THE , STUDY OF ROCKS AN ELEMENTARY TEXT-BOOK OF PETROLOGY BY FRANK RUTLEY, F.G.S. u H. M. GEOLOGICAL SURVEY LI UK A \\ \ D. APPLETON AND CO. NEW YORK 1879 EARTH SCIENCE^ LIBRARY PREFACE, THE rapid advance of Petrological study during the last few years has rendered it imperative that some English text-book should be written for the guidance of students in this branch of science. Several good manuals of petrology have recently been published on the Continent ; but, hitherto, comparatively little has been done, in this country, to supply elemen- tary instruction in the systematic study of rocks. The application of the microscope, in this special branch of geology, has of late years afforded more precise information, concerning the mineral constitu- tion and minute structure of rocks, than it was pos- sible to acquire by the older methods of research ; and, in this book, I have endeavoured to give a clear explanation of the method of preparing sections of rock for microscopic examination, as well as a des- cription of the microscopic characters of the most important rock-forming minerals, upon the identifi- cation of which the determination of the precise character of a rock is necessarily based. I have been compelled to make very free use of foreign vi Preface. works on petrology, especially those of Zirkel, Rosen- busch, and Von Lasaulx, and I have also extracted much information from other foreign and British publications. In all instances I have endeavoured to indicate the sources from which the information has been derived, and in this respect I trust that no injustice has been done to any author. I am also greatly indebted to Professor A. Renard, of the Royal Museum in Brussels, for the revision of one or two chapters and for many useful hints. I would especially thank Professor John Morris for his kind- ness in voluntarily undertaking the revision of the entire work ; indeed I cannot adequately express my obligation to him for his friendly and valuable cri- ticisms. Professor Zirkel has kindly given me permission to copy some of the microscopic drawings from his works, and has also assisted me by sending me some of his publications. To Professors J. W. Judd, A. H. Green, and T. G. Bonney, and to Messrs. H. W. Bristovv, J. A. Phillips, H. Bauerman, S. Allport, W. Chandler Roberts, Trenham Reeks, E. Best, T. V. Holmes, and others, I am also indebted for information or for specimens which have helped me in the prosecu- tion of my work, while from the Editor, Mr. C. W. Merrifield, I have received a multitude of useful suggestions. My thanks are likewise due to Mr. E. T. Newton, Preface. vii Assistant Naturalist to the Geological Survey, for some interesting notes on his method of preparing microscopic sections of coal, and to Mr. J. B. Jordan, of the Mining Record Office, for the revision of the chapter on the preparation of microscopic sections, and for the use of the wood-block representing the construction of his section-cutting machine. In the classification I have to some extent deviated from the systems commonly adopted ; and, in the general treatment of the different subjects, original ideas and observations are more or less plentifully interwoven with the information derived from books. From the limited size of this work I have neces- sarily been compelled to treat certain portions of the subject with brevity, but I trust that nothing of importance to the student has been omitted. I have intentionally avoided anything more than casual references to localities where particular rocks occur, and have preferred to devote additional space to the descriptions of the typical rocks themselves. If I appear to have entered too much into micro- scopical details, I can merely observe that nature makes no difference between great and small ; that the great features which diversify the earth's surface, and which Appear stupendous to our finite perceptions, are absurdly trivial when compared with the dimen- sions of the globe itself, while the latter, in relation to the sun, is a mere speck. Minute structure and gross are alike governed in their development by the viii Preface. same natural forces, which are giants commanding legions of atoms, and these hosts of pigmies constitute the world. If the present power of assisting vision were amplified thousands of times, we should probably find similarly perfect results, governed by the same laws, the general principle of which seems remotely hidden in those fields of inquiry, which fade away on every side into the regions separating human reason from Omniscience. In conclusion, I may add that petrology cannot be learnt merely by reading, and that this little work does not pretend to be more than a rudimentary guide to the subject. F. R. CONTENTS. PART I. THE RUDIMENTS OF PETROLOGY. CHAPTER I. METHOD OF RESEARCH, ETC. PAGE Introductory remarks on methods of petrological research, and on the scientific and practical value of observations on the chemical and mineralogical constitution of rocks I CHAPTER II. ROCKS DEFINED AND THEIR ORIGIN CONSIDERED. Rocks are mineral aggregates Conditions of aggregation Preli- minary considerations bearing on the origin of rocks . . 6 CHAPTER III. DISTURBANCES OF THE EARTH'S CRUST STRUCTURAL PLANES SEDIMENTARY ROCKS STRATIGRAPHY. Subterranean forces Vulcanicity and Seismology Evidence of the internal heat of the earth Fissuring and displacement of rock-masses Structural planes Faults Joints Laminar fission Cleavage Columnar and spheroidal structure Rocks divided into sedimentary and eruptive groups Stratification Contents. PAGE Mode of formation of sedimentary rocks Clays, sands, lime- stones, &c. Classification of sedimentary rocks Dip and strike of beds Flexure of beds The influence of dip and erosion on the geological map of a country Measurement of the thickness of beds Denudation, marine and atmospheric Cliffs and escarpments Weathering of rocks dependent on relative hardness Hills and valleys Erosion of limestones Weathering of eruptive rocks Formations Palseontological considerations Unconformities . . . . -9 CHAPTER IV. GENERAL CHARACTER AND MODE OF OCCURRENCE OF ERUPTIVE ROCKS. Volcanic and plutonic rocks Basic and acid rocks Origin of slaty cleavage Foliation Metamorphism Definition of a volcano Volcanic phenomena . . . . ... 32 CHAPTER V. THE COLLECTING AND ARRANGEMENT OF ROCK SPECIMENS. On the collecting, dressing, labelling, and arrangement of rock specimens .......... 39 CHAPTER VI. PRELIMINARY EXAMINATION OF ROCKS. Hardness tests Description of needful apparatus, &c. . . 44 CHAPTER VII. THE MICROSCOPE AND ITS ACCESSORIES. Microscopes suitable for petrological work Points essential in the construction of, and needful apparatus for such microscopes Goniometric measurements Stauro-microscope of Prof. Rosenbusch Inverted microscope . . . .46 Contents. X1 CHAPTER VIII. METHOD OF PREPARING SECTIONS OF MINERALS AND ROCKS FOR MICROSCOPIC EXAMINATION. PAGE Description of needful materials and apparatus Preparation of chips and slices Preliminary grinding Cementing to glass Grinding slab Second grinding process Final grinding Re- moval of section from grinding slab Cleansing and mounting process Preparation of sections of very soft rocks Mr. E. T. Newton's method of preparing thin sections of coal . . 59 CHAPTER IX. ON THE EXAMINATION OF THE OPTICAL CHARACTERS OF THIN SECTIONS OF MINERALS UNDER THE MICROSCOPE. Phenomena of polarisation Stauroscopic phenomena . . -74 CHAPTER X. THE PRINCIPAL ROCK-FORMING MINERALS : THEIR MEGA- SCOPIC AND MICROSCOPIC CHARACTERS. Species of the felspar group Nepheline Leucite Scapolite and Meionite Sodalite, Hauyne, and Nosean Olivine Hyper- sthene Enstatite Bronzite Species of the pyroxene group Species of the amphibole group Species of the mica group Chlorite Talc Tourmaline Epidote Sphene Species of the garnet group Topaz Zircon Andalusite and Kyanite Apatite Rutile Cassiterite Calcspar Quartz, &c. Magne- tite Titaniferous iron Hematite Limonite Ironand copper pyrites Zeolites Viridite Opacite, &c 86 xii Contents. PART II. DESCRIPTIVE PETROLOGY. CHAPTER XI. THE CLASSIFICATION OF ROCKS ERUPTIVE ROCKS, CLASS I. : VITREOUS ROCKS. PAGE Description of structures developed in vitreous rocks Obsidian- Pumice Perlite Pitchstone Tachylyte . . . .174 CHAPTER XII. ERUPTIVE ROCKS, CLASS II. : CRYSTALLINE ROCKS. Granitic group Syenite group Trachyte group Phonolite group Andesite group Porphyrite group Diorite group . . 202 CHAPTER XIII. ERUPTIVE ROCKS, CLASS II. : CRYSTALLINE ROCKS (cont.). Diabase group Gabbro group Basalt group Rocks of excep- tional mineral constitution, including garnet rock, eklogite, Iherzolite, &c. Volcanic ejectamenta Altered eruptive rocks. 244 CHAPTER XIV. SEDIMENTARY ROCKS. Unaltered series : Arenaceous, argillaceous, and calcareous groups Altered series, including porcelain-jasper : chiastolite- and Staurolite -slates, quartzite, gneiss, granulite ; mica-, chlorite-, talc-, hornblende-, and other crystalline schists Conglo- merates, breccias, tufas, and sinters Mineral deposits con- stituting rock-masses ........ 274 APPENDIX . . . . 307 ERRATA 312 INDEX ........... 313 Ll HK;:-/- ,;- : NlVKi^lTV'oK STUDY OF ROCKS. PART I. THE RUDIMENTS OF PETROLOGY. CHAPTER I. METHODS OF RESEARCH, ETC. THE means at the disposal of the older petrologists for iden- tifying the mineral components of fine grained or minutely crystalline rocks were so primitive, that we wonder, not so much at the little that was known about them, as at the quantity of information amassed by such simple methods, and at the truth or comparative accuracy of many of their statements bearing directly upon this subject. The pocket lens was one of their most important implements in this work, and was indeed the only means they possessed for distinguishing minute structure; for although compound microscopes were known and used for physiological pur- poses, still the idea of slicing and grinding down fragments of rock into thin sections had not at that time occurred to anyone, or, if it had done so, had at all events never been carried into practice. Chemical analysis and simple tests of hardness, specific gravity, &c, such as have been given in treatises on mineralogy for considerably more than half a B 2 The -Riidiments of Petrology. century, \vere the other, methods which they were enabled to call in to their assistance ;' but. chemical analysis of aggregates of minute and undetermined minerals served only to throw a very imperfect light upon the precise nature of the com- ponent minerals themselves, and in this way rocks which differed widely in minute structure and in mineral composi- tion often yielded almost identical results so far as their ultimate chemical composition was concerned, while a know- ledge of the physical conditions which governed them at the time of their deposition or solidification were matters which could only be inferred from observations made in the field, upon their mode of occurrence and relation to other .rocks. Vague speculations were discussed with an energy which shows how deeply these pioneers of geology were interested in this branch of their science, but from the difficulties which attended the successful prosecution of these inquiries, especially so far as the eruptive rocks were concerned, the study of petrology as a special geological subject seemed to lapse from a state of misplaced energy into one of hopeless torpor. The first steps which in this country tended to beget fresh ardour in this direction were the publication of a paper by H. C. Sorby, in the ' Quarterly Journal of the Geological Society of London,' vol. xiv. p. 453, 4 On the Microscopical Structure of Crystals, indicating the Origin of Minerals and Rocks,' and a short article by the late David Forbes, in the ' Popular Science Review,' Oct. 1867, entitled 'The Microscope in Geology.' The obser- vations recorded in these papers were based upon the microscopic examination of thin sections of minerals and rocks ; and although Mr. Sorby appears to have been the first to apply this kind of examination to purely mineralogical and petrological questions, still the method of grinding such thin sections for microscopic work was first practised by H. Witham in 1831, when conducting researches on the minute internal structure of fossil plants. The great advantage derived from the examination of thin sections of minerals Microscopic Examination of Rocks. 3 lies in the circumstance, that in many cases a mineral which in ordinary hand-specimens, in thick splinters or in thick slices, would appear to be opaque, is rendered more or less translucent or transparent, the transparency increasing with the thinness of the section, so that structures which in many instances could not be discerned by reflected light are rendered apparent when the specimen is thin enough to be moderately translucent, while, in conjunction with the micro- scope, the polariscope, spectroscope, and goniometer may be used, and additional facilities are thus given for examin- ing the optical properties of the mineral. Such advantages accrue from the preparation of thin slices of rocks, the com- ponent minerals of which may be studied microscopically, their crystallographic, optical, and other physical properties noted, even in rocks whose texture is so fine that examina- tion with ordinary hand lenses is insufficient to give any clear insight as to the .nature of their components, while opportunities are thus offered for acquiring much information about the paragenesis of minerals, the physical conditions under which rocks have been formed, and the changes both physical and chemical which they have subsequently under- gone. It is therefore manifest that the preparation of thin slices of minerals and rocks has led the geologists of the present day into a vast and hitherto unexplored field of inquiry, in which new questions will propound themselves, and old ones, in time, find their solution. The benefits to science which are likely to accrue from a steady prosecution of such studies are, from the foregoing remarks, sufficiently obvious, but many people may be prone to think that little practical advantage is likely to be derived from this branch of microscopic research. Such objectors, if asked whether the texture and quality of build- ing stones, bricks, and mortars were matters of conse- quence in architectural work, would probably reply in the affirmative, and would then most likely add : * Such matters can be decided by experience, by noticing buildings both B 2 4 The Rudiments of Petrology. old and new, and seeing how the different materials of which they are built have resisted the ravages of time and expo- sure, and if we want further information we can have the materials analysed by a chemist, and he will be able to tell us all we need to know.' Such a remark as this would embody a great deal of truth. The chemist could furnish an analysis of the stone itself, and we should thus learn how much lime, or magnesia, or silica, or carbonic acid, &c., &c., was contained in it; but if the stone happened to be a very fine-grained one, and although in freshly quarried samples apparently homogeneous, yet when exposed to the weather it suffered unequal decomposition or disintegration, it would be clear that atmospheric agencies detected weak spots better than the chemist, and better than the practised eyes of geologists, architects, stonemasons, and quarrymen. It would then be evident that more practically useful informa- tion could be derived from inspection of buildings than from the sources just enumerated; but at last the awkward ques- tion arises, * How is it that stones which often have the same, or almost the same, chemical composition, and which also closely resemble one another in appearance, have dif- ferent powers of resisting the effects which result from exposure ? ' It is true that the stones most affected may not have been judiciously laid; it is true that the atmosphere of large towns is more prejudicial to building stones than the purer atmosphere in country places ; but, given similar conditions, why is it that apparently similar stones wear differently ? Those who have examined the minute struc- ture and mineral composition of rocks well know how little dependence is to be placed on outward similarity, and even in chemical analysis ; for changes go on within rocks which often produce little or no definite chemical change so far as the aggregate is concerned, but beget numbers of little interchanges in the chemical composition and molecular arrangement of the component minerals. Although in the present state of our knowledge but little Practical Applications of Petrology. 5 practical use has yet been made of the recorded observa- tions which now constitute merely the small nucleus of what will no doubt eventually become a huge pile of information, still we may look forward to the time when a knowledge of the minute structure of rocks will be recognised as indis- pensable to the right understanding of the changes which building stones undergo, and when not merely the few but the many will be benefited by this branch of scientific inquiry. A general knowledge of petrology will always be found useful by those who may have to deal with architec- ture or with mining enterprises, and it is to be hoped that some day, as science progresses, a definite connection may be found to exist between metalliferous lodes and the min- eral composition of the rocks in which such lodes occur. Questions of water supply hinge mainly upon the porous or impervious character of rocks, upon their mode of occur- rence, and upon the structural planes or planes of dislocation by which they are traversed, so that matters of this kind can be best dealt with by the field geologist. In concluding this brief introduction, it seems needful to caution the student not to regard petrology from a narrow point of view ; not to confine his attention' solely to observations in the field; nor to devote himself exclusively to microscopical or che- mical research^ The disadvantage under which the spe- cialist labours is, that he frequently takes infinite trouble to unravel a question in his own special way, when by adopting some other method he might arrive at his result in far less time, and often with greater certainty. At times a penknife will be found more useful than a blowpipe, and a blowpipe than a microscope ; at other times a microscope will tell more than a complete chemical analysis. 6 The Rudiments of Petrology. CHAPTER II. ROCKS DEFINED AND THEIR ORIGIN CONSIDERED. IF we examine a fragment of rock, we find it to consist as a rule of crystals, the edges and angles of which may either be sharply denned or rounded, and which are cemented together either by crystalline or amorphous mineral matter, or we may find it composed of large or very minute angular fragments of mineral matter, or of rounded grains, or of a mixture of both angular and more or less rounded grains also bound together by mineral matter, which may either be amorphous or may possess a crystalline structure. Sometimes, however, the grains simply cohere without any perceptible cement, as in some of the new red and other sandstones. These kinds of rocks may be denned as mineral aggregates, but the term 1 rock ' in its geological signification does not merely imply a coherent mass but also loose incoherent mineral matter, such as blown sand, and in these cases, as in the preceding ones, the materials, whether they consist of fragments of one mineral only or of several different minerals, may still be regarded as mineral aggregates. There are, however, ap- parent exceptions, for some rocks appear to the naked eye to be perfectly homogeneous. Some quartzites may be so regarded, but we know that passages have been observed between quartzites and fine grained sandstones. A casual observer might also take such a rock as Lydian-stone or Hone-stone to be quite homogeneous, but examination of a thin slice under the microscope would show it to consist of numerous lenticular particles, which, from their form alone, imply the necessity for and the consequent existence of a cementing matter which differs from the particles themselves. Again obsidians, pitchstones, and other vitreous rocks, would be assumed by the general observer to be perfectly homo- geneous, but here again the microscope demonstrates that they contain fine dusty matter, microliths, and crystals in The Origin of Rocks. 7 great quantity ; so that even those rocks which are appa- rently exceptions to the general definition are found in reality to conform to it, and thus we may with considerable truth define all rocks as mineral aggregates. The state of aggregation of a rock depends upon the way in which it was formed and the changes which it may subsequently have undergone. We do not know what was the character of the rock or rocks which formed the crust when the surface of the heated mass which originally con- stituted the globe first solidified. The older geologists were Lnd of speculating upon this subject, and many of them believed that the primeval crust was granite, and they further- more believed that all granite could claim this high antiquity. These were, however, mere speculations, and granites are now known to be of various ages, some of them having been formed in comparatively late geological times. Great mis- conception seems to exist among geologists even at the present day about the origin of rocks, and these misconcep- tions usually resolve themselves into bickering about terms and the way in which they should be employed. It therefore seems desirable, at the very outset, to lay before the student a few preliminary considerations which may help him to think and work in a systematic manner. 1 i ) Assuming that the earth was originally a molten mass revolving in space, and that after considerable radiation had taken place solidification of the surface ensued, it is clear that that primitive crust was the first rock formed. (2) Assuming that radiation has constantly been going on from the earliest times up to the present day, it is evident that the whole mass of rock which now constitutes the crust of the earth represents the entire work of solidification, no matter what may be the character (whether sedimentary, erup- tive, 6^.) of those rocks now, unless, through changes in the interior of the globe, chemical action has been called into play, and has thereby generated additional heat, neutralising to some extent the work of solidification by fusing again the rocks which came in contact with these highly heated masses, 8 The Rudiments of Petrology. or, unless through fracture and disturbance of the already solidified crust (by the expansive force of gases generated by chemical changes, or by physical changes, such as the conversion of water into steam), portions of the already so- lidified crust were faulted down or depressed so as to come within the range of these heated portions of the earth, in which case, as in the preceding, some of the pre-existing work of solidification, from loss of heat, by radiation, would be undone, and the total mass of solid rock now constituting the crust would then represent the total amount of rock formed by solidification through radiation, minus the amount of rock re-fused. (3) The rocks now called plutonic, and which have solidified at various depths beneath what was the surface of the earth at the time of their solidification, may be regarded as the result of the solidification of the earth's magma through loss of heat by radiation the rocks now called volcanic, and which have been erupted through the crust and solidified on what was once, or on what is now, the surface of the earth (whether subaerial or marine), may be regarded merely as outwardly-trending phases of the plutonic rocks, and are therefore also the products of solidification through loss of heat by radiation ; the rocks now called aqueous or sedi- mentary are the result of the degradation of land by sub- aerial or marine denudation, and because the first land must of necessity have been portions of the crust formed by the solidification of the earth's surface through radiation. Lastly, the rocks now called metamorphic are merely either eruptive (i.e. plutonic and volcanic rocks) or else sedimentary rocks, which have subsequently undergone changes either physical or chemical, and are therefore only the products of the so- lidification of the earth's magma through radiation, in an altered condition. Consequently no sedimentary rock has ever been formed except from materials which, in the first instance, were supplied in a solid form through the radiation of heat from the globe. Vtdcanicity and Seismology. CHAPTER III. DISTURBANCES OF THE EARTH'S CRUST. STRUCTURAL PLANES. SEDIMENTARY ROCKS. STRATIGRAPHY. THE study of forces existing in the interior of the earth, and the phenomena attendant upon their exertion and affecting the earth's crust, constitute the branches of physical geology known as vulcanicity and seismology, the former relating to volcanic phenomena and the latter to earthquakes. These forces are doubtless due to chemical and physical reactions and changes, resulting in the development of intense heat and the generation of gases. It is also assumed, however, that this high subterranean temperature is mainly owing to the original heat of the globe when first developed as an indivi- dual molten mass, only part of this heat having been dissi- pated by radiation into space. The consequent loss of heat having taken place from the exterior of the globe, and most affecting the superficial portion of it, has given rise to the formation of its crust, by solidification of the once molten matter. It seems reasonable to suppose that such radiation took place equally over the entire surface, and that the phenomena consequent upon the cooling of the mass, although doubtless affecting the centre of the globe, diminish in their intensity from the surface inwards. These physical changes probably extend inwards towards the centre with a certain amount of regularity, so that the globe might pos- sibly be regarded as a spheroidal mass consisting of a series of zones varying in temperature and augmenting as the central portions are reached. Assuming this to be the case, there would be a zone situated at some depth beneath the surface, whose temperature would be so high that any known rock matter could no longer exist in a solid state. The depth at which this zone of fusion is supposed to occur has been variously estimated by different writers. From twenty- IO TJie Rudiments of Petrology. five to thirty miles is about the smallest estimate which has been given ; while some, as Hopkins, have inferred that solid matter extends to much greater depths and may even exist at the centre, the loss of heat and consequent solidification having taken place in an irregular manner, and having thus converted the deeper portions of the globe into a somewhat honey-combed mass, the cavities still retaining matter in a molten condition, and constituting the reservoirs from which the eruptive rocks are derived. So many theories upon this subject have from time to time been started, and they embody such diversity of opinion, that a description of them would be out of place in so small a book ; and as they cannot be regarded as more than speculations, often based upon a little tangible truth and more or less tangible and intangible error, students, although doing well to make themselves acquainted to some extent with these theories, would do better in giving their attention to matters which are more readily demonstrable. That the forces just spoken of as existing in the interior of the earth exercise considerable influence upon its crust we have ample evidence. It is, indeed, solely from such evidence that we infer the existence of the forces. The evidence which we have of the internal heat of the earth may be summed up under the following heads : (1) In descending the shafts of mines a gradual rise of the thermometer takes place after the descent of the first sixty feet. Down to this point it remains stationary ; below this point there is a rise of one degree Fahrenheit for every sixty feet descended. How far this regular increase of temperature continues to take place has not yet been determined. (2) Flows of molten lava and of hot mud, the ejection of lapilli and ashes from volcanic vents through the genera- tion of steam or the evolution of gases, and the occurrence Dislocations of tJie EartJis Crust. 1 1 of geysers and thermal springs, are also evidences of the internal heat of the earth. (3) Earthquakes, the elevation and depression of large tracts of land, giving rise to changes in coast lines, as in Greenland, Sweden, and South America, at the present day; the fractures produced in the crust by subterranean forces, and the relative displacement of rock masses along such lines (faulting) ; the bending and contortion which stratified rocks undergo, and the chemical and physical changes (meta- morphism) which they sometimes experience, are all evi- dences of the internal heat of the earth. That the fissuring and displacement of portions of the earth's crust is frequently due to the exercise of the subter- ranean forces just mentioned there is not the least doubt, since they have occurred frequently, not merely within his- torical times, but men now living have been eye-witnesses of many remarkable changes which have been brought about in the structure of districts by seismic action, as in Calabria and in the country at the mouth of the Indus. 1 Moreover, many of the disturbances of the earth's crust which have happened in even very remote geological periods are pre- cisely such as we should attribute to the same forces which have produced, and are still producing, the modern dis- turbances. We may conveniently classify them all under certain heads, so far as they relate to structural planes ; either on a large 'scale, as affecting the general geology of a country, or on a smaller scale, as influencing not merely the gross but the minute and even microscopic structure of rock masses. It may here be remarked that the scenery and general configuration of a district is often due rather to the facilities offered for the weathering of rocks along small and closely disposed planes of fission than to the presence of long lines of fracture and faulting. The latter tend to produce 1 For descriptions of these and other kindred phenomena the student should consult Lyell'o Principles of Geology, in which accounts are given of the most important earthquakes and volcanic eruptions which have occurred within historical times. 12 The Rudiments of Petrology. lithological diversity of surface rather than diversity of con- tour or relief; atmospheric agencies apparently producing little or no effect upon one colossal divisional plane, since a fault seldom developes a feature in any landscape, other than perhaps a difference in its vegetation. Atmospheric degradation, however, along innumerable divisional planes of very trivial dimensions gives rise to outlines which often enable a practised observer to discriminate between different formations merely from the aspect which they have derived from weathering. The following table gives a rough and rudimentary classi- fication of these structural planes : CLASSIFICATION OF STRUCTURAL PLANES. 1. Irregular (' Earthquakes and their attendant phenomena, fissures \ producing fracture of the crust by pressure and faults. ( from within directed outwards. " Shrinkage on consolidation of sediment by dry- ing and consequent contraction, producing fracture of the crust usually along more or 2. Jointing , ,. , ' H less parallel lines. Shrinkage on consolidation of eruptive matter by cooling and consequent contraction, producing fracture in directions more or less parallel. C Pressure, exerted by contiguous rock masses, producing (often by re-arrangement of par- ticles) planes of weak cohesion, along which Laminar fission readily takes place in parallel direc- fissionand-^ tions. cleavage. A. Coincident with bedding planes (laminar fission or flaggy cleavage). I B. Deviating from the direction of the bedding I. planes (slaty cleavage). It is possible that many of the lines along which faulting has taken place may in the first instance have been simply fissures due to shrinkage (Class 2) but in other cases faulting Origin of Faults. 1 3 has occurred along planes produced by seismic action (Class i). The parallelism which so often characterises a system of faults cannot, however, be adduced as proof that those faults have occurred along joint planes (Class 2), since parallel fis- sures might be produced by the upheaval of rocks along a certain line or rather along a definitely trending area (Class i) on either side of which relative displacement of strata or of eruptive rock masses may have subsequently occurred. It is also quite possible that the displacement itself originated synchronously with the line or plane along which it runs, in which case the plane would again belong to Class i. All that we can therefore safely say about the origin of faults is, that they are relative displacements of the earth's crust, caused by subterranean forces upheaving masses of rock along lines of least resistance, which may either be pro- duced at the time of upheaval or may have pre-existed simply as fissures or cracks, and that in some cases depres- sion of rock masses has caused faults, the subsidence of the downthrow having occurred by the mere gravitation of the mass between two outwardly diverging planes of fracture, as in the case of ' trough faults.' l Faults may also arise from an unequal horizontal shifting of undulating beds along a fissure, or from partial flexure or bagging down of strata upon one side only of a fracture, the beds on the other side remaining horizontal. 2 The fissures and cracks, therefore, along which faulting has taken place, may be due either to volcanic or seismic action (for earthquakes and volcanic phenomena are so inti- mately related, and are apparently so indisputably due to similar or identical causes, that they may safely be classed together), or to the shrinkage of sedimentary rocks by loss of 1 Vide Students Mamtal of Geology, by J. Beete Jukes (1862), p. 260. - Ibid. pp. 254-5. 14 The Rudiments of Petrology. moisture and that of eruptive rocks by loss of heat during solidification. Another class of small structural cracks which occur in some eruptive rocks and also exceptionally in argillaceous beds which have undergone considerable desiccation, is, in the opinion of many observers, due, in the case of the erup- tive rocks, such as basalt, to contraction on cooling. These planes intersect one another in such a manner as to divide the mass of rock into a series of closely packed prisms varying at times in the number of their sides and in the measurement of their angles. This structure is especially characteristic of the basalts in some districts (Staffa, Giant's Causeway, Unkel on the Rhine, parts of Auvergne, and many other localities.) The prisms are generally cut trans- versely by numerous divisions, which are sometimes flat, sometimes either convex or concave, while occasionally, as in the celebrated Kasekeller, they consist of superposed spheroidal lumps or balls, which have a concentric shaly structure. An analogous structure on a very small and often purely microscopic scale is to be met with in vitreous rocks such as perlite. 1 According to the theory held by Sir Henry De la Beche and others, mountain chains owe their origin in many cases to immense fractures and dislocations of the earth's crust caused by unequal contraction of the crust in zones, the inner zones contracting and leaving the outer and already solidified zone unsupported, so that in places it cracked, large masses subsided on to the lower zone, and thus caused immense ridges and depressions. Such mountain -forming fissures, colossal though they may be, are, however, hypothetical rather than demonstrable. Enough has now been said to show that structural planes and divisions occur in rocks ranging from those of gigantic size to others of quite microscopic dimensions. Some of them occur in rocks of eruptive origin, some traverse sedi- 1 Bonney, Q. J. G. S., vol. xxxii. p. 140 ; Allport, Q. J. G. S., vol. xxxiii. p. 449 ; Rutley, Trans. 7u Mic. Soc., vol. xv. p. 176. Sedimentary and Eruptive Rocks. 1 5 mentary deposits, but in all cases they facilitate the weather- ing and disintegration of the rocks in which they occur, and consequently exercise a more or less marked effect upon the scenery of a district. The rocks composing the crust of the earth may be con- sidered mainly to belong to two great divisions, viz., (A) the aqueous, sedimentary, fossiliferous, or stratified rocks, which have been deposited as sediment in beds, or strata beneath water, each bed or stratum having successively formed the floor of a sea, or of a lake ; and (B) igneous or eruptive rocks, which have formed intrusive bosses, or- dykes, or have been poured out from volcanic vents, as lava flows. The former usually contain organic remains which may be identified with a marine or a lacustrine fauna, and con- sequently afford a tolerably safe clue to the circumstances under which the beds were deposited. It may be safely assumed that all such beds were originally spread out in an approximately horizontal position, and that any strong deviation from the horizontal position which may be shown by planes of bedding is due to subsequent disturbance of those beds. Sediments may sometimes, however, be somewhat irre- gularly deposited; for example, a number of thin beds may thin out completely, overlie one another, and the whole of them may overlie a perfectly horizontal bed upon which their thinned-out ends appear to rise unconformably, and this kind of arrangement may be repeated again and again through a considerable thickness of deposits. This irregular kind of stratification is called ' false bedding. ' Fig. i represents a good example, occurring in the lower greensand, at Frith Hill, near Godalming. In such a case the inclination of the beds is not due to any disturbances during or subsequent to deposition, but simply to the overlap of successive deposits as they are thrown down in shallow water. The sediments which i6 The Rudiments of Petrology. constitute stratified rocks result from the wear and tear which takes place from the action of rain on land sur- faces, and, in the beds of rivers from attrition, each eroded fragment serving as a tool with which other fragments are ground away from the rock. When a turbid river empties itself into the sea or into a lake, the materials held in suspension become deposited according to their relative specific gravities, the heavier fragments sinking first, while the lighter particles are carried to a greater distance from the FIG. i. river's mouth. The relative sizes and shapes of the frag- ments also exercise some influence on the sorting process which takes place. Fragments which have undergone but little attrition are usually more or less angular in form, while those which have been carried long distances, and which have been rubbed together for a length of time, become sub- angular or perfectly rounded. 1 The rounded form of the pebbles on sea-beaches is due to the incessant grinding 1 Except in instances where the fragments have been transported by ice. Stratified Rocks. 