvallii, Leveille jgp., 268 -' gigas, D'Orb., 266 spinigerum, D'Orb., 264 Ancylus villetia. Sow., 54 624 INDEX ANDALUSITE Andalusite, characters of, 605 Andesine, characters of, 602 Andesites, 462 and Propylites of Western Isles of Scotland, 490 Animals, classification of, 608 Animike strata, 437 Annularia sphenophylloides, Zenk, 361 Anodonta Cordieri, D'Orb., 53 Jukesii, Forbes, 379 latimarginata, Lea, 53 Anoplotherium commune, Cuv., lower molar tooth, 205 Anorthite, characters of, 602 Anthracite, composition of, 32 formation of, 558 Anticlinal strata, 79 Antrim, leaf-beds of, 213 relation of Plutonic to Volcanic rocks in, 521 Tertiary volcanoes of, 490 First Period of, 490 Second Period of, 491 Third Period of, 492 Antwerp Crag of Belgium, 228 Apiocrinus rotundas, Mill., 297 plate encrusted with Serpula and Bryozoa, 298 Aplite (Haplite), 516 Aporrhais Soicerbyi, Mant., 218 Appalachian Mountains, 33 type of Mountain chains, 564 Aptychi of Ammonites, 294 Aqueous rocks, 16 Aquitanian Series, 247 of Switzerland, 230 Araucaria Sphferocarpa, Carr., 301 Aravalli System, 438 Arbroath flags of Forfarshire, 389 Arcestes multilobatus, Brown sp., 313 Archaean, 433, 437 Archcegosaurus minor, Goldf., 366 Archveopteryx macrura, Ow., head, 286 tail and feathers, 285 AIICHIAC, D', 221, 230 Arctic Eocene Flora, 201 Arctic Regions, Jurassic strata of, 332 Arcto- Pacific life-province, 331 Ardnamurchan, granites and gabbros of, 530 Tertiary volcano of, 490 Arenaceous rocks, 25 Arenicolites linearis, Hall, 414 Arenig beds of Wales, 418 Argile plastique, 221 Argillaceous rocks, 27 Argillite, 28 ARGYLL, DUKE OF, 212, 213 Arietites Bucklandi, Sow. sp. Arkose. 26 ARNOLD, Dr. J., [34] Arran, dykes in, 476 : granite of. 522 Armorican, 411 Arthur's Seat, Edinburgh, 504 Artinsk Etage, 393 Arvicola intermedium, E. T. Newt., 153 Arvonian strata, 434 Asaphus tyrannus, Murch., 414 Ashburnham Beds, 272 Ashdown Sand, 272 BASAL Aspidura loricata, Ag., 312 Astarte borealis, Chem. sp., 148 Omalii, Laj., 174 Asteropnyllites foliosus, Lindl. et Hutt., 361 Astian Series, 246 Astronomical Cycles, attempt to correlate with Geological periods, 593 Astropecten crispatus, E. Forbes, 216 Atherfield Clay, 267 Atlantic Islands, volcanic rocks of, 499 Atrypa reticularis, L., 401 Aturia ziczac, Brown, 218 Augite, characters of, 604 Augite-Andesites, 463 Augite-diorite, 517 - syenite, 517 Aulacoceras sulcatum, Hauer, 314 Auricula, 55 Auriferous deposits, age of, 586 Australia, Cave-breccias of, 241 , former glaciers of, 241 Australian White Coal, Microscopic struc- ture of, 60 Auvergne, Oligocene of, 223 Chain of Puys in, 469 younger volcanoes of, 493 : older volcanoes of, 498 Aviada cygnipes Phil., 307 incequivalvis, Sow., 307 Aviculopecten papyraceus, Sow. sp., 353 sublobatus, Phil., 353 Axiolitic structure, 461 AYMARD, 225 Aymestry Limestone, 406 Azores, volcanic rocks of, 501 BACILLARIA paradoxa, Gmel., 49 Baculites anceps, Lam., 261 Faujasii, Sow., 328 Bagshot beds, 213 BAKEWELL, R., 67, 592, [31] Bala beds, 415 thickness of in North Wales, 417 Limestone, 415 BALL, SIR R., 593 Baltic provinces of Gertnrny, Older Palaeo- zoic rocks of, 430 BALTZKR, PROP., 582 Banded structure, 460 Banksia (?) Deickieana, Hr., 200 Barbadoes earth, 51 BARRAXDE, 398, 410, 416, 419, 426, 429 Barren Island, Bay of Bengal, 468 BARROIS, M. 0., 264, 266, 275, 438 Barton clay, 207 Bartonian series, 427 Basalts, 465 Basalt-glass, 465 Basaltic columns, bent, 481 of various dimensions, 480 varieties of, 481-481 of Bertrich-Baden, 482 Basaltic plateaux of Antrim, 491 of India, 501 of Western Isles of Scotland, 491 of Western Territories of U.S., 501 Basaltic tuffs, 465 ' Basal wrecks ' of volcanoes, 475 INDEX 625 BASIC Basic lavas, 464 Plutonic rocks, 518 BASTEROT, DE, 142 Bastite-serpentine, 519 Bath, hot springs of, 489 heat liberated by, 489 Bear Island, flora of 394 BECKER, PROF., 417 BECKER, DR., 582 Bedding, peculiarities of, 37 Beinn Shiant, volcano of 492 Belemnites of the Cretaceous, 251 327 hastatus, Blain., 295 of the Jurassic, 295 Puzosianus. D'Orb., 295 Pelemnitella mncronata, Schloth., 327 tlellerophon costatus, Sow 354 Belonites, 460 Brtoxpia sepioidea, De Blainv., 216 Bembridge series, 204 Bending of strata, how effected, 90 Bent strata of &. J ean de ^ Pyrencegj of Sicily, 91 BERGKR, 478 BKRTRAXD, M., 584 Better-bed coal, microscopic structure of, composition of 33 BEYRICH, 144, 191, 228 Bharwar system, 438 Bijawar system, 438 BIXXEY, MR. E. W., 504 Biotite, characters of, 603 Birds of Jurassic, 279 Cretaceous, 252 Eocene, 196 Birkhill shales, 405, 419 BISCHOFF, 31 542 Black Crag of Belgium, 228 Blackdown beds, 264 Black Jura, 275 BLADTVILLB, 221, 227 BLAKE, PROF., 435, 438 7 J V) WWAj t/UV Blown sands, 25 Bog-iron ore, origin of, 48 Bognor beds, 214 Bohemian, 398 Bohemia, Cambrian of, 429 Ordovician of, 429 Silurian of, 429 Bone-beds, 30 Bone-bed of Ludlow, 403 405 fishes of, 403, 405 BOXELLI, 142 BOXXEY, PROF., 165, 434, 438 482 549 Bordeaux, Miocene strata of 2-7 ' BORY DK ST. VIXCEXT, 471 Bos taurus, L., 153 BOUCHER DE PERTHES, 156 Boulders in Chalk, 256 Boulder clay, distribution, 166 BURROWS I Boulder clay, nature of, 166 Bournemouth beds, 210 flora of, 210 B^veyTracey^ligmtes and clays of, 212 BOWERBAXK, 214,' 260 Box-stones of East Auglia, 189 Bracheux, Sables de <>20 Brachiopoda of the Cretaceous, 257-258, 263- Jurassic, 298, 302, 307 Trias, 312 Permian, 339 - of Carboniferous, 352 - of Devonian, 376 - of Silurian, 398 of Ordovician, 414 of Cambrian, 421, 424 - aberrant forms of, 339 Bracklesham beds, 208 Bradford clay, 297 Breaks in succession of strata in Europe, Breccias, 26 Brick-earths, 162 Bridget series, 242 Bridlington drift, 168 BRISTOW, 218 British strata, chronological sequence of, Brittany, Pre-Cambrian of, 438 .nrixnam cavern, 160 BROCCHI, 148 Brockenhurst marine group, 207 BUODERIP, 300 BRODIE, REV. P. B., 288, 292 309 BHOGGER, 518, 535 Bronteusflabellifer, Goldf., 378 Brontosaurus ?rn>ivue ATr,i, -^ i . skeleton Bronze, Age of, 171, 240 Brora, Oolite of, 302 Brown coal, composition of, 37 Lower, 228 Jura, 275 BROWX, ROBERT, 214, r361 Bruxellian of Belgium 2*2 Bryozoa of the Crag, 188 Chalk, 250 Permian, 344 ?!5:X- v '! 3 Vn 74 ' 533 ' 573 J, 7, 160, 168, 217, 256, 261, 289, BUCKMAX, S., 326 BuUmus ellipticHs, Sow 205 lubricus, Mull., 56 BUXBURY, 500 BUX.SEV, 488 Banter, origin of name, 310 of Britain, 320 Bunter-sandstein of Germany, 324 B'lprestis (V), Elytron of, 299 tfurdie-House limestone, 370 Burdigalian scries, 247 BURMEFSTER, 416 Burniuland, Fife, section at 475 Burrows and tracks, 43 BURROWS, 247 SS 626 INDEX CADEB CADER IDRIS, volcanic rocks of, 506 Cainozoic, defined, 127 Caithness flags, fossil fish of, 389 Catamites Suckowii, Brong., 360 restored, 360 radical termination of, 360 Calamophyllia radiata, Lamouroux, 297 Calcaire a Nerinees, 295 coquillier, 311 de la Beauce, 223 de St. Ouen, 222 grossier, Upper, 222 Middle, 222 Lower, 221 of Mons, 220 Calcareous grit, 295 rocks, 28 sandstone, 26 Calceola sandalina, L., 376 Calciferous sandstone, 371 ' Calciphyres,' 556 CALLAWAY, DR. C., 425, 434, 435, 438 Calymene Blumenbachii, Brong., 403 Cambrian strata, nomenclature of, 419 system, British representative of, 425 strata, classification of, 420 fauna, number of species, 420 Upper, 425 Middle, 426 : Lower, 427 - of Bohemia, 429, 431 of North America, 430, 431 of Scandinavia, 430, 431 Cambridge Greensand, 262 Cambro-Silurian, 411 Camptonite, 517 Canada, Archaean plutonic rocks of, 535 Canaries, volcanic rocks of, 500 CANHAM, REV H., 187 Cannel coal, 373 Caradoc group, 415 Sandstone, 415 Carbonaceous rocks, 31 Carboniferous, classification of, 349 nomenclature of, 348 strata, mode of formation of, 369-71 - Amphibians, 365 Brachiopoda, 352 Cephalopoda, 354 corals, 350 crinoids, 351 fish, 356 foraminifera, 350 Gastropoda, 354 insects, 365 LameUibranchiata, 353 Myriapoda, 364 plants, 357-63 Pulmonata, 364 rain-prints and cast of, 368 limestone, 369 of Derbyshire, 369 of Ireland, 370 of Scotland, 370 of Russia, 393 Plutonic rocks, 533 shales, 369 slate, 391 Carcharodon angustidens, Ag., teeth, 211 Cardiocarpum Lindleyi, Carr., 358 CHALK Cardiocarpum Ottonis, Gutb., 342 Cardita (Venericardia) planicosta, Lam., 209 sulcata, Brand., 208 Cardium dissimile, Sow., 293 striatulum, Sow., 293 CARXE, 571 Carnsilver Cove, Cornwall, granite veins at. 526 CARPENTER, DR., 74 Carrara, Marble of, 582 Carrock Pell, 527 CARRUTHERS, Mr. W., 214 Carstone, 266 Caryophyllia Bowerbankii, Ed. and H., 351 Cants, formation of, 69 internal and external. 69 Catastrophism in Geology. 594, [31] ( 'aulopleris primceva, Lindl., 360 Cave-bear, teeth of, 1 53 Cave breccias of Australia, 241 Caverns, mode of formation, 158 in limestone, 158 stalagmite on floors, 1 58 human and animal remains in, 158 - in Belgium, 158 Cavities in minerals, 512 CAYEUX, M., 438 Cementing materials in rocks, 65 Cement-stone group, 370 Cenomanian, 261 Cephalaspis Lyrllii, Ag., 380 restoration of, 380 Cephalopoda of the Cretaceous, 251 - Jurassic, 278 Trias. 313 Permian, 340 Carboniferous, 355 Devonian, 378 Silurian, 402 Oniovician, 414 Cambrian, 422 Ceratites nodosux, Schloth., 313 Cerithium confavum, Sow., 206 funatum, Mant., 55 plicalum. Lam., 203 portlandicttm, Sow. s^ ., 292 Cerviis alces, L., 152 Cestracion Phillippi, Cuv., 260 Chalcedony, characters of, 602 ' Challenger ' ridge, 501 Chalk, 29 nature of, 255 microscopic structure of, 50 origin of, 50, 255 foraminifera of, 255 boulders in, 256 strata, area covered by, 254 thickness of, 254 zones of, 265 grey, 261 marl, 262 phosphatic, 261 Red, of Hunstanton, 263 rock, 261 - shallow-water representation of, 326 freshwater representatives of, 326 Upper, 257 Middle, 261 Lower, 261 INDEX 627 CHALK Chalk of Faxoe, 327 of Meudon, 327 of Southern Russia, 254 Chama squamosa, Sol., 207 Champlain period, 170 series of North America, 245 Chara elastica, Amici, 58 medicaginula, Brong., 58 tube re ulatus, Lyell, seed-vessel, 205 CHARLES WORTH, 215 Charnockite, 515 Cheirotherium, footprints of, 321 Chellean period, 246 Chemically formed aqueous rocks, 23 Cheviot Hills, volcanic rocks of, 506 Chiastolite slate, 555 Chillesford and Aldeby bed?, 184 Chimera monstrosa, L., 279 China-clay rock, 516 Chlorites, 'characters of, 606 Chlorite schist, 561 Chlorite slate, 560 Chloritic marl. 262 Christiania, granite and limestone near, 552 Chronological groups of strata 132 CHURCH, PROF. A. H., 27 Cinder bed, 288 Cinnamomum polumorimum, Ad. Brong., 181 Rossmassleri, 200 Cipolinos, 556 CLARKE, MR. F. W., 10 CLARKE, W. B., 247 Classification of animals, 608 of plants, 607 of Silurian, 404 strata, 397 of strata necessarily local, 146 of Trias, 311 Clastic rocks defined, 15 Clausilia bidens, Drap., 56 Clays, cause of colour of, 27 Clay slate, 559 ironstone of Coal-measures, 373 Cleavage in slates, 513 in beds of different hardness, 543, 545 in curved strata, 543 origin of, 545-546 experiments illustrating origin of, 574 and joints, 543 Cleveland dyke, 477 iron ore, 308, 568 Climate of Crag Period, 190 Clymenia lined ris, Miinst., 377 Coal, varieties of, 31 composition of, 32 rate of formation of, 63 formation on land, 37 purity of, 62 origin of, 31 underclays of, 371 - - trees in, 60 of Brora, 302 basins, how found, 349 measures, meaning of term, 349 in North of England, 368 in Scotland, 368 fields, 349 CORDIEB I Coal-field of South Wales, 371 COCCHI, 234 Coccoliths, 50 Coccospheres, 50 Cochliodus contortus, Ag., 356 Ccelacanthus granulatus, Ag., 346 COHX, PROK., 24 COLE, PROF. G. A. J., 521 Collyrites (Dysaster) ringens, Ag., 303 Colly weston slates, 301 Columnar structure in basalt, 480 in dykes, 476 in granite, 523 Comb structure in veins, 572 Comley sandstone, 428 Comparative thickness of strata of Europe, 441 Composite dykes, 527 of Arran, 492 Concretionary structures, 65, 601 CONDAMINE, DE LA, 217 Congerian strata, 235 Conglomerates, 26 Connecticut valley, Newark strata of, 332 Confonnability, 99 Conocoriiphe striata, Emmr. sp., 423 CONRAD, MR., 334 Consolidation of strata, 64 Contact metamorphism, 538, 551 minerals formed by, 605 rocks formed by, 555 extent of, 553 in dykes, 477 age of rocks formed by, 578 and regional metamorphism, analogy of rocks formed by, 578 Contemporaneous erosion, 38 veins, 528 volcanic rocks, 479 age of, 485 Contorted drift, 41 Contortion of strata, 82 produced by tangential pressure, 84 of strata, illustrated by e.rj>eriment,%Z Conularia ornata, D'Arch., 377 Con us deperditns, Brug., 211 CONYREARE, 478 Coomhola grit, 391 COPE, PROF. E., 197 Copper, Age of, 171 Coprolite of fish from Chalk, 261 ' Coprolite ' beds of Crag, 187 Coral reefs, upraised, 165, 601, 602 rag, 294 Corals, modern and ancient types of, 351 growth on electric cables, 47 of Older Tertiaries, 194 of the Cretaceous, 250 Jurassic, 277 Trias, 311 Permian, 339 Carboniferous, 360 Devonian, 375 Silurian, 398 Ordovician, 413 Corallian, 294 Coralline Oolite, 294 or White Crag, 187 Corbula pisum, Sow., 203 CORDHCR, 465 ss2 628 INDEX CORNBRASH Cornbrash, 296 Cornwall, mineral veins of, 571 Correlation of strata in different areas, difficulty of, 146 of Tertiary strata, 246-47 . of Mesozic strata, 336 of Newer Pakeozoic strata, 395 of Older Palaeozoic strata, 431 Corsite (Orbicular diorite), 514 Cosmoceras Elizabethce, Pratt sp., 295 Jason, Rein., 295 Cosmogony, relation to Geology, 5 Cotopaxi, 472 Coutchiking strata, 437 Crags, relations of to London Clay, 186 Crag, origin of name, 182 fauna, relation to that of existing seas, 190 'Crag,' defined, 172 Antwerp, of Belgium, 228 Black, of Belgium, 228 Crania parisiensis, Defr., 258 attached to Ananchytes (Echinocorys), 46 Crassatella sulcata, Sow., 208 Craters, nature of, 472 origin of, 472 ' Craters of elevation,' 474 Crater-lakes, 474 Crater-lake of Gustavila, Mexico, 473 CREDNER, PROF. H., 247, 341 Cretaceous system, 248 flora of, 327 zones of, 265 Plutonic rocks of, 531 volcanic rocks of. 507 Criuoids of the Cretaceous, 250, 257 Oolites, 297 Lias, 307 Trias, 312 Carboniferous, 351 Devonian, 376 Silurian, 399 Crioceras Duvallii, Leveille, 268 Croatia, Oligocene of, 234 CROLL, 119, 125, 593 Cromer, forest-bed of, 183 Cross-bedding, 37 Crushed and impressed pebbles, 91 Crust of globe, defined, 8 dimensions. 8 physical characters, 9 chemical composition, 10 Cryptodon (Axinus) angulatum, Sow., 216 Crystals, negative, 512 skeleton, 460 Crystalline rocks defined, 15 limestone, 556 Crystallites, 460 Cuddapah system, 433 Cuilin Hills, Skye, 521, 530 Culm facies of Carboniferous in Britain, 366 of European Carboniferous, 392 CULVERWELL, MR., 593 Cumberlandite, 518 Current bedding, 37 CUVIER, 141, 222, [82] Cyathocrinus caryocrinoidea, M'Coy, 852 ptanus* Mill., 352 DENUDATION Cyathophyllum flexuosum, Goldf., 351 Cycas circinalis, L., 291 Cyclas (Sphwrium) corneus, Sow., 53 Cyclolobus Oldhami, Waagen, 340 Cyclopean Isles, dykes and contortions in, 494, 495 Cyclostoma elegans, Mull., 56 Cylindrites (Actonon) acntus, Sow., 298 Cypridea spinigera, Fitton, 272 in Weald 'day, 272 Cyprina Morrisii, Sow., 218 Cyprus swamp of Mississippi, 63 Cyrena cuneiformis, Sow., 217 153 (Corbicula)Ji t iminalis,W\A\., 149, semittriata, Desh., 203 Cyrtoceras precox, Salt., 422 Cystoidea of the Silurian, 399 Ordovician, 412 Cambrian, 421, 424 DACHSTEIN limestone, 329 Dacites, 463 Dadoxylon, Endl., fragment of, 357 DAINTREE, 586 Dakota group, 335 DALL, W. H., 247 ' Dalradian ' rocks, 434 Damuda beds of India, 396 DA\A, J. D., 74, 123, 247, 337, 410, 440, 443, 444, 458, 566, 568, 591, 592, 601 DAXA, PROF. E. S., 601 Danian series, 327 Dapedius monilifer, scales, Ag., 278 DARWIX, 7, 77, 116, 123, 154, 165, 168. 318. 442, 4/a, 646, 549, [25], [45] PROF. G., 503, 594 ; Dr. F., [25], [26] DAUBENY, 542, [36] DAUBREE, 25, 484, 540, 549, 550 DAVIS, Mr. E., 426 PROF. W. H., 125 DAWKINS, PROF. W. BOYD, 160, 171 DAWSON, SIR J. W., 74, 364, 385, 394 DEBEY, 327 Deccan ' trap,' 501 DECHEN, VON, 525 D'Esino beds, 329 Deep-sea deposits, 50 Deep-water formations at great heights, 124 DELESSE, 113, 482, 541, 550 Deltas, formation of, 109 Denbighshire grits, 409 Dendritic markings, 73 Density of earth, 10 of earth's crust, 10 Denudation, 102 marine, 112 submarine, 112 subaerial, 107 aided by rise of land, 107 affected by subsidence, 107 rate of, 121 effects of on contorted strata, 82 in mountain making, 5(i6 an action of hypogene forces, 122 compensated for by action of hypo- gene forces, 121 of Carboniferous in North Amerntv 3G4 INDEX 629 DENUDATION Denudation, effecis of in Mendips and South Wales, 109 Deposition, rate of, indicated by fossils, 44 vast in subsiding areas, 118 Derbyshire, mineral veins in, 571 Derived fossils, 137 Desert sands, 25 DESMAREST, 4, [42] DKSNOYERS, 142, 223 Deuterozoic strata, 128 DEVILLE, C. ST. CLAIRE, 575 Devitrification, primary and secondary, 461 Devonian, nomenclature of, 374 classification of, 386 Brachiopoda, 376 corals, 375 Stromatoporoids, 375 Trilobites, 378 Eurypteridce, 379 Fish, 380-382 plants, 383 Upper, of Devonshire, 386 Middle, of Devonshire, 386 Lower, of Devonshire, 386 strata of Brittany, 392 . of the Eifel, 391 of North America, of Kussia, 392 Devonshire, volcanic rocks of Triassic age in, 503 Diabases, 518 Diabasic structure, 461, 518 Diallage, characters of, 604 Diatomacefe, formation of rocks by, 48 Diatomaceous earths, 52 ooze, 52 Diceras tondsalii, Sow., 267 Didelphys Azarae, Temm., part of lower jaw, 300 Didy mograpt us geminus, His., 412 Murchuonii, Beck, 412 pristis, His., 412 Didy moheli.r f err uginea, Ehb. sp., 49 Diestien of Belgium, 228 Dikelocephaliis minnesotensis, D. Ovv., 423 Diluvium, 112 Dimetian strata, 434 Dinotherinm giganteum, Kaup., 177 Dior it e, 517 Dip, defined, 84 measurement of, 86 Dip slopes, 101 Diplograptus (Phyllog rapt us) folium, His., 412' Dirt-bed of Purbeck, 290 Disintegration of rocks, 103 Distribution of forms of life, 131 Dogger, 275 bank, 114 Dolerites, 518 Dolomite, 29, 555 Dolomites of Alpine Trias, 329 Dolomitic Conglomerate, 319 reptiles, 320 Domites, 464 Double overfolding, 565 Downthrow, side of fault, 94 DREW, F., 40, 272 Drift, 112 EMMONS DUFREXOY, 553 DUNCAX, DR. P. M., 364 Duuites, 518 Dura Den, Fife, fossil fish of, 389 Durness limestone, 428 DUROCHER, 488 DUTTOX, CAl'T., 125 Dyas, origin of term, 338 Dykes, volcanic, 475 volcanic, connection with lava-flows, 475 dimensions of, 