1 7 which they undergo against one another during the advance and retreat of every wave that washes the shore. Deposits mainly composed of angular fragments are termed breccias, while those consisting of rounded pebbles are called conglomerates. In indurated rocks of this kind the coarse frag- ments, or pebbles, are generally cemented together by a finer material, often consisting of carbonate of lime or silica. There are frequently other substances, however, which act as a cement. Aqueous or sedimentary rocks may be conveniently classed as follows : (1) Clays, which when indurated become mudstones, and, when cleaved, slates^ When they merely exhibit a fissile character in the direction of the lamination, or bedding, they are called shales. Clays, slates, and shales are mainly com- posed of hydrous silicate of alumina. There are arenaceous and calcareous clays, slates, and shales ; calcareous clays are termed marls. (2) Sands. These when indurated constitute sandstones, and when more or less coarse-grained, and composed of angular or sub-angular grains of sand (frequently with an admixture of fragments of other minerals), they are then termed grits. The term ' grit ' is, however, very loosely used, and it would be difficult to give it a sharp definition owing to the great variation in the physical and mineralogical cha- racters of the rocks to which this name has been applied. Sands and sandstones are usually composed of fine grains of quartz cemented either by carbonate of lime, carbonate of iron, oxides of iron, or silica. There are calcareous and argillaceous sandstones. (3) Limestones. These may vary from soft and earthy, to hard, compact, and even finely crystalline rocks. Some limestones may be merely eroded granules of pre-existing limestone carried mechanically in suspension in water, and ultimately deposited as a sediment. Some may have re- sulted from the precipitation of carbonate of lime from water holding the bicarbonate of lime in solution. In this c 1 8 The Rudiments of Petrology. case the deposit may be considered to have a chemical origin. The cause of precipitation would be the elimination of one atom of carbonic anhydride from each molecule of bicarbonate of lime. Travertine, calcareous tufa, and pisolite are rocks formed in this manner. The last consists of rounded grains like shot or peas, whence the name pisolite or peastone ; these little pellets consist of a series of concentric coats of carbonate of lime which sometimes have a small grain of sand as a nucleus. Limestones are also at times composed in great part of the shells of minute animals called 'foraminifera.' These organisms, whose remains constitute the very earliest record of life of which we have any knowledge, have peopled the waters of various geological epochs with their descendants, and at the present day the foraminifera have numerous living representatives. The animals themselves are little more than small shapeless masses of animated jelly, but they have the power of sepa- rating carbonate of lime from solution in water, and of building up the material into shells of very variable and extremely beautiful forms. Some are perforated by immense numbers of minute holes through which the gelatinous occu- pants can protrude their filamentous processes. To these holes, or foramina, the order owes its name. Corals also have the power of secreting large quantities of carbonate of lime, and some limestone rocks are in great part due to the secretions of these polyps. The shells of the mollusca, which have originated from a similar secretive faculty, also at times contribute largely towards the formation of some limestones. This secretive process can merely be regarded as a chemical process performed through the intervention of the animal; and when we speak of such rocks as having an organic origin, we must be careful not to imply that the animal had actually formed the calcareous matter instead of having merely secreted it. Pure limestones consist simply of car- bonate of lime. A compound of the carbonates of lime and magnesia constitutes magnesian limestone, or dolomite. Those limestones which contain a certain amount of clayey Stratified Rocks. 19 matter are termed argillaceous limestones, and those con- taining sandy impurities are styled arenaceous limestones. The changes which sedimentary rocks undergo may be regarded as physical, as chemical, or as the result of phy- sical and chemical agencies acting either simultaneously or at different periods. CLASSIFICATION OF THE SEDIMENTARY ROCKS. Clay. Composed of hydrous silicate of alumina, usually with mechanical admixture of sand, iron oxides, and other substances. Marl Clay containing calcareous matter. Shale. Indurated clay, fissile in direction of bed- ding. Slate. Indurated clay, fissile in parallel planes other than those of bedding. f Sand. Chemical composition silica. Mineral com- ponents quartz or flint. Sandrock* Coherent sand. Sandstone. A more or less strongly coherent and often highly indurated sand. Grit. A coarse-grained and somewhat coherent, or at times a fine-grained and very hard and compact sandstone, frequently containing fragments and granules of other minerals beside quartz, flint, or chert. Calcareoiis Sandstone. Sandstone cemented by carbo- nate of lime. Ferruginous Sandstone. Sandstone cemented by an oxide of iron or by carbonate of iron. Conglomerate. Rounded pebbles of flint, chert, jasper, quartz, &c., cemented either by siliceous, calcareous, or ferruginous matter. Siliceous Breccia. A rock similar to the above in com- position, but differing from it in contain- ing angular fragments instead of rounded pebbles. Siliceous Sinter. Silica deposited in a more or less loose or spongy form from waters holding silica in solution, c 2 2O The Rudiments of Petrology. Many of the slates, sandstones, and grits afford good building stones. Limestone. Chemical composition, carbonate of lime. Limestones vary greatly in their physical characters ; some being earthy, soft, and friable, as chalk ; others hard and crys- talline. The principal limestones used for building purposes are the Devonian, the carboni- ferous, and the magnesian limestones ; many of the oolitic limestones, especially the Bath stone, Portland stone, and Pur- beck limestone. Limestones which are capable of receiving a polish are called marbles. They vary so greatly that it is not possible to describe even the leading kinds in a -small space. Bands of chert occur in the carboniferous and in some other limestones, as the Port- land, and bands and nodules of flint are met with in the upper chalk. Magnesian Limestone. Chemical composition, carbo- nates of lime and magnesia. This rock is also called Dolomite, after Dolomieu. The proportions of carbonate of lime to carbonate of magnesia vary greatly in different localities. Argillaceous Limestone. Limestone containing some clayey matter or hydrous silicate of alu- mina. When this reaches a certain pro- portion the rock is termed an hydraulic limesitone ; such limestones are used for the manufacture of cements which set under water (hydraulic cements). The lias limestone is a good example of an argillaceous limestone. Arenaceous or Siliceoiis Limestones represent a transi- tional condition between limestone and chert. Some of them, such as the Kentish rag, afford good building stones. Dip and Strike. 21 It has been already stated that the sedimentary rocks occur in beds or strata (hence they are also called stratified rocks). This arrangement has, in the first instance, been an approximately horizontal one, and, in most cases, where there is any marked deviation from horizontally, the deposits have been disturbed by the action of subterranean forces. When any such disturbance has taken place, so as to communicate an inclination to the beds, this inclina- tion is termed ' dip.' If we assume a long strip of paper to represent a horizontally deposited bed or stratum, and then fold it lengthwise as in FlG 2 fig. 2, we have two dips, one in the direc- tion of a and the other in an opposite direction, b. The direction be or cb is termed the ' strike ' of the beds. The strike is always an assumed horizontal direction, so that if we tilt our strip of paper on one end the strike will still be a horizontal line as de (fig. 3). The dip is always reckoned at right angles to the strike. It is somewhat difficult to render this apparent in a diagram ; but if we represent one side, of our strip of paper to be dipping vertically, i.e. at 90, as in fig. 4, it will render the FIG mutual directions of dip and strike as seen in plan : a b ^^ J representing the strike, and cd & the dip. In nature it is not usual to find beds bent in the acute manner indicated in the preceding figures. They generally describe curves which represent the arcs of circles sometimes many miles, sometimes only a few feet in extent. In the latter case this small crumpling is spoken of as ' contortion.' When flexure of strata occurs in an upward direction the result FIG. 3. The Rudiments of Petrology. FIG. 5. is spoken of as an anticlinal flexure, curve, or ridge ; while, on the other hand, when the curve is directed downwards in a basin -shaped manner, it is termed ' synclinal. \ The dip of strata exercises a marked influence on the scenery of a country. If no disturbance of stratified deposits had ever taken place in a district no knowledge of its geology could be obtained, except along valleys which had been scooped out by the action of rain, rivers, and general atmospheric agency, or in railway cuttings, quarries, and other excavations, and in borings such as wells and the shafts of mines. Let AB (fig. 5) represent the surface of a country in which the strata have never been disturbed* and therefore lie horizontally just as they were deposited. Let us also sup- pose that the surface has not ; - ; /.'... :.>:".: ; been carved out into hills and valleys, but is a level, un- broken surface. It is evident that an observer walking across such a district would meet with no diver- sity in the character of the soil or of the rocks over which he passed, unless indeed the same deposit ex- hibited slight lithological change in its own horizon such as a passage from clay into sandy clay, and the geological map of the district repre- sented by the section A B would, if coloured, be merely painted over with one uniform tint, or stippled thus (fig. 6). If such a country were scooped out by the action of rain, rivers,. &c., the section AB (fig. 5) would undergo con- siderable alteration, as shown in the accompanying fig. 7 ; FIG. 6. Geological Maps and Sections. FIG. 8. while its geological map would now indicate the exposure not merely of the uppermost stratum, but also the outcrop of several underlying beds some- what in the manner shown in _____ fig. 8, with probably a river R R \". = ' ? ".'~Vr'""T""~?' T . " vr^ --. running along the bottom of the valley. Maps of the oolitic and liassic districts of England will be seen to resemble this in their geological boundary lines, and these lines usually follow the general con- tours of the country. Let AB (fig. 9) represent the surface of a country in which the stratified rocks have been disturbed and tilted upwards ; in other words, a country in which the strata have a definite dip. It is highly improbable that such a country would have a level surface, as the unequal hardness of the different rocks exposed to the action of the atmosphere would tend to beget considerable irre- gularity, the harder ones being worn away less easily than the soft ones. Still, supposing the surface of the country to have been planed off to a perfect level, the geological map of the district would nevertheless pre- sent great diversity in its colouring or shading. A man walking across the country from east to west would pass over several different lormations. In this case ss, ss (fig. 10) would represent the strike of the beds, and the arrows would show the direction of tlieir dip, at right angles to the strike. FIG. 9. i B The Rudiments of Petrology. If the beds were repeated, dipping in an opposite FlG . I0 . direction, i.e. if they had an an- ticlinal arrangement, some esti- mate could be formed of the amount of rock which had been denuded by restoring the curves as indicated by the dotted lines in fig. n, although this would probably represent only a portion of the total amount of matter which had been removed. The thickness of beds should in all cases be measured at right to the planes of bedding, whether they be undis- FlG. II. turbed and horizontal or disturbed and inclined. Thus, if A A (fig. 12) represent the surface of the ground, and B an inclined bed, then the thickness of B should be measured along the dotted line xx, and not along the surface A A. Enough has now been said to denote the way in which the disturbance of sedimentary rocks influences the surface of a country, and, when aided by denudation, promotes our knowledge of its geology, by bringing to view subjacent deposits which would otherwise have been accessible only by excavation and boring, thus affording us the means of selecting from various sources materials of industrial Denudation. 2 5 importance such as building stones ; bringing within work- able distance various mineral deposits, and diversifying the surface of the land in a manner which affects agri- culture and water-supply, civil engineering, and last, but not least, the sanitary condition of its inhabitants. The manner in which rocks are worn away is spoken of as denudation. Denudation may be regarded under two heads : (1) Marine denudation. (2) Atmospheric or subaerial denudation. The tendency of all denudation is to wear away existing land to lower levels until it reaches the level of the sea. The wearing process then stops, because the sea can only act destructively in planes situated between high and low water-mark. The breakers do all the work of marine denu- dation ; and when they can no longer act upon rocks because they have planed them down so far as they can plane them, that is, to their own level, the process of denu- dation of course ceases ; and when such a stage of degrada- tion is reached, subaerial denudation also becomes inert. Marine denudation may therefore be denned as the de- grading influence exercised by the sea on a level with the breakers. This degradation is effected not merely by the force of the breakers dashing and pounding against their barriers, but the detached fragments are hurled again and again against the rOcks, which thus continually supply ammunition for their own destruction. The tendency of all this is to cut back the coast into cliffs ; and the process, so far as the sea is then concerned, becomes an undermining one. This is well shown in Shakespeare's Cliff at Dover, where the sea has undercut the chalk at the base of the cliff (fig. 13) ; but it is usual to see few signs of excavation at the foot of a cliff, because subaerial denudation is also con- tinually going on ; and when its effects upon certain rocks are more marked than those of marine denudation, the face of the cliff will be seen to slope backwards more or less. 26 The Rudiments of Petrology. If marine denudation acted more powerfully than atmospheric denudation, then the cliffs would become undermined, and the overhanging masses would eventually give way, falling on to the shore and forming a talus, or heap of broken frag- ments banked up against the lower portion of the cliffs. A talus may also be formed by the fall of fragments dislodged by frost and other atmospheric denuding agents, so that under any circumstances it is common to find the foot of a cliff so protected, except where the shore is very steep or where the scour of the tide is considerable. FIG. These natural barricades prevent the sea from attacking the base of the cliff for a time, but after a while they are cleared away and the undermining process recommences, to be again temporarily retarded by successive falls of rock from above. When the stratification of the rocks which form a coast dips inland, the sea acts very destructively, large frag- ments being dislodged with comparative ease ; but when the dip is seawards, the breakers run up inclined planes of bedding instead of dashing against an abruptly raised barrier, and so the work of degradation takes place much more Cliffs and Escarpments. 27 slowly. Sometimes steep slopes, forming features which to a certain extent resemble cliffs, are seen far inland. These are called escarpments. They owe their origin to subaerial FIG. 14. / > and not to marine denudation, and they differ from true sea-formed cliffs in that the escarpment will be found to trend in a direction parallel to the strike of the beds (fig. 14), Flc. 15, FIG. 1 6, 'while a coast line backed by cliffs runs in an irregular manner, which bears no relation whatever to the strike (fig. 15). It is therefore evident that the configuration of a coun- try, so far as its coast-line is concerned, does not depend upon the strike of the rocks along that coast -line except so far as the strike determines the position along that coast-line, of relatively hard and soft rocks, or of rocks whose chemical or physical characters render 28 The Rudiments of Petrology. them relatively difficult or easy of disintegration, as in fig. 1 6, where H represents rocks comparatively difficult, and E those which are comparatively easy, to wear away. 1 The difference in the weathering of rocks, dependent upon their relative hardness, does not merely influence the coast-line of a country, but it affects its inland configuration to a greater or less extent, as shown on the east and west coasts of England. lf r for example, a country consist partly of granite and partly of slate, it will usually be found that the granite constitutes the high ground, while the slates occupy the lower portions of the district. It does not, how- ever, necessarily follow that all the valleys of a country should be scooped out in the softer rocks, while the harder ones only form the hills. If this were the case, the drainage of a country would be determined by the general strike of the rocks, and all the valleys would trend in the direction of the strike. This, however, is not always the case. In the Lake District of England, for example, most of the valleys run across the general strike of the Upper Silurian rocks, instead of coinciding with it. It seems, indeed, that the directions of the rivers and valleys of a country are often determined rather by its initial slope than by the relative resistances offered to erosion by its rocks. As a rule, how- ever, it is probable that neither of these considerations can be utterly ignored, and that the truth involves them both : first, the initial slope of the district ; and secondly, the relative resistance which the rocks offer to atmospheric denudation. Such resistance does not merely influence the large features of a country ; it renders itself evident in the actual shapes assumed by the hills. 1 The Geological Observer and How to Observe Geology, by Sir Henry De la Beche, are good works for the student to consult upon these points ; also a Paper on ' Subaerial Denudation and on the Cliffs and Escarpments of the Chalk and the Lower London Tertiary Beds,' by W. Whitaker, Geol. Mag. vol. iv. pp. 447-483. London. 1867. The phenomena of denudation are also well described in Prof. Ramsay's Physical Geology and Geography of Great Britain. Weathering of Rocks. 29 The forms of hills and mountains are mainly due A. If composed of sedimentary rocks (1) To the lie of the beds, whether horizontal or in- clined. (2) To the presence or absence, the paucity or multitude, of the structural planes which traverse those beds. (3) To the physical characters of the beds. (4) To their chemical composition. B. If composed of eruptive rocks (1) To the presence or absence, paucity or profusion, of structural planes, (2) To the physical character of the rocks, (3) To their mineral constitution. (4) To their chemical composition. The average rainfall of a district of course has more or less influence on the weathering and disintegration of rocks, while the water which filters into them, especially along joint planes and fissures, expands, on its conversion into ice, in frosty weather, and greatly facilitates their degradation, by forcing apart fragments and blocks of rock which become detached as soon as the ice thaws. The process of disintegration is carried on to a great extent by the rain, which, in its passage through the atmosphere, ab- sorbs a considerable quantity of carbonic acid. This acts upon compounds of lime, potash, soda, &c. In limestone districts a great amount of matter is annually removed in this way by the conversion of the carbonate into the soluble bicarbonate of lime. The water of rivers, and the water from swamps and peat-mosses also, contains more or less carbonic acid, especially in the latter cases, where vegetable matter by its decomposition gives off considerable quantities of this gas. Limestones are by no means the only rocks acted upon by waters containing carbonic acid. 30 The Rudiments of Petrology. The felspars, which constitute so great a portion of many eruptive rocks, suffer decomposition through this cause. If the felspar be a lime-felspar the lime is the substance first removed; the potash and seda, which most felspars contain, are next carried off as carbonates, although they are some- times removed in the form of soluble silicates. This gradual removal of the alkalies, &c., ultimately results in the forma- tion of kaolin or china-clay, a hydrous silicate of alumina, 2^iSi 2 +2H 2 - Should the waters contain magnesian salts in solution, the lime or soda of the felspars may be replaced by the isomorphous magnesia, thus giving rise to steatitic matter so long as the alumina of the felspar is not involved in the change. Through similar causes the various species of mica become at times partially decomposed and converted into steatite, serpentine, compounds allied to chlorite, and possibly some other hydrated minerals. The alterations and replacements of minerals, by the infiltration of water charged with acids and soluble salts through the rocks in which they occur, are so numerous, and the pseudomorphs to which they give rise are so interesting, that special mention of them will be made in those parts of this work where the rocks in which they are found are described. The interchanges which take place in the formation of these pseudomorphous minerals are in great part due to the isomorphism of the replaced and the replacing substances, although at times the original mineral may be entirely removed and its place subsequently occupied by matter bearing no such relation to the components of the mineral replaced. It has already been stated that the sedimentary rocks occur in layers, technically termed strata or beds. A number of such beds, deposited at the same time, under approximately the same conditions, in one or in many distinct areas (areas which during deposition may once have been united, or in which deposition may have gone on independently), are termed ' formations.' The formations are usually more or less fossiliferous, and separate formations are distinguished Formations. *j* 3* '//. v , by characteristic fossils, which bear testimony to A< certain'/ . sameness in the animal and vegetable lif^/vhich 'exjsted from the deposition of the lowest to that of the highest beds in the formation. The geological age of any bflfc^may' ^ therefore be determined either by its stratigraphical fi6ri^6ji or by the fossils which it contains. The lithological cha-y racter of a bed sometimes varies, and, therefore, less depen- dence is to be placed upon it than upon palaeontological evidence. Furthermore, the slight variation in the litho- logical characters of sedimentary rocks often renders it very difficult to assign them to any particular horizon in the absence of fossils ; sandstones, slates, shales, and limestones of different geological ages often bearing a close resemblance to one another. Again, rocks differing widely in lithological character may have been deposited at the same time, as in the case of the Devonian and old red sandstone rocks, the former having been thrown down in the sea, and the latter in lakes, as proved by the fossils which they respectively contain. Yet both formations occupy a position intermediate between the Upper Silurian rocks and the lowest members of the carboniferous series. The grouping of sedimentary rocks into formations is, of course, more or less arbitrary. Some genera, and frequently species, which occur in a lower formation are often represented in the succeeding deposits of a newer formation, and probably, if the truth were known, it would be found that all the formations, which we now recognise, pass from one into another. Because an uncon- formity, i.e. a break both stratigraphical and organic, occurs in one limited district, it does not necessarily follow that this break extended over the entire globe. Allowances must be made for relative distributions of land and water, which we have often no means of realising, and no doubt the universal application of limited knowledge often does more harm than good in this branch of geological inquiry. 32 The Rudiments of Petrology. CHAPTER IV. ERUPTIVE AND METAMORPHIC ROCKS. THE eruptive or igneous rocks differ entirely from those of sedimentary origin in their mode of occurrence (except in the case of volcanic ejectamenta, presently to be explained, in the interbedding of lava-flows, and in the intrusion of sheets of eruptive rock between planes of bedding, as in the case of the Whin Sill of Northumberland * ). They bear, except in such instances as those just cited, no definite relation to the sedimentary rocks, but form irregular masses, often of very great extent, from which vein-like prolongations or tabular and wall-like masses (dykes) are often sent off into the surrounding rocks. They also emanate from volcanic vents in the form of molten viscous lava, forming flows or coulees, which are forced over the edge of the crater, frequently breaking it down on one side, creeping down the sides of the cone, and often spreading for many miles over the surrounding country. 2 Sometimes they are erupted in the form of large and small fragments of rock (lapilli), of peculiar spheroidal molten masses (volcanic bombs), and of finely comminuted and dusty mineral matter (ashes). These fragmentary ejecta- menta are often thrown high into the air. Part of them fall back again into the crater to be again and again thrown up, so that by constant attrition they become more or less rounded. Part fall on and outside the rim of the crater, thus helping to build it up higher. Part may be carried by 1 Vide paper 'On the Intrusive Character of the Whin Sillof Northumberland,' by W. Topley and G. A. Lebour, Q. J. G. S., vol. xxxiii. p. 406. 2 Intrusive sheets may be distinguished from true lava- flows, which have been subsequently overlaid conformably by sedimentary strata, by the fact that the rocks both above and below the intrusive sheets are altered at the contacts, while in the case of lava-flows the rocks over which they ran have been altered, but the deposits above them show no trace of metamorphism. Plutonic and Volcanic Rocks. 33 the wind and showered down over the adjacent country, and if in the state of very fine dust may be transported immense distances by the wind, a passage of between 700 and 800 miles having been recorded. There are also some craters, usually of comparatively small dimensions, which pour out liquid mud, frequently accompanied by an outpouring of water also. The water is in some cases boiling, in others cold, and bitumen has also been seen to exude from some of them. The hot water ejected by the Geysers of Iceland, and that of the thermal springs of Roto-Mahana, near Lake Taupo, in New Zealand, carry a large amount of silica in solution, which on the evaporation of the water leaves a deposit or incrustation of white siliceous sinter. Besides the lavas, ashes, &c., which emanate from volcanoes, steam and gases, such as carbonic anhydride, sulphurous acid, hydro- chloric acid gas, sulphur vapour, &c., are also emitted. It is still customary to divide the eruptive rocks into two classes the plutonic and the volcanic, the former class including those rocks which have solidified at considerable depths beneath the earth's surface, and which are now only exposed because the rocks which once overlaid them have been removed by denudation. The volcanic rocks, on the other hand, although likewise originating at considerable depths, have been forced up until they not merely reached but in many cases have overrun the surface. From the mode of occurrence of the rocks belonging to these two classes it is by no means easy to affirm with positive certainty that they are merely different phases of the same eruptive rock-forming matter, emanating from the same source. Still, on comparing the rocks of the one class with those of the other, a tolerably continuous chain of evidence can be adduced to show that they graduate into one another, and that this, like all other classifications, which are necessarily arbitrary, is more hypothetical than real. The quartz-porphyries or elvans resemble the granites more or less in mineral composition, and are known to emanate D 34 The Rudiments of Petrology. from granitic masses ; but the mica, which is plentiful in granites, is only poorly represented or is totally absent in the quartz-porphyries. The porphyritic felstones resemble the quartz-porphyries, except that they contain no definite and well-marked crystals and blebs of quartz. The felstones which are not porphyritic are often identical with the magma of quartz- porphyry. The trachytes vary in their affinities, some (such as the quartz-trachytes) inclining more towards the granites and felstones in mineral composition, while others (as sani- dine-oligoclase-trachytes) occupy an intermediate position, passing into or becoming allied to the basalts, dolerites, &c., in the oligoclase trachytes, the andesites, and the trachy-dolerites. 1 If, then, such close resemblances in mineral constitution can be discerned in the rocks which come near the boundary line drawn between the two classes, and since rocks containing under 60 per cent of silica (basic, of Bunsen), and over 60 per cent, of silica (acidic, of Bunsen) are found in both classes, it seems reasonable to suppose that close resemblances in mineral constitution are almost equivalent to observed passages. Again, both acidic and basic rocks are known in some instances to have emanated at different periods from the same volcanic vent, Durocher, in the ' Annales des Mines,' vol. xi., 1857, enun- ciated the theory that all eruptive rocks have been derived from one or other of two magmas which occur in distinct zones beneath the solid crust of the earth, the one poor in basic materials, but containing over 60 per cent, of silica, the other rich in basic matter, but holding less than 60 per cent, of silica. The former magma having a less specific gravity than the latter, is assumed to float, as it were, upon it, the 1 Mr. J. Clifton Ward states that some of the rocks occurring in the English Lake District are intermediate in mineral composition between felstones and dolerites, and in describing them he designates them 'felsi-dolerites.' Q. y. G. S. t vol. xxxi. p. 417. In examining some of the eruptive rocks from the Silurian districts of North Wales the author has met with similar examples, and can fully endorse Mr. Ward's conclusions. Cleavage. 3 5 difference in specific gravity of the rocks derived from these magmas being from one and a-half to twice as great as between oil and water. Taking all these things into consideration, we may be justified in assuming that the difference between the plutonic and volcanic rocks of Bunsen's acidic class lies wholly in the fact that they have solidified under different conditions, but that their differences do not sufficiently warrant their separation by a hard and conventional boun- dary-line, nor debar us from the inference that they may often have arisen from the same deep-seated sources. The same may be said of Bunsen's basic class. If it could be demonstrated that they have done so, nothing would remain but to admit that the plutonic rocks are the roots and stem, the volcanic rocks the branches and twigs, of a great petro- logical system. The sedimentary rocks, as already mentioned, occur in beds, or strata which were originally deposited in an ap- proximately horizontal manner. Furthermore, the beds generally exhibit lamination (or still finer bedding). The mineral particles of which FIG. 17. these rocks are composed are, when inequiaxial, ar- ranged with their longer axes parallel with the la- mination or bedding (fig. 17), in which BB represents the direction of the planes of bedding. Fig. 18 indicates the change of direction which r IG. To. these minute and usually microscopic particles as- sume when the rock has undergone great lateral pressure. A fissile struc- ture, called cleavage, is then set up in some direc- tion other than that of the bedding planes B B ; as, for example, in the direction c c, so that the rock splits more or less D 2 36 The Rudiments of Petrology. readily in that direction. This change in the minute struc- ture of rocks which have been subjected to strong pressure was first demonstrated by Sorby, and fully explained by D. Sharp in the ' Quarterly Journal of the Geological So- ciety,' and by Professor John Phillips in 'Report of the British Association, 1856.' It also affects any fossils which the rock may contain, squeezing and distorting them to a considerable extent. True schistose fission and slaty cleavage are seldom or never met with in rocks of eruptive origin, except sometimes in beds of volcanic ash, and occasionally in some of the older lavas, as shown by J. Arthur Phillips, neither does lamination occur in eruptive rocks, but a structure slightly resembling it is often to be noticed in sedimentary rocks which have undergone such great change (meta- morphism) that they approach to, or are identical with, true eruptive rocks in their mineral constitution. This foliation consists in the segregation of any one mineral component of the rock along a more or less regular plane, and the result is a differentiation of the rock into a series of alternating layers of different mineral composition. These layers are often very thin, and at times scarcely to be discerned with the naked eye. Hornblende schist, for example, consists of alternating layers of hornblende and quartz ; gneiss of layers of quartz, felspar, and mica. Gneiss may therefore be regarded lithologically as a foliated granite. Foliation has often been found to coincide with the original planes of bed- ding, as noticed by Ramsay, Darwin, Sterry-Hunt, and other observers, but this is not invariably the case. Metamorphic rocks form, as it were, a connecting link between the sedi- mentary and the eruptive classes, their pseudo-eruptive characters having been superinduced by the contact or proximity of highly heated eruptive matter. Thus, where basalts come in contact with limestones, the latter frequently become crystalline for some distance from the contact. Such metamorphism affects rocks sometimes on a small, sometimes on a large scale, occasionally influencing only a Volcanoes. 37 few inches, at other times extending for miles. It consists sometimes merely in physical, at others in chemical and physical changes, which frequently involve complicated atomic interchanges (chemical reactions), and symmetrical molecular rearrangements (crystallogenesis) . By these means minerals are developed in a rock which it did not previously contain ; and this process may take place without any accession of fresh elementary substances, analyses of the unaltered and the metamorphosed rock being sometimes nearly identical. The presence of hygrometric water, or quarry water, greatly facilitates such changes in the mineral constitution of rocks. This fact is very ably dwelt upon by Mr. John Arthur Phillips in several papers on the petrology of Cornwall published during the last few years in the ' Quar- terly Journal of the Geological Society.' These matters will, however, be more fully discussed in the sections specially devoted to the changes which rocks undergo. Although many admirable descriptions of volcanoes are to be found in most manuals of geology and in works specially devoted to the geology of volcanic districts, 1 yet it may be well to give here a brief description of the general structure of volcanic vents. An active volcano may be denned as a passage or pipe which affords to deep-seated mineral matter, in a state of fusion, the means of transmission through the earth's crust, and of egress at its surface. A passive or extinct volcano is one in which this communication is obstructed, either by a plug of solidified lava, or by accumulations of fragmentary matter, a dissipation, temporary or permanent, of the eruptive energy, permitting the solidification of the molten matter. Should an augmentation of the eruptive force occur, the plug will either be shattered, and ejected in the form of lapilli 1 The student may consult the works of Lyell, Scrope, Darwin, Daubeny, De la Beche, &c. with advantage ; also some very interest- ing papers by Prof. J. W. Jucld, entitled ' Contributions to the Siudy of Volcanoes,' published in the Geological Magazine, and the chapters devoted to this subject in Geology for Students' by Prof. A. H. Green. 38 The Rudiments of Petrology. and ashes, or re-melted and poured out as lava, but, if it be unable to re-open the old passage, new vents may be pro- duced, either within or without the lip of the crater. Lava transmitted through a fissure or pipe and extruded at the surface may give rise to hills of a dome-shaped cha- racter. Ashes and lapilli ejected from a vent become piled up around it, and in time form a conical hill on what may once have been a level surface. As, however, they are loose incoherent deposits, the hill will gradually acquire a slope at which they are no longer stable. On measuring the superficial inclination of hills composed of such ejectamenta, it has been found that the slope is usually about 30. * If we pour sand upon a level surface, it forms a complete cone, but the loose volcanic materials come from below in the first instance, and since the pipe of the volcano is open, and ashes and lapilli are ejected from it and fall around it, the cone can have no apex. Some of the ejected matter, which is not carried away by the wind, showers down again upon the hill and around the orifice of the vent, but the law which governs the stability of these loose accumulations again pre- vents them from resting upon a very steep slope, and they are found to dip inwards towards the orifice as well as out- wards down the slopes of the hill. The boundary line between these two slopes, which of course represents their greatest altitude, assumes a more or less annular form, and the inner slopes which dip towards the vent constitute a cup- like hollow, termed a ' crater.' Volcanoes, however, pour out lava as well as eject ashes, and these phenomena usually alternate. Lava in a viscous, pasty condition, rises through the pipe into the crater, where, after perhaps surging up and down for a time in a state of ebullition, it rises to the lip of the crater and runs over it down the sides of the hill and for some distance over the adjacent country. Sometimes the mass of molten lava carries away one side of the 1 Further observations upon this subject have been made by Prof. J. Milne in the Geological Magazine. Decade II. vol. v. p. 337. Hints on Collecting. 