477 volcanic, age of, 484 - - altering strata, 477 action of denudation on, 476 columnar, structure in, 476 composite, 477, 527 of Anglesea, 477 of Antrim, 477 of Arran, 476 of Cleveland, 492 of Eskdale, 492 in Madeira, 476 of Palagonia, Sicily, 495 in Stye, 476 rocks, defined, 511 Dynamo-metamorphism, 538, 542 Dynamo-metamorphic action, experimen- tal illustrations of, 549, 550 EARTH'S crust, defined, 7 proportion to whole globe, 597 Earthquakes and faults, 96, 609, 610 EBELMEN, 27 Echinoconus conicus, Breyn., 257 Echinocorys vulgaris, Breyn., 257 with crania attached, 46 Echinodermata of the Cretaceous, 250 Jurassic, 277 Trias, 311 Carboniferous, 352 Silurian, 399 Ordovician, 413 Echinosphcerites balticus, Eichw., 412 EDEN, CAPT., 487 Edge-coal of Scotland, 371 EDWARDS, F. E., 204, 207 , MILNR, 225 Egeln, marine clays of, 228 EGERTON, SIR P., 207, 347 EHREXBKRG, 48 Eifel, Pliocene volcanoes of, 496 Eifelian, 374 Eigg, pitchstone porphyry of, 492 Ejected fragments, volcanoes made up of, 497 , of Lava. 456 } Elba, Tertiary granites and gabbros, 531 Elephants, ancestry of, 604 Elepnai antiquus, Falc., 151 meridionalis, Nesti, 151 primigenius, Blumeub. (Mammoth), teeth of, 151 Elevation-craters, 474 proofs of, 75 Elgin, reptiles of, 318 Elk, teeth of, 152 Emarginula (Rimula) clathrata, Sow., 399 EMMOXS, 419 630 INDEX ENCRINUS Encrinus liUiforrtiis, Schloth., 312 Eustatite, characters of, 604 Audesites, 463 Entomis serratostriata, Sandb. sp., 377 Eocene, defined, 144 Flora of Arctic Regions, 201 - - of United States, 241 - - of Western Territories of United States, 242 Eogene, defined, 145 ' Eolithic ' flint implements, 158 Eozoic strata, 437 Eozoon canadense, Daws., 74, 437 Eparchian strata, 437 Epidiorite, 517 Epigene rocks, defined, 15 Eppelsheim strata, 229 Equisetum arenaceum, Scliimp., 315 Equus cabal lus, L., 152 Erect fossil trees, 60 Brian formation, 396 Erratics near (Jhiehester, 168 Escarpments, 101 difference from sea-cliffs, 108 Eschara disticha, G-oldf., 259 oceani, D'Orb., 260 Extheria minuta, Alberti sp., 317 - ovata, Lea sp., 333 ETHERIDGE, 210, 386, 387 Etna, 494 ETTIXGHAUSEN, BARON VON, 214 Eucrite, 518 Euomphalus pentangulatus, Sow., 354 Eurypterida of Devonian, 378 EVANS, SIR J., 171 EVERETT, PROF., 13 Evolution in Geology, 595, [25] Exogenous growth illustrated, 358 Exogyra virgula, Defr., 293 Experimental illustration of cleavage, 546 Experiments illustrating metamorphic action, 549-550 Extinction of species, 448, 603 Extracrinus Briareus. Mill, sp., 307 FALCONER, 160, 288, 289 Falls of Niagara, age of, 592 False-bedding, 37 Faluns of Bordeaux, 226 of Touraiue, 226 Falunian, defined, 172 Fan structure, 565 Fan-taluses, 40 Farewell rock (Millstone grit), 368 Fascicularia aurantium, M. Edw., 188 Fault-rock, 94 Faults, how produced, 94 produced by repeated movements, 98 and earthquakes, connexion of, 96 amount of throw of, 97 ordinary, 94 reversed, 94, 565 shifting veins, 570 concealed by denudation, 98 Faunas, defined, 137 Fauna of American Devonian, 394 of Lower Cambrian, 424 of Cambrian, number of species, 420 of Forest bed of Cromer, 184 FORTIS Favositfs gothlandica, Lam., 399 (I'achypora) cervicornis, Blainv., 375 FAVIIK, 581 Faxoe beds, 327 Felis tigris, L., 153 Felspars, characters of, 652 Felspathic sandstone, 26 Felspathoids, characters of, 602 Fenestella retiformis, Schloth. sp., 344 Ferro-magnesian silicates, 603 Fifeshire, volcanic rocks of, 503 Fire clay, 27 Fishes, age of, 447 Fish, classification of, by forms of tail, 346 of the Older Tertiaries, 196 Cretaceous, 260 Jurassic, 278 - Trias, 314 Permian, 345 Carboniferous, 356 Devonian, 380-382, 607 of Silurian, 403 of Ordovician (?), 415 Fissures in whjch veins are formed, 569 FITTON, 266, P73, [20], [30] Flagstone, 2G Flat coals of Scotland, 371 ' Flecktschiefer,' 555 Flint, nature of, 255 origin of, 255 tabular, 255 implements. Neolithic, 157 . Palaeolithic, 156 Floras, defined, 137; origin of, COS Flora of Bear Island, 394 marine Cambrian, 421 of Devonian, 384 of Carboniferous, 357 of Permian, 341 of Cretaceous in Arctic regions, 332 of Upper Cretaceous, 327 of Newer Tertiaries, 179 of Forest bed of Cromer, 183 F LOWER, SIR W., 187 Fluidal structure, 460 Fluvio-marine Crag (Norwich Crag), 185 Foliated structure, 2u Foliation, nature of, 547 Folkestone beds, 266 Footprints, 43 in Sandstone, 321 and snow-cracks in Carboniferous Sandstone, 367 of Dinosaurs (?), Connecticut, 332 Foraminifera of Newer Tertiaries, 176 of Older Tertiaries, 195 of Cretaceous, 250 of Carboniferous, 350 of Ordovician, 412 FORBES, DAVID, 547, 549, 587 , EDWARD, 7, 147, 189, 190, 202, 204, 206, 212, 218, 267, 287, 530 Forest-bed of Cromer, 183 Forest marble, 296 Forfarshire, anticlinals and synclinals of, 80 Formation, defined, 16 of rock basins in glaciated regions, 169 FORSYTH-MAJOH, Dn... 234 FORTIS, 481 INDEX 631 FOSSILISATION Fossilisation, 68 preservation of minute structures, 71 Fossils, denned, 17 derived, 137 discussions as to nature of, 5 imperfection of, 450 wood, sections of, 70 oldest known, 438, 440 variation in number of indifferent kinds of rocks. 439 in Pre-Cambrian strata, 438, 440 FOSTER, DR. C. LE N., 272, 576 FOCQUE, PROP., 22, 466 FOURXET, M., 542, 573 Fox, REV. DARWI.V, 205 Fox-Hills group, 335 Fragments, derived, in Plutonic rocks, 527, 529 FRESHFIELD, MR. D., 169 Freshwater deposits, 51 strata, comparative rarity of, 439 alternation of with marine, 59 bivalves, 53 fish, 58 univalves, 54, 55 fossil plants, 57 FRITSCH, DR. A., 341 Froddingham ironstone, 307 Frost, effects of, 104 F ucoid beds, 425, 428 Ftilyar caniculatus, L. sp., 176 Fuller's Earth, 27, 302 Funafuti Boring, 601, 602 Fundamental gneiss, 434 Fungia patellaris, Lam., 351 Fii&ulina cylindrica, Fisch., 350 Fusus confrarius, Sow., 173 quadricostatus, Say, 1 76 GARRROS, 518 Tertiary of Scotland and Elba, 530, 531 Galeocerdo latidens, Ag., 211 Galerites albogalerus, Lam., 257 Gallionella distant, Ehb., 49 ' Garbenschiefer,' 555 GARDIXER, Miss L, 579 GARDNER, MR. STARKIE. 211, 213, 214, 491 Garnets, characters of, 605 Garnet-bearing rocks, 556, 561 Gas-cavities, 512 G istropoda of the Pleistocene, 148 Newer Tertiary, 174, 176 Older Tertiary, 208 Cretaceous, 251 Jurassic, 278 Trias, 312 Permian, 340 Carboniferous, 353, 354 Devonian, 377 Silurian, 402 Ordovician, 414 Cambrian, 421 GAUDIN, 211, 234 GAUDRY, PROP. A., 227, 236, 247 Ge-anticlinals, 566 GKIKIE, PROP. J., 102, 168, 171 SIR A., 116, 120, 125, 247, 370, 374, 388, 3*9, 391 . 43 i, 438, 475, 479, 492, 504, 505, 608, 547, 549, 566 GKANITES GEINITZ, 342 Gelinden, flora of, 220 GEMMELLARO, 338 Generalised types, 449 Geognosy, use of term, 14 Geology, origin of name of science, T and Cosmogony, 600 and Historv, 600 Geological enquiry, limits of, 596, 600 history, supposed limitation of period covered b'v, 5!)8 ; and supposed Primeval state of globe, 599 periods, relative duration of, 591 absolute duration of, 592 attempt to correlate with astro- nomical cycles, 593 time, duration of, 589 , 610 measures of, 592 record, imperfections of, 130, 601 map, first oi England and Wales, 136 of world, 432 Georgian strata. 420 Geo-synclinals, 566 Germany, Baltic Provinces, Older Paleo- zoic rocks of, 430 Gervillan anceps, Desh., 2f 6 (Avicula) socialis, Schloth., 313 GILBERT, G. K., 125, 169, 592 Girvan district, Cambrian of, 428 Ordovician strata of 418 Glacial deposits, 112 of North America, 245 - of Russia, 237 epoch, time since, 593 Period, 165 sands and gravels, 1G8 Glaris, slates of, 579 Glass-cavities, 512 Glauconie grossiere, 221 Glauconite sands, 27 Glenkiln shales, 419 Glen Tilt, Scotland, granite veins of, 524 Globiform pitchstone of Ponza Island, 483 Globigerina ooze, Globular structure in lavas, 482, 483 Globulites, 460 Glyptostrobus europceus, Heer., 183 Gneiss, 561 augen, 561 granites, 515 hornblendic, 561 micaceous, 562 pyroxene, 562 structure of, 547 GODWIN-AUSTEX, 138, 168, 257, 267, 590 Gold deposited by hot springs of Nevada, 569 origin of, in South America, 586 in California, 586 Goniatites crenistria, Phil., 355 Listeri, Mart., 355 GOPPERT, 71, 372, 394 Gothlandian, 398 Graham's Isle, 234, 474, 499 Grand Canary, volcanic rocks of, 500 Granites, with two micas, 515, 516 muscovite, 515 pyroxene, 515 hypersthene, 515 graphic, 514 632 INDEX GRANITES Granites, micropegmatitic, 516 mode of weathering, 522 of Arran, 522 of Land's End, Cornwall, 523 - uiicropegmatitic, of Mournc Moun- tains, 491 of Skye, 522 Tertiary, of Skye, Mull, &c., 491 of Tertiary age in Hebrides and Antrim, 491 Tertiary of Scotland and Elba, 530, 631 junction of, with Silurian Norway, 534 hornblende, 515 'Granitic' structure, denned, 513 Granitic veins, 524, 525, 526 sand, 26 Granitite, 515 ' Granopfiyre,' 516 Granulite, 561 Graphic structure, 514 Graphite, formation of, 558 Graptolites of Silurian, 398 of Ordovician, 412, 607, 608 of Cambrian, 421 Gravels, 26 Great Oolite, 296 GREEK, PROF. A. H., 63, 102, 116, 374 Green River series, 242 Greenland, sinking of west coast of, 77 cretaceous strata of, 332 Greensand, 27 Upper, 264 of Cambridge, 262 of New Jersey, 334 ' Greenstone,' 517 GREENWOOD, COL., 116 Gres bigarre, 325 de Beauchamp, 222 de