39 crater, forming a great breach through which successive streams of lava are poured. The eruptions of lava may be succeeded by fresh ejections of lapilli and ashes, and these again may be followed by more lava streams, the hill eventually consisting of stratified fragmentary accumulations with interbedded flows of lava. Occasionally the lava is also forced through fissures in these deposits, forming dykes or wall-like masses which intersect them in various directions, usually, however, assuming a somewhat radiate disposition around the cone. When the volcano has done its work as a safety-valve the eruptions may cease for a time, and the vent may become plugged in the manner already described. Should a fresh eruption occur, it may force a new vent. Ashes are again showered out, lava is again poured forth, and a new cone is erected within the old one, or little cones and fumaroles are formed on the sides of the hill and dotted over the surrounding country. CHAPTER V. THE COLLECTING AND ARRANGEMENT OF ROCK SPECIMENS. IN collecting specimens of rocks, it should be borne in mind that small pieces of compact and fine grained rocks answer the collector's purpose just as well as large ones, and often belter, should he have but a limited space in which to store them. Small specimens are easier to get than larger ones, and in the course of a day's work, a much greater number of small specimens can be carried. In collecting for a mu- seum, where there is plenty of available space, of course large specimens are best. About 5 inches by 4 inches square will be found a convenient size when they are properly dressed, but if it should not be advisable to dress them in the field, larger pieces should be collected, as it frequently happens that a roughly broken block is reduced to half its original 40 The Rudiments of Petrology. size before it is properly dressed. When a rock presents any large structural peculiarities, it will of course be necessary to collect proportionally large specimens in order to show the structure clearly. For ordinary private collections, specimens about 4 in. by 3 in. or 3 in. by z\ in. square are convenient sizes, or, if space be very limited, pieces about two inches square will suffice. A hammer with a tolerably heavy head made of Swede-iron, with steel ends welded on and well tempered, but not so highly as to be brittle even when used on the hardest rocks, will be found to be best suited for collecting. The shaft should not be less than 13 or 14 inches in length ; a tough wood such as ash answers very well for this purpose, and care should betaken in the selection of the wood. The eye into which the shaft is fitted ought not to be less than i inch in length by at least \ inch in breadth, and the head, which should have one end wedge-shaped, ought to be filed away slightly around the under opening of the eye to reduce the chance of breaking the shaft, as the fracture almost always takes place just under the head of the hammer. The author has been in the habit of using hammers with very heavy heads and with shafts long enough to serve as walking-sticks. Much heavier blows can be struck with a hammer of this kind than with a short-handled one, and their use does not necessi- tate such continual stooping. In some cases, however, a short- shafted hammer has its advantages, while for the purpose of dressing specimens one with a very long shaft is perfectly useless. When, therefore, a long- shafted hammer is taken into the field it is well also to carry a light dressing hammer. Sh rt- shafted hammers are most easily carried in a small leathern frog with a flap, on the back of which are fixed one or two little vertical straps through which a waist belt is run ; and this belt can also carry pouches for a compass and a clinometer. A strong canvas bag of tolerable capacity is necessary for carrying the specimens in, and it should have a little tab by which it can be loosely attached to a button on the back of the coat to pre- vent it from slinging forward when the wearer stoops. A good supply of paper should also be carried in which to wrap the specimens, and on the inside of each wrapper the precise locality from which the specimen is derived should be recorded. Hints on Collecting. 41 These may seem trivial details, but neglect of them often causes disappointment and inconvenience. With regard to the best shape for the crushing end of the hammer-head, some prefer it flat and square, and others rounded. When it is slightly rounded the hammerer is less liable to be struck by splinters of stone, but for chipping purposes a flat square face is best, and the dressing hammer should always have one such termination. In collecting, one of the first and most important things is to procure specimens which are unweathered or which have suffered as little as possible from atmospheric agency. Sometimes it so happens that a weathered surface of rock shows structural peculiarities which are especially worthy of note, owing to the different power which its component minerals possess of resisting disintegration and decomposi- tion. Interesting specimens of this kind should always be collected so long as their transport will not lessen the number of more interesting unweathered specimens. It is only from the latter that a true knowledge of the normal mineral and chemical composition of rocks can be derived. The writer lays especial stress upon this, as he has at times been greatly troubled by being requested to determine rocks from badly selected specimens in an advanced stage of de- composition. Where quarries occur there is no excuse for collecting such rubbish. In other cases it is often a matter of difficulty to get unweathered samples, and sometimes any specimen is better than none ; still, as a rule, the collection of weathered chips is time wasted, and it is far better to take a little extra trouble in order to get good and typical pieces. The specimens should ultimately be dressed with a small hammer, the piece of stone being held in the palm of the left hand, while with the right successive flakes and chips are struck off by sharp blows with the hammer. When very tough rocks are operated upon in this way it is by no means uncommon for the novice to end by getting a more or less rounded mass, covered all over with powdery, crushed sur- faces, resulting from the bruises made by the hammer, and which do not show the character of the rock. With practice 42 The Rudiments of Petrology. he will, however, soon ascertain the directions in which his blows will prove effective, and those where no amount of ham- mering will avail. When the specimen is properly dressed, one, or, still better, two little labels carrying numbers should be affixed to it. The use of two labels is desirable, since, if one becomes detached, the specimen can still be identified, so long as the other adheres. Ordinary strong gum answers fairly well, but in some cases, especially when the rock is soft and earthy, glue will be found preferable. Specimens are sometimes numbered with red sealing-wax varnish, but the figures are often difficult to find, and, when painted on rough surfaces, are not very legible. Of course, where num- bers are used, the specimens must be carefully catalogued. In other cases labels an inch or more in length may be affixed, with the name and locality of the specimens written on them, but they have the disadvantage of covering a larger space than the little numbered tickets. Should labels be used, care should be taken to stick them on the worst dressed and least interesting parts of the specimens. When rocks are very soft and earthy, numerals may be scraped on them and form more permanent records than labels, which often become detached, even without handling. In arranging rock specimens, various systems of classification may be adopted. When they are intended to illustrate the petrology of any particular district or country they have merely a topographical arrangement which seldom admits of any really scientific classification, since eruptive and metamor- phosed rocks have to be placed beside the sedimentary ones with which they are associated, and, so far as eruptive rocks are concerned, this does not always mean a chronological arrangement. The latter is certainly the right system to follow in dealing with sedimentary rocks, but with those of eruptive origin it is often very uncertain, and is of compara- tively little value, at all events, in the present state of our knowledge ; since eruptive rocks, almost identical in mineral composition, range from very early geological periods up to Arrangement of Petrological Collections. 43 the most recent times. For the purpose of teaching petrology a classification based upon the mineral constitu- tion of the specimens is doubtless the best, although, for general geology, the topographical arrangement is a useful one. When this is adopted in museums there should also be another small collection of rocks classified according to their mineral composition. A classification based upon structure is also to some extent to be commended, since it serves more or less as a grouping according to the condi- tions under which the rocks have been formed. Collections so arranged as to be illustrative of the stones used for build- ing and ornamental purposes also have their advantages, but the former should always include weathered examples of the rocks to illustrate iheir powers of resisting disintegrat- ing agencies, and, with this end in view, it would be most desirable to have an accompanying suite of weathered and partially decomposed stones taken from buildings, with the date when the building was erected recorded on them, or, if they have merely been used in the restoration of those edifices, the date of those repairs should be affixed to them, to show how much disintegration the stone has undergone since its surface was dressed and exposed in the building. In private collections, where rock specimens are usually arranged in cabinets, the drawers should not be less than two and a quarter inches in depth (inside measure), and deeper drawers are often very convenient. In museums, where space is ample, table-cases are best suited for the dis- play of the specimens. Wall-cases are objectionable, because the specimens on the higher and lower shelves cannot be seen with any degree of comfort, while, if the rooms be badly illuminated, the wall-cases are almost certain to be worse lighted than any others. Glass cases standing away from the walls and fitted with shelves which range from the height of an ordinary table up to about five or six feet, and so arranged that the higher ones gradually recede more and more from the glass, are very good for purposes of study. 44 The Rudiments of Petrology. Under any circumstances the receptacles for the specimens, whether cases or drawers, should be well fitted so as to ex- clude dust as much as possible. For eruptive rocks to be properly displayed in museums they require quite as good or even better illumination than minerals. This is also desirable for sedimentary rocks, but is of less importance as a rule. CHAPTER VI. PRELIMINARY EXAMINATION OF ROCKS. FOR the more general and preliminary examination of rocks the following implements will, if judiciously used (their use being backed by a moderate knowledge of mineralogy), be found sufficient for simple investigations. A stout-bladed penknife for testing hardness, &c. Small fragments of the minerals which constitute Mohs' Scale of Hardness. The diamond (No. 10 in this scale) may be omitted, as rock- forming minerals seldom have a hardness exceeding 8. A pocket magnifier, one of the ordinary pattern, having two or three lenses, will suffice for most purposes, but a good Cod- dington lens is also useful at times. There is, however, a disadvantage attending the use of these lenses when they are applied to the examination of rocks. This lies in the difficulty experienced by the observer when he attempts to examine the streak of minerals under the lens, especially when the minerals occur in very minute crystals or patches, as it is scarcely possible to hold a specimen, with a lens over it in focus, in one hand, and to work with a knife in the other. Laying the specimen on a table, and using a lens in one hand and a knife in the other, is a most unsatisfactory process, while the use of a lens fixed on an adjusting stand is scarcely better. To obviate this difficulty the author has devised a small lens with a clip, which can be worn on the nose like an eye-glass, and both hands are then at liberty / , Preliminary Examination of, other fingers being applied to the top, b c, of the fixed tube. The fine adjustment consists of a micro- meter screw, shown at a. The unattached portion of the inner tube is steadied in the outer one by means of a spring and three little screws set horizontally and capped with scraps of parchment. The arm of the microscope carries two screws with milled heads, one of which is shown at h. These are set at right angles to p^ one another, and serve to centre ' ' the tube in a manner presently to be described. Each eye-piece carries two cobwebs within it, which intersect at right angles in the centre of the field. To the outside tube of each eye-piece a small peg is fixed, which slides into a corresponding slot in the top of the inner movable tube of the microscope. This arrangement prevents any rotation of the eye-pieces, and so keeps the cobwebs in a fixed posi- 5 6 Tlie Rudiments of Petrology. tion. An analyser,/, fitting in a brass cap, slides easily over the top of the eye-piece. The bottom of the cap is sur- rounded by a bevelled flange, g, which is graduated to 5. An index mark on the plate, a e, serves to record the angle through which the prism may be rotated. The stage, /, of the microscope has a circular form, and a circular plate or object-table, /, is arranged so as to revolve horizon- tally on it. This table is graduated on its margin, and an index to record the amount of the revolution which may be imparted to it is attached to the front of the fixed stage at t. Beneath the stage is set an easily displaceable polariser, consisting of a Nicol's prism, which revolves within its external tube by means of the disc, v, which is graduated to 10, and has its index marked on the fixed outer tube, z. This polariser does not turn when the object- table is rotated, but remains unaltered in position. A plate of quartz, 375 millimeters thick and mounted in a little brass fitting, is shown at q. It slides into a corresponding slot, situated close to the lower end of the inner microscope tube and above the objective, which is omitted in the figure. A small plate of calcspar for making stauroscopic measure- ments is also supplied with the microscope, together with a brass ring for fixing it above the eye-piece and beneath the analyser. The following directions for using this microscope are extracted from Professor Rosenbusch's paper. 1 If any particular spot in an object, such as a granule of magnetite, be brought exactly under the point of intersection of the eye-piece-cobwebs, i.e. in the middle of the field of vision, and the object-table be then turned in its horizontal plane, the inner tube of the microscope will be found to hang neither vertically nor concentrically without the inter- vention of the centering screws, while the spot under observation will not remain in the centre of the field, and 1 Ein nenes Mikroskop fur mineralogiscJie und petrographische Untersuchungen. H. Rosenbusch. Neues Jahrbuch fiir Mineralogie. 1876. Prof. Roscnbusctis Microscope. 57 FIG. 22. under the point of intersection of the cross-bars a a and /3 /3, (fig. 22), but will describe an eccentric circle somewhat in the manner shown in fig. 22. The tube of the microscope must therefore be placed vertically ; in other words, the instrument must be centered by means of the two centering screws, one of which is shown at h in fig. 21. It will be seen by fig. 22 that the optical axis of the microscope is at o, and not at t v, and by the other in the direction fj. When these adjustments have been properly made, the spot should be brought exactly under the inter- section of the cross-bars in the eye-piece, and should remain there during the revolution of the object-table. When this operation is once properly performed, any other spot or part of an object which may be brought into the field will, upon rotation of the object-table, be found to revolve concentrically, so long as the same eye-piece and the same objective are used, but if one or other of them be changed, it will usually be necessary to re-centre the instrument. This will, however, generally entail only a slight alteration of the centering screws. The movement imparted to the microscope tube by these screws tends to throw the analyser slightly out of position with regard to the polariser, but the inventor finds that this produces scarcely any appreciable error. In testing the pleo- chroism of a mineral, the object-table bearing the sec- tion may be revolved above the fixed polariser, or the polariser may be turned beneath the stage, the graduations 58 The Rudiments of Petrology. affording facilities for determining the position of the pleochroitic maxima. The principal directions of vibration in a mineral sec- tion may be determined by inserting it in the maximum extinction of light between crossed Nicols, but since the eye is incapable, under these circumstances, of appreciating in certain cases very slight differences in the transmission of light by depolarisation, a calcspar plate is inserted in the stauroscope between the analyser and the section of the mineral under examination. The interference figure of' the calcspar appears distorted, until a direction of principal vibration in the section coincides with that of the polariser. In stauroscopic measurements very precise results may be arrived at by the employment, not of ordinary white light, but of monochromatic light derived from a coloured gas flame ; this method, however, although useful when an ordinary stauroscope is used, is inapplicable to microscopic research. In microscopic examinations a plate of quartz 375 millimeters thick is used instead of a calcspar plate ; where this is employed a monochromatic field is procured. When the principal direction of vibration in the analyser is turned at a different angle to that of the polariser, the field will become changed to various colours, where doubly refracting bodies are situated in the field of view, and their principal directions of vibration do not coincide with that of the polariser. By turning the object-table until such coincidence is arrived at, a purely monochromatic field will be produced ; very slight movement of the object will again suffice to destroy the monochrome. The employ- ment of such a quartz-plate is most useful when very feebly double-refracting media are being examined, and also for detecting isotropic particles in rocks with admixtures of amorphous paste of a doubtful character. In addition to the apparatus here described this microscope as manufac- tured by Fuess of Berlin (Alte Jacob Strasse, 108) is supplied with three eye -pieces and three objectives of Hart- nack's make, which give a range of amplification varying from Preparation of Sections. 59 about 90 to 1,150 diameters, and by the use of the different eye-pieces affording a series of 9 different powers. An eye-piece micrometer and an apparatus for heating objects under examination, and recording the temperature by means of a thermometer, are also supplied with this instrument. Some useful modifications of Prof. Rosenbusch's microscope have been made by Prof. Renard and described by him in the ' Bulletins de la Societe beige de Microscopic/ tome iv. 1877-78. In the 'Neues Jahrbuch fur Mineralogie, &c.,' 1878^.377, Professor A. von Lasaulx has described methods of con- verting ordinary microscopes so that they can be employed for the examination of minerals in convergent polarised light. Another paper by the same author, op. cit. p. 509, written a couple of months later, describes the construction of a polariscope, suitable for purposes of demonstration, which consists in part of the tube and Nicol's prisms of an ordinary Hartnack's microscope. A microscope devised for mineralogical and chemical purposes was devised some years since by Dr. Leeson. It was manufactured and improved by Mr. Highley, by whom it is described in the 1 Quarterly Journal of Microscopical Science,' p. 281. Another microscope, specially constructed for the examination of substances in hot acid solutions and corrosive fluids, has been devised by Dr. Lawrence Smith. In this instrument the stage is placed above the objective, and the object is viewed from the under surface of the slide. 1 CHAPTER VIII. METHOD OF PREPARING MINERALS AND ROCKS FOR MICROSCOPIC EXAMINATION. THE preparation of thin sections of minerals and rocks for microscopic examination, although effected by simple means, 1 American Journal of Science, 2nd series, vol. xiv. 1852. The instrument is figured in How to Work -with the Microscope, by Dr. Lionel S. Beale. 60 The Rudiments of Petrology. presents numerous difficulties to those who have had no previous experience in work of this kind. The object of the present chapter is to supply plain instructions concerning the needful appliances and the methods of manipulation by which such sections may be successfully made. It is true that sections may be prepared by lapidaries, 1 and that the student is thus spared considerable labour and loss of time ; but he will find, at all events in the earlier stages of his work, that there are certain advantages which he will derive from the preparation of his own sections. These advantages consist mainly in the facilities which he will have for testing the hardness of minerals, their deport- ment with chemical reagents, and the different appearances which they present when examined at intervals while the process of grinding them thinner and thinner is being carried on. The apparatus needful for such work is of a very simple kind, but more or less complex appliances for cutting and grinding will be found advantageous. As it is desirable to lessen the labour of grinding as much as possible, the first thing to be done is to procure a thin chip or a thin slice of the mineral or rock about to be examined. A square inch is a convenient size for the chip FIG 2 or slice, as such a piece will often undergo considerable diminution before it is reduced to a sufficiently thin state. Chips may be procured by using a small hammer, but fre- quently a number of flakes have to be struck off before one of suitable size, thinness, and flatness is got. When the specimen is very small, and difficult to hold in the hand while the hammer is used, a satisfactory chip may often be procured by holding the fragment in a suitable position on the edge of a cold chisel either let into a block of wood (fig. 23), or screwed into a vice, but then the 1 Mr. F. G. Cuttell (52 New Compton Street, Soho) prepares ad- mirable sections. Preparation of Sections. 6 1 operator must take care of his fingers. In the chipping of very hard rocks it is also advisable to protect the eyes, especially when the hammerer is not well practised in stone- breaking. For this purpose a pair of wire-gauze spectacles will be found useful. When cleavable minerals are to be dealt with it is best to avail oneself of the cleavage, but also to note in which direction of cleavage the plate is struck off, and, if it be desirable to make a section in some other plane than that of cleavage, a slitting or sawing process, hereafter to be described, is the only way in which such a section can be procured. When a suitable chip has been struck off the specimen, the first thing to be done is to grind one side of it perfectly flat. This may be accomplished either by grinding it by hand on a flat cast-iron plate with moderately fine emery and water, or by using a machine with a revolving leaden lap, similarly charged for the purpose. The former method is the more tedious, and, although preferred by some people, is far less convenient than the latter, supposing the operator to have a suitable machine at his command. There are various forms of machines which have been devised for this purpose, some of them being worked by a treadle and others by hand ; the latter are the more portable, but the former are usually considered easier to work. Machines of both kinds are manufactured by Fiiess, of Berlin, and other makers. Good treadle machines, devised by Mr. J. B. Jordan, of the ' Mining Record ' Office, may be procured from Messrs. Cotton & Johnson, 21, Grafton Street, Soho, and will be found to be well suited for the purpose. These machines are supplied with slitting discs for sawing off thin slices of rocks or minerals, and with laps for grinding them down to the requisite degree of thinness. The following details of construction, extracted from the Journal of the Quekett Microscopical Club, together with the use of the illustration, have been kindly furnished by the inventor : ' As will be seen from the diagram below, 62 The Rudiments of Petrology. this machine consists of a wooden frame-work, a a, support- ing a crank-axle and driving-wheel, the latter being two feet in diameter ; the top part of this frame is formed of two cross-pieces, #', fixed about an inch apart, as in the bed of an ordinary turning-lathe ; into the slot between them is placed a casting, B, carrying the bracket for the angle- pulleys, c ; this casting is bored to receive the spindle, D, FIG. 24. which, by means of a treadle, can be made to revolve at the rate of 400 or 500 revolutions per minute ; it is also fitted with another spindle, E, having a metal plate, F, fixed on the top, for carrying the small cup, H, to which the specimen is attached by means of prepared wax. This method of mechanically applying the work to the slicer is far preferable to holding it in the hand in the ordinary way ; Preparation of Sections. 63 the requisite pressure against the cutting disc is regulated by the weight, G, and the thickness of the slice by the thumb screw, K, on which the spindle rests. By this means, it is possible to cut thin and parallel slices the thinness of course varying according to the strength of the rock which is being operated upon. The slitting disc is made of soft iron, eight inches in diameter, and about ^ of an inch in thickness, and it is fixed on the spindle, D, between two brass plates or washers, four inches in diameter, by means of the nut, n. The lap or grinding disc is eight inches in diameter, of lead or cast iron about f of an inch thick in the centre, and having rounded edges arid slightly convex sides ; this form facilitates the grinding of uniform thinness, there being always a tendency on a flat surface (which soon wears hollow) for the edges of the section to grind away before it is sufficiently thin towards the centre.' In using such a machine for slitting off slices the edge of one of the thin iron discs should be charged with diamond dust. This should be worked into a paste on a slab or in a small watch-glass (a stand for which may be made with the rim of a pill-box cover), with a little sweet oil, and the mixture taken up in small quantities on the end of a crow-quill suitably cut, and it should then be applied carefully to the edge only of the disc, the disc being slowly turned by hand for a short distance, say an inch or two, and afterwards rubbed in hard with a short but thick piece FIG. 25. // ^ I it if '"'^'T/'Y )/, v of glass cylinder about half an inch long, fitted to a prongdd > J handle in such a manner that it acts as a roller (fig. 25). This process should be continued until the entire edge of the disc has been well charged. The piece of stone to be slit 64 The Rudiments of Petrology. should then be fixed firmly in the small metal cup which is afterwards to be clamped in the movable arm or plate pro- vided for its reception. The chip or fragment of stone may be fixed in the cup by means of Waller's wax, otherwise known as red cement. Small fragments of the cement should be placed in the cup and the whole held over a spirit lamp or a Bunsen's gas-burner until the wax is fairly melted. The piece of stone, previously warmed, is then pressed firmly into the wax. It is well to press the cement while yet warm closely round the fragment, which is best done with a cold metal point. It may then be allowed to remain until it is quite cold. After fixing the cup into its arm or plate, the latter should be adjusted to the proper height for the disc to make the first slice. Suitable pressure should Jthen be applied either by hand, by a pulley and weight, or by an elastic spring, fitted by one end to an upright rod on the table of the machine and by the other to a stud fixed on the carrying arm. Under any circumstances it is better to assist the regulation of pressure by hand. It should also be observed that, in commencing the process of slitting, the edge of the disc should first be brought into contact with a comparatively flat or rounded surface of the fragment of stone and not with a sharp edge, as, in the latter case, the diamond dust will probably be stripped off during the first revolution of the disc. This is a point to be very carefully observed : indeed it is better for the first few revolutions to apply the pressure entirely by hand. Oil-of- brick, or a mix- ture of soft-soap and water, should then be applied to the edge of the disc at a spot just in front of the stone, so that it may be properly lubricated before it traverses the hard surface, and on no account ought the disc to become dry while cutting, or the diamond edge will instantly be lost. The application of the lubricant may be made either by means of a brush held in the left hand or by a dripping apparatus, such as a tin pot with a very small tap fixed in Preparation of Sections. 65 the bottom. The disc should next be set in motion and a steady pressure and constant lubrication kept up until the slice is cut off. The carrying arm should then be raised, say one-eighth or one-sixteenth of an inch, according to the tenacity of the mineral or rock which is being dealt with, and a like process should be repeated until the second cut is finished, when the slice is ready for grinding. The pro- cesses connected with the grinding of a sawn-off slice and of a hammer-chipped fragment are identical. The leaden lap should be substituted for the slitting disc. Two pots or saucers should be at hand, the one filled with moderately fine emery and the other with water. A house-painter's brush (a small sash tool as it is technically termed) should then be dipped in the water and afterwards in the emery, and the resulting paste smeared all over the upper surface of the leaden grinding-lap. The machine should then be set in motion and the slice or fragment be firmly pressed on the surface of the lap by the fingers of one or of both hands, care being taken to keep the finger-tips clear of the revolving lap. A little practice will soon teach the operator the best way of doing this. When a good, even surface is procured in this way, the slice or chip ought to be carefully washed and wiped to free it from all adhering particles of coarse emery, and then the some- what rough surface should be rendered as smooth as possible by grinding the fragment by hand on a flat brass slab, or on a slab of thick plate glass, about six or seven inches by four or five inches in diameter, smeared with the finest flour-emery and water. The motion of the hand in grinding should be a circular one, and it should be carried systematically all over the plate, so that the latter may not become unequally worn. When a perfectly smooth surface is procured the process must be stopped and the fragment again thoroughly washed and cleansed from all adhering emery. The next process consists in cementing the smooth F 66 The Rudiments of Petrology. FIG. 26. surface of the stone to a small slab of plate glass about two inches square and about a quarter of an inch or more in thickness, the edges being previously ground roughly on the lap to avoid the risk of cutting the fingers, in case it should slip when pressed on the revolving disc. One of these small glass slabs should be placed upon an iron, brass, or copper plate, sup- ported either on a tripod or by other means, over a Bunsen's gas jet (fig. 26) or a spirit lamp, 1 and a few scraps of the oldest and driest Canada balsam which can be procured should be laid upon the top of the glass, the piece of stone to be cemented also being laid on the iron plate (but not on the glass slab) with its smoothly ground surface uppermost. The jet or lamp should now be lighted and the gradual liquefaction of the balsam carefully watched. As soon as the balsam liquefies (it ought on no account to be allowed to approach ebullition) the piece of stone should be taken up, reversed, and its smooth surface pressed into the balsam and on to the surface of the glass slab (fig. 27). The slab should then be pushed to the edge of the hot plate so 1 A good form of spirit lamp (fig. 28) made for cooking purposes, but admirably adapted for microscopical mounting, is sold by Scott & Son, 42 Bedford Street, Strand, London, and costs less than two shil- lings. It is made of tinned iron ; the wick is stuffed into a tinned iron cylinder an inch in diameter, and a cap with a brass collar screws over the burner when it is not in use. On the sides of the lamp are three small sockets which carry bent iron wires, thus forming a strong tripod upon which a copper or iron plate can be placed for mounting. FIG. 28. Preparation of Sections. 67 that it can be conveniently removed by a small pair ot tongs. The tongs best adapted for this purpose consist of an ordinary wine-cork cut in half, the separate halves being fixed to a jointed arm, such as an old pair of compasses or dividers (fig. 29). With these the corner of the slab should be firmly held, and it ought then to FIG ^ be placed upon a piece of wood or thick paste- board laid on the table and the piece of stone firmly squeezed and held down on the glass slab until the balsam begins to harden. For this purpose the cork ends of the tongs may be used, as the stone is usually too hot to be fingered. All these processes must be rapidly performed. The slab should then be taken up and examined from its under side to see that no air-bubbles have been included between the glass and the stone. Should they be present in any quantity, the slab must be again placed on the hot plate, the balsam liquefied, the stone detached, and both stone and glass cleansed from balsam by means of tur- pentine or benzol, and the whole process of cementing repeated, as otherwise there would be great risk, if not cer- tainty, of the stone becoming detached from the slab in the later stages of grinding, and it is far better to undergo a little additional work in this way than to run the chance of sacri- ficing all the labour previously devoted to the preparation of the chip and that subsequently expended on the second stage of grinding. It is needful to impress the operator with the importance of using hard or dry balsam in cementing the chip to the slab, otherwise failure is almost sure to ensue. When the balsam is fairly hardened so that scraps of it surrounding the stone can be scraped off and rolled, almost without adhesion, between the fingers, the slab ought to be taken up, reversed, and the adhering stone ground again with coarse emery and water on the iron plate or on the leaden lap of the machine. When the stone has been re- F 2 68 The Rudiments of Petrology. duced sufficiently to transmit light, great care must be taken, and, if the section be very thin (i.e. if the stone be naturally rather opaque in thin plates), the pressure upon it should be diminished. The process of grinding with coarse emery must not be carried too far, as, when the section is ex- tremely thin, it may often be entirely removed by one or two turns of the lap. The slab and its adhering section should then be thoroughly washed and freed from all traces of emery, and the final grinding conducted by hand on the brass plate or glass slab with flour-emery and water. In the very latest stages a few drops of paraffin may advan- tageously be used in order to diminish friction. During this final grinding the section should be frequently examined under the microscope, but must be thoroughly washed and cleansed prior to each examination and a drop of turpentine placed on its surface to increase its transparency. Now is the time to apply tests to the different component minerals, if the section be a rock or an impure mineral, and doubt exist as to the nature of any of the substances present. Some operators employ rouge for polishing the section in its very last stage of preparation, using a piece of parchment as the surface on which to polish it. The slab, with the adher- ing section uppermost, must once again be placed on the hot plate, while a watch-glass containing turpentine and placed on the rim of a pill-box or other support should be near at hand. When the balsam is thoroughly liquid, the opera- tor should take the slab off the hot plate with the cork tongs, and by means of a blunt-ended wire (fig. 30), held in the right hand, gradually push or slide the section off the slab into the turpentine contained in the watch-glass. He ought then to hold the watch-glass by means of a wire ring, or a pair of crucible-tongs or forceps, over the lamp or gas jet suf- ficiently long to heat the turpentine or even to make it boil, but the watch-glass must be kept at a suitable distance from the flame to prevent ignition. It may then be replaced on its support, and the section should be very gently turned Preparation of Sections, 69 over and washed in the turpentine by means of one, or two, small camel's-hair brushes. A glass slide (those in ordinary use for microscopic preparations are perhaps as good as any for size ; i in. by 3 in.) must then be placed on the hot plate, which, however, should have been allowed to cool beforehand ; a drop FlG . 3 1 1 5 / /' more or less translucent matter (fig. 65), xbut income very j small crystals no clear area is visible in the Centre, ' /,', even in excessively thin sections, the whole GT/S tal being nearly or quite opaque. The striae the quadrangular sections cross one another at right angles. Crystals of hauyne from other localities possess a clear border with sharply-defined internal ' and external boundaries, the inner portion of F the crystal being crossed by striae which some- times intersect at right angles (fig. 68), and at others follow three directions, each set of striae forming a series of parallel lines which run either at right angles to the opposite faces of the crystal (fig. 67), or pass in directions which would corre- F spond with lines drawn between opposite angles of the six-sided section (fig. 66). In both cases the striae do not intersect, but are divided either by dark or clear lines which radiate from the centre of the crystal ; in the former case joining opposite angles, and in the latter passing from the centre to the middle of the faces. The striae are found under high amplification to consist of opaque granules, gas pores, and minute glass inclosures. Sodalite occurs in rocks either in an uncrystal- lised condition, or in crystals which yield six- sided or else qua- drangular sections ; the latter are very frequently distorted, so that the alternate or opposite sides of the section are unequal. They usually present a yellowish, grey, or blue colour. The rhombic-dodecahedral cleavage is represented, under the microscope, by undulating cracks. The crystals sometimes appear to be remarkably pure, at others they are crowded with various inclosures ; the most common being steam pores. A singly-refractive substance containing fixed bubbles also occurs in some sodalite and this singly-refracting matter, which frequently has a dark border, forms well defined I 2 1 1 6 The Rudiments of Petrology. rhombic dodecahedra which contain them; they do not, however, contain any bubbles. Fluid lacunae which change on the application of heat are found sparingly in the sodalite of Somma (Vesuvius)^ while complete included crystals often contain inclosures of nepheline, augite, meionite, and biotite. OLIVINE * I is a common constituent of many eruptive rocks, in which / it occurs sometimes in the form of crystals whose angles ' 1 frequently appear more or less rounded, and sometimes as 1 rounded granules which in some cases form rounded aggre- gates, occasionally showing traces of crystal faces. The faces most commonly presented by crystals of .olivine are those of -the rhombic prism oo P (giving an angle of little over 130), the rhombic pyramid P, the macropinakoid oo P oo, the brachypinakoid oo f oo, the macro- and brachydomes P oo and oo and the basal plane o P. Olivine has two di- / rections of cleavage, one parallel to the macro- the other to | the brachypinakoid, the former being very imperfect and the Ljatter rather distinct. Olivine varies in hardness from 6 '5 ^to 7. Its fracture is conchoidal and its colour is not only I olive or bottle-green, but at times brownish and yellowish. Where, therefore, olivine occurs in rounded imbedded gra- nules, and displays neither its characteristic form of crystal nor its common green colour, its hardness and its conchoi- dal fracture cause it greatly to resemble quartz, which also occurs in rounded granules in certain rocks of eruptive > J?