Fontainebleau, 223 Greywacke, 26 Grey-wethers, 27 GRIFFITH, SIR R., 391 Grits, 26 Gryllacris lithanthraca, Goldenb., 365 Gryphcea incurva, Sow. (G. arcuata, Lam.), 54, 306 with Serpulae attached, 45 GUMBEL, 247, 517 GUNN, MRS., 256 GUTBIER, VON, 342, 347 GUTHRIE, DR., 550 Gwalior system, 438 Gyps bengalensis, Gm., tail, 286 Gypseous series of Montmartre, 222 Gypsum, fountain of, 31 HADE, defined, 94 HAGUE, MR. ARNOLD, 501 Hakea (?) salicina, Heer, 182 saligna, R. Brown, 182 HALL, CAPT. BASIL, 476 , SIR JAMES, 81 Hallstadt beds, 328 and St. Cassian areas, 328 Halysites catenularice, L. sp., 39! HAMILTON, SIR W., 473 Hamites spiniger, Sow., 264 Hampshire Basin, 270 HOOKE Hampshire and London Basin, table of strata in, 194 Hamstead (Hempstead) series, 203 BARKER, MR. A., 22, 466, 527, 536, 554, 563 HARKNESS, 322, 347, 438 Harlech grits, 427 Harpactor maculipes, Heer, 183 HARPE, DE LA, 211 HARRIS, MR. D., 247 , MAJOR, 322 HARTUNG, 500, 501 Hastings sands, 272 HATCH, DR., 22, 466, 536 HAUER, VON, 247, 387 HAUGHTON, DR. S., 74 Haiiyne basalts, 465 Headon series, 206 HEBERT, 220, 221, 247, 275 Hebrides, Tertiary Plutonic rocks in, 530 HEER, 63, 178, 201, 202, 211, 213, 231, 232, 394, 491, 500 Heersian of Belgium, 220 HEIM, PROF. A., 102, 565, 582 Heliolites porosa, Goldf., 375 Heliophyllum Halli, E. and H., 375 Helix hispida, Mull., 162 labyrinth-lea, Sow., 206 occlusa, F. Edw., 205 turonensis, Desh., 56 Helvetian series, 247 Hemare, meaning of term, 326 Hemicidaris purbeckensis, E. Forbes, 287 ' Hemicrystalline rocks,' denned, 513 Hemiptera of Newer Tertiaries, 182 Hcmitelites Brotmii, Gopp., 302 Hempstead (Hamstead) series, 203 HENRY, DR., 541 HENSLOW, 187, 290, 477 Hesperornis regalis, Marsh, 252 Heterocercal and Homoce real fish, 346 HICKS, DR. H., 409, 418, 419, 425, 426, 427, 428, 434, 437, 535 High-level gravels, 161 plateau gravels, 163 Hildoceras bifrons, Brug. sp., 304 Walcottii, Sow. sp., 304 Hils-thon, 268 Himalaya, fan-taluses of, 40 HINDE, DR. G., 419 HISE, MR. C. R. VAN, 437 Hipparion of Pliocene, 178 Hvppopodium ponderosum, Sow., 306 Hippopotamus major, Nesti. 152 Hippurite Limestone, 330, 331 Hippurites organisans, Desmoulins, 330 Histioderma hibernica, Kill., 422 History of Geology, 5 compared to Geology, 1, [50] HOFF, VON, [37], [38] HOLL, 435 HOLLAND, SIR T. H., 515 Hollybush sandstone, 42S Holocrystalline structure, defined, 510 Holoptychius nobilissimus, Ag., scale of } 382 restored, 381 Homalonotus armatus, Burm., 378 delphinocephalus, Green sp., 403 Homocercal and Heterocercal fish, 346 Homotaxy, defined, 138 HOOKE, 406 INDEX 633 HOOKER HOOKER, SIR J., 214 Hoplites Deshayesii, Leym. sp., 268 noricus, Schloth. sp., 269 Horizons, geological, 135 Hornblende, characters of, 604 Hornblende- Andesites, 463 HORNE, MR. J., 428, 438, 566 HORNES, PROF., 235 Hornstones, 462 Horse, ancestors of, 178, 04 - teeth of, 152 HOWELL, MR. H. H., 374 HUBBARO, PROF., 530 HUDLESTOX, MR. W. H., 438 HUGHES, PROF. T. McK., 427, 543 HULL, PROF., 322, 369, 374 ' Human period,' 147 ; antiquity of, 603 HUMBOLBT, 473 HUME, DR. W. P., 275 HUMPHREYS and ABBOTT, 120, 125 Hunstanton, lied Chalk of, 263 HUNT, ROBERT, 373 , STERRY, 540 Huronian strata, 436 BUTTON, 5, 595, [26], [30] HUXLEY, 14, 61, 138, 14*6. 281, 318, 322, 347, , [25], [50] Hycena spelcea, Goldf., 153 381, 382, 383, 595, 596, Hybodus reticulatus, Ag., spine of, 279 - - teeth of, 279 Hydrothermal action, 538, 540 Hymenocaris vermicauda, Salt., 424 Hypersthene, characters of, 604 Hypersthenite (Hyperite), 518 Hyperodapedon Gordpni, Huxley, 318, 319 Hypocrystalline structure, defined, 513 Hypogene rocks, defined, 15 --- - uniformity of character in, 584 Jlypsiprimnus (Potorous), tooth, 288 Hythe beds, 266 ICE-SCRATCHED surfaces, 167 Ichthyodorulites, 279 Ichthyornis victor, Marsh, 253 Ichthyosaurus cominunis, Conyb., skeleton restored, 280 IDDI.VUS, PROF., 490, 501, 601 Igneous rocks, analyses of, 536 Jgtianodon Bernissartensis, Boulenger, skeleton, 271 Mantelli, Meyer, teeth, 270 Ilfracornbe group, 386 Impressed and crushed pebbles, 91 Inchnadamff, Sutherland, section near, 436 Included fragments in strata, 133 as a test of age in volcanic rocks, 487 Inclusions in Plutonic rocks, segregative, 527 Incompleteness of the geological record, 442 Incrustations, 72 India, glaciers of, 241 Pre-Cambrian of, 438 Tertiary strata of, 236 Inferior Oolite, 301 of Yorkshire, 302 Inland sea cliffs, 115 Inliers, 101 KEUPER Inner Hebrides, volcanic phenomena of, 490 Inoceramus Lamarckii, Park., 258 Insects of Newer Tertiaries, 183 of Older Tertiaries, 201 Jurassic, 309 of Carboniferous, 365 Devonian, 385 Interbedded volcanic rocks, 479 age of, 485 Intermediate lavas, 462 Plutonic rocks, 516 International Geological Congress, scheme of nomenclature of strata, 146 Intrusions, volcanic, age of, 4S4 Inversion of strata, 92 Invertebrata, Age of, 447 Iron, Age of, 171 Iron and nickel in rocks, 519 IRVIXG, PROF. R. D., 437 Isastrcea oblonga, M. Edw., and J. Haime, 293 Ischia, volcanoes of, 493 Isle of Portland, dirt-bed in, 291 - of Wight, Upper Greensand of, 264 Isoclinal overfolding, 565 Isogeothermal lines, 13 Italy, Tertiary volcanic rocks of, 496 JOINTS, nature, 67 mode of formation, 68 Jointed basaltic columns, 480 JUKES, 68, 510 JUKES-BROWNE, Mr. A. J., 37, 275 Jura, Alpine blocks on, 239 structure of, 87 Jurassic, origin of name, 275 subdivisions of, 276 zones of, 325, 326 fauna, 277 fish, 278, 279 flora, 286 of Arctic Regions, 332 strata of Central Europe, 325 of Russia, 331 Plutonic rocks, 532 life-province, 328, 331 KAOLINISATION of felspars, 516 Karoo beds of South Africa, 331 KARPINSKY, PROF., 338, 396 KARRER, 235 Kasegrotte, Rertrich- Baden, basaltic co- lumns of, 482 KAUP, 320 KAYSER, 235, 247, 431 Keewatin strata, 437 KEILHAU, 547, 551 Kellaways Rock, 296 KELLER, DR. F., 240 KELVIN, LORD (Sin W. THOMSON), 12, 593 Kentish Rag, 266 Kent's Hole Cavern, 160 Kersantite (Kersantone), 517 Keuper, origin of name, 310 of Britain, 317 of Germany, 324 634 INDEX KEWEENAWAN Keweenawan strata, 437 KEYSERLJNG, Vox, 586 Kiltorcan beds, 391 Kimeridge Clay, 292 KING, 347 Kitchen-middens, 158, Klein-Spauwen beds, 229 KOEXEX, Vox, 207 KOXIXCK, DK, 229 Koninckia Leonhardi, Wissm., 312 Kbssen beds, 329 KOTO, B., 96 Krakatoa, eruption of, 467 Kupferschiefer as au ore deposit, 568 LABRADOR series, 437 Labradorite, characters of, 64, 602 Labyrinthodon, tooth of, 314 Jaegeri, 0\v., section of tooth, 314 LACROIX, M. A., 22 Lagena Moorei, Dav., 307 Lake-District, Ordovician strata in, 418 Lake-dwellings of Switzerland, 240 Lake-ores of Sweden, 568 origin of, 48 Lakes in glaciated region^, 169 LAMARCK, 57, 259, [39], '[53] Lamellibranchiata of the Pleistocene, 148 Newer Tertiary, 174 Older Tertiary, 208-18 Cretaceous, 251 Jurassic, 277 Trias, 312 Permian, 340 Carboniferous, 353 Devonian, 376 Silurian, 402 Ordovician, 412 Cambrian, 421 Lamination of rocks, 35 LAMPLUGH, 268, 275 Lamproyhyres, 517 Land, average height of, 122 barrier of Central England in Carboni- ferous times, 368 surfaces, ancient, 290 plants, order of appearance of, 448 Lapilli, 457 LAPPAREXT, PROP. A. DE, 247, 337, 398, 411 LAPWORTH, PROP. C., 410, 411, 419, 420, 424, 427, 428, 429, 437, 438, 549, 559, 566 Laramie formation, 335, 338 Lariosaurus BaUami, Curioni, skeleton, 315 LARTET, 221, 227, 240 LASAULX, Vox, 22, 521 Laxtcera stiriaca, Ung., 199 Laurentian strata, 436 Upper, 437 t- Lower, 437 Lavas, nature of, 456 structure of, 459 chemical composition of, 458 acid, 461 intermediate, 462 basic, 464 solfataric action on, 458 streams, ropy surface of, 456 LODES Lava- cones, nature of, 471 LAWSOX, PROF., 527 Leaf beds of Antrim, 213 of Mull, 212 LE COXTE, 584 Leda amyydaloidea. Sow., 213 Desfiayesiana, Duch., 225 lanceolate Sow., 148 truncata, Brown, 14t) LEE, J. E., 407 LEHMANX, J., 550 LKIDY, 197 Lenticular forma of strata, 39 Leperditia injiata, Murch. sp., 355 Lepidodendron corriigalttm, 384 Griffithsii, Brong., 383 Sternbergii, Brong., 3G2 Lepidostrobus ornatus, Brong., 363 Lepidotus gigas, Ag., scales, 278 Mantelli, Ag., 273 Leptcena depressa, Sow., 401 Leucite, characters of, 603 basalts, 465 Leucitite, 465 LEVY, M. MICHEL, 22, 463 Lewisian strata, 434 Lias, Upper, 304 Middle, 304 Lower, 305 White, 308 LlEBIG, 31 Life forms, accident in discovery of, 444 order of appearance of, 443 predominance at different periods, 446 provinces of past times, 449 Lignite, composition of, 32 Lignitic formation, 335, 337 Lima giganlea, Sow., 306 Hoperi, Sow., 263 Limestone, alteration of by Granite, 552 Limestones, 28 Limncea caudata, F. Edw., 206 fun if or mis, Sow., 205 longiscata, Brong., 54 LIXDSTROM, PROF., 410 Lin a la Crednerii, Gein., 345 .sumortieri, Nyst., 189 Lewisii, Sow., 402 flags, 425 Lingulella Davisii, M'Coy, 421 ella, H. and W., 424 LINK, M., 321 Liparites, 461 Liquidambar europwum, var. trilobatum, A. Brong., 181 Liquid cavities, 512 Lithoidites, 461 Lithostrotion basalt if or me, Phil, sp., 35] Littoral deposits, 113 Lihiites gigantem, J. Sow., 402 Living tree ferns allied to those of Car- boniferous, 359 Llanberis slates, 427 Llandeilo beds of England, 417 Llandovery formation, 404 Upper, 409 Lower, 409 Loch Coruisk, Skye, 521 Lodes or metalliferous veins, 56 INDEX 635 LODGE LODGE, PROF. 0., 594 Loess of Northern Asia, ]64 of Rhine, &c., 163 Loo AX, SIR W., 393, 436, 437 London /iasin, section of, 193 London Clay, 214 flora of, 214 marine shells of, 215 and Hampshire Basins, table of strata in, 194 Longmyndian strata, 435 LOXSDALE, 136, 244, 374 Lonsdaleia floriformis, Mart, sp., 351 LOSSES, 582 Lorn, SIGXOR, 530 Low-level gravels, 1C1 Lower Greensand, 249, 265 Lower Lias Clay and Limestone, 305 Lower London Tertiaries, 194 Lncina serrata, Sow., 211 Ludian series, 247 Ludlow formation, 403 Upper, 405 Lower, 406 Lustre-mottling, 461 Lutetian (Parisian) series, 247 Luxullianite, 516 LYCKTT, DR., 298 Lycopods, composition of, 33 Lycopod spores, composition of, 33 Lycopodiiim denmim, Labill., 362 Lydian stone (Lydite), 28 Lyntou group, 387 MAARE of Eifel, 497 MACCULLOCH, Uu. Jonx, 435, 525, 530, [33] McCiTj.Lonr, .Mit. JAMKS, 437 McGEE, 592 MCMAHOX, GKX., 535 Madeira, dykes in, 476 leaf beds of, 499 volcanic rocks of, 499 Mn< fax pianila. Sow., 257 Vkigdalenean Period, 246 Magma basalts, 465 Magnesian limestone, 29, 343 concretionary structures in, 66 : fossils of. 341 Magnetite, characters of, 605 MALLET, R., 456, 567 Malm, 275 ' Malvernian ' rock?, 435 ' Mamelon ' in Bourbon, 471 internal structure of l 471 Mammals, Age of, 447 ; ancestry of, 600 of Older Tertiaries, 197 of Pur beck, 288 of Stonesfield slate, 300 Mammalian fauna of Pikermi, 236 of the Sivalik Hills, 236 of the Western Territories of the United States, 242-245, 605 of the Limagne, 224 of Montmartre, 223 of the Mesozoic, 334 Mammoth, Age of, 170 MAXKREDI, 119 MAXTKLL. I)K., 205, 269. 272, ( [34] Mantrllia nidiformis, Brong., 291 MICAS Map of Eocene areas in North-Western Europe, 192 MAHCOU, 325, 419, 432 Maryarites, 460 Marine beds of the Coal-measures, 373 currents, effects of, 114 denudation, 112, 602 strata more j.i'equenWy preserved than freshwater, 439 Marl-slate, 345 fossils of, 345 Marlstone, 304 rock-bed, 304 Marnes irisees, 311 Marnes a virgules, 294 MARR, MR., 419, 425, 429 MARSH, PROF. 0. C., 197, 242, 244, 251. j{J2, 253, 254, 283, 284, 335, 337 MARSHALL, PROF., 63 Marsupites Milleri, Mant., 257 Mastodon arvernensis, Croin. et Job., 177 Mastodonsaurus Jaegeri, Meyer, 314 Mayeuce basin, Oligocene of, 229 Klein Spauweu, 229 MAYER, KARL, 500 May-Hill series, 404 Measures of geological time, 592 Mechanically formed aqueous rocks, 24 Mediterranean life-province, 328, 331 series, 247 Medlicottia Wynnei, Waagen, 340 Megalodon cucttllalus, Sow., 377 Afegalosattnu Bucklandi, Meyer, skeleton restored, 282 Melania (Melanatria) inquinata, Defr., 217 turritissima, Forbes, 204 Melanopsis buccinoidea, Fer., 55 Melaphyres, 465 Melbourne rock, 261 Melilite, 603 basalts, 465 Mencviau beds, 425 Mesozoic, denned, 127 strata, correlation of, 336 Messinian series, 247 Metamorphic rocks, 21 nature of, 537 derived from aqueous, 563 derived from igneous, 563 disturbed condition of, 578 analyses of, 588 order of succession of, 583 tests of age of, 577 different ages of, 577 of Pre-Cambrian age, 583 of Older Paleozoic age, 583 of Newer Palaeozoic age, 582 of Mesozoic age, 582 of Tertiary age, 581 Metamorphism, different kinds of, 538 contact, extent of, 553 - produced by dykes, 477 general or regional, 538 Metasomatic changes in rocks, 14 Meteorites, 519 Meudon, chalk of, 327 marls of, 220 MKYKR, HERMANN vox. 325, 365 MIALL, PROF., 63 Micas, COS 636 INDEX MICA Mica andesites, 463 schist, 561 slate, 561 syenite, 517 traps, 517 Micaceous minerals, 606 sandstone, 26 sandstones and shales, formation of, 35 Micraster cor-anguinum, Leske, 257 with Serpula attached, 46 Microcline, characters of, 602 Microconchus (Spirorbis) carbonarius, Murch., 355 Microlestes antiquus, Plien., 309 Microlites, 460 ' Microliticfeit,' 461, 463 Middle (or marine) molasse, 231 Lias, 304 Permian of Britain, 343 Midford sands, 301 MilioUna seminula, L. sp., 195 Miliolite limestone of Paris Basin, 222 MILL, DR. H., 14 MILLER, HUGH, 322, 382, 383, 389, 391, 392, 492 Millet-seed sands, 25 Millstone grit of South Wales, 368 in Staffordshire, 369 in Derbyshire and Yorkshire, 369 of Scotland, 369 MILNE, PROP. J., 104 Minerals which replace substance of fossils, 72 produced by contact metamorphism, 554 Of heavy metals, 606 occurring as ores, 606 rock-forming, 601 secondary, 605 Mineral composition as a test of age in volcanic rocks, 487 veins, different kinds of, 568 infilling of, 573 varying width of, 573 cause of varying width of, 574 'riders ' and ' Horses ' in, 574 relative ages of, 585 age of, in Ireland, 585 Cornwall, 586 Glamorganshire, 586 Somersetshire, 586 Bohemia, 586 Minette, 517 Minute organisms building up rocks, 48 Miocene, defined, 144 absence in England, 191 of Italy, 232 of Vienna Basin, 235 of Western Territories, 243 Mississippi, material brought down by, 120, 121 MITCHELL, S., 212 - REV. HUGH, 384 SIR T., 241 Mitra scabra, Sow., 208 Modiola acuminata. Sow., 344 Moel Tryfaen deposits, 168 MOJSISOVICS, PROP, 328, 331, 337 Molasse, Upper, 231 Lower, 230 NEOCENE Molasse, Lower, flora of, 231 Mollusca in high-level glacial deposits, 168 fossil, value of to geologist, 14o of Mediterranean and Red Seas, 132 ' Monian ' rocks, 435 Monoclinal folds, 81 mountain chains, 564 Monograplus priodon, Gein., 398 Monte Nuovo, near Naples, 468, 493 Montian strata, 220 Montmartre, gypseous series of, 222 Monts Dome, Auvergne, volcanoes of, 474 MOORE, CHARLES, 571 Moor-rock, 369 Morea, Cretaceous volcanic rocks in, 507 MORTILLET, 246 MORTOX, DR., 334 Moss-agates, 72 Mountain-chains, different kinds of, 564 origin of, 566 sculptured by denudation, 567 age of, 584 of Archaean age, 585 of Palaeozoic age, 585 of Mesozoic age, 584 r- of Tertiary age, 584 Mountain limestone, 369 meal, 48 MOURLO.V, M., 327 Mourne-Mountain granite, 521 Mousterian Period, 246 Mudstones, 28 Mull, granites and gabbros of, 530 leaf -beds of, 212 plant-beds of, 491 Tertiary volcano of, 490 MURCHISON, 7, 136, 146, 317, 322, 338, 342, 388, 390, 392, 397, 398, 405, 406, 407, 410, 411, 434, 435, 438, 499, 503, 544, 447, 549, 566, 581, 586, [42] Murchisonian, 398 Murchifonia gracilis, Hall, 414 Murex vaginatus, Jan., 174 MURRAY, DR. J., 63 Muschelkalk, 311 of Germany, 324 Muscovite, characters of, GJ3 Musical sands, 25 Mutability of continents and oceans, 123 Myliobatis Edwardsi, Dix., palatal teeth oi', 210 Mylonites, 559 Mylothrites (Vanessa) Pluto, Heer,201 Myriapoda of Carboniferous, 364 Mytilus septifer, King, 344 NATHORST, PROP., 73 Natica clausa, Brod. & Sow., 148 Natica helicoides, Johnst., 176 Nature of fossils, 5 Natural gas, 373 oils, 373 Nautilus ce.itralix, Sow., 215 danicus, Schloth., 328 plicatus, Sow., 266 truncatus, Sow., 306 'Needles,' formation of, 115 Neocene, defined, 145 INDEX 637 NEOCENE Neocene of Western Territories, 244 Neocomian, 249 Upper, 265 Middle, 268 Lower, 268 strata of Alps, 330 Neogcne defined, 145 Neolithic Period, 240 implements, 157 Nepheline, characters of, 603 basalts, 465 syenite, 518 Nerinwa Goodhallii, Sow., 294 Nerinaean limestone, 295 Nerita conoidea, Lam., 221 costulala, Desh., 299 granulosa, Desh., 55 Neritina concava, Sow., 206 globitlns, Defr., 55 NKUMATE, 286, 331, 337 Neuropterous insect, wing of, from Lias, 309 Nevada, hot springs in depositing gold, 569 Nevadites, 461 (and Rhyolites) of Tardree, Antrim, 491 Newark strata of New England, 332 NEWP.EHUY, DK., 213, 335 Newer groups should be studied first, 141 Tertiaries defined, 148 strata, nomenclature, 171 classification, 173 fauna, 173 flora, 179 Palaeolithic Age, 170, 239 Pliocene of Sicily, 233 of Val d'Arno, 234 Newfoundland bank, 115 New Red Marls, 317 Sandstone, 317 NEWTON, E. T., 153, 184, 218, 318, 322 New Zealand, glaciers of, 241 NICHOLSOX, PROP. H. A., 419 NICOL, JAMES, 436, 438, 565, 583 Nipudites ellipticus, Bow , 214 Nofyyvrathia ciineifolia, Brong., 342 NOKTLINfJ, DR., 438 No Man's Lands, formation of, 115 Nomenclature of Pleistocene deposits, 147 of Trias. 310 of Silurian strata, 397 of Ordovician strata, 411 Norian scries, 437 Norite, 518 North Downs, Pliocene strata of, 189 North-West Highlands of Scotland, sections in, 436 Cambrian of, 428 Norwich Crag (Fluvio-marine Crag), 185 Novacuh'te, 28 Nucnla Cobboldice, 173 Nurnmulitic formation, 229 NummnUtfs Itpoigatus, Lam., 209 Puschi, D'Arch., 194 variolarius, Lam., 208 OBERMITTWEIDA Conglomerate, 435 Oblique lamination, 37 Obolella crassa, Hall sp., 424 OBE Obolus Apollinls, Eichw., 413 Obsidian, 460, 461, 462 Odonlopteryx toliapicus, Ow., 19 Oeningen strata, flora of, 232 strata of, 231 OKYNHAUSEN, VON, 525 Ogygia Buchii, Burm., 414 Oil shales, 28, 373 Old Red Sandstone, general characters, 388 . of England, 390 of Scotland, 388 Upper, of Scotland, 385 of Scotland, Middle and Lower, 389 of Ireland, 390 Older Tertiaries defined, 145 Tertiary Strata, Nomenclature of, 191 Classification of, 191 Faunas, 194 insects, 201 _ Floras, 199 - Pa'feolithic period, 170 Pliocene, flora of, 233 of Italy, 232 of Greece, 236 Oldhamia antiqua, Forbes, 73 radial a, Forbes, 73 Oldhamina decipiens, De Kou., 339 Oldhaven Beds, 216 Olenellus Beds, 427 armalHS, Peach, 424 Callavei, Lapw., 424 Lapworthi, Peach, 424 Olenus beds, 425 micrurus, Salt., 423 Oligocene, defined, 144 origin of term, 228 strata of the Hampshire Basin, 202 of Italy, 232 of Croatia, 234 of Western Territories. 