^ origin, but a mistake of this kind is only likely to occur in the hasty examination of hand-specimens in the field. The other minerals, however, with which the granules are asso- ^ciated usually give some clue to their probable nature. Powdered olivine is easily decomposed by hot hydrochloric . acid or by sulphuric acid ; separation of gelatinous silica occurring in either case. Before the blowpipe alone, only j the highly ferruginous varieties are fusible, forming a black Olivine. u/ magnetic globule ; but with borax the ordinary olivine may J be fused to a clear green bead/ The foregoing reactions ~ certainly serve to distinguish olivine from quartz, but the microscopic characters of the two minerals are sufficiently distinct to prevent any chance of mistake. The general formula of olivine is (MgO, FeO) 2 SiO 2 . It sometimes contains some lime, alumina, and protoxide of manganese, traces of titanic, phosphoric, and chromic acids, and the pro- toxides of nickel and cobalt. Potash, soda, and small quan- tities of water have also been detected in olivine. / Olivine"} frequently occurs in meteorites, often forming a large pro- 1 portion of them. In some of these meteoric olivines traces of arsenious acid, fluorine and oxide of tin have been dis- covered. The alterations which olivine undergoes are either due to peroxidation of the iron, or to its removal by water charged with carbonic acid, in which case some of the magnesia may also be removed. Changes of the former] class cause the mineral to assume a reddish or brownish^ colour, and sometimes render it iridescent. This process when further advanced sets up a micaceous structure in the mineral. 1 Changes of the latter class, viz., by water charged with carbonic acid, give rise to the formation of serpentine, steatite, &c. Many of the Wurtemberg basalts, especially those from H6\ven, are little more than serpentine rocks containing some magnetite, since the olivine and -augite which composed the basalt are changed into serpentine. 2 Under the microscope olivine appears almost colourless") when in very thin sections, and of a light greenish tint in those of moderate thinness. Olivine shows very feeble dichroism, even when thick sections are examined/ In very thin sections the dichroism is scarcely perceptible. It is doubly refractive, polarising when in a fresh, undecomposed state, in moderately strong colours, but these are much more feeble than those displayed by quartz. The surfaces of sec- 1 Dana, System of Mineralogy, 5th edition, p. 258. 2 Letter from Dr. M. G. Reinhold Fritzgartner. 1 1 8 The Rudiments of Petrology. \tions of olivine are nearly always rough, since the ordinary grinding is never capable of imparting a smooth polished face to the section, and these roughened surfaces, which, when ex- amined under the microscope, present an appearance some- what like that of ground glass, are clearly perceptible in sections mounted in Canada balsam and covered in the usual way. Lines of accretion, such as those which form zones cor- responding with the boundaries of sections of augites and felspars, have not yet been observed in sections of olivine, even when they occur in rocks in conjunction with zoned crystals of those minerals. 'Olivine is not known to occur in the form of very minute crystals or microliths, the crystals and granules being usually sufficiently large to be distinguished without the assistance of a lens. Crystals of olivine generally afford six- oreight-sided sec- tions, the angles of which are often rounded, a fact regarded by some observers as indicative of a secondary fusion. Quadrangular sections are not uncommon, and unsymmetri- cal forms frequently result from obliquely-cut crystals. Olivine is often traversed by strong and irregular fissures which bear no relation to the form of the crystals, and some- what resemble the fissures visible in suddenly cooled glass. The alteration of olivine into serpentine commences along these fissures and also along the boundaries of the crystal. It appears under the microscope as a finely fibrous green fringe, the fibres lying at right angles to the surfaces from ' which they originate. As the alteration proceeds, this fibrous structure extends further inwards until the whole crystal is converted into a mass of interlacing and contorted, or radially disposed, fibres, and no longer displays any of the optical characters which formerly belonged to it. r Gas /pores, fluid lacunae, consisting generally of liquid carbonic acid, glass lacunae and crystals, granules and microliths of magnetite are the most common inclosures which occur in olivine. The gas pores are often arranged in rows. The Hyper sthene and Enstatite. 119 microliths frequently assume peculiar forms, being sometimes zig-zag, sometimes claviform or acicular. Inclosures of augite, felspar, leucite, &c., are never met with in olivine. HYPERSTHENE. v Crystalline system rhombic. Usually occurs, in rocks, in a crystalline condition or in granules, but seldom in actual crystals. The mineral shows a strongly marked cleavage parallel to the brachypinakoid, very imperfect cleavage parallel to the macropinakoid, and a tolerably perfect pris- matic cleavage. The colour is black, greyish, brownisrA green, or pinchbeck-brown, and presents in certain directionsj a glistening appearance or bronze-like lustre. --' The chemical composition is (MgO, FeO) SiO 2 . Before the blowpipe it fuses to a black enamel, and, on) charcoal, to a magnetic mass. When tested with a single Nicol, thin sections of hyper- sthene appear strongly pleochroic. The colour is mostly grey- ish-green in the direction of the principal axis ; in the direc- tion of the macrodiagonal it is reddish-yellow, and in that of the brachydiagonal it is hyacinth-red. Hypersthene is not, however, so strongly pleochroic as hornblende, but it pos- sesses this character in so marked a degree that it at once serves to distinguish it from diallage, bronzite, and enstatite. The hypersthene of Labrador shows numerous imbedded lamellae which were regarded by Vogelsang as possibly diallage, but A. von Lasaulx thinks this improbable, since similar imbedded lamellae have been observed in diallage from several localities. These plates have been referred by different authors sometimes to gothite, to specular iron, to brookite, &c. Their true nature is, however, as yet unde- termined. Minute granules of magnetite are of frequent] occurrence in hypersthene, often following one another inj definite directions parallel to the principal axis./ Boficky mentions the occurrence of minute plates of calcite imbed- I2O The Rudiments of Petrology. ded in the direction of principal cleavage in the hypersthene of a rock from Tellnitzthal. 1 ENSTATITE. Crystalline system rhombic. Cleavage parallel to the faces of the rhombic prism oo P. The cleavage planes make by their intersection angles, according to Descloizeaux, of 88 and 92 ; according to Kengott 87 and 93. Enstatite also affords two other directions of less perfect cleavage, one parallel to the macro-, the other parallel to the brachy- pinakoid. The cleavage surfaces sometimes have a fibrous appearance, and usually a rather pearly or vitreous lustre, while in some varieties it is metalloidal. In colour the mineral ranges from greyish- or yellowish-white to green and brown. It occurs massive and lamellar as well as crystal- lised. It is almost infusible before the blowpipe, the edges of only very thin splinters becoming rounded. It is insolu- ble in hydrochloric acid. Enstatite becomes altered to schiller-spar or bastite, talc, &c. The formula of enstatite is (MgO, FeO) SiO 2 . Enstatite is strongly dichroic, the greatest colour differ- ences being clear brownish-red and pale sea-green. Enstatite occurs in Iherzolite and certain gabbros, and, according to Professor Maskelyne, some of the diamond- bearing rocks of South Africa contain more or less of this mineral. BRONZITE. This mineral crystallises in the rhombic system. It is very difficultly fusible, the edges only of very thin splinters becoming rounded before the blowpipe. It is insoluble in acids. In its microscopic structure it is more closely allied to hypersthene than to enstatite. It resembles the latter in its strongly fibrous structure, and the former in the character of its numerous inclosures which occur in the form of lamellae, or, as in the Kupferberg bronzite, of ' elongated 1 Petrographische Studien an den Basalt gesteinen Bohmens, von Dr. E. Boficky, p. 16. Prag, 1874. Pyroxene. 1 2 1 stripes of an acicular character set perfectly parallel to the fibrous structure and varying in colour from dark honey yellow to deep brown.' 1 Some bronzite is very feebly dichroic, or scarcely di- chroic at all. In this respect bronzite differs both from ensta- tite and hypersthene- The directions of cleavage in bronzite are, parallel to the brachypinakoid, highly perfect ; parallel to the right rhombic prism, imperfect ; parallel to the macropinakoid, most imperfect. The formula of bronzite is (MgO, FeO) SiO 2 . Some analyses show the presence of lime and alumina. PYROXENE GROUP. All the minerals of this group crystallise in the mono^ clinic system and are bi- silicates of different protoxides, having the general formula RO, SiO 2 , the protoxides often being converted into sesquioxides. The protoxides are lime, magnesia, sometimes potash and soda, and the protox- ides of iron and manganese. The different members of this group vary more or less in composition, a fact due to the respectively isomorphous character of their protoxides and sesquioxides. The most important difference in the chemical composition of the pyroxenes and the amphiboles, both of which groups have the same general formula, lies in the fact that in the former lime is in all the varieties of the group an important constituent, while in some of the varie- ties of the latter group it is either totally absent or occurs only in very small quantity. Chemically, the pyroxenes may be divided into those which are poor in, and those containing from three to over nine per cent, of, alumina. The diallages in this respect range between the two divisions. Some of the so-called diallages (metalloidal diallage), belong rather to enstatite than to pyroxene, since the crystallisation is rhombic. Des- cloiseaux regards the diallage occurring in the Cornish serpentines as probably a form of enstatite. 2 1 Mik. Beschaff. d. Min. u. Gest., Zirkel, p. 187. 2 System of Mineralogy, J. D. Dana, 5th edition, 1871, p. 209. In 122 The Rudiments of Petrology. Augite crystals frequently occur twinned, the plane of ! twinning being parallel to the orthopinakoid. The angle formed by the oblique rhombic prism in augite is 87 5'. The corresponding angle in hornblende crystals is 124 30'. This great discrepancy in their angular measurements serves to distinguish the minerals of the one species from those of the other, and even under the microscope, when the planes of sections often lie obliquely to the principal crystallographic axis, a rudely approximate measurement frequently enables the observer to discriminate between hornblende and augite. Another means of dis- . languishing between these two minerals is afforded by the strong dichroism which usually characterises hornblende, and the very feeble dichroism presented by sections of augite under the microscope ; the latter giving only a succes- sion of slightly different tints, while in hornblende an actual .difference in colour is to be observed. It sometimes happens, however, that very thin sections of hornblende exhibit only feeble dichroism, and that thick sections of augite may show this character in a rather marked manner. Still the test of dichroism may, as a rule, be accepted with considerable safety. This character is best shown, in thin sections, by removing the analyser from the microscope, and by rotating the polariser. Such dichroism as augite sections may exhibit is always, according to Mr. Allport's observations, 1 connected with a purple tinge, and he has noted this fact in augites occurring in rocks of very different ages and derived from widely separated localities. He also states that dichroism is more strongly elicited from those augite crystals which by ordi- nary illumination exhibit a variation in colour, i.e. when one end or side of a crystal appears purplish-brown and Descloiseaux's Mamiel de Mineralogie, t. i., 1862, there is, however, no mention of this ; the Cornish serpentines being there, and at that time, cited as containing diallage. 1 ' On the Microscopic Structure and Composition of British Car- boniferous Dolerites,' S. Allport, Q. y. G. S., vol. xxx. p. 536. Augite. 123 shades off into yellowish-brown towards the opposite end or side. By ordinary transmitted illumination thin sections of! augite appear of a greenish or yellowish-brown colour. ; Wedding has noted the occurrence of greenish and brown- ish layers, sharply defined, in the augites of the Vesuvian lavas, and concludes that the latter colour is not due to oxi- dation of ferrous compounds constituting the greenish por- tions of the crystals. Similar bands, frequently well marked and of variegated colour, are often to be seen by polarised light in sections of augite and represent twin lamellae /which lie parallel to the orthopinakoid. Even by ordinary transmitted light they may sometimes be recognised as exceedingly delicate striae. Sections of hornblende and augite, when cut at right angles to the principal axis, may usually be distinguished by the augite giving eight-sided sections, as a rule ; while those of hornblende are generally six-sided. In some sections of augite crystals striae are visible under the microscope, which correspond with the boundaries of the section, or, in other words, with the external form of the crystal. These lines lie one within the other, constituting zones of variable width, and are often rendered still more apparent by granules of magnetite, microliths, and small cavities, which follow or coincide with these lines in a re- markably regular manner. In polarised light the concentric bands exhibit different tints and appear more strongly marked. They no doubt represent lines of accretion. The inclosures in augite crystals consist, so far as they \ have yet been observed, of acicular microliths of augite itself. Triclinic felspars and leucite also occur in minute crystals imbedded in augite, and the sides of augite crystals are often penetrated by crystals of apatite. Magnetic and titan iferous iron are of frequent occurrence in augite, some- times almost entirely replacing it. Furthermore, augite often contains inclosures of amorphous glass, sometimes 1 24 The Rudiments of Petrology. (spherical, sometimes irregular in form, and occasionally cavi- Jjies containing fluid have been observed. In some excep- tional cases the augites of certain basalts are merely repre- sented by a thin line or border of augitic substance, the interior of the crystal being filled with the admixture of minerals which constitutes the ground-mass of the rock. (The nature of the minerals, &c., inclosed in augite crystals idepends of course very materially upon the mineral compe- tition of the rocks in which the augites occur. This mineral does not merely occur in rocks in the form of well-defined crystals, but also as acicukr microliths which are generally greenish or yellowish-brown, unless they are exceedingly minute, in which case they are almost colourless. These microliths vary considerably in form, some being straight, some curved, while occasionally they are forked at the ends, or assume club-like forms. Sometimes they lie independently scattered through the matrix of the rock, at others they are clustered together, especially around the margins of crystals. Diallage, which is a common constituent of some rocks, is usually regarded as a variety of augite. It exhibits a highly perfect cleavage parallel to the clinodiagonal, thus differing from augite, in which the corresponding cleavage is imperfect. Both of these minerals also possess a perfect, or tolerably perfect, cleavage parallel to the faces of the oblique rhombic prism (PI. II. p. 173).- Diallage resembles augite in displaying very weak dichroism, so weak that it gives rise to no marked dif- ference of colour, and may thus be distinguished from horn- blende and hypersthene. Thin plates of diallage when ex- amined microscopically are often seen to include numerous little tabular crystals and acicular microliths. The tabular crystals mostly run in definite planes parallel to the ortho- and to the clinopinakoids of the diallage and also to a third plane which lies obliquely to that of the orthopinakoid. ' Sections taken parallel either to the ortho- or clinopinakoid always exhibit the broad sides of those little imbedded Diallage. 125 crystals which are conformable to the face under examina- tion, together with the line-like transverse sections of similar crystals which lie with their broad sides parallel to the other face/ 1 The acicular microliths in some diallages also follow definite directions and appear to cross one another obliquely, forming a somewhat lozenge-shaped net work. Fluid lacunae have likewise been observed in the diallage of a Silesian gabbro. Diallage sections usually exhibit a pale greenish or brownish-yellow tint by ordinary transmitted light. Asbestus is in some cases a fibrous form of augite, but most asbestus is hornblendic in its affinities. Breislackite is a capillary or woolly form of pyroxene, occurring in the lavas of Vesuvius and Capo Di Bove. Although definitely known to be a pyroxene, the precise species to which it belongs has not yet been determined. Diallagic Augite. A form of pyroxene intermediate in character between augite and diallage has been noticed by Professor E. Boficky, of the university of Prag, in his ' Petrographische Studien an den Melaphyr Gesteinen Bohmens,' p. 19. He describes it as an augite whose sec- tions may be distinguished from ordinary augite by the occurrence of straight and parallel fissures or striae which, in longitudinal sections of the crystals, cross the coarser cleav- age planes at angles of from 70 to 90. Professor Boficky considers these lines to represent the edges of sections of - delicate lamellae which he regards as twin-like intergrowths, most probably lying parallel to the basal planes of the crys- tals. He states that in the melaphyres of Neudorf near Lomnitz the diallagic augite sections are broad, irregularly bounded, and contain numerous bubbles and stone lacunae or 'stone cavities' (Schlackenkornchen] which are often ranged in lines, either parallel to the fissures or actually along them. Such fissures or striae often occur in one part only of an individual crystal. The mineral is not dichroic, and polarises in strong colours, the crystal sections sometimes presenting 1 Mikroskop. Beschaff. der Min. und Ge.st., Zirkel, p. 181. 126 The Rudiments of Petrology. iris- coloured margins. He has noted their occurrence in the melaphyres of Hofensko, Lomnitz, Neudorf near Lomnitz, and Zdiretz in Bohemia. ALTERED CONDITIONS OF PYROXENE. It is important that the microscopist should have some knowledge of the alterations which pyroxene undergoes, in order rightly to understand the nature of the pseudomorphs after the different members of this group which so frequently occur in eruptive rocks. The simplest kind of alteration which pyroxenic minerals experience is hydration. It is frequently accompanied by a loss of silica. The lime and iron contained in these minerals also undergoes considerable diminution, or is totally removed by water charged with carbonic acid or holding carbonates in solution. The following are some of the results of the alteration of pyroxene. TABLE SHOWING THE APPROXIMATE CHEMICAL COMPOSI- TION OF THE VARIOUS MINERALS WHICH RESULT FROM THE ALTERATION OF PYROXENE. Si a & *e Fe Mn Mg Ca Na * Ti H Augite . Pycrophyll . Pyrallolite . Schiller-spar . 51 494 43 3 2 4 2 - 6 6 ii 3 i 36 24 - - 10 12! Rammelsberg Rose Runeberg Kohler Epidote 46 el _ JO 8 12* 9 5 Streng Mica . 43 15" 23* io| i i 5 Kjerulf Uralite . 49 I 12 "4 i Rath Glauconite . 5 J 4 7 21 _ 6 2 6 6| Delesse Serpentine . 4 1 2 2 42 13 Scheerer Steatite *4 3 1 - 5 Richter 64 I . 28 Tengstrom Palagonite . 42 16 7 7 2 12^ Waltershausen Hematite 100 Limonite 2 4 16 Ulmann Magnetite . Titaniferous ) Magnetite j - - 69 22 68 31 30 2 - - - - 2 - Knop Michaelson The list is headed by a typical analysis of unaltered augite from Schima in Bohemia. The decimals in the Amphibole. 127 original analyses have been omitted, and the fractions rather differently distributed, in order to simplify the columns for reference. In addition to these analyses may be cited one of a highly siliceous pseudomorph after pyroxene, by Rammels- berg, consisting of 85 per cent, silica, \\ alumina, 2 protox- ide of iron, 2 magnesia, 2\ lime, and 6 water. Serpentine, steatite, and limonite are probably the most common of these alteration-products in British eruptive rocks Epidote seldom gives direct evidence of its deriva- tion from pyroxenic minerals, since it generally occurs in characteristic crystals or radiating tufts along minute lines of fracture, and not in pseudomorphs bounded by the outlines of pyroxenic forms. Nevertheless, it does not follow that it has not, in these cases, resulted from the decomposition of pyroxene. Pseudomorphs after pyroxene of quartz, opal, and calcite are also of occasional occurrence, but these are more probably due to subsequent infiltrations, than to a partial' removal of the basic constituents, or, in the case of opal, to the hydration of a residuum of silica. The usual mode of deposition of hydrous silica, opal hyalite, &c., in vesicles and cavities, and on the surfaces of joint planes in eruptive rocks, tends to support this view. 1 AMPHIBOLE GROUP. The minerals of this group closely resemble pyroxene in chemical composition, while they also crystallise in the same system (monoclinic). They differ, however, as already pointed out, in the angular measurements of the oblique rhombic prism, which in hornblende is 124 30' and in augite from 87 5' up to 92 55'. They are all bi-silicates 1 According to Mitscherlich, Berthier, and G. Rose, tremolite and actinolite (both varieties of amphibole) yield, when fused in a porcelain furnace, forms similar to those of pyroxene. Crystals of augite are of common occurrence in blast furnace slags, sometimes even associated with hornblende crystals, but the latter are less frequently met with ; and it is stated that Mitscherlich and Berthier, although able to produce artificial crystals of augite, failed to procure any of hornblende. 128 The Rudiments of Petrology. of protoxides and sesquioxides, the former being lime, mag- nesia, soda, potash, and the protoxides of iron and manga- nese, while the latter are represented by alumina and the peroxides of iron and manganese. Crystals of amphibole differ from those of pyroxene, not merely in the -angular measurements of their oblique rhombic prisms, but also in the angles at which their cleavage planes intersect. In both groups the relation of the cleavages to the respective faces of the crystals is the same, but they differ in their respective facilities ; the cleavage parallel to the faces of the oblique hombic prism in hornblende being more perfect and more trongly marked than the corresponding cleavage in augite, while the cleavages parallel to the pinakoids are on the other hand less strongly marked in hornblende than in augite. Furthermore, the discrepancy in the angles of the two respec- tive oblique rhombic prisms begets a corresponding discre- pancy in the angles at which the prismatic cleavages of the two different species intersect. This circumstance is of con- siderable value to the mineralogist, since it is often difficult or impossible to measure the angles of the actual crystallo- graphic faces, but it is generally possible to measure the angles of cleavage, and since these cleavages respectively coincide with the plane co P, the results deduced from cleavage are as good as those derived from the actual faces of the crystals. The crystals of minerals belonging to the amphibole group usually exhibit a fine longitudinal striation. The very feeble dichroism of augite and the strong di- chroism of hornblende has already been mentioned, and, although exceptional cases may occur, it must nevertheless be regarded as a most valuable test, especially in the micro- scopic examination of rocks which contain minerals pertain- ing to one or other of these groups. The amphiboles, like the pyroxenes, may be divided into two sub-groups, viz., those which contain little or no alumina, and those which are rich in that base (sometimes Hornblende. 1 29 containing over 15 per cent.) The similarity in the chemi- cal composition of amphibole and pyroxene begets a simi- larity in the minerals which result from their alteration, so that the alteration-products tabulated at the end of the pyroxene section may be taken as fairly representative of the changes which amphibole undergoes. In the microscopic examination of thin sections of hornj blende, the forms are often so irregular, that it is difficult to arrive at any sound deductions from their contours. Colour also affords no safe means of discriminating between pyroxene and amphibole, since the members of both groups exhibit greenish and brownish tints. /The augites and j hornblendes which occur in basalt are mostly brownish in colour. The hornblende in syenite is also generally brown, but that which occurs in phonolite is mostly of a greenish tint while the augite in leucite-lavas is, as a rule, also green. The most important microscopic characteristics of common hornblende may be summed up in the following manner. Transverse sections of crystals show two sharply' denned sets of cleavage planes, each set corresponding with the opposite and alternate faces of the oblique rhombic]'; prism/and intersecting at an angle of 124 30' (when the section is at right angles to the principal crystallographic axis). In sections taken parallel to the chief axis of the n crystal a fine longitudinal striation, indicative of a fibrous structure, is often to be observed. Furthermore, the dichroism of hornblende is very strong, although the clear green varieties, as pointed out by Zirkel, show only very feeble dichroism and might be mistaken for augite. 1 Some of the small crystals, such as those often occurring in phonolites, appear to be made up of fine parallel rods or elongated microliths, the margins being frequently very irregular. Microliths of hornblende are of common occurrence in 1 Zirkel, Mik. Beschajf. d. Min. u. Gesteine, p. 169. K 1 30 The Rudiments of Petrology. I eruptive and metamorphosed rocks ; when very minute they are often colourless. They seldom present any recognisable crystalline form, but frequently exhibit longitudinal fibrous structure, often appearing to be frayed out at the ends. Zone-like bands of accretion, similar to those which are sometimes visible in sections of augite, also occur in horn- blende sections, t Magnetite, apatite, nepheline, biotite, j quartz, &c., occur, enclosed in hornblende crystals ; glass lacunae are also frequently met with in the hornblendes of Ltrachytes, phonolites, pitchstones, &c. The hornblendes of syenites, diorites, &c, contain as a rule fewer microscopic inclosures than those of volcanic rocks, such as trachytes, phonolites, basalts, &c. Crystals of hornblende frequently exhibit a dark granular ^border of magnetite, and at times the latter mineral has been observed greatly to dominate over the hornblende, the horn- blendic matter merely appearing as little interstitial specks between the magnetite granules. In such crystals, one may almost say pseudomorphs, the outlines of the sections still clearly denote the form of hornblende ; and it seems, as sug- gested by Zirkel, that in these cases even a very small pro- portion of hornblende, in spite of an almost overwhelming percentage of magnetite, is capable of asserting its crystal- line form, just as in the well known Fontainebleau sandstone a trivial proportion of calcite develops rhombohedral crystals which contain as much as 65 per cent, of sand. 1 In this latter case the sand, as stated by Delesse, was formed prior to the crystallisation of the calcite ; in other words the silica and the carbonate of lime did not crystallise at the same time. It is therefore probable that the granular magnetite j was also developed prior to the crystallisation of the horn- Lblende, and was simply taken up by it. The minerals of the amphibole group frequently show a tendency to develope long blade-like crystals. One of the 1 Delesse, ' Recheiches sur les Pseudomorphoses, ' Annales des Mines, t. xvi. 1859. (Separately printed extract, p. 35.) Hornblende. 131 principal varieties, actinolite, shows this tendency in a very marked manner, the crystals arranging themselves in radiate groups. Actinolite is a magnesia-lime-iron amphibole ; it is usually of a dark green colour. Under the microscope the crystals mostly appear pale green by ordinary transmitted light, and are often traversed by numerous transverse fractures, frequently accompanied by displacement of the crystal on either side of these cracks. Some of the long radiating crystals in a serpentinous rock from Cannaver Island, Galway, Ireland, display magnificent variegations of colour under polarised light. Tremolite is a magnesia-lime amphibole (CaO, MgO) Si0 2 . In colour it is generally white or greyish. Nephrite or jade is in part a tough compact fine-grained tremolite, having a tinge of green or blue, and breaking with a splintery fracture and glistening lustre. It usually occurs associated with talcose or magnesian rocks.' ! Asbestus or amianthus is a fibrous variety of pyroxene occurring in white silky fibres, which are matted together, but are easily separable. Byssolite is more compact in aggregation, the fibres are coarser as a rule and are not easily- separated, the structure more resembling that of wood<, while the colour is usually dark green or greenish grey. It may be regarded as an iron-manganese amphibole, All these varieties, viz., actinolite, tremolite, jade, as- bestus, and byssolite belong to the non-aluminous, or almost non-aluminous, sub-group of amphibole. Ordinary hornblende, its variety pargasite,- and smarag- dite, which is a foliated grass-green form of hornblende, somewhat resembling diallage in appearance, are varieties of the aluminous sub-group of amphibole. Hornblende is frequently twinned along a plane parallel to the orthopinakoid. This gives rise to a difference in the terminations of the crystals, one end exhibiting four faces (hemidomes), and the other two faces (basal planes). 1 Dana, System of Mineralogy, 5th edition, p. 233. K 2 132 The Rudiments of Petrology. Both hornblende and augite sometimes occur together in the same rock ; but as a rule the former mineral is found in those rocks which contain a large percentage of silica, the associated minerals being usually quartz and orthoclase, while augite is generally found in rocks of a basic character containing triclinic felspars, and with little or no free silica. Augite and its variety, diallage, sometimes occur together in the same rock, and in such cases the petrologist often has difficulty in the precise determination of the pyroxenic constituents ; the diallagoid augite * of Boricky sufficiently evinces, merely from the name which has been given to it, the intermediate character which such pyroxenic minerals may sometimes possess. MICA GROUP. The minerals of this group, most commonly occurring as constituents of rocks, crystallise either in the hexagonal or in the rhombic system. They have a highly perfect cleavage parallel to the basal plane, and the thin laminae procured by cleavage are not merely flexible but also elastic, springing back, when bent, into their original position. The hexagonal species are uniaxial, the rhombic biaxial. The micas mostly have a pearly or sub-metallic lustre. Their hardness is very variable, some of them ranging as low as i, while one species, ottrelite, is sufficiently hard to scratch glass. In chemical composition the micas vary considerably, and cannot well be represented by a general formula. They are silicates of alumina, with silicates of potash, magnesia, or lithia, and protoxides of iron and manganese. Some species are hydrous, but this condition implies alteration in most instances. It is now considered probable by some mineralogists that isomorphous mixtures of the different species may occur amongst the micas, just as they do among the felspars. 1 Diallagahnlicher Atigit. Boricky, Petrographische Studien an den Melaphyrgcsteinen Bohmens, p. 19. Prag, 1876. Rhombic Micas. 133 RHOMBIC MICA SECTION. A Muscovite. A potash mica, optically biaxial, crystallising in the rhombic system, and usually occurring in rhombic or six-sided tabular crystals ; the lateral faces are sometimes those of pyramids, sometimes of the rhombic prism, afford- ing in the latter case an angle of about 120, the opposite acute angles often being truncated by faces of the brachy- pinakoid. In many rocks the crystals are but poorly deve- loped, or only represented by irregularly-shaped scales, which occasionally, but rarely, exhibit a slight curvature. Cleavage basal and very perfect. Hardness = 2 to 3. Colour, mostly silvery white ; seldom, but occasionally, dark brown or black. The percentage chemical composition of muscovite may be regarded typically as : Before the blow-pipe it whitens and fuses on thin edges to a grey or yellow glass. Muscovite is not decomposed by sulphuric or hydrochloric acids. Under the microscope sections of muscovite appear transparent, and exhibit clear colours. Plates, viewed at right angles to the basal plane, show tolerably strong chromatic polarisation, in this respect differing from the uniaxial micas * which, under similar conditions, become dark' between crossed Nicols. When tested for dichroism it shows but little change of actual colour ; as a rule merely displaying a change from light to dark shades of the same colour. Two systems of striae are often visible under the microscope on the surfaces of thin plates, the one set running parallel to oo P and oo P co , the other less perfect following oo 3 and oo P oo . These directions of striation bear a definite relation to the optical axes. Fluid inclosures occur in some micas,' but they contain no bubbles, and, as suggested by Dr. A. von Lasaulx, are doubtless merely infiltrations which have 134 The Rudiments of Petrology. (occurred along cleavage planes. Tourmaline, apatite, garnet, /quartz, magnetite, and undetermined microliths have been ,observed as inclosures in muscovite; but, as a rule, the mineral contains few foreign matters. Newton's rings are I often visible between the laminae. Small crystals and, at j times, interlaminations of biotite occur in muscovite. Fuchsite is a variety of muscovite containing about 4 per cent, of chromic acid. ^Lepidolite (lithia mica) corresponds cry stall graphically and physically with muscovite, for which it is frequently a substitute in granites. It usually occurs in fine scaly or granular aggregates, rather than definite crystals. The colour is generally violet, rose-red, or violet-grey, and occa- sionally white. In chemical composition lepidolite may be regarded as muscovite, in which the potash is partially replaced by lithia. An analysis of lepidolite from Rozena in Moravia gave silica = 50-32, alumina = 28-54, peroxide of iron.= 0-73, magnesia = 0*51, lime=roi, lithia = o-7o, fluoride of lithium = 0-70, fluoride of potassium = 12*06, fluoride of sodium=i77, rubidia = o'24, csesia = traces, water = 3'i2. Lepidolites from Uto in Sweden have yielded over 5 -5 per cent, of lithia. Before the blowpipe lepidolite colours the flame purple-red. After fusion before the blowpipe, it is completely decomposed by acids ; but otherwise it is only imperfectly soluble. Damourite and Seriate are hydrous potash micas usually occurring in scaly aggregates, but their crystallographic system has not yet been determined. Sericite occurs in undulating scales which have a fibrous structure. These wavy scales often run through the schistose rocks in which they occur in tolerably parallel directions ; at other times they anastomose or form a mesh-work. The fibrous struc- ture distinguishes it from mica. Each fibre has an in- dividual polarisation, a character which is very constant. It may be distinguished from chlorite by the absence or feeble- Hexagonal Micas. 135 jjess of dichroism. By ordinary transmitted light thin sections of sericite are colourless, yellowish, or greenish, but never of so strong a green colour as chlorite. 1 Paragonite is a hydrous soda mica. Margarodite, crystallographically and optically, is almost identical with muscovite, and results from the hydration of that mineral. Chemically it approximates to damourite. The colour is generally silvery white, and it has a more pearly lustre than muscovite. Phlogopite crystallises in the same system, and has the same cleavage as muscovite. The divergence of the optical axes is from 5 to 20. The colour is usually brown or copper-red, sometimes white. Thin laminae often show a stellate figure when a candle flame is looked at through them ; this asterismus is due to the presence of included microliths or small crystals, which follow three definite directions. The chemical composition is essentially silicate of alumina, magnesia, potash, and frequently soda. Mag- nesia is sometimes present from 20 to 30 per cent It is a mica which mostly occurs in crystalline limestones, dolomites, and serpentines. The dichroism of this mineral is strong. Its microscopic character is not, however, sufficiently well marked to enable it easily to be distinguished by this means from muscovite or lepidolite, chemical analysis being the only trustworthy method of discrimination in this case. HEXAGONAL MICA SECTION. Biotite crystallises in the hexagonal (rhombohedral) system. 2 Colour, black or dark green. Very thin laminae, appear brown, greenish, or red by transmitted light Che- mical composition, silicate of magnesia, potash, iron, and 1 Memoire sur les Roches dites Phitoniennes de la Belgiqiie, Poussin and Renard. Bruxelles, 1876, p. 164. 2 According to Descloizeaux some specimens of biotite are optically biaxial, but the observed divergence of the optical axes is veiy slight. In these cases the mineral must be regarded as rhombic in crystallisa- tion, and closely related to, if not identical with, phlogopite. 136 The Rudiments of Petrology. lalumina. The percentage composition of the mineral varies considerably. The basal cleavage is highly perfect, and the laminae are flexible and elastic, as in other members of the mica group. The mineral is only slightly acted upon by hydrochloric acid, but is decomposed by sulphuric acid, ^leaving a residue of glistening scales of silica. Under the microscope, crystals which lie parallel to the basal plane appear dark between crossed Nicols, but sec- tions taken in other directions through the crystals show _yery strong dichroism. As observed by A. von Lasaulx, hornblende, tourmaline and epidote are the only other minerals which exhibit equally strong dichroism. Sections transverse to the basal plane show a fine striation, which represents the cleavage, and the crystals often appear frayed \out at the ends. Lepidomelane occurs in small six-sided tabular crystals, or in aggregations of minute scales. Colour, black. Lustre, adamantine or somewhat vitreous. Easily decomposed by hydrochloric acid, leaving a fine scaly residue of silica. In chemical composition it is an iron-potash mica. An analysis of lepidomelane from Carlow Co., Ireland, by Haughton, gives SiO 2 = 35-55. A1 2 O 3 = 17-08. Fe 2 O 3 = 237o. FeO = 3'55. MnO = 1-95. MgO = 3'o/. CaO = o-6i. Na 2 O This mica occurs in some of the Irish, and, according to Allport, in some of the Cornish granites. It is also found in certain Swedish rocks. Lepidomelane is usually regarded as hexagonal, and consequently as optically uniaxial, but some crystals have been observed which appear to be biaxial, a very slight separation of the axes being discernible. CHLORITE. Crystalline system hexagonal. It occurs sometimes in a granular form (' Peach '), sometimes in small green crystals and scales. Its formula is 2 (2 RO, Si O 2 ) + A1 2 O 3 , sH 2 O in which RO signifies MgO and FeO. The average per- Chlorite. Talc. Tourmaline. 