243 Oligoclase, characters of, 602 Olivaflammulata, Lain., 174 Olivine, 604 gabbro, 518 rock, 518 Omphyma subturbinata, E. & H., 399 Onchus tenuistriatus, Ag., spine, 403 Oolite (roestone), 29 Oolites, Upper, 286 Middle, 294 Lower, 296 Great, 296 Bath, 296 Inferior, 301 Oolitic grains, forming at present day, 30 structure, 30 Opal, characters of, 602 Ophites, 518 Ophitic structure, 461, 518 OPPEL, 325 Orbicular structure, 515 ORBIGNY, A. D', 249, 275, 397, 411 Orbitoidal limestone of United States, 24 Ordovician, origin of name, 41 1 of Bohemia, 429, 431 of Scandinavia, 430, 431 of North America, 430, 431 Plutonic rocks, 533 Ore deposits, nature of, 568 638 INDEX ORE Ore deposits, classification of, 576 hypogene origin of, 568 theories of origin of, 576 Ores, minerals occurring as, 606 Oreodaphne Heerii, Gaud., 181 Organic remains, 130 that have died oiit do not appear again, 130 used in identifying strata, 136 in volcanic deposits, 486 Organically formed aqueous rocks, 24 Origin of Forest bed of Croruer, 184 Orohippus (ffyracotherium) of Eocene, 178 Orthis elegantula, Dalrn., 401 tricenaria, Conrad, 413 vespertilio, Sow., 413 Orthoceras (Endoceras) duplex, Wahlcnb., 414 laterale, Phil., 354 ludense, Sow., 402 Orthoclase porphyry, 517 Orthophyre, 517 ' Osborne and St. Helen's Series,' 206 Osteolepis restored, 381 Ost radon, defensive spine of, 210 Ostrea acuminata, Sow., 301 carinata, Lam., 263 (Exogyrd) columba, Lam., 263 deltoidea, Sow., 293 distorta, Sow., 287 expansa, Sow., 293 gregaria, Sow., 294 'Marshii, Sow,, 304 vesicularis, Lam., 258 Otodus obliquits, Ag., 211 Ottrelite slate, 556 Outcrop defined, 87 Outliers, 101 Ova of Crustaceans with plants from Devonian, 380 Overfolded strata, 565 Overfolding of strata, 92 Overlap, 41, 100 ef Chalk on lower strata, 102 Overstep of strata, 100 Overthrusts. 565 OWEN", 7, 154, 210, 218, 221 285, 289, 299, 321, 324, 331 Ox, teeth of, 153 Oxford clay, 295 Oxfordian, 295 Oxides found as rock-forming minerals, 605 PAGE, D., 380 Palceaster atperrimut, Salt., 413 Palcechinus gigns, M'Coy, 352 Palwocoma (Ophioderma) tenuibranchiata, E. Forbes, 305 Palaeolithic Period in Western Europe, 239 implements, different types of, 156 valley graves, 161 alluvial deposits, 161 Palceoniscns comptus, Ag., 346 elegans, Sedgw., 346 glaphyrus, Ag., 346 restored, 345 Palasontological evidence of duration of Geological time, 590 methods, limits of, 138 PHILLIPS Palaeontological method, cautions in use of, 136 history, summary of, 448 Paheophis typhceiis, 0\v., 209 Palieopteris hibeimicus, Schimp., 383 Palasozoic, defined, 127 : floras, (506 Palagonia, dykes of, 495, 41)6 Palagonite tuffs, 466 Paleocene beds, 220 Paludina lenta, Brand., 203 - - orbicularis, F. Edw., 205 vivipara, Brand., 54 ' of gravels and peat, 48 ' Pan PANDER, 381 Paradoxides beds, 425 bohemicus, Barr., 423 Davidls, Salt., 423 Paramoudra of Chalk, 256 J'arasmilia centralis, Mont. sp.,351 Parka decipiens, Flem., 379 PARRY, SIR E., 148 Patella rugosa, Sow., 299 PAVLOV, PROP., 272 PEACH, MR. B., 428, 438, 566 Pea-grit, 29 Peat, composition of, 32 rate of growth of, 63 mosses, 1 58 Pebidian strata. 434 Pecopteris elliptica, Bunb., 360 Pecten Beaceri, Sow., 258 cinctus, Sow., 268 crassitesta, Rom., 268 ulandicus, Mull., 148 Jacobceus, L., 175 (Janira) quinquecoslatua, Sow., 263 Pegmatite, 515 Pele's Hair of Hawaii, 477, 4G5 Pelites, 27 Penarth beds, 308 Peneroplis cylindraceus, Lam. sp., 195 PENGELLY, 160, 212 Pentacrimts Briareiis, Mill., 307 Pentamerus Knightii, Sow., 400 oblongus, Sow., 400 Pentremites cllipticus, Sow., 352 Peridotites, 518 Periods, geological, relative duration of, 443 Perlitic structure, 460 Permian, origin of term, 338 classification of, 338 marine fauna and flora, 338 strata of Central Europe. 393 of Ural Mountains, 393 of Alpine district and Sicily, 393 Permo-Oarboniferons, 339 Perna Mulled, Desh., 267 PKRRY, PROF. J., 593 Persistence of oceanic areas, 123, 602, 003 I Persistent types, 449 I Petrographical provinces, 490, 608 Pet rophiloides Richardsoni, Bowerb., 215 j Phacops (Asaphus) caudalus, Brong., 403 latifrons, Bronn., 378 I Phascolotherium Bucklandi, Brod. sp., lower jaw, 300 ! 'JIM. i, i PS, PROP. J., 10, 127, 309, 369, 408, 507, 544 J. A., 25, 527, 569, 576, 587 INDEX 639 PHLEBOPTERIS Phlebopteris contigua, Lind. et Hutt., 302 Pholadomya cuneata, Sow., 218 fidicula, Sow., 302 Phonolite, 463 of Wolf's Rock, 464, 491 Phorus extensm, Sow., 215 Phosphatic nodules of Crag, 187 J'hrtHjmoceras ventricosum, J. Sow., 402 Phyllites, 29, 560 Physa Bristovii, E. Forbes, 287 columnaris, Desh., 55 hiipnorum, L., 54 Physical evidence of duration of geological time, 589 Pier it es, 518 Pikerrni strata, 236 mammalian fauna of, 236 PINGEL, 77 Pipe-clays, 27 Pisolite, 29 Pitchstorres, 462 Pitchstone porphyry of Sandy Braes, Antrim, 491 Placodus (jiijas, Ag., 316 Plaisancian series, 246- llayiaulax Becklesii, Falc., tooth, 288 Falc., lower jaw, 288 Planorbis euomphalus, Sow., 206 discus, F. Edw., 205 Plants, classification of, 607 action of, in forming calcareous rocks, 24 ; flowering, origin of, 605 of Mull and Antrim, 491 of Stonesfield slate, 301 Platanus aceroides, Gb'pp., 181 Platystoma Suessii, Homes, 313 Pl,AYKAIH,5, 93, 119 Plectrodus mirabilu, Ag., jaw, 403 Pleistocene, defined, 147 deposits, nomenclature of, 147 fauna and flora, 148 subdivisions of, 164 mollusca, greater longevity than inol- lusca, 150 fauna, Arctic forms, 149 Southern forms, 149 mammalia, forms of teeth, 150 relation to existing forms in same area, 152 of Europe, 150 of South America, 154 of Australia, 152 birds of New Zealand, 154 Marine deposits, 164 Estuarine deposits, 164 Lacustrine deposits, 164 Plesiosanrus dolichodeirus, Conyb., skeleton restored, 280 Pleurotoma attenuata, Sow., 211 exorta, Brand., 57 Pleurotomaria carinata, Sow. sp., 353 granulata, Sow. sp., 303 ornata, Sow. sp., 303 Pliocene, defined, 144 Period, climate of, 190 strata of North Downs, 189 of France, 227 of Germany, 229 of Vienna basin, 235 - of Western Territories, 244, PRIMEVAL Plombieres, metamorphic action in baths of, 541 Plutonic rocks, 21 relation to volcanic, 510 analyses of, 536 - structure of, 510 ultra-acid, 518 . acid, 514 intermediate, 516 basic, 518 ultra-basic, 518 origin of, 509 segregative action in, 527 inclusions in, 527 ages of, 528 tests of age of, 529 mineral composition as a test of age in, 529 of Pre-Cambrian age, 535 of Ordovician age, 533 of Carboniferous age, 533 of Triassic age, 533 of Jurassic age, 532 of Cretaceous age, 531 of Tertiary age, 530 Pluvial period, 170 Plymouth Limestone, 386 Podocarya Bucklandi, Ung., part of fruit, 301 Polishing of rocks by ice, 167 Polypterus of Nile, &c., 381 Pontian series, 246 Ponza Islands, pitch stone in t 483 Porcellanite, 28 Porphyrites, 464 Porphyritic structure, 459 Porphyroides, 461 Portland Oolite, 292 sand, 292 Portrush, altered rocks of, 478 Post-pliocene, defined, 147 of Northern Europe, 237 - classification of, 237, 238 of North America, 245 Post Tertiary, defined, 147 Potamides cinctus, Sow., 55 Potsdarnian strata, 430 Potstones, 225 POWRIE, 384 PRATT, 205 Pre-Cambrian strata, Nomenclature of, 433 period, proofs of existence of life in,432 fossils in, 438 strata, 432 area occupied by, 432 of Brittany, 438 Plutonic rocks, 535 Pre-Glacial period, 171 Pressure, distribution of, in earth's crust, 14 in earth's crust at great depths, 539 effects of, in producing consolidation, PRKSTWICH, SIR J., 13, 111, 125, 156, 157, 163, 171, 187, 189, 190, 191, 215, 218, 221, 247, 595 Priabonian series, 247 PRICE, MR., 264 Primaeval state of globe a question not within limits of geological enquiry, 599 640 INDEX PRIMARY ' Primary ' rocks, 127 ' Primordial ' strata, 419 PRINZ, 584 Productus limestone of the Salt Range, India, 396 horridus, Sow., 345 semireticulatus, Mart, sp., 353 Propylites, 464 Proteaceous plants (?) of Eocene, 215 ' Proterozoic ' strata, 128, 437 Psammites, 25 Psammodus porosus, Ag., 356 Pseudo-bombs, 457 fossils, 72 Pseudocrinites bifasciaCus, Pearce, 399 Psilophyton princeps, Daws., 384 Pteranodon longicrps, Marsh, 251 Pterichihys restored, 282 Pterodactylus antiquus, Somm., 284 Pterygottis anglicus, Ag., 379 portion of back, 379 Ptychodus decurrens, Ag., 260 Pulmonata of Carboniferous, 364 Pumice, 462 Pumiceous structure, 460 Punfield series, 274 Pupa muscorum, Mull., 162 tridens, Drap., 56 vetusta, Daw., 364 Purbeck, dirt-bed of, 290 Upper, 287 Middle, 287 Purpura tetragona, Sow., 173 Purpuroidea nodulata, Y. and B., 298 Puy de Gome, Auvergne, 494 de Tartaret, Auvergne, 493 ' Puys ' of the Western Isles of Scotland, 492 Pygope (Terebratula) diphya, Col., 329 Pyrenees, foliated rocks in, 548 local metamorphism in, 553 Pyromerides, 461 Pyroxene granulite, 561 Pyroxenes, 604 Pyroxenites, 518 Pyrula reticulata, Lam., 189 QUARTZ, characters of, 601 andesites, 463 diorite, 517 felsite, 516 fel sites, 461 pantellerites, 462 ' Quartz porphyr,' 516 Quartz rock, 555 trachytes, 461 Quartzite, 