137 centage of magnesia is about 34 and that of water over 12. It is essentially a product of the decomposition of other minerals. When heated in a glass tube it gives off water. Before the blowpipe it is fusible with difficulty on thin edges. It is decomposed by HC1 and completely soluble in hot H 2 SO 4 . Chlorite is optically uniaxial, consequently a thin crystal of chlorite, when lying with its basal plane at right angles to the axis of vision, exibits no dichroism, but sections of chlorite crystals taken either obliquely, or at right angles to the basal plane, show feeble dichroism, giving a change from pale to deep tints of green when examined with a single Nicol, although, as a rule, no actual difference of colour is discernible. Chlorite often contains fluid lacunae. It frequently forms fibrous and radiate aggregates. Mag- netite and radiating nests of actinolite are commonly associated with chlorite. Clinochlore, which is monoclinic in crystallisation and optically biaxial, also frequently occurs in admixture with chlorite. TALC. Crystalline system rhombic ? or, according to some au- thors, hexagonal. Has a highly perfect basal cleavage. Cleaved plates are flexible, but not elastic : by this character it may be distinguished from mica. Hardness = i. Crystals are rare. Colour silvery- white to various shades of green. Lustre pearly. Unctuous to the touch. Formula (Mg O) 6 (Si O 2 ) 7 . Before the blowpipe it turns white and exfoliates. It is, neither before nor after ignition, soluble in either hydro- chloric or sulphuric acids, thus differing from chlorite. It also differs from chlorite in showing no dichroism. Under the microscope it appears as imperfectly developed scales, with a fibrous structure: these often have frayed margins. Steatite may be regarded as a massive variety of this mineral. TOURMALINE. Crystalline system hexagonal (rhombohedral). The crys- tals commonly occur in long prismatic forms. The develop- 138 The Rudiments of Petrology. ment is often hemihedral, thus giving rise to triangular prisms. The terminations of tourmaline crystals are frequently com- posed of a great number of faces, and the one termination differs from the other. The crystals, when heated and freely suspended, exhibit polar electricity, a phenomenon which usually accompanies hemimorphism. Cleavage rhombohedral but very imperfect. Crystals generally striated longitudinally. They often form radiate groups. The chemical composition of tourmaline is very variable. Tiae general formula is 3R 2 O 3 . SiO 2 + aRO . SiO 2 . All the varieties contain silicate of alumina with about 4 to 10 per cent, of boracic acid, some contain protoxide, some peroxide, and some both protoxide and peroxide of iron. Magnesia, soda, lime, phosphoric acid, lithia, and fluorine also occur in some varieties. The fusibility of some varieties is effected with more or less difficulty before the blowpipe. Others fuse with comparative ease; in some cases the fusion is accompanied by intumescence. Heated with powdered fluor-spar and bi-sulphate of potash, all the varieties impart to the oxidising flame the green colouration indicative of boracic acid. Tourmaline is strongly doubly-refractive and strongly dichroic. In thin section, by ordinary transmitted illumination, crystals, or portions of crystals, frequently ex- hibit a deep blue colour : this is especially the case when the tourmaline is associated with quartz. Transverse sections of prisms often afford triangular forms. Although in its strong dichroism it resembles hornblende, the latter mineral may be distinguished from it by its well-marked cleavage-planes. Biotite is not likely to be mistaken for tourmaline owing to the lamellar structure of the former mineral: a structure almost always visible in those sections of biotite which are not cut parallel to the basal plane and which exhibit dichroism. Tourmaline contains, as a rule, few inclosures, the most common being fluid lacunae. The black variety of tourmaline is termed schorl, and is of frequent occurrence in certain rocks, especially in granites, which generally Epidote. 139 become schorlaceous near their contact with other rocks. In such cases it is not uncommon to find the schorl segre- gated into nests, OT small spheroidal masses. EPIDOTE. Crystalline system monoclinic. Cleavage parallel to the orthopinakoid perfect, parallel to the basal plane, very perfect. The cleavage planes intersect at ai\ angle of 115 24'. Colour usually green or yellowish-green, sometimes brown. Epidote occurs at times in a granular or massive condition; the crystals commonly form radiate or fan-shaped groups, and either form nests, or line fissures and cavities. The mineral is essentially an alteration-product, consisting of silicate of alumina, lime, and peroxide of iron, with variable amounts of oxide of manganese and water. Before the blowpipe it usually gives an iron or manganese reaction with fluxes. There are several varieties, one containing about 14 per cent, of manganese oxides. Under the microscope epidote exhibits strong pleo- chroism. When tested with a single Nicol it does not, how- ever, show this property so strongly as hornblende. Inclosures are rare in epidote; nevertheless fluid lacunae have been observed in it. The way in which epidote occurs in rocks forming little fan-shaped aggregates of radiating needles or fibres along fissures, &c., its bright yellowish- green colour, and its strong pleochroism all of these charac- ters taken in conjunction serve, as a rule, to distinguish it from other minerals. Epidote often forms little fringes around hornblende crystals. Chlorite is one of the minerals which most resembles it in its mode of occurrence. In doubtful cases, the respective strength of dichroism, the absence of dichroism in plates of chlorite viewed perpen- dicularly to the basal plane, the hexagonal form of such chlorite plates, and the difference in the colour of the two minerals, are points which should be looked for and taken into consideration. 1 40 The R ndimen ts of Petrology. SPHENE (Titanite) crystallises in the monoclinic system ; common form, the oblique-rhombic prism. The crystals are usually thin and have sharp wedge-like edges. Colours, brownish-grey, yellow, ; green, and black. Transparent to opaque. Approximate chemical composition. Silica=3o 35, titanic acid= 33 43, lime = 21 33 P er cent. The formula jnay be written : CaO, 2SiO 2 + CaO, 2TiO 2 . Under the microscope, sections of sphene generally appear, by transmitted light, brownish yellow, yellow, some- jimes red, reddish brown, and colourless ; 'they show distinct though not strong, pleochroism, except when the sections are very clear and colourless or of extreme thinness, j Infiltration products, consisting either of hydrous or anhy- i drous oxide of iron, are sometimes seen between the planes L-of cleavage in sphene. Intergrowths of sphene and horn- blende have been observed by Groth, in the syenite of the Plauenschen Grund, near Dresden. / Frequently the crystals of sphene appear cloudy or imperfectly translucent. The sections usually present very characteristic wedge-shaped forms. As a rule sections of sphene appear to be remarkably free from inclosures of other minerals. GARNET GROUP. All the members of this group crystallise in the cubic system, the common forms being either the rhombic dode- cahedron or the icositetrahedron. The cleavage is parallel to the faces of the dodecahedron. The garnets vary in hardness from 6'5 to 7*5. They have a subconchoidal or uneven fracture, and they all afford an approximately white streak. Before the blowpipe most of them fuse easily, but, in accordance with the different chemical composition of the various species, they give different blowpipe-reactions with the fluxes, in which the iron reactions dominate. In chemical composition the garnets are essentially unisilicates of different sesquioxides and protoxides. The Garnet Group. ^141 '*/, \/ t - sesquioxides are those of aluminium, iron, and chromium ; / sometimes also of manganese, while the protoxides ar-0'those of iron, calcium, magnesium, or manganese. *-/jr > The principal sub-species have the following 6qmposi- J tion : 1. Lime-alumina garnet, 6CaO, 3SiO 2 + 2A1 2 O 3 , 3SiO 2 . f 2. Magnesia-alumina garnet, 6MgO, 3SiO 2 + 2A1 2 O 3 , 3SiO.,. 3. Iron-alumina garnet, 6FeO, 3SiO 2 + 2A1 2 O 3 , 3SiO 2 . 4. Manganese-alumina garnet, 6MnO, 3SiO 2 + 2A1 2 O 3 , 3SiO 2 . 5. Iron-lime garnet, 6CaO, 3SiO 2 + 2Fe 2 O 3 , 3SiO 2 . 6. Lime-chrome garnet, 6CaO, 3SiO 2 -f 2Cr 2 O 3 , 3SiO 2 . The percentage of silica in the various sub-species is tolerably uniform, ranging from 35 to 40 per cent. For descriptions of the general characters of the different sub-species of the garnet group, the student may refer to any manual of mineralogy, since only the microscopic cha- racters of the group will be described here. . In thin sections of rocks, garnets frequently appear under the microscope merely as rounded granules, somewhat resembling spots of gum, generally colourless or of clear reddish tints, but sometimes, as in the case of picotite, of a deep brownish or reddish-brown colour, and forming irre- gular gummy-looking streaks. When definitely formed crystals occur, they afford, as a rule, four-sided, six-sided, and eight- sided sections. They are mostly traversed by irregular cracks. The crystals occasionally appear to have formed around a nucleus of some other mineral, such as quartz, epidote, &c. The minerals usually inclosed in garnets are magnetite, hornblende, tourmaline, quartz, occa- sionally apatite and augite, and colourless microliths, forming narrow prisms, whose nature has not yet been determined. Garnets also contain at times cavities in the form of the rhombic dodecahedron (negative crystals). Although garnets occur in some eruptive rocks, yet they are most plentiful in those which have undergone strong metamorphism, and 142 The Rtidiinents of Petrology. they are especially common in granulite, gneiss, talcose, and chloride slates, and other schistose metamorphic rocks. They also occur in serpentines, and granular and crystalline limestones. The Belgian hone-stones (coticules] consist in great part, according to Prof. Re'nard, of the manganese garnet spessartine. 1 Since they crystallise in the cubic system they exhibit single refraction. Some garnets have, however, been observed to possess double refraction, but these anomalous examples have not yet been fully investi- gated. Idocrase, or Vesuvian, is in its chemical composition closely allied to the lime-alumina garnets, but crystallises in the tetragonal system. According to Sorby it often contains fluid lacunae in which very numerous, but undetermined crystals occur. TOPAZ is not a mineral of common occurrence in rocks. It only attains importance as a rock-component in the ' Topazfels ' of Schneckenstein in Saxony ; still it is occasionally met with as an accessory constituent of certain rocks. Its crystallisation is rhombic, but it also occurs in a granular condition. It is infusible before the blowpipe and insoluble in acids. Under the microscope it is strongly doubly-refracting ; but, although it is dichroic, the dichroism is so weak that in thin sections it is scarcely perceptible. Fluid lacunae are common in crystals of topaz; the liquid is sometimes a saline solution, sometimes liquid carbonic acid. The fluid inclosures in topaz were first investigated by Brewster in 1845, an( ^ nav e subsequently been examined by Sorby and Vogelsang. Rosenbusch notices the inclu- sion in topaz of scales of hematite, and of black flecks and granules which show no metallic lustre, and suggest the idea 1 * Memoire sur la Structure et la composition Mineralogique du Coticule,' A. Renard. Bruxelles, 1877, tome xli. des Mhnoires couronnes de r Academic royale de Belgique. Zircon. Andalusite. 143 of carbonaceous matter; but, as they undergo no change when heated, the notion is regarded as erroneous. 1 ZIRCON. This mineral is met with in some lavas and volcanic ejected blocks, also in zircon- syenite. It crystallises in the tetragonal system. Sections of some varieties of zircon display strong dichroism, while in others it is scarcely perceptible. In chemical composition it is essentially silicate of zir-, conia and since it contains no protoxides it is but little liable to undergo alteration. It, however, at times becomes j hydrous, loses silica, and becomes partly replaced by other I substances. ANDALUSITE. Crystallographic system, rhombic. The common form is a combination of the rhombic prism oo P, the macro- domes P oo and the basal planes oP. The crystals, especially the larger ones, are usually incrusted with and penetrated by scales of mica. At times they are feebly translucent and very rarely transparent. The cleavage is very indistinct. Andalusite sometimes occurs in a granular condition. Its chemical composition is represented by the formula A1 2 O 3 , SiO 2 . Under the microscope thin sections polarise strongly and show well-marked pleochroism. Alteration is denoted by the development of a fibrous structure which usually runs in definite directions. In some crystals of andalusite carbo- naceous matter occurs ; but, as a rule, the mineral is very free from inclosures of foreign substances. It is most com- monly met with in mica schist and slaty rocks. The variety chiastolite or made, so named from the spots and cruciform markings which occur in the interior of the crystals, is met with only in slates which have undergone 1 Rosenbusch, Mik. Physiog. d. Min. p. 282. Stuttgart, 1873. 144 The Rudiments of Petrology. alteration from proximity to eruptive rocks, as in the Skiddaw slate where it nears the granite. The changes which take place in this district, as described by Mr. J. Clifton Ward, 1 consist first in the faint development of oblong or oval spots in the slate, together with a few crystals of chias- tolite ; the latter then become quite numerous and well developed, constituting the true chiastolite slate. This passes into a harder, foliated, and spotted rock which Mr. Ward regards as knotenschiefer, the spots being imperfectly developed chiastolite crystals, accompanied by more or less mica and quartz, while, in the immediate neighbourhood of the granite, the rock passes into mica-schist. Crystals of chiastolite afford sections which vary considerably in form, some giving rhomboidal outlines with a dark nucleus in the centre of the section. The boundaries of the crystal section are usually sharply defined, but the nucleus which, under the microscope, may generally be resolved into a mass of dark flakes and granules, is not very distinctly separable from the more or less translucent surrounding matter of the crystal, appearing to have a hazy boundary. Zirkel states that in such nuclei a linear arrangement of granules or flakes may sometimes be discerned passing from the centre to the angles of the section (i.e. to the lateral edges of the rhombic prism), but he adds that this arrangement is seldom so clearly perceptible in microscopic individuals as in the larger crystals in which such divisional markings are visible to the naked eye. KYANITE. Crystalline system triclinic. Chemical composition similar to that of andalusite. The crystals are generally long prisms, which appear broad in one direction and narrow 1 Memoirs of the Geological Survey of England and Wales (The Geology of the northern part of the English Lake District), pp. 9-12. See also Abhand. zur Geol. Specialkarte v. Elsass-Lothringen. Die Steiger Schiefer u. ihre contactzone an den Granititen v. Barr-Andlau u. Hochwald. H. Rosenbusch, pp. 210-215. Strassburg, 1877. Apatite. 145 in the other. The cleavages are prismatic and basal, the former being tolerably distinct parallel to the broad faces, but less so in the direction of the narrow ones. The basal cleavage enables the prisms to be easily broken in a trans- verse direction. The crystals are seldom terminated. It also occurs in radiating or interlacing fibrous conditions. It is transparent, with a vitreous lustre, and is coloured blue, or greyish-blue and white, individual prisms often showing a succession of blue and colourless bands which graduate into one another. The mineral is pleochroic, but this is not perceptible in thin sections. Inclosures of other minerals are rare in kyanite. APATITE. Crystalline system hexagonal. The crystals are usually 1 combinations of the hexagonal prism and basis, sometimes modified by faces of the hexagonal pyramid. Although crystals of apatite several inches in length, and sometimes of much greater size, are occasionally found, still the majority of those which enter into the composition of rocks are of microscopic dimensions. These little prisms are usually very long in proportion to their breadth. The hardness of apatite is 5. /It contains from 90 to per cent, phosphate of lime, arid its formula is either 3 (Ca 3 P 2 8 ) + CaCl 2 or 3 (Ca 3 P 2 O 8 ) + CaF 2 , according to whether the mineral contains chloride, or fluoride of calcium. It frequently contains both. The detection of phosphoric acid in rocks is best effected by finely pulverising a tolerably large sample, digesting it in hydrochloric acid, filtering off the solution, and treating it with molybdate of ammonium. If phosphoric acid be present, a yellow precipitate will be formed, and the pre- cipitation, which usually takes place very slowly, may be accelerated by frequent stirring with a glass rod. / Most] of the phosphoric acid which exists in rocks probably) 146 The Rudiments of Petrology. (occurs in the form of apatite ; except in cases where, instead of being minutely disseminated, it occurs as phos- phorite. Under the microscope apatite appears in elongated hexagonal prisms, which, when cut longitudinally, afford .rectangular sections, and transversely, hexagonal ones. /The i boundaries of the crystals are always sharply denned. Since the mineral is uniaxial, the sections taken at right angles to ^the principal axis appear dark between crossed Nicolsr^In | the colourless apatite crystals, which usually occur in rocks, j no dichroism can be discerned, although in some coloured /' varieties of the mineral it is quite perceptible. /Apatite crys- ^tals often contain light greyish or yellowish dusty matter, the nature of which has not yet been determined, although, from an examination of large crystals containing somewhat similar impurities, it has been inferred that the dust may consist partly of magnetite granules, and partly of acicular microliths, together with inclosures of glass and of fluid, the former showing motionless, and the latter movable, bubbles. In examining some large pellucid apatite crystals from the Val Mayia, Fritzgartner found them to contain small elongated hexagonal prisms and pores filled with liquid. The latter varied in form and size, but were mostly round. The hexagonal prisms lay with their longer axes parallel to the basal plane of the containing crystal, and appear to follow irregular curves, and to be arranged in no directions corresponding with the other axes of the crystal which con- tained them. 1 Apatite crystals sometimes envelope a black' opaque substance which corresponds in its boundaries with the boun- daries of the containing crystal, the latter often forming little more than a clear, narrow margin around this dark nucleus. Zirkel notes the occasional symmetrical disposition of six small apatite crystals around a larger one. 1 Private communication from Dr. R. Fritzgartner. Apatite. Rutile. 147 Minute crystals of apatite may be distinguished from those of felspars by their hexagonal transverse sections. They may usually be distinguished from nepheline by oc- curring on a much smaller scale, and being of much greater length in proportion to their breadth, so that they afford rectangular sections which are generally much longer, and hexagonal sections which are much smaller, than the cor- responding ones derived from nepheline. The student should also be on his guard against mistaking small apatite needles for colourless microliths of hornblende, augite, &c. Apatite crystals seem to be rather gregarious, often colo- nising in certain portions of a rock, and being nearly absent in others. ,/ It is of all minerals one of the most widely? distributed, occurring in a vast number of rocks of very| diverse mineral composition, often being present only inj very minute quantity/ It is even regarded by Zirkel as of more common occurrence than magnetite. Apatite crystals usually remain clear and fresh long afterj the other mineral constituents of a rock have decomposed.} Although so comparatively invulnerable to the natural agents which decompose rocks,/it is soluble in hydro-chloric} acid. Asparagus-stone and moroxite are names given to yellow- ish green and blueish green varieties of apatite. In Canada the latter mineral occurs in a bed ten feet thick passing from North Elmsley into South Burgess. Three feet of this bed consist of pure sea-green apatite, while the remainder is made up of apatite and limestone, in which crystals of pyroxene and phlogopite also occur. 1 RUTILE (TiO 2 ), which crystallises in the tetragonal system, ap- pears deep red or brown when seen in thin section under the microscope. It is not very strongly dichroic. The crystals are often seen t be traversed by thin plates or 1 System of Mineralogy', Dana, 5th edition, p. 533. L2 148 The Rudiments of Petrology. strise, and by included crystals which follow the direction of the principal axis, and that of the twinning plane co P. A good figure, showing this structure along the plane of geniculation, is given in Rosenbusch's * Mikroscopische Physiographic, 5 vol. i. p. 187, to which work the student is referred for further particulars respecting the microscopic character of this mineral. CASSITERITE, (SnO 2 ) crystallises in the tetragonal system. The crystals, when examined in thin section under the microscope, appear by transmitted light of a honey-yellow colour. CALCSPAR (Calcite). Crystalline system rhombohedral. The crystals vary greatly in the combinations which they present. The most common forms are rhombohedra, scalenohedra, and hexa- gonal prisms, terminated by basal planes or by planes of a rhombohedron. The simple rhombohedral forms with angles of 105 5' and 74 55' are those which chiefly occur in rocks, and the cleavage corresponding with this form is commonly to be recognised in granular aggregates of carbonate of lime. Chemical formula CaCO 3 . The mineral frequently contains some magnesium or iron replacing part of the calcium. In some cases it. is im- pregnated with sand, as in the well-known crystals from Fontainebleau, which sometimes contain over 60 per cent, of that material. When treated with acids, calcspar effervesces, giving off carbonic anhydride. It is easily scratched with a knife, since it only has a hardness of 3. Before the blowpipe it is infusible, becoming strongly luminous ; and, giving off its carbonic anhydride, it is reduced to quick-lime, the crystal, if previously transparent, becoming opaque, white, and pulverulent. Under the microscope, sections of calcspar show very strong double refraction, which may be observed by using Calcspar. Quartz. 149 the analyser alone. The planes of cleavage which intersect one another are also generally visible, while by polarised light it is common to find that the separate granules which constitute crystalline aggregates are composed of numerous lamellae which polarise in different colours, and which denote a system of twinning parallel to the face \ R. The lines of demarcation between the lamellae are sharply denned, and run parallel to one another in the same individual or granule ; but the planes of twinning in any one granule observe no relation to those belonging to adjacent granules. This twin structure may be well seen in crystalline limestones, statuary marble, &c. It is very characteristic of calcspar, and serves as a rule to distinguish the mineral from dolomite, which seldom shows any such structure. Reusch has demonstrated that a similar twin structure may be artificially produced in calcspar by pressure. Inclosures of other minerals are common in calcspar. Iron pyrites, native copper, copper pyrites, copper glance, and a large number of other minerals, are at times met with in calcspar, but most of these inclosures are visible without the aid of a microscope. The fluid inclosures which sometimes occur in calcite are generally regarded either as water, or as water containing carbonic acid or bicarbonate of lime. QUARTZ. )\ Crystalline system hexagonal ; or, as indicated by the occasionally occurring tetartohedral faces, rhornbohedral. The usual forms are either hexagonal pyramids, or combi- nations of the hexagonal pyramid and hexagonal prism. In the former case the sections parallel to the principal axis yield rhomboidal figures, in the latter elongated hexa- gons, while in both instances the sections transverse to the principal axis are regular hexagons. Twinning is common, sometimes giving rise to geniculation, sometimes ' producing cruciform arrangements, at others causing irre- gular interpenetration of dissimilar parts of the crystal, the 150 The Rudiments of Petrology. positive rhombohedral faces being irregularly penetrated by the negative, and vice versa. The chemical composition of quartz is Si O 2 , with occasional impurities, such as iron oxides, titanic acid, &c. Quartz is infusible before the blowpipe, insoluble in all acids except fluoric acid. It is also more or less acted upon by a hot solution of potash ; in the case of the purer crystallised varieties, but very slightly. In the compact and crypto- crystalline conditions its solu- bility in this reagent is, however, according to Rammelsberg somewhat greater. 1 The hardness of quartz is 7, the point of a penknife producing no effect upon it, unless it be in a finely granular condition, when the point may simply rake up and detach a few granules, upon which, however, it is unable to make any impression. In this instance, as in others, it behoves the student to be on his guard in testing the hardness of granular or finely crystalline substances, to distinguish between the disintegration of granular struc- tures and true streak. The pressure, however, required to scrape off -the smallest trace of dust from a quartzite, or from granular conditions of quartz, is very considerable. The fracture of quartz is conchoidal. The specific gravity = 2*65. Sections of quartz appear clear and pellucid under the microscope. They show circular polarisation, and exhibit, in some sections, magnificent variegations of colour. The plane of polarisation is sometimes right-handed, sometimes left-handed, in its rotation, and both the right-handed and left-handed phenomena of rotation are at times seen in the same crystal. For further information upon this subject the student is referred to the works of Brewster, Descloizeaux, Groth, and others. Quartz is seen frequently to contain inclosures of other substances, sometimes as crystals, sometimes in the form of lacunae filled with liquids, &c. These inclosures are often visible to the naked eye, but the microscope commonly reveals their presence in vast numbers. The crystals of * Ann. cxii. 177. Quartz. 151 most frequent occurrence are those of rutile and chlorite ; crystals of kyanite are also occasionally met with. Lacunae of glass, others filled with gas, and others containing por- tions of the rock matrix in which the crystals are imbedded, are by no means uncommon in some rocks. The most numerous lacunae, however, are those containing liquids. The liquids are frequently pure water; sometimes water hold- ing carbonic acid in solution, sometimes liquid carbonic acid, sometimes a supersaturated solution of chloride of sodium, minute crystals of rock salt being visible within the lacunae under tolerably high powers. These lacunae are at times completely filled with the fluid, at others they are seen to contain bubbles which vary in magnitude and are re- garded as representing the diminution of volume which the fluid in the cavity has undergone during the cooling of the rock mass in which it occurs. Deductions based upon the relative volumes of the fluids, and of the va- cuities in such cavities, may be found in the paper com- municated to the Geological Society by Mr. Sorby in 1858. It is, however, deserving of note, as pointed out by Mr. John Arthur Phillips, that, in the same crystal, cavities may be found, some completely filled, while others contain vacuities whose relative volumes to those of their sur- rounding fluids vary very considerably. The bubbles in these lacunae are often moveable, being displaced either by simply turning the crystal, or more usually by heating it, in which case the bubble undergoes diminution of volume, or even disappears. In some small lacunae diminutive bub- bles, which have a spontaneous motion, are visible under high magnifying powers. These bubbles are, however, so minute that they appear frequently as mere specks, and it needs careful and steady watching to see their motion, which looks like a tremulous gyration, often resulting in a com- paratively well-marked change of place, followed perhaps by a pause, to be again succeeded by the oscillations and gyrations already mentioned. 152 The Rudiments of Petrology. Tridymite is a form of silica discovered in 1866 by Vom Rath. It occurs in very small six-sided tabular crystals. The system to which these crystals belong has not yet been satis- factorily determined. The specific gravity is 2-2 to 2-3, the same as that of opal. The crystals occur in compound groups, mostly composed of three individuals, whence the name. It is found in the sanidine oligoclase trachyte of the Drachen- fels, in a volcanic porphyry from near Pachucha in Mexico, in a porphyry from Waldbockelheim, in some Hungarian liparites, in the hornblende-andesites of Dubnik, 1 in a trachytic-looking phonolite from Aussig, 2 in the wolf-rock (phonolite) described by Allport, in several phonolites de- scribed by Mohl. 3 It has also been mentioned as occurring in some Irish rock, but the author is unable either to recall the precise locality, or to find the reference. According to Vom Rath, the discoverer, it is doubly refracting and opti- cally uniaxial. A globular condition of silica has been lately described by Michel Levy 4 as occurring in the euritic porphyries of Les Settons, and similar globular conditions of silica have also been observed and noticed by M. Velain in a quartz- trachyte from Aden. The former author regards this con- dition as intermediate between the crystallised and the colloid forms of silica. The following extract from M. Michel Levy's paper will convey an idea of the microscopic characters of these glo- bules : ' Between crossed Nicols one is surprised to see such regular globules, the centre of each appearing to be a pole of symmetry with four extinctions situated at right angles for every total revolution of the section ; one is therefore forced to conclude that they are composed of a crystallised sub- stance, and are, moreover, orientated in an unique manner, 1 H. Rosenbusch, Mikroskop. Physiogr. d. massigen Gesteine. Stutt- gart, 1877, p. 301. 2 Ibid. p. 225. 3 H. Mohl, Basalte u. Phonolithe Sachsens. Dresden, 1873. ' Bull. Soc. Geol. de France, 3* serie, t. v. 1877, no. 3. Magnetite. 153 Sometimes the extinction is simultaneous over an entire globule, sometimes it is different for two or more segments ; but the most curious peculiarity, exhibited by the concentric- ally-zoned globules, lies in the fact that two adjacent zones do not always undergo extinction simultaneously. Such a globule will undergo extinction in its central portion and will, at the same time, present a perfectly regular narrow border which is still illuminated ; if then the observer con- tinue to turn the section, this border will become dark while the spherical central nucleus will in its turn become clear in a homogeneous manner.' MAGNETITE. Crystalline system cubic. It usually occurs in the form of the octahedron, sometimes in that of the rhombic-dodeca- hedron, also granular and massive. Cleavage parallel to the faces of the octahedron. Colour black. Streak black. Strongly magnetic and often displays polarity./ The chemical formula of magnetite is FeO, Fe 2 O 3 , or Fe 3 O 4 . The approx- imate percentage composition is Fe 2 O 3 = 69. FeO = 31. Magnetite is frequently titaniferous. / It [is very difficultly! fusible before the blowpipe. When pulverised it is com- I pletely soluble in hydrochloric acid. Even in the thinnest ' sections magnetite appears opaque under the microscope ;.. x nevertheless, when it has undergone alteration either into hematite or limonite it appears feebly translucent at times and of a reddish or brownish colour. The sections of magnetite crystals, which are of most common occurrence in rocks, present square forms which represent sections passing through opposite solid angles of the octahedron, or triangular forms which result from sections taken parallel to one of the faces of the octahedron (fig. 69). / Twin I crystals sometimes occur, the twinning taking place on a plane parallel to a face of the octahe- _ ^ dron. The superposition of one crystal on another sometimes gives rise to cruciform figures. Magnetite 1 54 The Rudiments of Petrology. very usually occurs in a granular condition in rocks, 'some- times in coarse irregular grains, sometimes as a fine dust, while at others these granules form segregations which give . rise to rod-like forms. ''Occasionally, as in some basalts, magnetite crystals are grouped in a very regular manner, following lines which are frequently disposed at right angles. Such groupings are not merely to be met with in volcanic rocks, but also in furnace slags, and their arrangement often seems to imply the rudimentary stages of aggregation which might eventually result in the formation of a large crystal from the contiguous development of smaller ones. TlTANIFEROUS IRON. Crystallisation rhombohedral. It mostly occurs in tabular forms with the basal planes largely developed and with hexagonal boundaries. Titaniferous iron is opaque and black, with a semi-metallic lustre. Chemically it may be a combination of titanium and peroxide of iron, or a combination of titanic acid with protoxide of iron, plus a variable amount of the peroxide. The different varieties of titaniferous iron depend mainly upon the relative proportions of iron and titanium which they contain. When heated alone before the blowpipe titaniferous iron is infusible. Heated in concentrated sul- phuric acid it affords a blue colouration but is insoluble. When pulverised it is, however, soluble in nitro-hydrochloric acid. When occurring in microscopic preparations it is often difficult to distinguish it from magnetite, except when it affords well-marked rhombohedral sections (fig. 70), It may, however, often be recognised by the peculiar greyish- FIG o wmte alteration-product which is often developed within it, and which frequently follows definite crystallographic directions. When this alteration is far advanced, merely a dark skeleton or a few dark specks of the unaltered mineral remain, surrounded by Hematite. 1 5 5 the white decomposition product. The precise nature of this white substance has not yet been ascertained, but it is generally assumed to be either titanic acid or some silicate of titanium. l Both titaniferous iron and magnetite frequently occur together in the same rock. HEMATITE. The crystallised variety, specular iron or ironglance, belongs to the rhombohedral system, and mostly occurs in six-sided, thin, tabular crystals in which the basal planes are largely developed, while the boundaries are formed either by faces of the rhombohedra R and JR, or by faces of a hexagonal prism. Crystals of this kind may be easily procured by dissolving a fragment of the mineral carnallite, when the residue will be found mainly to consist of beauti- fully-developed thin tabular crystals of specular iron, which are translucent, and of a clear red or orange-red colour. Thicker crystals appear black or iron-grey, and, as in some of the specimens from Elba, show beautiful superficial iri- descence. Sometimes the crystals are only imperfectly de- veloped, or merely form irregularly- shaped scales. In this scaly condition it is spoken of as iron-mica or micaceous he- matite (Eisenglimmer). Hematite also occurs in a granular state, sometimes earthy as reddle, while reddish stains of ferric oxide are of common occurrence in rocks, especially in those which have undergone weathering. The botryoidal or mammillated forms of hematite mostly occur in pockets or cavities in rocks into which they have been subsequently introduced, or else line drusy cavities, but hematite in this form does not occur as a common rock constituent, although in a compact and massive condition it is often met with in lodes. In some cases, however, the massive and micaceous forms of hematite may almost of 1 It has since been examined and described by Giimbel, under the name of leucoxene. ! 156 The Rudiments of Petrology. themselves be regarded as rock masses ; a hill in the state of Missouri (the Pilot Knob, 700 feet high) consisting almost exclusively of hematite. Hematite gives a blood-red or cherry-red streak. It is feebly magnetic. Its chemical composition is Fe 2 O 3 when pure, but it is often rendered impure by admixtures of sand, clay, &c. Before the blowpipe it is infusible, but becomes black and strongly magnetic when heated in the reducing flame. Under the microscope ;t is usually seen to occur in irre- gular flecks and scales, distinct crystalline forms not being of common occurrence in rocks. It exhibits no dichroism, and shows red tints of various intensity by transmitted light. By reflected light it also usually appears red, especially when in an earthy or finely granular condition. LIMONITE. This is a hydrated peroxide of iron having the formula 2Fe 2 O 3 , 3H 2 O. It is essentially a decomposition product, resulting from the alteration of protoxides, or of anhydrous peroxides of iron, which have previously existed as consti- tuents of other minerals, or in the latter case sometimes simply as hematite itself. Limonite occurs in stalactitic, mammillated, pisolitic, or earthy, conditions. It is com- monly blackish-brown or yellowish-brown, in an earthy or ochreous state often yellow. The streak is yellowish-brown. In thin sections of rocks it is often seen to occur, forming pseudomorphs after crystals of various ferruginous silicates, and as irregularly-shaped blotches. It appears opaque under the microscope, or occasionally, in very thin sections, it is feebly translucent, and of a brownish colour. IRON PYRITES. Crystallises in the cubic system, the most common form being the cube. The faces of the crystals are frequently Iron Pyrites. Copper Pyrites. 