556 ' Quaternary,' defined, 147 QUEXSTEDT, PROF., 325 Quicksands, 25 RADIOLARIAN ooze, 51 Radiolites foliaceus, D'Orb., 330 Mortoni, Mant., 259 radiosa, D'Orb., 330 Rain, effects of, 104 and hail prints, 43 ROCK Rain prints and casts in Carboniferous, 368 Raised beaches, 123 RAMSAY, 7, 97, 108, 109, llfi, 134, 169, 322, 347, 390, 417, 426, 438, 489, 503 RANCE, MR. DK, 264 Range of Animals and Plants in Geological time, 445 Rannoch, Moor of, granite, 522 Rastrites peregrinus, Ban-., 412 READE, T. MELLAHD, 125 ' Recent period,' 147 Record, geological, imperfection of, 139, 440 Red Crag, 185 unconformity to White Crag, 188 Sandstones, origin of, 321 Redruth, Copper lode at, 572 Regional metamorphism, 538 rocks formed by, 557-559 age of rocks formed by, 579 REICHER, 549 REID, C., 184, 218 Reindeer Period, 170, 239 Relative duration of Geological periods, 591 REXARD, PROF. A., 63 REXEVIER, PROF. E., 582 Reptiles, Age of, 447 Reptilia of Permian, 342 of Trias, 315 of Jurassic, 279 of Cretaceous, 254 of Eocene, 196 Retinites, 462 REUSCH, PROF., 560, 580 Rhabdoliths, 50 Rhabdospheres, 50 Rhsetic strata, 308 Rhamphorhynchus Muensteri, Goldf., re- stored, 284 Rhine, Prussia, volcanoes of, 497 Rhinoceros leptorhinus, Cuv., 152 megarhinus, Christol, 152 tichorhinus, Cuv., 152 RhynchoneUa navicula, So^-., 401 octopUcata, Sow., 257 plica/His, Sow., 257 spinosa, Sow., 302 Wilsoni, Sow., 401 Rhyoliten, 461 of Arran, &c., 492 of Tardree, 521 Richmond, Virginia, coal-beds of, 333 RlCHTHOFEN, Vox, 164, 488, 490 Richthofenia Lawrenciana, De Kon. sp., 339 Ripple marks, 42 Rissoa Chastelii, Nyst., 20;> ROBERTS, DR. R. D., 440 Rocks, defined, 8 classification of, 15 chronology of, 126 produced by contact metamorphism, 555 formed by regional metamorphism, 557-559 Rock-forming minerals, 601-606 classification of, 514 Rock-salt, origin of, 322 INDEX 641 ROCK Rock-sections, use of, 22 method of making, 22 ROGERS, H. D., 33, 84, 564, 566 , W. B., 564, 566 Ropy surfaces of lava streams, 456 ROSK, GUST A v, 515 ROSEXBCSCH, PROP., 22, 422, 514, 515, 516, 517, 518, 521, 536 Rostellaria (Hippocrenes) ampla, Brand., 215 Rotalia armata, D'Orb. sp., 184 ROTH, 236, 536 ROTHPLETZ, 30 Roth-todt-liegende, origin of term, 338 RtrcKER, PROF. A., 63 Rudistes in Chalk, 259 Rum, granites and gabbros of, 530 Running water, action of, 105 Rupelian of Belgium, 225 RUSSELL, PROP. J. C., 337 Russia, glacial deposits of, 237 Jurassic strata of, 331 Older Palaeozoic rocks in, 430 RUTLEY, Mil. P., 22, 466, 536 SABAL major, Ung., 200 Saccharoid limestone, 30, 556 Saddles and troughs, 81 St. Abb's Head, curved strata of. 81 St. Cassian and Hallstadt areav, 328 beds, 329 St. Davids, Cambrian of, 426 St. Erth's, Cornwall, Pliocene strata of, 190 St. Kilda, Tertiary volcano of, 490 granites and gabbros of, 530 Saliferous system, 310 SALTER, 418, 426 Salterella grit, 428 pulchella, Bill., 424 Salt-Range (India) formation, 396 Samotherium Boissieri, Forsyth Major, 179 SANDBERGER, DR. F., 229 Sands, composition of, 26 rare minerals in, 26 Sandstones, induration of, 26 Sandy Braes, Antrim, pitchstone porphyry of, 491 Sao hirsuta, Barr., 423 Sarmatian series, 246 Sarsen stones, 27 SAUSSURE, DE, 79 Saxicava rugosa, Lam., 148 Scandinavia, glacial deposits of, 237 Cambrian of, 430, 431 Ordovician of, 430, 431 Silurian of, 430, 431 Scaphites cequalis, Sow., 262 Scelidosaurus Harrisoni, Ow., skeleton re- stored, 281 Scelidotherium leptocephalum, On., 155 SCHEERER, 541 SCHEUCHZER, 231 Schiller structure, 460 Schist spoliation in, 548 Schistose structure, 20 Schizodus Schlotheimi, Gein. .344 hinge of, 344 Schladebach boring, 13 SCHMERLING, 160 SODA Scoliostoma, 313 Scoriae, 457, 458 Scoria-cones, nature of, 470 on lava streams, 470 Scotland, Ordovician strata of, 418 SCOTT, DR. D., 243, 608 SCOTT-RUSSELL, 113 SCKOPE, 116, 472, 473, 474, 480, 482, 483, 547, [38], [50] SCUDDER, PROF. S... 385 Sea cliffs, difference from escarpments, 108 . inland, 115 Secondary minerals, 605 ; rocks, 127 Section of London Basin, 193 SEDGWICK, 7, 66, 136, 33S, 397, 409, 411, 415, 418, 419, 425, 435, 477, 542, 543 SEELKY, PROF. H. G.,263, 317,331, 334, 337 ' Segregation veins,' 528 Segregative action in Plutonic rocks, 527 SKXARMOXT, 541 Senonian, 257 Septaria, 66 Septarien-Thon of Germany, 228 Sequence of volcanic rocks, 488 Sequoia Langsdorfli, Ad. Brong., 200 Serpentine, 604 rock, 519 Serpula attached to Micraster, 45 ' Serpulite grit,' 428 Shale, defined, 28 Sharp Tor, Cornwall, granite of, 522 SHARPS, D., 544, 547 Sheets, volcanic, 475 age of, 484 Shell mounds, 158 beds of the Canaries, 500 of Madeira, 500 Shifting of veins by faults, 570 Shineton shales, 425 Sicilian series, 246 Sicily, Newer Pliocene of, 234 Siebengebirge, 497 Sigillaria losvigatus, Brong., 363 Silica minerals, forms of, 601 ' Sills,' volcanic, 475 Silurian strata, nomenclature of, 396 Upper, 398 Lower, 411 of Bohemia, 429, 431 of Scandinavia, 430, 431 of North America, 430, 431 Siluro-Cambrian, 411 Siphonotreta unguiculata, Eichw., 41 3 Sicatherium giganteum, Falc. et Cautl , 179 Skaptar Jokul, eruption of, 467 Skeleton crystals, 460 Skelgill shales, 405 Skye, dykes in, 476 granite of, 522 granites and gabbros of, 530 Tertiary volcano of, 490 Slaty cleavage, nature of, 543 Slickensides, 91, 93 Small isles (Rum, Eigg, Muck, and Canna), Tertiary volcanoes of, 490 Smilax sagittifera, Heer, 182 SMITH, WILLIAM, 7, 136, 286, 296, 308, 309, 325, 397, [31] Snowdon, volcanic rocks of, 506 Soda rhyolites, 462 TT 642 INDEX SOISSONNAIS Soissonnais, Lignites of, 221 Solenastrcea cellulosa, Dune., 195 Solenhofen slate, 283, 285, 286 Solfatara, near Naples, metamorphic action at, 542 SOLLAS, PROF. W. J., 73, 528 SOLMS-LAUBACH, PROF., 74 SOPWITH, 88 Sopwith's models. 88, 89, 90 SORBY, H. C., 33,' 512, 541, 544, 545, 546,548 South America, rise of land in, 77 Wales coalfield, 33, 371 Sparagmites, 430 Sparmacian series, 247 Specialised types, 449 Species, appearance of new, 448, 604, 605 rapidity of change in, 449 extinction of, 448, 608 Speeton clay, 268, 269 SPENCER, PROF. J. W., 125, 169, 592 Sperenberg boring, 13 Sphcerexochus mums, Beyr., 403 Sphcerulites agariciformis, Blainv., 330 Sphenophyllum, erosum, Lindl. et Hutt., 361 Sphenopteris gracilis, Fritton, 274 Spherulitic structure, 460 incipient ', 461 Spirifera alata, Schloth., 345 ditiuncta, Sow., 376 glabra, Mart, sp., 353 mucronata, Hall, 376 trigonalis, Mart, sp., 353 Spondylus spinosus, Sow., 258 Sponge in flint, 260 Spore coals, 31 microscopic structure of, 61 Spotted slates, 28, 555 SPRING, PROF. W., 67, 549, 550 Staff a, columns of, 481 Stalactites, 24 Stalagmite, 24 Statical and dynamical pressure, 540 Statuary marble, 556 Stauria astrceiformis, M. Edw., 351 Steam-coal, composition of, 32 Step-faults, 96 Stephanoceras Braikenridgii, Sow. sp., 304 Humphriesianus, Sow. sp., 303 macrocephalus, Schloth. sp., 298 STEVENSON, 113 Stigmaria ficoides, Brong., 363 attached to Sigillaria, 363 surface of, 364 Stiper-stones, 418 Stone, Age of, 170 Stone-cavities, 512 Stonesfield slate, 299 mammals of, 299 plants of, 301 STOPPANI, PROF., 329 STRACHEY, SIR R., 18 STRAHAN, MR. A., 218 STRANGWAYS, MR. C. Fox, 309 Strata, mineral characters of, 129 superposition of, 128 lenticular forms of, 39 thinning out of, 124 consolidation of, 64 bending of, 90 fracture of, 93-95 TACHYLITE | Strata, inversion of, 92 overfolding of, 92 conf 'or mobility of, 99 un conf or mobility of, 99, 134 overstep of, 100 overlap of, 100 tests of age of, 128 included fragments in, 133 age of, 128 equivalent, 135 identified by organic remains, 136 characterised by fossils, 130 chronological sequence, 135 groups of, 133 breaks in succession of, 133 Stratification, defined, 17 forms of, 34 irregularities in, 35 Stratum (pi. strata), defined, 16 Striation of rocks by ice, 167 STRICKLAND, 317 Stricklandinia lirata, Sow., 401 lens, Sow., 401 Strike, defined, 85 measurement of, 86 Stringocephalus Burtini, Defr., 377 Stromatopora of Silurian, 398 of Devonian, 375 Strophomena depressa, Sow., 401 grandis, Sow., 413 Structure, axiolitic, 461 banded, 460 columnar, 480 ball-and-socket in basaltic columns, 480 diabasic, 461, 518 fluidal, 460 jointed columnar, 480 ophitic, 461, 518 orbicular, 515 perlitic, 460 pumiceous, 460 porphyritic, 459 spherulitic, 460, 461 of lavas, 459 STUDER, 581 Subapennine strata, 232 Sub-Carboniferous, 396 Submarine denudation, i.13 volcanoes, 474 Subsidence, necessary, for formation of thick deposits, 116 Succinea amphibia, Drap. (S. putris, L.), 54 oblonga, Drap.!. 162 SUESS, PROF., 235, 602 Sun, action of, in disintegrating rocks, 105 Sun-cracks, casts of, from Wealden, 273 Superga, strata of, 232 Sussex marble, 271 Sweden, rise of land in, 77 Syenite, 516-517 Syenite-porphyry, 517 Synclinal strata, 79 forming ridges, 90 Synthetic types, 449 TABLE of strata, abridged, 145 of Fossiliferous strata, British and Foreign, 450-454 Tachylyte, 465 INDEX 643 TACONIC 'Taconic ' strata, 419, 420 TA IT, PROF., 593 Talc schist, 561 slate, 561 Talchir beds of India, 396 Tarannon shale, 408 Tardree, Antrim, rhyolites of, 491, 521 Tealby series, 268 TEALL, MR. J. J. H., 22, 466, 536 Tellina balthica, L., 149 calcarea, Chem., 149 obliqua, Sow., 173 Temnechinus excavatus, Forbes, 189 Temperature of Earth's crust, 11,. 