157 striated, the striae on one face lying at right angles to those) on the adjacent faces. Pyrites also occurs massive, in no- dules which have internally a radiating structure, (many of these may no doubt be referred to marcasite), while in some rocks it exists in a granular or finely-disseminated state/sometimes forming pseudomorphs after other minerals. /Fossils are at times entirely replaced by pyrites. It is" mostly of a pale brass-yellow colour, gives a greenish or brownish-black streak and a conchoidal or uneven fracture. It has a strong metallic lustre, strikes fire with steel, and fuses before the blowpipe to a metallic globule which is attractable by the magnet. When heated it gives off sulphur. When fused with carbonate of soda, the assay, if placed on a clean silver surface, and moistened with a drop of water, produces a dark stain on the silver. Its chemical com- position is iron = 467, sulphur = 53-3, giving the formula FeS 2 . Under the microscope, in thin sections of rocks, pyrites appears perfectly opaque. The ground surfaces look glis- tening and yellowish by reflected light, and this partly serves to distinguish it from magnetite. The sections are those resulting from cubes or dode- cahedra sliced in various directions, except in cases where the mineral is pseudomorphous after some other mineral. Occasionally pyrites occurs in minute elongated rod-like forms. Marcasite resembles pyrites, except that it crystallises in the rhombic system. Twinning is common in this species. COPPER PYRITES (CHALCOPYRITE). This mineral is occasionally met with in rocks such as diabase, some granites, gneiss, argillaceous schists, &c. It crystallises in the tetragonal system ; the crystals, however, closely approximating to cubic forms. It usually has a deeper yellow colour than iron pyrites, from which it may be distinguished by its inferior hardness, being sectile, while 1 5 8 The Rudiments of Petrology. iron pyi es cannot be cut with a knife. Copper pyrites does noi emit sparks when struck with steel. Before the blowpipe it colours the borax bead blue in the oxidising flame, but to get this colouration the assay should not be previously reduced, for, if so, only a deep green co- louration will be procured. The blue colour is probably due to sulphate of copper, and a previous roasting of the assay of course expels the sulphur. It is soluble in nitric acid, with the exception of the contained sulphur, forming a green solution which changes to a deep blue on the addi- tion of ammonia in excess. The chemical composition of copper pyrites is copper = 3 2 '5 34- Iron= 2975 31-25. Sulphur = 34 36 per cent. The formula is Cu 2 S, Fe 2 S 3 . Under the microscope it appears opaque. By reflected light it shows a somewhat metallic lustre on ground sur- faces, and is generally rather deeper in colour than iron pyrites, but not much reliance can be placed upon this appearance, and its presence should be confirmed by che- mical examination. ZEOLITES. Want of space precludes more than a brief mention of the microscopic characters of a few of the most common zeolites. They may all of them be regarded as alteration products, and in all probability never form normal con- stituents of rocks. They usually occur either lining or completely filling cavities in vesicular and other volcanic rocks, and also occupy fissures and small cracks; occasionally they are developed in crystals of other minerals which have undergone more or less alteration. They often occur in spherical crystalline aggregates, with a radiating structure, in which case they exhibit a black cross when examined between crossed Nicols, the arms of the cross coinciding with the planes of vibration of the Nicols. The section may be horizontally rotated while the crossed Nicols Natrolite. Analcime. 159 remain stationary, yet, although the object revel, es, the dark cross does not move. This is explained by G^oth as being due to the principal directions of vibration of the doubly-refracting fibres lying parallel and at right angles to their longer axes, and bearing a similar relation to rays which undergo extinction between the crossed Nicols. If the analyser be turned through 10 or 20 the dark cross becomes somewhat faint, and a second imperfectly developed cross appears, which makes an angle of 5 or 10 with the fixed cross. It will therefore be seen that it travels at only half the rate of rotation of the analyser. When the analyser has been so far turned that the two Nicols stand parallel, the first cross disappears and the second imperfect cross attains its maximum intensity. This phenomenon is met with in all doubly-refracting, radiate crystalline aggregates ; and, since zeolites frequently occur in this condition, its presence in certain rocks often suggests that such aggregates are zeolitic. Natrolite, which crystallises in the rhombic system, pos- sesses weak double refraction, and polarises in vivid colours. It very commonly occurs in crystalline aggregates, which almost invariably have a radiate structure, and then show, especially when in rounded masses, the interference figure characteristic of such aggregates. At times, also, natrolite is seen filling minute fissures. In this case crystallisation commences on either side of the fissure, and the crystals meet in the middle, their termination giving rise to a zig-zag median line which divides the two growths. Nepheline crystals at times become partly altered into natrolite, a meshwork of little prisms of natrolite, in some instances, almost completely filling the crystal. 1 Analcime, so far as is yet known, crystallises in the cubic system, but, although regarded as cubic, it exhibits some rather exceptional optical properties, first pointed out by Brewster, and subsequently investigated by Descloizeaux, 1 Rosenbusch, Mik. Physiog. (Min.) vol. i. p. 285. Stuttgart, 1873. 160 The Rudiments of Petrology. Rosenbusch, and other observers. According to Des- cloizeaux, sections cut parallel to any one of the faces of the cube, when viewed in the direction of one of the axes by parallel polarised light, appears between crossed Nicols perfectly dark in the direction of the two other axes, while, in the direction of the diagonals of the cube- face, a faint bluish, distorted cross appears. Analcime is seldom or never a normal constituent of rocks. Tschermak, however, regards it as an essential component of the rock which he terms teschenit, which consists of plagioclase, hornblende, analcime, magnetite, biotite, and apatite. In a leucitophyr from Rothweil, near the Kaiserstuhl, analcime occurs pseudomorphous after leucite ; at .all events, the leucite crystals contain fibrous and granular aggregates of analcime, which at times almost totally replace them. Heulandite has been observed to contain various micro- scopic inclosures such as minute orange-yellow coloured acicular crystals, irregularly- shaped or round granules and flecks of a reddish-yellow mineral, and, in one specimen, from the Faroe Isles, Rosenbusch noted the occurrence of innumerable perfectly-developed microscopic quartz crys- tals. The colour of heulandite is due to the reddish and orange-yellow inclosures just alluded to. They have been regarded as gothite, limonite, or hematite. Zirkel considers that they are hematite. Chabasite. This mineral appears from numerous obser- vations always to be devoid of fluid inclosures. Micro- scopic envelopments of quartz have been met with in chabasite. For further particulars respecting the large family of zeolites the student is referred to the various manuals and text-books of mineralogy. CRYSTALLITES. Under this head may be grouped a vast number of purely microscopic bodies, which, in their progressive de- Crystallites. 161 velopment, represent the various forms and conditions of mineral matter, from its departure from an amorphous state, to one of crystallographic completeness, such as may be correlated, if not identified, with the crystals of recognised mineral species. The forms belonging to the highest stage of this microscopic development are spoken of as rnicroliths, and they frequently present crystal faces sufficiently distinct to admit of goniometric measurements, and optical characters well enough denned to permit their correlation with recog- nised minerals. The less perfectly developed crystallites cannot however be referred to any particular species, and hence has arisen the necessity for the employment of various terms, more or less vague, and each of them embracing a vast multitude of different forms, but convenient, because indicative of structural types. Doubtless, as knowledge increases, these terms will give place to better ones with more precise significations, and the progressive developments which these minute forms display will be properly worked out, and afford a key to the important subject of crystallo- genesis. The crystallites may be ranged in a descending series as follows : Microliths. Crystalloids. Trichites. Globulites. The globulites represent the most embryonic stage of crystallogenesis, the most rudimentary change effected in amorphous matter. They are spherical in form, and by their coalescence give rise to variously shaped groups, according to the number of individual globulites which enter into their composition. Sometimes they arrange themselves in strings, and into other systems of disposition, implying more or less symmetrical grouping. They usually show a central speck or nucleus, and at times display concentric markings and indications of a radiate structure. M 1 62 The Rudiments of Petrology. Trichites (from fy><, a hair) are minute elongated bodies resembling small hairs or fibres ; sometimes they are straight, sometimes they cross one another in a more or less regular manner ; at others they appear bent in zigzags, or are curi- ously twisted, while occasionally a number of trichites emanate from a central granule around which they radiate or twirl like whip-lashes. Some trichites show regular or interrupted lines of granules attached to them, forming rows like beads either upon one or upon both sides of the trichite. The crystallites proper and crystalloids exhibit in many instances a much higher development, being bounded by curved or by straight lines, and often assuming crucial or stellate forms, which appear to result from the symmetri- cal grouping of individual crystallites. The crystalloids especially exhibit consider- able complexity of internal structure, while in external form they often approximate to crystals of recognised minerals. Some of them indeed show so close a resemblance to true crystals that one cannot help feeling impressed with the significance of their internal structure when contrasted with that of larger crystals. The accompanying figures convey a far better idea than any description could of the forms which these minute bodies present. Microliths. These again show a more complete phase of development than the preceding forms. They are some- times very imperfectly developed, but in all cases it is gene- rally considered that they exhibit a nearer approximation to true crystals. They very commonly display double refraction, occasionally show hemitropy, and frequently present suf- ficiently well-developed faces to enable the observer to Microliths. 163 measure their relative inclination. In some of the larger microliths dichroism may now and then be detected. It is therefore possible at times to determine with some precision the species to which a microlith belongs. Occasionally crystals are to be met with which are visibly built up of microliths, as in the case of the hornblende crystal, fig. 72, which is copied from a woodcut in Zirkel's FIG. 72. ' Mikroskopische Beschaffenheit der Miner- alien und Gesteine.' Microliths are to be found in most eruptive rocks, and in many metamorphosed sedimentary deposits. Glo- bulites, trichites, crystallites, and crystalloids may best be studied in sections of vitreous ., rocks such as obsidians, pitchstones, and perlites, also in artificially formed glasses and slags. Streams of microliths may commonly be seen under the microscope in sections of pitchstone and perlite. They often lie with their longest axes in one direction, which represents the direction of flow in the once viscid mass, for we not merely see micro- liths but also strings of vitreous matter, spherulites, &c., elongated and drawn out in the same direction. The microliths sweep in curves round any large crystals or frag- ments which may chance to lie in their course, and seem to have behaved just as planks or sticks do when floating down a stream. These appearances in a rock are spoken of as fluxion structure or fluidal structure. The development of microliths is one of the causes of devitrification in glassy rocks and in artificial glass. Micro- liths also occur as products of alteration, frequently filling or partially filling the interior of crystals which are undergoing decomposition. For further information upon the microscopic characters of crystallites, both of natural occurrence and of artificial for- mation, the reader is referred to ' Die Krystalliten J by the late Hermann Vogelsang. Bonn, 1875. M 2 164 T/ie Rudiments of Petrology, FLUID INCLOSURES, &c. When salts are allowed to crystallise from a saturated solution, it is by no means uncommon to find that the crystals, in the course of their formation, shut in small portions of the mother-liquor ; and should the temperature at which the crystals form be a moderately high one, the imprisoned fluid will, upon cooling, diminish in volume, so that a vacuity in the form of a bubble will also be seen to occupy a portion of the cavity originally filled by the liquid. The relative dimensions of these bubbles to the cavities which contain them have FIG. 73. been carefully studied by Sorby, Renard, Phillips, Ward, and other observers. The cavities which contain these fluids are of very vari- able form (fig. 73). Occa- sionally they are so large as to be distinctly visible to the naked eye, but usually they are of quite microscopic dimensions. In the excep- tionally large ones the bubble may be seen to move to different parts of the cavity by merely turning the crystal about in the hand. In the microscopic cavities the bubble can be made to move and the liquid to expand by the application of heat. This may be effected either by means of a voltaic current or by a blast of heated air. 1 The cavities containing air and gases, which are sometimes met with in crystals, present strong, dark out- lines, which serve to distinguish them from those containing fluids, while the differences in the refractive indexes of their contents also serve as another distinction. Furthermore, although some cavities occur completely filled with liquid, 1 W. N. Hartley, ' On Identification of Liquid Carbonic Acid in Mineral Cavities,' Trans. Royal Mic. Sac., vol. xv. p. 173, 1876. Glass and Stone Inclosures. 165 still the presence of movable bubbles in most of the fluid- containing cavities at once affords a means of distinguish- ing them. In some microscopic inclosures of fluid very minute bubbles, which have a spontaneous motion, may be seen under high powers. The liquids usually contained in such cavities are water, liquid carbonic acid, and aqueous solutions of salts, fre- quently of chloride of sodium; and occasionally cavities may be seen in quartz which, besides the liquids and bubbles, contain minute cubic crystals of rock salt (fig. 74). 1 Glass Inclosures are of common occur- rence in the minerals which are met with in vitreous rocks, or in rocks which contain a certain amount of interstitial glassy matter. They are spherical, sphe- roidal, fusiform, or of very irregular shape, or else they assume definite crystallogra- phic forms, corresponding as a rule with that of the crystal in which they occur. Such forms may be regarded as negative crystals. They either appear as singly refracting matter, or, when more or less devitrified, as doubly refracting. In the latter case they may be devitrified either by the development of crystalline granules or of microliths. Glass inclosures frequently contain bubbles, but these bubbles are fixed, and do not change their position when the section is heated. Stone Inclosures are analogous to the foregoing, except that they consist of portions of a rock's magma which has a crystalline and not a vitreous character. In the cases both of glass and stone inclosures small portions of the matrix have been taken up while still in a fluid or pasty condition by the crystals in which they occur, 1 Vide Memoirs sur les Roches dites Plutoniennes de la Belgique, De la Vallee Poussin et A. Renard. Bruxelles, 1876. Also 'The Eruptive Rocks of Brent Tor and its Neighbourhood,' Memoirs of the Geological Survey of England and Wales, F. Rutley. London, 1878. 1 66 The Rudiments of Petrology. and the crystal having developed itself before the solidifica- tion of the surrounding magma, these small portions have been shut off and imprisoned. Sometimes the severance of the little mass of glass or other matrix has not been perfectly effected, and it merely appears as a small pocket with a constricted neck, which opens out on the margin of the crystal. Such partial inclosures may frequently be seen in the quartz of quartz-porphyries and quartz-trachytes. PROVISIONAL NAMES APPLIED TO MINERALS. The following are terms used to designate provisionally certain substances which are sometimes met with in thin microscopic sections of rocks, and which, from occurring only in very minute quantities difficult of isolation, it has not as yet been possible to analyse. Their precise che- mical constitution and mineralogical affinities are, therefore, undetermined; and, to avoid erroneous descriptions of them, certain terms have been coined which merely imply sub- stances which present certain microscopical appearances, and whose mineralogical 'characters may vary more or less, and may embrace several distinct mineral-species under each .term. Viridite includes mineral matter which is probably re- ferable to different varieties of chlorite and serpentine. It appears under the microscope in the form of translucent green matter, either in little scales, or fibrous aggregates. It may always be regarded as a product of decomposition, and frequently represents the alteration of such minerals as hornblende, augite, olivine, &c. It is probably a silicate of magnesia and protoxide of iron. Opatite is the term applied to perfectly opaque, black, amorphous, microscopic granules, patches, or scales. It is usually present in rocks which contain magnetite; frequently it forms pseudomorphs after other minerals. It is regarded by Zirkel as representing earthy silicates, possibly allied to micas, and amorphous metallic oxides, especially hydroxides x v. Felsitic Matter. / j . 167 and oxides of titanium and manganese/ /In soto'ev cases it may be amorphous magnetite, or carbonaceous matter graphite, &c. Ferrite is amorphous red, brown, or yellow earth/ matter which is often pseudomorphous after ferruginous minerals/ Chemically it most likely represents hydrous or anhydrous^ oxides of iron, but the different kinds cannot be referreoT~ with any certainty to definite mineral species. FELSITIC MATTER. This substance, which is of such common occurrence in many rocks, and, in some, constitutes a very large proportion, forming the groundmass of quartz-porphyries and many other porphyritic rocks, and often representing, in the rhyo- litic series, the devitrification of glassy matter, has hitherto been described in a more or less vague manner by numerous observers. The student has consequently been left in a state of considerable doubt as to what felsitic matter really is, and, as a rule, the more he has read on the subject, the less able has he been to fix any satisfactory definition for the term. A masterly account of the various opinions which have been put forward on this subject will be found in Rosen- busch's ' Mikroskopische Physiographic der massigen Ge- steine.' Stuttgart, 1877, p. 60 et seq. In this place it will suffice to give the conclusions arrived at by Prof. Rosen- busch, since, although they represent in part the views of Prof. Zirkel, they seem to meet most of the objections to which other definitions are open, and possess a precision hitherto wanting in most descriptions of these difficultly determinable substances. To begin with ; these substances which cannot be pro- perly investigated, except under high magnifying powers, may be resolved microscopically either into a thoroughly crystalline aggregate, or into homogeneous, amorphous matter. 1 68 The Rudiments of Petrology. The former is designated groimdmass by Vogelsang and Rosenbusch ; and the latter magma. Zirkel employs the name basis for the amorphous sub- stance, or ' unindividualised ground-paste/ as he terms it, and Rosenbusch also adopts the term basis. Zirkel's definition of a micro-felsitic basis is: 'that it is amorphous, that it shows, in thin sections, no independent contours. Its boundaries are moulded upon the forms of the crystalline constituents, and it forms roundish creeks or inlets in the latter. Its true nature is variable, and not easy to render in words. It represents a devitrification product in which, indeed, a hyaline aspect is utterly wanting, but which, on the other hand, is not separable into true indi- vidualised parts. It generally consists of quite indistinct, often half-fluxed granules, or indistinct fibres which consti- tute the micro-felsitic mass. Between crossed Nicols it becomes, in its typical development, perfectly dark, but also, at times, transmits a very feeble, fluctuating light. The little fibres and granules often show a decided or a rough ten- dency to radial arrangement. In thin section micro-felsitic matter appears very clear, and either light-greyish, yellowish, reddish, or quite colourless. It is often speckled with little dark granules which in certain spots show a crude radial arrangement, or it may be sprinkled with brownish-yellow and brownish-red granules of a ferruginous mineral. An ultimate glass magma may be present in many micro- felsitic masses, although not clearly to be recognised as such. Experience shows that weathered felspars may be represented by micro-felsitic matter. ! Rosenbusch states that in many cases felsite, or the groundmass of porphyries, consists of a microscopically fine-grained aggregate, formed of minerals which can be identified with those constituting granitic rocks, often in the same combinations as those in which they occur in rocks 1 Mikroskop. Beschaff. d. Min. in Gest., Zirkel. Leipzig, 1873, p. 280. Felsitic Matter. 169 of the granitic group. One or other of these minerals is often absent, and of these mica is the one which is generally missing. So long as the granules of such aggregates, which differ in no essential respect from many vein-granites, or the groundmasses of many granite-porphyries, are not of too minute dimensions, one can recognise the mosaic-like aggregate polarisation and the sharp boundaries of the indi- vidual crystalline grains. Diminution in the size of the grains naturally renders the recognition of the individual particles more difficult, and often impossible. The individual granules do not merely lie side by side, but also in various planes one over another, and the various refractions, reflec- tions, and interferences which ensue from these overlaps tend to render any deductions concerning the optical characters of the constituent granules highly untrustworthy, and give rise to the generally vague transmission of light which cha- racterises these aggregates when they are examined between crossed Nicols. Rosenbusch goes on to state that if we accept Groth's definition of a crystal as a compact body in which the elas- ticities differ in different directions, and, if we furthermore allow that external boundaries are immaterial so far as the foregoing definition extends, it follows that if double refrac- tion be the result of any mechanical conditions of tension, or strain, the expression ' non-individualised,' used by some authors in reference to micro-felsitic matter, is either mean- ingless, or that it indicates, at best, that external form does not entail special internal conditions. From this point of view Rosenbusch designates as crystalline all those parts of felsites which are doubly refract- ing, so long as it cannot be demonstrated that their aniso- tropy is in any way the result of anything resembling conditions of strain which are not related to molecular structure. Basing his nomenclature upon these considerations, Rosenbusch describes those parts of porphyritic ground- 1 70 The Rudiments of Petrology. masses which are aggregates of mineralogically-recognisable elements as micro-crystalline, while those parts which are simply crystalline aggregates, without any definite character being discernible in the constituent particles, he terms crypto-crystalline. Those portions of porphyritic groundmasses in which no double refraction can be recognised must be regarded as amorphous, although, as Rosenbusch remarks, that, ex- cluding isometric crystals, there are those belonging to other systems in which the elasticity-differences in some directions become too insignificant to afford any perceptible phenomena o/ double refraction when thin sections of them are examined. He also cites certain alteration-products after pyroxene and amphibole, in which their anisotropic character can only be distinguished by their pleochroism. In very many cases micro- or crypto-crystalline matter contains an intimate admixture of fine films, stripes, and flecks of a perfectly structureless and almost invariably colourless substance which remains dark in all positions between crossed Nicols. It may be absolutely homoge- neous, or it may contain excessively minute granules and trichitic bodies of various kinds. This substance Rosen- busch designates glass or glassy-basis. The condition in which it contains the granules and trichites he regards as a phase of devitrification. In most instances this impure or devitrified matter is opaque, or so feebly translucent, and occurs in such minute films or grains, that a determination of its isotropic or anisotropic character is seldom possible. Instead of a true glassy-basis, matter of a somewhat different kind is very often present, forming excessively thin films which appear interwoven with the micro- or crypto- crystalline aggregates. This substance is perfectly isotropic, colourless, greyish, yellowish, or brownish, but, unlike true glass, it is not structureless, but appears to be made up of extremely minute scales, fibres, granules, or aggregates of granules, together with other developed forms and interstitial Cleavages. 171 matter. It differs also from micro- and crypto-crystalline aggregates in its want of any action upon polarised light. This substance is the micro-felsite or micro -felsitic basis of Rosen - busch. It is not micro- or crypto-crystalline, and it is not amorphous in the sense in which those terms are employed. Rosenbusch adds that it yet remains to be shown whether micro-felsitic matter is inert upon polarised light, owing to exceptional conditions of tension, or whether it should be regarded as a fibrous, scaly, or granular glass, or as some- thing else. The observations of Leopold von Buch, Delesse, Stelz- ner, Wolff, Vogelsang, Allport, Kalkowsky, and other writers, are commented upon in the review of this subject, which Rosenbusch gives in the work from which these statements have been extracted. 1 TABLE SHOWING THE CLEAVAGES OF THE MOST COMMON ROCK-FORMING MINERALS. Cubic System. Dodecahedron Garnet Hauyne Nosean Sodalite very imperfect rather perfect Tetragonal System. Prism(2nd order) Prism (ist order) Basis Pyramid(ist ord.) 00 P 00 OOP oP P Leucite most imperf. most imperf. Scapolite rather perfect less distinct Zircon imperfect imperfect Melilite rather perfect 1 Miltroskopische Physiographic. Stuttgart, 1877, bd. ii. p. 60 et seq. 172 The Rudiments of Petrology. Hexagonal System. Rhombohedron Prism Basis R 00 P o P Biotite highly perf. Apatite imperfect imperfect Tourmaline very imperf. . f ooP2 very) 1 imperf. J Calcspar very perfect Nepheline imperfect imperfect / traces have \ Quartz l most imperf. j been ob- I _- ( served ) Rhombic System. Brachy-pina- Macro-pina- koid koid Prism Basis 00 P 00 00 P 00 00 P oP Olivine rather distinct very imperf. Enstatite imperfect distinct Bronzite very perfect traces imperfect - Hypersthene M very imperf. distinct Andalusite traces traces not very dist. Muscovite imperfect highly perfect Monoclinic System. Ortho-pinakoid Clino-pinakoid Prism Basis oo -E oo 00 00 OOP oP Orthoclase very perfect very imperf. very perfect Augite imperfect imperfect j more or ) 1 less perf. J Diallage perfect 5) rather perf. Hornblende very imperf. very imperf. very perfect Epidote perfect very perfect TricUnic System. Brachy-pina- Macro-pina- koid Prism Basis 00 P 00 00 P 00 00 P oP Labradorite rather perfect very perfect Oligoclase i hemipr. im- j perf., also \ perfect I pr. imperf. ) Anorthite perfect Albite f hemipr. im- } 2 1 perf. 1 Cleavage in quartz is rare. 2 Also tetarto-pyram. imperfect. Cleavages. 173 PRISM S7 Q 5' PLATE II. PRISM 87 5' PRISM 024 30' HORNBLENDE PRISM 87 5' CLEAVAGE 124 30 f CO P GO HYPERSTHENE U RALITE CLEAVAGES. MMMR Highly perfect. Perfect. - Imperfect. - Very imperfect. The observer is supposed to be looking down directly on the basal planes of the crystals. Except in fig. f, the angles of the prismatic cleavage correspond with those of the prisms. Figs, a, b, c, and f represent monoclinic crystals. Figs, d and e represent rhombic crystals. 174 PART II. DESCRIPTIVE PETROLOGY. CHAPTER XL THE CLASSIFICATION OF ROCKS. THE classification of rocks involves considerable difficulty, and no scheme has yet been propounded which is not more or less open to objection. Our knowledge is not at present extensive enough to enable us to speak with certainty regarding the origin of all the different rocks with which we are acquainted, and we are not as yet in a position to assert how far the mineral constitution and the physical charac- ters of rocks afford clues to their origin. The following points have to be considered in framing a petrological classification : (i.) The chemical composition of the rocks. (ii.) Their mineral constitution. (iii.) Their physical characters. (iv.) Their mode of occurrence. (v.) Their order of sequence in time. The chemical examination of a rock shows us what elementary substances enter into its composition, and may afford some clue to its mineral constitution and to its origin. The mineralogical and physical examinations teach us how those elementary substances have combined, and, Classification of Rocks. 175 in some cases, the conditions under which those combina- tions have been effected, the various minerals which enter into the composition of the rock, the crystal! ographic and other physical peculiarities which the component minerals present, the relative order in which those minerals have sometimes crystallised, the arrangement, if any, which the individual crystals, granules, scales, or fragments observe towards each other, and the general state of aggregation of the crystals or mineral particles of which the rock is composed. The microscope, furthermore, affords the means of ex- tending these investigations by enabling the observer to see structural peculiarities which unassisted vision fails to detect. The following classification has been framed for the purpose of bringing certain important typical rocks promi- nently before the student's notice, these type-rocks consti- tuting, as it were, the nuclei of their respective groups. Since the groups of each class merge into one another more or less in mineral constitution, no sharp boundary lines can be drawn between them ; the type-rocks of the different groups therefore serve as milestones by means of which the student may ascertain in what part of the great series to class any particular rock ; the types holding a relation to the whole petrological series somewhat analogous to that which Frauen- hofer's lines bear to the spectrum. CLASSIFICATION OF ROCKS. ERUPTIVE ROCKS. I. Vitreous. Obsidian \ Pumice } including hyaline rhyolite. Perlite Pitchstone Tachylyte. 176 Descriptive Petrology. II. Crystalline. * /-Granite group Felstone Syenite Trachyte ' including rhyolite proper. Phonolite A. Typical groups \ Andesite Porphyrite Diorite > included under the old -Diabase [ term 'greenstone' in -Gabbro [ its original and broadest -Basalt ) signification. B. Rocks of exceptional mineral constitution. III. Volcanic Ejectamenta. IV. Altered Eruptive Rocks. SEDIMENTARY ROCKS. I. Normal Series. Arenaceous group (sands). Argillaceous (clays). Calcareous (limestones). II. Altered Series. A. With no apparent crystallisation. B. With sporadic crystallisation. C. Crystalline I ^ ^on- foliated. [o. foliated and schistose. III. Coarse Fragmental Series. Breccias and conglomerates. IV. Tufas and Sinters. V. Mineral Deposits consisting Rock-Masses. Vitreous Rocks. 177 ERUPTIVE ROCKS. CLASS I. VITREOUS ROCKS. The vitreous rocks are characterised by their generally homogeneous aspect, their more or less glassy lustre (which, however, is sometimes only feebly glassy, greasy, or dull when the rock is partially devitrified), by their conchoidal fracture, and, optically, by the single refraction which they exhibit, except when more or less crystalline structure has supervened. They may, like the crystalline eruptive rocks, be divided into two sub-classes, the highly- silicated or acid (those containing over 60 per cent, of silica), and the basic (or those which contain less than 60 percent). Some of those usually occurring in the former sub-class vary somewhat in the amount of silica which they contain, and at times appear to belong rather to the basic sub-class j the pumice from some localities, for example, having less than 50 per cent, of silica, while that from others contains over 62 per cent. The vitreous rocks may be conveniently arranged in the following order : I. Containing over 60 per cent. SiO 2 : Obsidian. Pitchstone. Pumice. Perlite. II. Containing less than 60 per cent. SiO 2 : Tachylyte. Pumice. The vitreous rocks of the first or highly-silicated sub- class closely resemble the liparites, trachytes, andesites, and other highly-silicated eruptive rocks in their chemical com- position, while the minerals which are developed in many of them also imply a similarly close relationship. So close, indeed, is this relation that some petrologists include obsidian, pitchstone, perlite, pumice, and certain quartzi- N 1/8 Descriptive Petrology. ferous trachytic lavas, under the terms rhyolite and liparite. 1 The student should therefore bear in mind the fact that the separation of the vitreous from the crystalline rocks refers merely to physical differences which the members of these two sub-classes respectively present, and does not imply any special difference in their chemical composition. These physical differences depend upon the conditions under which solidification was effected, whether gradual or rapid. In the former case the molten mass would develop crystals, in the latter it would remain amorphous : it would, in fact, result in a more or less perfect glass. In these natural glasses it is, however, common to find crystallites and crystals, the former usually developed very completely, the latter less perfectly formed as a rule, since they generally' present rounded boundaries, or their angles, if any exist, also appear rounded. The cause of these rounded boun- daries does not, as yet, seem to be satisfactorily deter- mined. It is known that fragments of rock and individual crystals become rounded by constant attrition during their ejection from, and their returning fall into, the throat of a volcano ; and since, in rather rare instances, the microscope shows that some of these vitreous lavas contain not merely rounded crystals, but also well rounded fragments of other vitreous rocks of a quite distinct and different character from that of the matrix in which they are enveloped, it seems possible that in such cases the rounded crystals and fragments of rock represent ejectamenta, which, rounded by attrition, and lying within or around the crater, have been taken up by the viscous mass of lava as it welled over them. If this were the case, we might at first be tempted to think that the rounding was due to the superficial fusion of the 1 The name rhyolite, from uo| (a lava stream) and \idos, was in- troduced by v. Richthofen in 1860, and included certain Hungarian quartz-trachytes, which showed strong evidence of viscous fluxion, and the highly silicated vitreous rocks just mentioned. A year later Justus Roth applied the term liparite to similar crystalline and vitreous rocks occurring in the Lipari Islands. Vitreous Rocks. 179 fragments or crystals, just as fragments of minerals become fused in a borax bead before the blowpipe ; and it may be that such a supposition is not wholly incorrect, since, although in the borax bead the minute fragment as it fuses becomes surrounded by visible tortuously-twirling strings of its own molten substance, before these fused products become per- fectly incorporated with the borax glass; still we must remember that this fused, ropy matter is visible, because it differs in density from the fused borax, while in the case of vitreous rock fragments, felspar crystals, &c., fusing in a highly heated vitreous magma, the respective specific gravi- ties do not differ sufficiently to render the phenomenon of imperfect incorporation apparent. The sp. gr. of borax is 171 obsidian 2-4:1 2-57 perlite 2-25 sanidine 2*56 2*6 plagioclase 2-56-276 It should, however, be borne in mind that the substance of felspars, which are the principal rounded crystals in vitreous rocks, is approximately colourless, so that in a colourless magma the phenomena of imperfect mixture would not be apparent. Such phenomena are, however, distinctly visible in some obsidians and pitchstones, in which, under the microscope, tortuous lines of glass, differing markedly in colour or tint from the glass in which they lie (fig. 75), denote, no doubt, a difference in the relative specific gravities of the two glasses. 1 Such included glass lines and bands in hyaline rhyolites, although they show us that the glass is not homogeneous, do not furnish us with any clue as to the source of the material which differs from its matrix. It cannot well be imagined that the rounded crystals and 1 This may be seen in the obsidians from Tolcsva in Hungary, Truckee Ferry in Nevada, and other similar rocks. Kindred phenomena may be seen on mixing liquids of different specific gravities. N 2 1 80 Descriptive Petrology. fragments in these vitreous lavas were showered down on the surface of the viscous lava stream, since that would imply a synchronous eruption of lava and ejection of ashes, dust, sand, &c., from the same vent, for, where two craters are situated near one another, one is generally at rest while the other is active. If neither of the foregoing hypotheses be adopted there seemingly remains but one other, namely, that the crystals have been developed during the solidification of the rock, and that the rounded contours which their sections present are due to aborted crystallisation, such as that pointed out by Poussin and Re'nard as occurring in the orthoclase of the porphyrite of Mairus in Belgium. 1 We may, perhaps, admit with safety that in many instances the crystals have been developed in the rock ; but, if we admit it in all cases, how are we to account for the included fragments of rock which may occasionally be noticed in microscopic sections of these lavas? It is also worthy of note that the same section may exhibit crystals with well- developed angles and also rounded crystals of the same mineral. Certain structural peculiarities and inclosures, many of which can only be observed microscopically, are character- istic of the vitreous rocks. These structures or inclosures do not always individually characterise these rocks, since it is not uncommon to find crystals, crystallites, microliths, spherulites, &c., all developed in the same specimen. The following is a descriptive list of the principal struc- tures which occur in vitreous rocks. Homogeneous. This condition is more hypothetical than real, since, when examined microscopically, scarcely any of the most homogeneous-looking obsidians are seen to be free from inclosures of microliths. If, however, these microliths and other inclosures be put out of the question, the glassy 1 Memoire sur les caracttres mineral ogiques et stratigraphiques des Roches dites Plutoniennes de la Belgique et de P Ardenne Fran$aise. De la Vallee Poussin et Renard. Bruxelles, 1876. Banded and Damascened Structures. 181 .Fie. 74 A. matrix in which they lie may be regarded as homogeneous, or as approximately homogeneous, although, under high powers, it often shows included dusty matter, which might exhibit some definite characters if still higher powers were employed. It may, however, be stated that, as a rule, all of these natural glasses contain fine dust and micro- liths. Banded. This structure is rendered evident by the inter- lamination of glasses which differ in tint, or by the segrega- tion of granular mat- ter in strings or layers. The bands are seldom continuous for any distance, and usually exist merely as elon- gated lenticular streaks. Fig. 74 A shows the banded appearance of a sec- tion of black obsidian from the Island of Ascension, magnified 50 diameters. Da mascened. The author suggests this term as a convenient one by which to describe the struc- ture shown in some obsidians, in which streaks or threads of glass are contorted in a confused manner, which some- what resembles the markings on Damascus sword-blades or the damascening on gun-barrels. Fig. 75 represents part of a section of a red obsidian, from Tolcsva, in Hungary, mag- nified 50 diameters, in which the damascene structure is well shown. These twisted threads of glass are of a different tint or colour to that of the glass in which they lie. The appearance which they present when seen in thin sec- tion under the microscope suggests that which two liquids 182 Descriptive Petrology, of different density exhibit when they are imperfectly mixed and slightly agitated, as pointed out on page 179. FIG. 75 . Perlitic. A struc- ture especially cha- racteristic of the rocks termed peril tes, but sometimes deve- loped in other vitreous rocks such as trachylyte, &c. This structure lias been described as a phenomenon attend- ant upon contrac- tion, first by Pro- fessor Bonney and subsequently by the author, who was at the time ignorant of Professor Bonney's conclusions. These conclusions have since been admirably demonstrated by Mr. Allport's examination of some ancient perlites occurring in Shropshire. The structure consists in the deve- lopment of numerous minute cracks which exhibit varying curvature, and produce somewhat concentric and approximately spheroidal or elliptical figures, but the lines which bound these forms do not coalesce as a rule, so that the structure may be described as an imperfect, concen- tric, shaly one, which, on a large scale, finds a parallel in the spheroidal structure developed in some basalts. The spheroids in perlite are almost invariably found to lie packed between minute rectilinear fissures which traverse the rock in all FIG. 76. Spherulitic Structure. 183 directions, but which are seldom seen to cut through the spheroids. Fig. 76 represents a section of perlite from Buschbad, near Meissen, Saxony, magnified about 10 diame- ters. The latter are, however, often seen to be traversed by more or less parallel streams of microliths which bear no relation, or observe no relative disposition, to the spheroids, thus showing that the perlitic structure had no existence when the rock was in a state of fluxion, but was developed on the solidification of the rock. Spherulitic. This is a structure totally distinct from that just described and may be regarded as concretionary, or as resulting from incipient crystallisation around certain points or nuclei. The nuclei, when they exist, consist either of a granule or a minute crystal or crystallite, but most commonly no nucleus is discernible. Spherulitic structure in its most rudimentary form seems to consist in the segregation, in spots, of glassy matter, of a different colour to that which constitutes the matrix, and often contains a considerable quantity of very fine dust, the nature of which has not been ascertained but which is pro- bably magnetite. The glass constituting the spherulites is usually of a deep yellowish- brown colour and, in very perfectly developed spherules, generally forms a broad zone, within which lies clear light- coloured or colourless glassy matter having a radiate struc- ture, due to imperfect crystalli- sation, while, at the central spot, from which the crystals, rods, or fibres emanate, a few doubly refracting granules may sometimes be observed. In some instances, as in the obsidians of the Lipari Islands, a per- FIG. 77- 1 84 Descriptive Petrology. fectly clear, colourless, but very narrow outer zone surrounds the brown glassy envelope of the spherulite, as in fig. 77. (Magnified 150 to 200 diameters). In vitreous rocks from the last named locality, in those from the Island of Ascen- FIG. 7 8. sion, and in many other ex- amples, the spherulites occur in definite layers or belts, and. have, in many cases, been elongated in the direction in which the lava- stream flowed ; at times they even coalesce and form more or less continu- ous bands, as in fig. 78 (mag- nified 22 diameters), which represents part of a band of coalesced spherulites in the obsidian of Rocche Rosse, Lipari. Occasionally, but very rarely, spherulitic structure is so extensively developed in vitreous rocks that the whole mass consists of closely packed spherulites, between which only small patches of the glassy matrix can here and there be discerned, while the spherulites are so closely crowded together that their boundaries are no longer spherical, but, by compression, assume polygonal forms. Spherulitic structure is sometimes developed in artificial glass. A fragment of a plate-glass window, from a house which had been burnt down, exhibited colonies of spherulites, when examined under the microscope. Axiolitic. A structure is developed in some of the vitreous rocks of Nevada, U.S., and elsewhere, the indivi- dual components of which have been termed axiolites by Zirkel. 1 These appear to be somewhat analogous in structure to spherulites ; elongated lenticular and curved zones of brownish glass forming the envelope of a smaller corresponding mass of paler vitreous matter, in which 1 Microscopic Petrography, Zirkel, U. S. Exploration of the Fortieth Parallel. Devitrification. FIG. 79. incipient crystallisation or fibrous structure trends at right angles to the inner surfaces of the envelope towards a longitudinal median line. The great diversity exhibited by such structures is well shown in the work cited in the foot- note from which the accompanying figure (79) is copied. The figure represents the axiolitic structure visible in a rhyolite from N.W. of Wadsworth, Nevada, U.S. Porphyritic. This term is applied to vitreous, just as to other rocks, implying that isolated crystals distinctly visible to the naked eye occur in them. The appli- cation of this term has in " all cases a purely arbitrary limit, refers to to their occur a purely arbitrary limit, since it not merely the mode of occurrence of the crystals, but also size ; rocks in which very small isolated crystals only being spoken of as micro-porphyritic, simply because, from their small dimensions, they do not convey to the naked eye the blotched appearance which characterises the commonly recognised porphyries. Ttichitic and Microlitic are terms which might also be given to those vitreous rocks which contain multitudes of the bodies already described as trichrites and microliths ; but as nearly all vitreous rocks are more or less microlitic, and as the word 'trichitic' sounds inconveniently like the adjective ' trachytic,' which latter is often applied to rocks of this class, such terms as trichitic and microlitic are perhaps better left alone. Devitrified. This implies that the rock has undergone, to a greater or less extent, certain physical changes which cause it no longer to behave as a glass, its vitreous character being partially or completely destroyed by the development either of microliths, crystalline granules, or crystals. The 1 86 Descriptive Petrology. ultimate stage of crystalline-granular devitrification is felsitic matter, and, when a rock has undergone complete change of this kind, it is only possible to arrive at conclusions as to its once vitreous nature, by means of those structural peculiar- ities which indicate former fluxion, and, should those characters fail to be very well marked, it is, as a rule, most hazardous to jump at any conclusions concerning the original condition of the rock. Filiform. A condition occasionally, but rarely, met with ; as in the filiform lava of Hawaii, in the Sandwich Islands, known as Pe'le's hair, in which molten vitreous lava has been frayed out and blown by the wind into long and extremely slender glassy threads, which commonly terminate in little fused knobs or pellets. This structure is also produced artificially in blast-furnace slags. OBSIDIAN. C Obsidian results from the quick solidification of lavas which, if slowly cooled, would develope crystalline structure and assume the character of trachyte, Hparite, &c., rocks which contain over 60 per cent, of silica. In obsidian no crystalline structure is developed ; it is a true, natural glassT) Nevertheless, obsidians frequently contain microliths; ana, when spherulitic^ the spherulites commonly show a radial crystalline or fibrous structure. Obsidians present a very homogeneous appearance and a strong vitreous lustre. Their fracture is eminently conchoidal. ( They vary somewhat in colour, but are mostly black or grey. In thin splinters they are all more or less transparent, = N Obsidians also vary in chemical composition^ The silica may be estimated at from 60 to 80 per cent.) the alumina at 18 to 19 per cent., while the remaining constituents are potash or soda, lime, magnesia, peroxide of iron, and occa- sionally as much as 0-5 of water. Their specific gravity ranges between 2^4 and 2*5, and the hardness equals 6 to 7. Obsidian. 187 (Those which contain the largest proportion of silica are only slightly acted upon by acids^ (Before the blowpipe obsidian is fusible on the edges of thin splinters^ Sections of obsidian when placed under the microscope between crossed Nicols exhibit no double re- fraction, the field appearing quite dark ; but this dark field is usually thickly studded with bright doubly-refracting microliths, and under moderately high powers crystallites of varied forms, exhibiting structural peculiarities of excessive beauty and interest, may often be met with in great profusion. (There are, however, some obsidians, such as the pseudo- chrysolite or bouteillen stein, which occurs as rounded pebbles in sand] at Moldauthein in Bohemia, and in some of the tuffs near Mont Dore in Auvergne, which show no crystallites under the microscope, and equally pure obsidians occur in one or two localities in New Zealand and in Iceland. All of these, however, contain great numbers of gas pores. The crystallites which occur in obsidian vary so greatly in form that mere descriptions of them would be of little use to the student. The precise mineral species which they represent are in many cases undetermined, but it is probable that many of them are incipient felspar crystals. In some of the small crystals or microliths, which are so common in rocks of this class, it is interesting to note the gradual deve- lopment of structure which may sometimes be seen in a single microscopic section ; in one place simply a comb-like crystalline growth springing from minute tapering rods, which constitute, as it were, the visible axes of these little crystal- lites ; in another, a microlith, or small crystal, in which may be seen a structure identical with the preceding, and which seems to show the plan upon which it has been built up, to be in fact the framework upon which it has been developed. The little axial rods, if they may be so termed, are not always straight ; at times they have a somewhat sigmoidal flexure, at others they occur in pairs arranged like two bows set back to back. This disposition is occasionally coupled 1 88 Descriptive Petrology. with the intermediate development of a small, square, rect- angular, or rhomboidal mass, from the four corners of which the apparent homologues of these arcuate rods sprout out like horns, as in fig. 80, while the whole is surrounded by a FIG. 80. hyaline border, whose external boundary and occasionally striated structure indicate differen- tiation of the surrounding glassy magma and the incipient extension of crystalline develop- ment. Similar borders often surround crystal- lites which give sections like those which an ^350' * elongated pyramid would afford ; and cruciform groupings, which closely resemble aggregates of such pyramids, may also be seen at times in obsidians. (Besides the crystallites just mentioned, it is common to find spherulites developed in these lavas.) In their most rudimentary condition they occasionally seem to be repre- sented by blotches of a glass, of deeper colour than that of the surrounding matrix. Generally the most perfectly de- veloped spherulites have a somewhat broad border of glass, which appears reddish-brown by transmitted light, and sur- rounds a central spherule of clear and often almost colour- less glass, in which a radiate structure is developed, while, in some instances, the whole spherulite is surrounded by a narrow colourless envelope of clear glass. These spherulites are sometimes elongated in the direction in which the once viscous stream of obsidian flowed, and this elongation has occasionally taken place to such an extent that the spheru- lites have coalesced, and formed more or less continuous bands, of which the central portion consists of vitreous and, at times, micro-crystalline matter. This is cased in an outer envelope of glassy matter which appears reddish-brown by transmitted light, and generally snow-white or greyish by reflected illumination. Occasionally this is surrounded by a thin external coat of clear, colourless glass, which, unlike the clear absorption areas seen around the crystallites and dust segregations in some vitreous rocks, is bounded by a Obsidian. 189 FIG. 81. sharp line of demarcation from the glass which constitutes the matrix. Such spherulitic bands have sometimes, when the coalescence of the spherulites has only extended to their cortical zones, merely the aspect of beads closely strung together ; but, in cases where the coalescence has been more complete, the boundaries of the bands are approximately parallel straight lines, so that the structure of such a band or string may be diagram- matically represented as in fig. 8 1, a being the transverse and b the longi- tudinal section. It is not, however, to be supposed that, where vast multi- tudes of spherulites are developed on the same plane, transverse sections such as a (fig. 81) are invariably to be procured, and in such cases we may assume that the coalescence of the spherulites gives rise to sheets, rather than strings, the vertical sections through such sheets afford- ing in all directions a disposition corresponding with b (fig. 81). In some cases, as in the spherulitic obsidian of Rocche Rosse in the Island of Lipari, clear colourless rods of glass are seen to have been extruded through the cortical layers of the spheru- litic bands into the surrounding glassy matrix (fig. 82), which are further enlargements of the little rods shown in fig. 78, page 184. After emergence from the band they are frequently hooked FIG. 82. 190 Descriptive Petrology. FIG. 83. or bent, but not, as a rule, in any mutually definite direction, and in most sections they are seen either to terminate blindly in rounded ends, or to be cut off on the ground surfaces of the preparation. ( Small crystals and microliths, as already stated, are of common occurrence in some obsidians. In many cases they can be safely identified with recognised minerals^ such as sanidine, plagioclase, augite, hornblende, olivine, tourma- line, zircon, magnesian mica, specular iron, and magnetite. The felspars occur in small prisms. The augite and hornblende exist either as distinct crystals, similar to the ordinary forms, or as minute acicular bodies and spicular forms (' belonites ') which are often bordered by imperfectly radiate, fibrous or hazy, and almost dendritic tufts, which cause them somewhat to resemble the fronds of ferns. Beautiful crystallites of this description may be seen in some of the pitchstones of Arran (fig. 83), and have been identified by S. Allport as augite. The crystals of olivine are always of moderate size and no microliths of this mineral have as yet been detected. The occurrence of tourmaline and zircon has not been definitely determined, but certain prisms belonging to the hexagonal or rhombohedral system have been thought to be tourmaline, while some tetragonal forms are regarded as zircon. The magnetite occurs in opaque octahedral crystals or granules, and the specular iron in little yellowish-red or orange-coloured hexagonal tabular crystals. The magnesian mica forms small, deep, reddish-brown scales and crystals which resemble those of specular iron, but although they show no dichroism when their basal planes coincide with the planes of section, yet their sections are very strongly dichroic when cut transversely or obliquely to the basis. Obsidian occurs on a large scale as lava-flows, which Pumice. 191 frequently present very rough and jagged surfaces. That forming the Rocche Rosse in the island of Lipari is a good example. The accompanying figure of this obsidian stream, which has issued from the crater of Campo Bianco, breach- ing one side of the crater, and flowing over the white pumice tuffs, of which the cone is composed, is copied by permission of Prof. J. W. Judd from one of the sketches published in his ' Contributions to the Study of Volcanoes,' Geological Magazine, Decade II. Vol. ii. No. 2, p. 66. FIG. 83 A. fobsidian at times becomes vesicular./' In some of the obsidians of Hawaii the vesicles are quite'spherical ; in others they are elongated or otherwise distorted. Occasionally these rocks are very finely vesicular, as is the case with some of the obsidians of the Lipari Isles, and^the vesicles are at times so numerous that the rock acquires quite a frothy character, and passes into pumice./) \Obsidians occur in districts where trachytic rocks are common. PUMICE. (Pumice is a porous, vesicular glass ; the vesicles being frequently elongated) sometimes in a more or less definite 192 Descriptive Petrology. direction, while at others they anastomose, and give rise to an irregular network of fibrous, intervesicular matter. uPumice varies considerably in chemical composition, the percentage of silica ranging between 57 and 73.^} The alumina varies from 9 to 20 per cent., and the remainder consists of lime, magnesia, potash, soda, and peroxide of iron. Water is also present. The specific gravity of pumice varies from 1*9 to 2*5. /The fusibility before the blowpipe is greater in some specimens than in others) ("When examined microscopically some pumice appears to consist of interlacing or anastomosing vitreous fibresXew or no microliths or crystallites being developed in the glassy matter. In others microliths and small crystals occur in abundance, and frequently show the stream-like disposition which results from fluxion, or from the drawing out of small portions of viscous lava. The microliths, when they can be determined, are found in most instances to be felspars, both monoclinic and triclinic. Magnetite is also of common occurrence. ( Pumice is developed on the surfaces of obsidian streams, and in such cases can only be regarded as a highly vesicular, spongy, or fibrous condition of obsidian. The porphyritic development of felspar crystals in pumice begets the rock termed trachyte-pumice. Pumice also occurs in the form of loose ejected blocks and fragments. These ejectamenta sometimes constitute volcanic cones. ~\ In fig. 83 A the cone, which is partially broken down by me stream of obsidian, consists of pumice fragments. PERLITE (Pearlite, Pearlstone). Perlite is sometimes , quite glassy in appearance, but it more frequently exhibits a shimmering, pearly, enamel-like, or greasy aspect oh recently fractured surfaces. The colour is mostly pale-greyish, bluish-grey, or yellowish-brown. It often appears to consist in great part of spherical or round- Per lite. 193 ish grains, which have a somewhat concentric shaly structure. These, although commonly individualised, very frequently coalesce, the rock, in such cases, assuming a more homo- geneous character. In chemical composition the perlites approximate to the quartz-trachytes (the rhyolites proper, or the liparites of Von Richthofen 1 ). They are highly silicated, containing from 70 to over 80 per cent, of silica. When heated they give off water, the amount varying from 2 to 4 per cent. Their hardness is about 6. Perlite must be regarded as the vitreous condition of the felsitic rhyolites, and, like other vitreous rocks, plays a very subordinate part in the constitution of the hitherto explored portions of the earth's crust, when compared with those rocks of which it is the vitreous representative. It should, how- ever, be borne in mind that the originally vitreous character of many eruptive rocks has yet to be discovered, since, owing to devitrification, they frequently present appearances which, in the absence of microscopic investigation, afford no clue to the physical characters which they possessed at the time of their eruption. It is indeed more than probable that many of the so-called hornstones, felstones, and even rocks, which were mapped by the older geologists as green- stones, are merely rhyolitic rocks in a devitrified condition. In this country the researches of Professors Bonney and Judd, Mr. Allport. and the author, have already demonstra- ted the existence of rocks of a rhyolitic type, of which the real characters had previously been overlooked. When examined microscopically thin sections of perlite exhibit numerous fissures, and between these fissures great numbers of somewhat concentric cracks are visible, causing a sepa- ration of the rock into more or less regular spheroids. The cracks do not appear to join up in continuous ellipses, but % 1 The term liparite has been applied by Roth to th% whole of .the rhyolites. Von Richthofen, however, limits the use of the^rm to the felsitic rhyolites, or rhyolites proper. O IQ4 Descriptive Petrology. thin off and at times overlap. They nevertheless form ap- proximately concentric envelopes around the spheroidal nucleus of glass. It is worthy of remark that these bodies are seldom or never traversed by the straight cracks which run in various directions through the rock, but lie between them, often closely packed, distorted, and apparently com- pressed against the planes of the straight fissures. This indicates that the straight fissures were formed first and that the spheroidal or perlitic structure, as it may be termed, was subsequently developed. Streams of microliths commonly occur in these rocks, and they traverse the perlitic bodies without the slightest indication of deflection. They have, in fact, been quite uninfluenced in their direction of flow by the minute structural planes and elliptical cracks which occur so plentifully in the rock. Crystals, both megascopic and microscopic, occur in considerable numbers in some perlites. They consist prin- cipally of sanidine and plagioclastic felspars, magnesian mica, magnetite, and occasionally specular iron. The mi- croscopic inclosures consist mostly of felspar-microliths, trichites, and belonites. Spherulitic structures are also present at times. That the perlitic structure has probably resulted from the development of more or less concentric zones of con- traction on cooling has been pointed out, both by Professor Bonney, and by the author. Perlitic structure bears a some- what close relation to the larger spheroidal structure which is occasionally to be seen in basalt. Indications of perlitic structure may be observed in thin sections of vitreous rocks other than perlite, and the author has noted the incipient development of this structure in an Irish tachylyte (Journ. Royal Geological Society, Ireland, vol. iv., p. 230), thus showing that the structure, although characteristic of some highly silicated vitreous rocks, is not exclusively peculiar to them. l Pitchstone. 195 Professor A. von Lasaulx points out the fact [ that many spherulitic rocks were formerly regarded as perlites, but he adds that spherulitic and perlitic structures are totally diffe- rent, since the latter consist merely of little masses of glass, while the former are ' crystalline individualisations.' Spherulites may in fact be regarded as spots of devitrifi- cation, while perlitic structure is simply a phase of fission resulting from contraction on cooling, the glass included by the elliptical fissures in no way differing from the surround- ing glass. Pitchstone.t{\\& pitchstones may be regarded as vitre- ous conditions of trachyte on the one hand, and, on the other, of those rocks which range from granites to felstones, including the porphyritic varieties of those rocks) such as quartz-porphyry, porphyritic felstone, &c. It is true that pitchstone is not recognised as a vitreous condi- tion of granite, but since passages are known from granite into quartz-porphyry, and from quartz-porphyry into fel- stone, it seems that we may fairly be allowed to regard the vitreous equivalent of the felstones and quartz-porphyries as, at all events, an indirect vitreous phase of granite, which appears more directly to find its rhyolitic representative in the nevadites. The pitchstones are classed as trachytic and felsitic. The devitrification of the trachytic pitchstones is effected by the development of microliths, which for the most part consist of sanidine and hornblende, while the felsitic pitchstones become devi trifled by the setting up of a micro-felsitic or, as v. Lasaulx terms it, a micro-aphanitic structure. (The pitchstones have, as a rule, a perfectly conchoidal fracture, sometimes rather splintery, and a more or less greasy semi- vitreous, or pitch-like lustre") whence the name. <^ They are mostly blackish-green, dark olive green, or brown. Occasionally they are red or dull yellow.) 1 Elemente der Petrographie, p. 223. Bonn, 1875. O 2 1 96 Descriptive Petrology. vln chemical composition they resemble obsidian, but contain from four to ten per cent, of water/) If, however, the water be omitted, and the other constituents brought up to 100, the composition of pitchstone may be .regarded as approximately identical with that of trachyte. The silica ranges from 63 to 73 per cent., the alumina from 9 to 13, and the alkalies from 2 to 8 per cent. Before the blowpipe some varieties do not fuse so readily as others, but they all melt either to a frothy glass or to a greenish or greyish enamel, and yield water when heated in a tube. The pitchstones have a hardness of 5 to 6, and a specific gravity of 2 "2 to 2 "4. (jThey are not acted upon by acids) The pitchstones, as already stated, may be divided into two groups the trachytic pitchstones and the felsitic pitch- stones. The former are related both geologically and in mineral constitution to the liparites, while the latter are re- lated to the quartz-porphyries and felstones. With regard to the trachytic pitchstones, Rosenbusch observes 1 that no sharp line of demarcation can be drawn between them and the glass-magma-liparites which contain water. Sections of pitchstone, when examined microscopically, appear, like other glasses and amorphous substances, dark between crossed Nicols. (By ordinary illumination they appear either as perfectly homogeneous glass containing crystals and microliths, or they exhibit streaks of glass of a deeper or paler tint than that of the surrounding matrix^} which in their disposition at once suggest the presence of fluxion structure. In the latter cases, as well as in the former, crystals are of common occurrence, and microliths are usually developed in abundance. The crystals are mostly sanidine, hornblende, and mag- netite, but plagioclase is not uncommon. The microliths may often be recognised as consisting of the above-mentioned minerals, and in some pitchstones, especially in those from the Isle of Arran, microliths of 1 Mik. Phys. d. Massigen Gesteine, p. 160, Pitchstone. 197 The pitchstone of Corriegills in FIG. 84 augite are also plentiful. Arran shows these augite microliths or belonites in great profusion, forming more or less stellate groups of pale green- ish forms, which somewhat resemble the fronds of ferns, and which have been admirably figured and described by Zirkel, Allport, Vo- gelsang, and other microscopists. Fig. 84 represents a mag- nified section of Arran pitchstone, and is copied, by per- mission of Prof. Zirkel, from his ' Mikroskopische BescharTen- heit der Mineralien u. Gesteine.' The glass of most pitchstones appears, when light is trans- mitted through thin sections, of a pale yellowish, greenish, or brownish tint, but it is scarcely safe to venture a decided opinion upon the precise nature of the pigment. Some- times, but not in all cases, there appears to have been an abstraction of this pigment from those portions of the glassy magma which immediately surround the microliths of magnetite and- augite, this absorption of pigment pro- ducing a comparatively clear, colourless ring around these bodies. QSteam pores, and occasionally spherulitic struc- tures, are met with in pitchstonesp The microliths in these rocks usually lie in streams, which- sweep round the larger imbedded crystals, and indicate, in a marked' manner, the originally viscid condition of the matrix in which they occur. The felsitic pitchstones bear much the same relation to the felstones that the trachytic pitchstones bear to the 198 Descriptive Petrology. trachytes, while the porphyritic varieties, termed pitchstone porphyry, may also be regarded as the vitreous equivalent of quartz-porphyry. To these pitchstone porphyries Vogelsang has given the name vitrophyre. (The principal porphyritic crystals in these rocks are quartz, sanidine, plagioclase, augite, hornblende, magnetite, and biotite. They also at times contain inclosures of glass,) which are sometimes devitrified by conversion into micro- felsitic matter, at other times, by the development of microliths. The devitrification of the magma of the felsitic pitch- stones is as a rule micro-felsitic, while that of the trachytic pitchstone is effected by the development of microliths. Microliths, some doubly refracting and some singly refract- ing, occur however in the .glasses of nearly, if not all diffe- rent varieties of pitchstone. Inclosures of fluid, containing mobile bubbles, are of rare occurrence in these rocks. The devitrification of the porphyritic and felsitic pitch- stones does not, as a rule, take place uniformly, but occurs in a seemingly capricious manner, often being developed in irregularly- distributed and irregularly-shaped patches, some- times occurring in spots, sometimes in strings, which usually indicate fluxion-structure. Dark granules also occur in these rocks, which possibly represent magnetite, and these granules frequently assume a string-like or banded arrange- ment, while irregular strings or lenticular flecks of included glass also denote a former state of fluxion. In their most perfectly vitreous conditions it seems that no sharp line of demarcation can be drawn between the different varieties of pitchstone here described, for although differences may, and doubtless do exist, still our power of appreciating those differences is limited or nil, so long as devitrification has not supervened, and so indicated, by inci- pient crystalline development, the true petrological affinities of the glass. The crystals which occur porphyritically in the different varieties afford us a very imperfect clue to these relations, simply because they represent for the most part Tachylyte. ' / , 199 -" \" A identical mineral species. The characters of A\JQ rocks ivdth which they are associated, and of which they refe/s,ent the /) vitreous conditions, give us however a more /exact ncr&cjn, pf the places which they ought to occupy in our ossification) of them. '/,- v fc Indeed it may not be indiscreet to believe that in rnar^, if not in all instances, the crystalline equivalents of these . vitreous rocks do but represent an advanced phase of devitrification, and that all trachytes were once trachytic pitch stones, and that all felstones were once felsitic pitch- stones, either at the time of or prior to their eruption. Tachylyte. The rocks included under this name must be regarded as vitreous conditions of basic rocks, especially of basalts. These glassy basalts are termed basaltvitro- phyres by Rosenbusch, and he subdivides them into tachy- lytes, or those which are soluble in acids, and hyalomelanes or those which are insoluble in acids. The tachylytes occur mostly as salbands, or thin crusts at the sides or margins of basalt dykes, but the essentially vitreous basic lavas, such as those of Kilauea in the Sandwich Islands, which form actual flows of considerable magnitude, constitute, as pointed out by Cohen, independent rock masses. These Kilauea lavas are, as a rule, rich in olivine, and are for the most part highly vesicular. Tachylyte also occurs lining or filling vesicles or cavities in basalt. The tachylytes are black or brown glasses and somewhat resemble obsidian, but when struck with the hammer they do not usually afford large flakes and extensive conchoidal fractures like obsidian, but generally break up into small irregular fragments and splin- ters. They also differ from obsidian in point of fusibility and chemical composition, tachylyte only containing from 5 to 55 P er cent, of silica. The remaining constituents are alumina, protoxide of iron, sometimes peroxide of iron, lime, magnesia, potash, soda, and usually about 6 or 7 per cent, of water. From the few analyses of tachylyte which have hitherto been made, the composition appears to be rather 200 Descriptive Petrology. FIG. 85. variable, but the percentage of silica is pretty constant. Before the blowpipe tachylyte fuses quite easily * with intu- mescence to a dark slaggy glass. It is decomposed with gelatinisation in hydrochloric acid. Its hardness is about 6'5, and its specific gravity about 2*5. Under the microscope tachylytes vary greatly in appear- ance, some being comparatively translucent, at all events in places, while others are almost wholly opaque, even in ex- cessively thin sections. This opacity seems in many cases to be due to the pre- sence of fine opaque, black, dusty matter which pervades a great portion of the glass, but is more densely segregated around certain spots, these spots being frequently small cry- stals of magnetite. Fig. 85 represents part of a section of tachylyte from Slie- venalargy,Co. Down, Ireland (magnified 300 diameters), in which these dust ac- cumulations are well shown. Where these dust segregations are less dense, the sections often appear of a brownish colour, and this is possibly due to peroxidation, the magnetite probably being converted into martite, or some closely allied mineral. In the clearer portions of the same section from which fig. 85 was drawn, numerous opaque patches of irregular form are visible (fig. 86, magnified 55 diameters) ; their boundaries are sharply defined, while they Whence the name from TO.XVS, quickly, and Aurbs, fusible. TacJiylyte. 201 are bordered by the clear absorption spaces from which the dust, so finely dis- seminated through the other portions of the glass, appears to have been abstrac- ted. At A A in the same drawing la- cunse of greenish co- loured ferruginous glass are shown, which mineralogi- cally are possibly allied to glauconite. At B B B portions of fine rod -like bodies are delineated, and these also contain, or are fringed with, green matter. In some places, as in the central part of the figure, a. tendency to perlitic structure may be detected. 1 A tachylyte from Bobenhausen is described by Vogelsang 2 as a brownish- red glass, containing dark crystallites which are developed in fern-like forms, but which in most parts of his drawing appear to be slightly-fringed spherical or cruciform bodies, or irregular dark patches. He did not regard this substance as magnetite, arguing that, when magnified 800 to 1,000 diameters, the granules constituting the thinner portions and edges of these crystallites do not appear opaque like mag- netite, but somewhat translucent and of a brownish colour. The glass around them appears much clearer and of a yellow colour. Vogelsang, Zirkel, and Mohl all concur in regard- ing these bodies as similar to those which often occur in 1 Fuller details respecting this rock will be found in a paper by the author ' On Microscopic Structures in Tachylyte from Slievenalargy, Co. Down,' Journ. Royal Geol. Soc. Ireland, vol. iv. part 4, new series, p. 227. Die Krystalliten. Bonn, 1875, P- I1[I ' 2O2 Descriptive Petrology. blast-furnace slags. Both Vogelsang and Zirkel consider these crystallites to consist of ferruginous glass, while Mohl regards them as magnetite. The microscopic characters of several tachylytes are re- corded by Zirkel in his ' Untersuchungen iiber die Basalt- gesteine,' Bonn, 1870, p. 182. In one from Meinzereichen in Hesse, he cites the occurrence of fern-like developments around a transverse section of an apatite crystal. In a sec- tion of tachylyte from Some in the Isle of Mull, the whole of the section, although an excessively thin one, appears opaque or very feebly translucent on the edges, while em- bedded in it are minute transparent crystals which frequently show hexagonal sections, and which are probably apatite. Mohl, in his 'Basalte und Phonolithe Sachsens,' Dresden, 1873, figures and describes the occurrence of patches of tachy- lytic glass in sections of nepheline-basalt, which contain fern- like trichites, and occasionally small crystals of nepheline. Rosenbusch describes a tachylyte from Czertochin in Bohemia, as a greenish-grey glass full of strings of minute steam pores which occasionally anastomose. He states that this rock is very quickly and completely dissolved in hydro- chloric acid, without the application of heat. 1 CHAPTER XII. ERUPTIVE ROCKS. CLASS II. CRYSTALLINE ROCKS. GRANITE GROUP. Granite. The granites (from the Latin granum, a grain) are essentially, as their name implies, crystalline-granular rocks which may be regarded as consisting typically of 1 Mik. Physiog. Min. p. 139, fig. 15. Stuttgart, 1873. Granite. 203 orthoclase, quartz, and mica. There is, however, considerable variation in the mineral constitution of granites. In some the felspathic component is not merely orthoclase ; plagio- clastic felspars such as albite and oligoclase being frequently present, while the mica, which is usually muscovite or biotite, may at times be represented by lepidolite, lepidomelane, or other micas. Other minerals, which are not regarded as essential constituents of granite, are often present. When widely disseminated, these accessory constituents play only a very subordinate part ; but, in certain limited areas, they are often developed in sufficient quantity to impart a distinctive character to the rock. Thus, for example, schorl frequently occurs in considerable quantity in granitic masses at or near their contact with other rocks, sometimes to such an extent that the term schorlaceous granite is applied to the rock. Apatite and magnetite are also minerals of common occur- rence in granites. Kpidote and garnets are less common, but are often met with. Pyrites is common in many granites, and it seems doubtful whether, in some cases, it should be regarded as a mineral of secondary origin. Talc, beryl, iolite, andalusite, topaz, cassiterite, and hematite are oc- casionally met with in granitic rocks. Hornblende is of common occurrence, and, when tolerably plentiful, the rock is then termed hornblendic or syenitic granite. When quartz is absent, or only poorly represented, and the mica is replaced by hornblende, the rock is called syenite. 1 Chlorite, epidote, pinite, and several other products of the alteration of other minerals, are not of unfrequent occurrence in these rocks. Kaolin very commonly occurs in granites, and results from the decomposition of the felspars. Graphite is also met with at times. Granite rocks vary very considerably in texture and in structural characters. 1 It should, however, be remarked that the term syenite, as first employed by Pliny, and as used in most geological works, until within the last few years, implied hornblendic granite, such as that which comes from the quarries of Syene in Egypt. 