601 at great depths, 539, 601 Tentaculites annularis, Schloth., 402 Tephrrtes, 464 Terebellnmfusiforme, Lam., 208 sopita, Brand., 208 Terebratula biplicata, Brocchi, 258 carnea, Sow., 257 digona, Sow., 298 (Pygope) diphya, Col., 329 flmbria, Sow., 302 sella, Sow., 267 Terebratulina striata, Wahlenb., 257 Terebrirostra lyra, Sow., 263 Teredina borings in fossil wood. 47 personata, Lam. sp., shell and cal- careous tube, 47 Teredo borings in wood, 47 navalis, L., shell and calcareous tube, 47 Terraces in valleys, 111 Terrestrial deposits, 59 between Tertiary lavas of West- ern Isles of Scotland, 491 shells, 56 Tertiary rocks, 127 strata, classification of, 144 groups, discovery of, 142 volcanoes in British Isles, 490 in Europe, 492 - - Plutonic rocks in Western Isles of Scotland, 530 granites and gabbros of Western Isles of Scotland, 530 granite of Antrim, Mull, and Skye, 491 granites and gabbros of Elba, 530 granites, supposed, of Alps, 531 (?) metamorphic rocks, 581 Thamnastrcea arachnoides, Park, 294 Thanet sands, 21 8 Theca (Cleidotheca) operculata, Salt., 422 Thecodontosaurus, tooth of, 320 Thecodus parcidens, Ag., scales, 403 Thecosmilia annularis, Flem., 294 Theralite, 518 Thermo-metamorphism, 538 Thermometers for determining under- ground temperatures, 12 Thicknesses of strata of different ages in Europe, 441 THOMSON, Sm W. (LORD KELVIN), 12, 593 , DR. T., 230 THORPE, PROP., 32, 63 Throw of fault, 94 Thrusts ' and Thrust-planes,' 56 TROUGHS Thrust-planes in North-West Highlands of Scotland, 436 Thylacotherium Prevostii, Vale*nc., lower jaw and molar, 300 Tiger, teeth of, 153 Tilestones, 405 Till, nature and distribution of, 166 Time, geological measures of, 592 length of, 592 Tinoceras (Uintatherium) ingens, Marsh 198 Tithonian strata, 329 Toadstone of Derbyshire, 505 Tongrian of Belgium, 225 Torbanite, 28, 373 TORELL, DR. 0., 148 Torquay limestone, 387 Torridon sandstones, 435 Torridonian strata, 435 Tors of granite, formation of, 522, 523 Tortonian series, 247 Totteruhoe stone, 262 Touraine, Miocene strata of, 226 Trachyceras Aon, Munst. sp., 313 Trachytes, 463 ' Trachytic ' structure defined, 513 Tracks and burrows, 43 ' Transition ' rocks, 127 Transport of rock materials, 106 ' Trap rocks,' 20 TRAQUAIR, DR. R. H., 384 Trass, nature and origin of, 497 Travertine, 24 Tremadoc slates, 425 Trias, origin of name, 310 of Germany, 324 of the Alps, 328 of India, 331 of South Africa, 331 Plutonic rocks of, 533 Trichites, 460 Trigonia caudata, Ag., 266 gibbosa, Sow., 293 Trigonocarpum olivceforme, Lindl., 357 ovatnm, Liridl. et Hutt., 357 Trilobita of Middle and Upper Cambrian, 423 of Lower Cambrian, 424 Trilobites, organisation of, 416 - appendages of, 417 stages of growth in, 423 larval forms of, 423 of Permian, 342 of the Carboniferous, 356 of the Devonian, 378 of Silurian, 403 of the Ordovician, 414 Trinucleus concentricus, Eaton, 414 young forms of, 414 Trionyx, fragment of carapace of. 204 Tripoli, 48 Tritylodon Frassii, Lyd., 316 longcevus, Ow., 317 Trochoceras giganteus, J. Sow. sp., 402 Troctolite ('Forellenstein '), 518 Tronstadt Strand, Christiania, quartz vein at, 526 Trophon antiquum, Mttll., 173 clathratum, L., 148 Troughs and saddles, 81 644 INDEX TROWLESWOBTHITE Trowlesworthite, 516 Tuedian series, 371 Tuffs, volcanic, 458 fossils in, 458 rhyolitic, 462 andesite, 463 basaltic, 466 cones, nature of, 470 Tunbridge -Wells sand, 272 Tupaia Tana, Raff., lower jaw, 299 Turonian, 261 Turrilites costatus, Lam., 262 Turritella multisulcata, Lain., 211 TYLOR, A., 119, 170 TYNDALL, PROP., 546 Typhis pungens, Brand., 208 Tyrol, Plutonic rocks in, 533 ULTRA-ACID Plutonic rocks, 518 Uncites gryphus, Defr., 377 Unconformability, 99 significance of, 139 Uncomformable overlap, 100 Unconformity of strata, 134 Underclays of coal, 371 Underground temperatures, 1 1 UNGER, 63, 213, 231, 231, 394 Uniclinal folds, 81 , Uniformitarianism in geology, 595, [5 1 ] Unio littorulis, Lam., 53 . valdensis, Mant., 273 United States, Eocene of, 241 Tertiary volcanoes of, 501 UPHAM, W., 592 Upper barren measures of North America, 396 Greensand, 264 Lias clay, 304 sands, 3C4 Permian of Britain, 343 Upthrow side of fault, 94 ' Uriconian ' rocks, 435 Ursa stage, 396 Ursus spelceus, Blumenb., 153 USSHER, MR., 385, 391 V's formed by outcropping strata, 88, 89 Val d'Arno, Newer Pliocene of, 234 Valley gravels, 111 Valvata piscinalis, Mull., 54 VAX 'T HOFP, 549 Veins, contemporaneous, 528 formed in fissures, 569 granitic. 524 segregation, 528 shifted by faults, 570 quartz, 526 mineral, 568 infilling of, 573 varying width of, 573 Veinstones, nature of, 572 Ventriculites infundibuliformis, R. Wodw., 256 Vermilion-Creek Group, 242 VEUNKUIL, DE, 275, 342, 406, 586 Vertebrate fossils, order of discovery of, 444 Vertical strata, 79 Vesuvius, 493 VOLCANIC Vesuvius, summit of in 1767, 473 crater of in 1822, 473 and Somma, 468 Vicar y a Lujani, De Verneuil, 274 Vicentin, basaltic columns in, 481 Vienna Basin, Pliocene of, 235 Miocene of, 235 Vindhyan System, 438 Virginian Sands, 244 VlRLET, M., 507, 542 ' Vitrophyric ' lavas, 492 Vivarais, basaltic columns in, 480 VOGELSANG, 516 Vogesite, 517 VOGT, PROF., 527 Volcanic action, nature of, 455, 457 different kinds of, 466-467 effusive, 467 explosive, 467 activity, scene of, constantly shifting, 489 bombs, 457 dust, distance to which it is carried, 487 tuffs, 458 ' ash ' with trunks of trees, Arran, 501 dykes, 475 ' necks,' 475 sheets, 475 intrusive character of, 485, 48C - mountains, origin of, 467 form of, 467 rocks, contemporaneous, 479 interbedded, 479 order of appearance of, 488 rocks, 18 relation to other classes, 455 analyses of, 536 consanguinity of, 490 age of, 485 of Newer Pliocene age, 494 of Older Pliocene age, 496 of Miocene age, 498 of Oligocene age, 497, 498 of Eocene age, 499 of Tertiary age in United States, 501 in Australia, 501 of Secondary age, absence of, in Britain and Western Europe, 507 : of Cretaceous age in Greece, 507 of Jurassic age in Italy, 508 in Greece. 508 in India, 508 of Triassic age, 5U2 of Permian age in Scotland, 503 of Carboniferous age in Derby- shire, 505 in Ireland, 505 . in Scotland, 503 of Devonian age in Ireland, 506 in Scotland, 505 of Silurian age in Ireland, 506 of Ordovician age in North Wales, 506 in Lake District, 506 in Scotland and Ire- land, 506 of Palaeozoic age in Central Europe, 508 INDEX 645 VOLCANIC Volcanic rocks of Cambrian age in West of England, 507 of Pre-Cambrian age, 507 in Canada, 508 phenomena, age of, 483 tests of age of, 484 succession of, in Western Isles of Scotland, 490, '609 activity in British Isles, latest manifestations of, 488 in Tertiary times, 490 in Tertiary times in Europe, 492 Volcanoes, internal structure of, 469 basal wrecks of, 475 submarine, 474 of Anvergn/*, 469 Voltzia Jteterophylla, Brong., 315 Valuta ambigua, Sol., 208 at li Ma, Sol., 208 Lamberti, Sow., 174 nodosa, Sow., 215 selsiensis, F. Edw., 211 WAAGEN, DR. W., 338, 347, 396, 438 Wadhurst Clay? 272 WAUXKR, 236 Wahsatch Group, 242 Walchia piniformis, Schloth., 341 WALCOTT, MR. C., 407. 417, .431, 438, 592 WALLACE, MR. A. R., 595, [25] Warminster beds, 264 Water of Ayr stone, 556 WATTS, MR". W. W., 410 Weald clay, 270 Wealden Formation, 268 Weathering of rocks, 104 of granite, 522 WEBSTER, 142 Wemrnelian of Belgium, 222 Wenlock formation, 404 limestone, 407, 408 WERNER, 4, 127, 478, 572, [31] Western Isles of Scotland, Archfean Plu- tonic rocks of, 535 relation, of Plutonic to Volcanic rocks in, 520, 609 Wheeling (West Virginia) boring, 13 ZOOLOGICAL Whetstones, 28, 556 WHITAKER, 108, 191, 218 White or Coralline Crag, 187 Jura, 275 Lias, 308 WHITE, DR. C. A., 335, 337, 338, 396, 444 WHITNEY, PROP., 587 WHYMPER, MR., 201 WILKINSON, 587 WILLIAMS, H. S, 396 WILLIAMS, 571 WILLIAMSON, PROF. W. C., 302, 320, 357, 359, 609 WILLIS, MR. BAILEY, 568 Wind, effects of, 105 WITHAM, 70 Wolfs Rock, phonolite of, 464, 490 WOOD, SEARLES, JUN., 184, 185, 186, 189, 190, 191, 206, 207, 436 WOODWARD, MR. H. B., 191, 309 Woolhope Limestone, 408 Woolwich and Reading series, 217 WRIGHT, DR., 171, 309 WUNSCH, 503 XENOD1SCUS plicatus, Waagen, 340 Xiphodon gracilis, Cuv., 197 Xylobius Sigillarice, Daws., 365 YOREDALE series, 369 Ypresian series, 247 ZANCLEAN series, 233 Zechstein, origin of term, 338 Zmglodon cetoides, Ow., 199 beds of United States, 242 Zircon-syenite, 517 ZIRKEL, DR. P., 22, 4fil, 466. 530, 563 ZlTTEL, DR. K. VON, 612, [81] of Jurassic system, 325, 326 Zone of Avioala contorta, Portl., 306 Zones of Cretaceous, 205 ; of Jurassic, 325, 326 ; of Carboniferous Limestone, 607 Zonites (Conulus) prisons, Carp., 364 Zoological provinces, 131 Printed by Hazell, IVatson >k Viney, Ld., London and Aylesbury. THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW N INITIAL FINE OF 25 CENTS WILL BE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE. THE PENALTY WILL INCREASE TO SO CENTS ON THE FOURTH DAY AND TO $1.OO ON THE SEVENTH DAY OVERDUE. 8$ep ? 58C$i * " UIV APR 1939 REC'D LD APR SO 19 : 2 9 1962 LD 21-1 V n ' ' ' I b D 298291 UNIVERSITY OF CALIFORNIA LIBRARY