204 Descriptive Petrology. The granites, as a rule, are either coarsely or finely crystalline-granular in texture, and, when very fine-grained, and the mica is only poorly represented, or totally absent, pass into felstones of variable texture. 1 The granitic rocks are frequently porphyritic. Crystals of orthoclase several inches in length being of common occurrence in some, while, in others, the mica (usually muscovite) forms, by its con- spicuous development, the dominant mineral. When mica is scarce, and the rock assumes a felsitic character, it is common to find eicher orthoclase or quartz porphyritically developed. In the former case the rock would be styled a felspar porphyry, in the latter a quartz-porphyry, or elvan. 2 In true granites no microrcrystalline or amorphous paste is visible between the crystals and crystalline grains of which the rock is composed. Of these component minerals orthoclase is generally the most plentiful; next follows quartz, and then mica, in the usual, but not the invariable, order of quantitative importance. In their order of solidi- fication, or crystallisation, quartz appears to come last, and,- although it sometimes occurs in hexagonal pyramids or in combinations of the pyramid and prism, still its development, as a rule, seems to have been imperfect, and to have resulted mainly in irregularly shaped, angular, crystalline grains. The orthoclase in granites varies in colour. In some it is red, often of a flesh-red or pink tint, in others white, grey, or yellowish. It very commonly occurs in Carlsbad twins. The crystals are often several inches in length, as in some of the Dartmoor granites. When granites are weathered, the felspar crystals are converted into kaolin and the rock in course of time crumbles away. The kaolin or china-clay which remains 1 Felsitic matter, which constitutes the chief bulk of felstones, is a very finely crystalline-granular, or micro-crystalline, or crypto-crystal- line, admixture of orthoclase and quartz. 2 Elvan is a Cornish name, and is commonly applied by the Cornish miners to most of the dykes which occur in that county, irrespective of their mineral constitution. The term has, however, of late years been restricted to quartz-porphyries. Granite. 205 after the disintegration of granite frequently contains a large proportion of quartz grains, and this decomposed rock is known as china-stone. The crystals of orthoclase are not always well developed in granites ; they sometimes have very irregular contours, and occasionally their angles are rounded. Under the microscope they frequently present a more or less turbid appearance, and this greatly increases in proportion to the stage of decomposition at which the rock has arrived, until they ultimately become completely kaolinised and opaque. They occasionally, but rarely, contain fluid lacunae. Plagioclastic felspars, either albite or oligoclase, are of frequent occurrence in granites. They usually occur in smaller crystals than the orthoclase. Under the micro- scope they exhibit, when fresh, the characteristic twinning of plagioclase, but, as decomposition advances, a granulated structure also supervenes, which obliterates this distinctive structure, and renders it impossible to determine whether they were originally monoclinic or triclinic felspars. The quartz, as already stated, sometimes occurs in well- developed crystals, and sometimes in angular, crystalline grains. The former often exhibit a polysynthetic structure when examined in polarised light. Under the microscope, in thin sections, they appear quite glassy and clear, and are seen to contain numerous fluid lacunae, which are often so plentiful, as to impart an almost turbid appearance to the crystal or granule. The contained fluid is generally water or aqueous solutions of chlorides and sulphates of sodium, potassium, and calcium. Apatite crystals are also frequently visible in the quartz of granite. The micas in thin sections of granite appear either in well denned crystals, which, when the section is taken parallel to their basal planes, appear as six-sided tables, or in scales of irregular form. The potash micas appear clear and nearly colourless, while the magnesian micas are dark reddish-brown or black, and the latter show strong dichroism, when the planes of section do not coincide with the basal planes of the crystals. 206 Descriptive Petrology. When schorl occurs in granites it may usually be recog- nised by the strong bluish tint which it here and there shows, when examined under the microscope by ordinary transmit- ted light, and also by the approximately triangular transverse sections which the crystals frequently exhibit. When in thin sections of granite, magnetite and pyrites are present, they both appear opaque, and, to distinguish between them, it is necessary to examine them by reflected light, when their differences of colour and lustre become apparent. With regard to the origin of granite, there has been con- siderable discussion, in which most antagonistic opinions have been brought forward ; theories of its igneous, aqueous, and metamorphic origin having all been strongly advocated. Its eruptive character is inferred from the granite veins, which in certain localities traverse older rocks in a most irregular manner, while the dykes of quartz-porphyry which often emanate from, and can be traced to underlying granitic masses, and are, indeed, mere differentiations of granite, afford additional proofs of its eruptive character. According to Hermann Credner, however, the mineral matter of the granitic veins in Saxony is not derived from deep sources, but from the partial decomposition of the adjacent rocks by the infiltration of water, and he observes that the mineral characters of the veins are influenced by those of the rocks which they traverse. * With regard to the larger bosses and the huge granitic masses, from which such dykes and veins are given off, we can scarcely deny to the parent masses the origin which must be attributed to their offshoots, but, in the, absence of such veins and dykes, it is easy to understand how, with considerable show of reason, a metamorphic origin may be assigned to those masses which, though once deep-seated, are now exposed by the denudation of enormous thicknesses of once overlying rock, and the question rather naturally 1 ' Die Granitischen Gange des sachsischen Granulitgebirges,' Her- mann Credner, Zeitsch.d. deutsch. geol. Ces., Jahrg. 1875, p. 218. Granite. 207 arises whether they have not resulted from the metamor- phism of sedimentary deposits, once so far beneath the earth's surface that they lay within a zone of comparatively high temperature. Admitting this, it seems that we are ad- mitting no more than the conditions, or phases of the condi- tions, under which all eruptive rocks have been formed, and which are, therefore, just as fully entitled to the appellation of metamorphic rocks. That the passage sometimes observed from granite into gneiss is a proof that granite is the extreme phase of the metamorphism of sedimentary rocks does not always appear to be conclusive, since instances are known in which foliation is not indicative of bedding, and a few cases are recorded in which gneiss actually occurs in veins. In the present conflicting state of opinion upon this subject it behoves examination candidates to accept and cite the different opinions commonly held and set forth in the various manuals of geology. The student may afterwards judge of their respective merits from his own observations. One of the arguments against the igneous origin of granite is that in granite the quartz has a specific gravity of 2-6, identical with that of silica derived from aqueous solu- tion, while the specific gravity of fused silica is only 2*2. This observation, in conjunction with many others, appears to have influenced to some extent the deductions of Professor Haughton, in his annual address to the Geological Society of Dublin in 1862. After giving a table of the relative specific gravities of natural and artificially fused rocks, he concludes in the following words : ' It appears to me that the column of differences ' (in the specific gravities of natural and artificially-fused rocks) 'greatly strengthens the argument of those chemists and geologists who believe that water played a much more im- portant part in the formation of granites and traps than it has done in the production of trachytes, basalts, and lavas, and that they owe their relatively high specific gravity to its agency.' 'The only manner in which it seems possible to reconcile 2o8 Descriptive Petrology. the opposite theories of the origin of granite, derived from physical and chemical arguments, is to admit for granite what may be called hydro-metamorphic origin, which is the converse of what is commonly called metamorphic action, but which might more properly be designated pyro- metamor- phic action. The metamorphism of rocks might thus be assumed to be twofold. Hydro-metamorphism, by which rocks, originally fused, and when in liquid fusion, poured into veins and dykes in pre-existing rocks, are subsequently altered in specific gravity and arrangement of minerals, by the action of water acting at temperatures which, though still high, would be quite inadequate to fuse the rock ; and pyro-metamorphism, by which rocks originally stratified by mechanical deposition from water, come to be subsequently acted on by heat, and so transformed into what are com- monly called the metamorphic rocks.' ' Granite, it appears to me, although generally a hydro- metamorphic rock, may occasionally be the result of pyro- metamorphic action ; and such appears to have been its origin in Donegal, in Norway, and, perhaps, in the chain of the Swiss Alps.' l This may be a very just opinion, especially if Professor Haughton does not imply, in his pyro-metamorphic action, the total exclusion of water from any participation in the changes effected. The two conditions of metamorphism which he indicates, most likely represent; in the hydro- metamorphism, the presence of a large proportion of water and a moderately high temperature ; in the pyro- metamor- phism, a comparatively small portion of water and a much higher temperature. Such at least is a probable construction to put upon these conclusions ; but, so far as metamorphism in its vulgar acceptation is concerned, there seems no reason, apart from the distinctions just given, for regarding granite 1 ' On the Origin of Granite, ' an address delivered before the Geo- logical Society of Dublin, by the Rev. Samuel Haughton, F.R.S. Dublin, 1862. Granite. 209 as a metamorphic rock any more than basalt or trachyte, which have, in a certain sense, resulted from the extreme alteration of other rocks. The crystalline schists, gneiss, &c., are but phases of the conversion of sediments into true eruptive rocks ; and, if the degree of alteration be put out of the question, the crystalline schists, the plutonic rocks, and the volcanic rocks, all seem equally eligible for the term metamorphic. In questions of metamorphism, it appears that the nature of the change is the first thing to consider ; its cause, the next; its degree, the last. 1 The different varieties of granite and of granitoid rocks may be summed up under the following heads : Porphyritic granite, in which the felspar crystals are large and well developed, being frequently several inches in diameter, as in those of Cornwall, Dartmoor, Shap, &c. Various grades of texture occur between these granites and those which are termed fine-grained. When of the latter character, they pass into micaceous felstones. Felstone (eurite, 2 halleflinta, petrosilex), consists of felsitic matter, (viz., an intimate granular-crystalline, micro-crystal- line, or crypto-crystalline, admixture of orthoclase and quartz, in which crystalline granules of plagioclastic felspars not unfrequently occur.) In this felsitic base, which, typi- cally," constitutes the matrix of all felstones, felspar crystals, commonly orthoclase, are often developed ; and, like those in the porphyritic granites, are frequently twinned on the Carlsbad type. Such rocks are termed felspar porphyries . 1 The less the unqualified term metamorphism is used, the better ; since it merely implies change, without specifying the nature or extent of the change or the conditions under which the change took place. 2 The terms Felstone and Eurite are frequently used synonymously ; but eurite is stated by some authors to be more easily fusible than orthoclase, while the eurite snrsilicee of Cordier is more difficultly fusible. The name Eurite is due to d'Aubuisson. Kinahan's definition of eurite, as a basic felstone (Handy-Book of Rock Names, p. 48), might lead the unwary to regard it as a rock containing less than 60 per cent. of silica, but he is probably, to some extent, right in keeping the dis- tinction between eurite and felstone, although at times rocks of an intermediate character are met with. 2 1 o Descriptive Petrology. If, in such a felsitic matrix quartz occurs porphyritically either in crystals, but more usually in roundish blebs, the rock is termed quartz-porphyry ; but it is common to find porphyritic orthoclase crystals also developed in quartz- porphyries. It seems, however, probable that the ground - mass of true quartz-porphyries should, in many cases, rather be regarded as a very fine-grained or micro -crystalline granite. Rocks of this class are called elvans by the Cornish miners, and, indeed, in that district, the term elvan is very loosely applied. As, however, the dyke-forming rocks of Cornwall are mostly offshoots from the granitic masses of that district, the term elvan has for the most part been applied to more or less fine-grained or porphyritic granitoid rocks, and it is now, as a rule, regarded as a synonym for quartz-porphyry, or, as some authors term it, quartz- felsite. Granitite is a term given to those varieties of granite which contain a certain amount of plagioclase (oligoclase). The orthoclase, in the rock to which this name has been applied, is flesh -red, and this mineral and quartz are the two principal constituents. The mica is a blackish- green magnesian mica, but it is usually present only in small quantity. Since plagioclastic felspars exist, though in a subordinate capacity, in many granites, it seems that no line of demarca tion can be drawn between them and the granitites. Cor dier tie-granite is a variety occurring in certain localities in Norway, Greenland, and Bavaria. It is characterised by containing cordierite or iolite ; this mineral partially, and sometimes wholly, replacing the mica. A greenish oligoclase is often present in the rock. Luxullianite is composed of schorl, flesh-coloured ortho- clase and quartz. The schorl, which is black, or greenish- black, is distributed in irregular nests or patches, and contrasts strongly in colour with the other constituents of the rock. Boulders of this stone occur in the neigh- bourhood of Luxullian, in Cornwall ; and the late Duke Granulite. Greisen. Gneiss. 2 1 1 of Wellington's sarcophagus was made from one of them. The rock has not, however, been met with in situ. In a paper by Professor Bonney, published in No. 7 of the ' Mineralogical Magazine/ 1877, two varieties of tourma- line are stated to occur in this rock, and some evidence is adduced to show that this mineral is a product of alteration. Aplite or Haplite (from aTrXooc, simple), also termed semi"-granite (Halb-granif] or granitell, is a rock of limited occurrence, consisting of a crystalline-granular admixture of felspar and quartz. The so-called graphic-granite or pegmatite is a structural variety of this rock, in which the quartz is developed in such a manner that it roughly resembles Hebrew characters, a polished surface of the rock appearing closely inscribed, whence the name 'graphic.' Granulitt (Weiss-stein or leptinite) is also composed of felspar and quartz, the felspar being orthoclase. In structure it is more or less finely crystalline-granular, and frequently has a foliated or schistose character. It generally con- tains numerous garnets, which, in thin section under the microscope, appear as little irregular, roundish, singly- refracting grains, like drops of gum. Mr. John Arthur Phillips has observed double refraction in the garnets of some granulites. This rock, in its schistose structure and mode of occurrence, seems to bear much the same relation to felstone that gneiss bears to granite, and it may therefore be classed with the metamorphic rocks. It often contains schorl and hornblende microliths, and occasionally sphene. The variety called trap-granulite contains plagioclastic felspars, and is somewhat poorer in silica. Greisen (Zwitter, Stockwerks-porphyr} is a granular- crystalline rock, consisting of quartz and mica, the latter usually lithia-mica. Quartz is, however, the predominating constituent. When orthoclase occurs in it the rock passes into granite. Tinstone (cassiterite) is very commonly met with in greisen, either in small strings and veins, or in little p 2 2 1 2 Descriptive Petrology. crystals or granules. It is a rock of common occurrence in Saxony and Cornwall. 1 Gneiss. This term, in its proper sense, signifies foliated granite ; but foliated rocks, consisting to a very great extent of hornblende and quartz, have also been styled gneiss, although they should rather be termed schistose amphibolites. Indeed, the name seems to have been somewhat loosely applied to foliated crystalline rocks of variable mineral constitution. True gneiss differs in no way from granite, except structurally. A foliated structure is its essential peculiarity. It is sometimes interbedded with other rocks, and frequently exhibits stratification, which is often but not invariably coincident with the foliation. Darwin has shown that, in the gneiss of the Andes, the planes of foliation coincide with planes of cleavage. Sir R. I. Murchison pointed out that the foliation in some of the Scotch Silurian rocks cor- responds with the planes of bedding ; and similar observa- tions have been made in Anglesey, by Professor Henslow ; while, according to Professor Ramsay, and a host of other observers, the coincidence of foliation with bedding is of extremely common occurrence. To crystalline rocks, which exhibit this structure, the adjective gneissic is applied, a good practice, when the rock deviates in mineral composition from a true granite. Gneiss has been split up into numerous varieties, which, m the main, are identical in mineral constitution with the corresponding varieties of granite. Thus we have, in addi- tion to ordinary gneiss, oligoclase gneiss, a foliated rock corresponding with oligoclase granite, dichroite gneiss, adularia gneiss, garnet gneiss, syenitic gneiss, &c., &c. Protogine is a gneiss in which, in addition to the ordinary constituents of granite, a greenish, pearly, or silvery talcose 1 A paper, 'On some of the Stock works of Cornwall,' by Dr. C. Le Neve Foster, was read before the Geological Society, London, January 9, 1878. Deviations from the Granitic Type. 2 1 3 mineral is present. The rock, when not foliated, is termed protogine granite. Cornubianite (proteolite) is a compact granular- scaly con- dition of gneiss, which is met with, at times, at the contact of granites with slates. The accessory mineral constituents which occur in gneiss are very numerous, and are similar to those which occur as accessories in granites. Those rocks which in mineral constitution and in struc- ture more or less resemble granites, are spoken of as granitoid rocks. They range from the coarsely crystalline to the micro- crystalline or crystalline-granular varieties. Many of them have a felsitic matrix, their distinctive characters being due to the larger, or porphyritic, develop- ment of one or more of their mineral constituents. The felsitic matrix of these rocks consists of an intimate micro-crystalline, or granular admixture of felspar (mostly orthoclase) and quartz. The following table might be greatly extended so as to embrace, all the chief rocks, both basic and highly silicated. Thus, for instance, if plagioclastic felspar were substituted for orthoclase, then syenite would become diorite. The student, by constructing such tables, may thus, as his knowledge increases, see how far the classification of rocks is useful, and how they gradually pass from one type to another. Gneiss, granulite, and several other rocks, have been described in this place because they are closely related to granite in mineral constitution ; but they should, perhaps, in most cases, rather be classed with the metamorphosed sedimentary rocks. TABULAR VIEW OF SOME OF THE PRINCIPAL DEVIATIONS FROM THE GRANITIC TYPE. In this table the letter F indicates felspar of one or more 2 1 4 Descriptive Petrology. species, but mostly orthoclase. M represents mica of one or more species. Q = quartz. H = hornblende. Syenite . F H cryst-granular. Quartz- syenite . F Q H cryst-gran. Syenitic granite . F M Q H ) cryst-gran. Syenitic gneiss . F M Q H j foliated GRANITE . F M Q \ cryst-gran. GNEISS . . F M Q [foliated. Haplite . F Q | cryst-gran. Granulite F Q j schistose. Quartz-porphyry F O \ Felsitic matrix O porphyritic. Felspar-porphyry F Q F FeMone . F Q, Greisen . M Q cryst-gran. Ouartzite . . Q gran, compact. FELSTONE GROUP. Felstone (eurite, petrosilex, halleflinte, felsite). Fel- stone is a more or less compact rock, those varieties termed halleflinte and hornstone having a peculiarly flinty aspect, while, in other cases, the rock is either finely crystalline- granular or granular, sometimes porphyritic, often micro- porphyritic. In colour felstone varies very greatly brick- red, brown, grey, yellowish, and greyish-white tints being the most common. Many varieties have a more or less conchoidal fracture, and all of them, before the blowpipe, are fusible on the edges of splinters to a white or speckled enamel. The eurites proper are more easily fusible than the felstones or eurites sursilicees of Cordier. In the com- pact and in the non-porphyritic examples no definite mine- rals can be detected with the naked eye or with a lens, and the same may be said of the matrix in which porphyritic crystals occur. Sometimes, but not commonly, they present an imperfect schistose structure, as in the varieties termed felsite schist. They differ considerably in chemical compo- sition, the amount of silica which they contain varying from Deviations front Granite as a Type. PLATE 111. 215 2 1 6 Descriptive Petrology. about 70 to 80 per cent., and they frequently have about 5 per cent, of the alkalies. Under the microscope they are also seen to vaiy greatly in character. Sometimes they show a micro-crystalline structure, in which, by polarised light, the section breaks up into a many-coloured mosaic, and the individual granules may be distinguished and identified, some of them as felspars, others as quartz, in others there is a somewhat similar but less defined structure, crypto-crystalline, in which individual minerals cannot be recognised. In some rare cases a considerable amount of true vitreous matter may be detected, lying between the micro-crystalline or granular component particles, or constituting the entire paste. More frequently the rock is wholly micro-crystalline or micro- felsitic. In the latter case between crossed Nicols the sub- stance behaves as an amorphous mass. In this case the structure may be granular, fibrous, or microlitic, the granular and fibrous structure seldom presenting any definite character or individualisation of the constituent granules and fibres. Sections of such a rock dc not however always present total obscurity between crossed Nicols, but transmit a feeble light, in an irregular and fickle manner, as regards its distribution. Sections of felstone occasionally present a radial-fibrous structure under the microscope ; these constitute the varieties known as spherulitic felsite. Fluxion-structure is sometimes to be observed in felstones. It is probable, however, that many of the felsitic rocks which show this are more or less closely allied to the rhyolites. It is quite possible that in many cases the micro-crystalline or micro-granular struc- ture of felstones simply represents the devitrification of an originally glassy magma, but, as remarked by A. von Lasaulx, the felsite pitchstones frequency fail to present the microscopic structure so characteristic of felstones. It is nevertheless far from uncommon to find small patches in sections of pitchstones and other vitreous rocks, in which devitrification has resulted in the production of structure, strikingly analogous, if not identical, witli that of felstones. Syenite, 217 Hornblende, micas, sometimes potash, sometimes magnesian, magnetite, titaniferous iron, &c., are met with in felstones. Felspars are, also, often porphyritically developed, and the rock then becomes porphyritic fel stone or felspar porphyry. Quartz also occurs porphyritically at times, either in crystals, or in roundish grains ; the rock then becoming a quartz- porphyry or elvanite. Indeed, passages may occur from felstone into granite, syenite, and various rhyolites, Felstone is generally more or less porphyritic, and occurs in dykes, veins and inter- bedded sheets. SYENITE GROUP. Syenite, in the acceptation of the term, as first em- ployed by Werner, 1 is a crystalline-granular rock, con- taining from 55 to 60 per cent, of silica, and consisting typically of orthoclase and hornblende. In mineral consti- tution, therefore, it approximates to some of the trachytes. Sometimes the felspar is microcline, and plagioclastic felspars are nearly always present. Sometimes augite or mica take the place of the hornblende, and occasionally the rock con- tains more or less sphene and quartz. The syenites may therefore be divided into three groups, viz., hornblende syenite, augite syenite, and mica syenite. When quartz is present in any notable quantity the reck passes over to quartz syenite, and thence, when mica occurs, into syenitic granite. Hornblende Syenite. Orthoclase and hornblende are the chief constituents. Triclinic felspar is usually present in 1 For many years it has been a common practice to apply the name syenite to syenitic or hornblendic granite. At times there has been considerable difference of opinion about the application of the name, which was first used by Pliny (Syenites) for the rock quarried at Syene in Egypt. The stone occurring at that locality is hornblendic granite. Hornblendic granite seems, therefore, to have a decided priority of claim to the name Syenite, but petrologists have found it convenient to restrict its application to quartzless rocks, such as those here described. 2 1 8 Descriptive Petrology. variable quantity ; it may generally be referred to oligoclase, but the amount is as a rule comparatively small. The colour of the rock mostly depends upon the colour of the orthoclase, which varies considerably, red, brown, and white being the prevailing colours. The orthoclase crystals are frequently twinned on the Carlsbad type. The hornblende is usually greenish-black, and the crystals are generally, but not invariably, short ; long bladed or acicular crystals sometimes occurring. The mica is a dark magnesian mica, commonly biotite. Epidote, magnetite, sphene, and pyrites frequently occur as accessories in this rock. In structure the syenites as a rule greatly resemble granites, and they also occur in large eruptive masses, bosses, or veins. The gneissic syenite sometimes occurs in considerable beds, especially in the Laurentian series of Canada. The foliated rock of Cape Wrath in Suther- landshire, Scotland, is rather amphibolite schist than gneiss, and some of the so-called gneiss of the Hebrides may also be referred to hornblendic schists. Augite Syenite is composed of felspars, which, as a rule, are mostly orthoclastic, but the plagioclastic ones occa- sionally, though rarely, predominate. Augite is frequently plentiful, and sometimes a little hornblende occurs, which, as pointed out by A. von Lasaulx, is generally of a uralitic character, implying subsequent alteration of some of the pyroxenic constituents. Biotite, apatite, magnetite, and sphene are also of common occurrence as accessories in the composition of augite syenite. According to V. Lasaulx sphene is less plentiful in those varieties in which orthoclase is the predominant felspar. Analyses show that the augite syenites contain one or two per cent, less of silica than the hornblende syenites. Mica Syenite is by no means a common rock, Calabria being almost the only district in which it is met with to any considerable extent. It occurs mostly in the form of veins or dykes. The rock consists of orthoclase, sometimes Minette. 219 more or less plagioclastic felspar, biaxial magnesian mica, hornblende, occasionally some augite, which is often altered into pseudomorphs of chlorite or delessite, as in the minette of Seifersdorf in Saxony, while apatite, calcite, magnetite, and pyrites are also of common occurrence in these rocks ; but, as a rule, sphene is never met with in mica syenite. The calcite and pyrites are products of secondary origin. According to Rosenbusch, 1 Zirkel, and other petrologists, mica syenite and minette are intimately related if not identical. The former author also points out that some minettes are to be referred to the augite syenites. Minette, The matrix or paste of minette appears, under the microscope, as granular or granular-crystalline matter, in which microliths frequently occur ; the latter according to A. von Lasaulx, who classes minette with the felstones, consist of felspar and mica, and the preponderance of evi- dence shows that quartz is an exceptional constituent of the rock. Under these circumstances the matrix can hardly be designated felsitic, and upon this ground hinges the question whether minette should be classed with the sye- nites or with the porphyritic felstones and granites. If the absence of free silica in the matrix be proved, it is evident that the affinities of minette are closer to syenite than to granite, but minette occurs in veins and dykes in both of these rocks, and dykes of it are also met with in sedimentary deposits of Silurian and Devonian age. It seems in many cases that micaceous felstones approximate rather closely to minette. Kengott appears to entertain some such idea in his ' Elemente der Petrographie.' V. Cotta's definition of minette is ' a felsitic matrix containing much mica and sometimes distinct crystals of orthoclase or hornblende.' The true difficulty seems to lie in the imperfect knowledge which we as yet possess of what a felsitic matrix really is. If quartz be excluded from such a matrix, and it is generally stated that minette seldom contains that mineral, then 1 Mik. Phys. d. Mass. Gcst., p. 122. 22O Descriptive Petrology. minette is a mica-syenite with a micro-granular or micro- crystalline matrix ; if, on the other hand, quartz be present, minette may be closely allied to the felstones, micaceous felstone forming a transitional link. It is possible that both conditions occur, and, if so, it may become necessary to clas- sify the minettes. In seme of the mica-traps of the English Lake District the author has found both orthoclastic and plagioclastic felspars which, in addition to the magnesian mica, occur in well-marked crystals. In such cases the rock appears to hold a position intermediate between minette and kersantite. If minette represent a condition of the syenites which are rich in orthoclase, then kersantite is allied to those which are rich in plagioclastic felspars, and, in such instances, it may be questioned whether the affinities of kersantite are not more in the direction of diorite, especially of the mica- ceous varieties of that rock. Speaking approximately, minette is a rock which contains magnesian-mica and sometimes hornblende crystals in a micro-granular or micro-crystalline matrix in which felspar crystals, mostly orthoclase, are porphyritically developed, while kersantite is a somewhat similar rock in which the felspathic components are mainly plagioclastic. Both minette and kersantite occur, as a rule, in the form of dykes. B. von Cotta in his remarks on syenite says : ' Properly speaking there are no varieties of composition to adduce, unless we consider as such those transitions into granite and diorite which are occasioned by the occurrence of mica, quartz, and oligoclase/ l This seems a very just generalisation, implying a sharp definition of syenite. If the term be allowed the wide scope which some petrologists now accord to it, we might as well term basalt a mica-diorite. That instances may be found of rocks, which, in mineral constitution, form connecting links 1 Rocks Classified and Described^, by B. von Cotta. English transla- tion by P. H. Lawrence, 1866, p. 179, Trachyte. 221 between very many, if not all, of the eruptive rocks there is no doubt, but sharp, or moderate!} sharp, definitions con- stitute the basis of all classification, and, if these be aban- doned, petrological nomenclature, as it exists at present, becomes almost worthless. A good account of the minettes, kersantite, and kersan- ton, by Delesse, will be found in vol. x. of the ' Annales des Mines,' for 1857. TRACHYTE GEOUP. Trachyte. (The rocks which have been included under this name are exceedingly numerousytind the term in its present acceptation has still a very wide range. The usual constituents of trachyte are sanidine, oligoclase, hornblende, sometimes augite, magnesian mica, magnetite, titaniferous- iron, tridymite, and at times some other minerals, such as sodalite, hauyne, nosean, sphene, mellilite, leucite, and oli- vine, which may, for the most part, be regarded as accessories. Plagioclastic felspar is generally associated to a greater or less extent with the sanidine in these rocks ; hence Rosen- busch, 1 Zirkel, 2 and Von Lasaulx 3 consider that their division into sanidine trachytes and sanidine- oligoclase trachytes is of little or no account. The first author suggests that they may eventually be classified by determinations of the presence or absence of tridymite, and he thinks it probable that, by noting the relative occurrence of hornblende, augite, and magnesian mica, the trachytes may be arranged in a series homotaxial with that into which the syenites have been divided. The sodalite-, hauyne-, and nosean-bearing trachytes appear to some extent to be analogous to the phonolites. (The more highly-silicated trachytes are comprised in the group of rhyolites, and, in part at least, constitute the rhyo- lites proper, whose vitreous condition is met with in obsidian^ &c., as already pointed out. The name trachyte is derived 1 Mik. Phys. d. Massigen Gesteine, 1877, p. 200. ? Mik. Besch. d. Min. u. Gest. 1873, p. 382. 1 Elcm. d. Petrographie, 1875, p. 278. 222 Descriptive Petrology. from Tpct^vg (rough), in allusion to the rough, scraping sen- sation which the surfaces of these rocks usually convey when rubbed with the fingers. (Geologically, the trachytes have been divided into trachytes and trachytic lavas, but the characters, even mi- croscopic, of the one, have not been found to differ from those of the other.) The trachytes proper are mostly of tertiary or post-tertiary age. Some rhyolites are coeval with them, while others have a great geological antiquity ; but, as yet, comparatively little is known of these old rhyolites. The trachytes may be conveniently classified in the following manner : Rhyolites proper ( i. Quartz-trachytes ) allied to perlite and or liparites 1 ( ii. Sanidine-trachytes ) obsidian. "I allied to syenite and the quartzless por- Trachytes proper r 11 ' Quar ' zl f s - trachytes 1- phyries, such as P 1- anddormtes fc ,_ stone, &c. Analyses of these rocks show the following approximate variations of the amount of silica which they respectively contain. Quartz-trachyte or quartz-rhyolite 75 to 77 per cent, silica. Sanidine-trachyte or sanidine-rhyolite 74 to 78 Quartzless trachyte or trachyte proper 62 to 64 From this it will be seen that, although in some of the sanidine trachytes little or no quartz can be recognised, even microscopically, yet they contain a considerably higher per- centage of silica than the trachytes proper. It will be well to begin, in each case, with a brief state- ment of the microscopic characters of the ground-mass, base, or matrix of each of these three types, since the megascopic appearance of these ground-masses affords little or no insight as to their mineral constitution or structural peculiarities ; 1 The term lithoidite has also been applied to these rocks. v Rhyolite Proper. , v / /, 223 while, without a tolerably precise knowledge l of tjiese cha^ac-/ ters, it is often difficult to discriminate correctly beiw^en the /^ different types. -// r y. '';- QUARTZ-TRACHYTE (Quartz-rhyolite, Liparite):/ > > v The matrix is generally micro-aphanitic, and contains^/ moderate-sized grains of quartz. The red varieties contain more or less peroxide of iron, in a finely-divided state, or in thin films. A micro-crystalline-granular or micro-granitic condition is less common in the matrices of quartz-trachytes than in those of quartz-porphyries, but nevertheless it is often present. The matrix of quartz-trachytes appears, to the naked eye, as a compact, or very finely -granular, substance, often rough and porous ; it sometimes resembles hornstone, or porcellanite, while, at others, it has a dull, earthy or kao- linised appearance. It varies considerably in colour, brick- red, reddish-grey and yellowish- and brownish-white being some of the most common. The principal bodies porphy- ritically developed in this matrix are crystals and crystalline granules of sanidine and quartz, while plagioclastic felspars, hornblende, and magnesian mica (biotite) are often also well developed. The sanidine crystals in the quartz-trachytes very fre- quently show twinning on the Carlsbad type. These crystals often appear much fissured and fractured. Under the microscope they frequently exhibit a zoned structure indica- tive of successive stages of accretion, and they often show numerous inclosures of glass, gas and steam pores, well developed crystals of quartz, and microliths of various kinds. Plagioclastic felspars, although frequently present, occur, as a rule, only very sparsely in the quartz-rhyolites, and are often so altered that the characteristic twin lamellae are scarcely, if at all, perceptible, since they are sometimes completely converted into kaolin. The quartz occurs both in roundish grains and in definite crystals. These contain inclosures of glass, which are often 224 Descriptive Petrology. bounded by planes, corresponding to those of di-hexahedral crystals of quartz. Fluid lacunae are not yet known to occur in the quartz of quartz-rhyolites, except in one instance (in the island of Ponza 1 ). This general absence of fluid lacunae distinguishes the quartz of these rocks from that of granite in which fluid inclosures are so common. A little biotite frequently occurs in the quartz-trachytes. Hornblende is seldom plentiful. Tridymite and garnets are occasionally met with, but neither of these minerals are common accessories. Magnetite is generally present, but only in small quantity. Hard, vesicular varieties of quartz- trachyte occur in some localities, and are known by the name millstone-porphyry. The vesicles are often lined with chalcedony or quartz. Nodules or balls of chalcedony and opal are met with in the Hungarian rocks. Some of the quartz-trachytes show a fissile, slaty, or slabby structure, which sometimes originates in the varying character of different bands which exist in the rock, or else in a parallel arrangement of the sanidine crystals. SANIDINE-TRACHYTE (Sanidine-rhyolite). The matrix of this rock is usually of an aphanitic or micro-crystalline character. Under the microscope it is seen to consist almost wholly of little felspar crystals, mostly orthoclastic, but among which plagioclastic felspars are seldom absent. The felspar crystals are usually interspersed either with glass or micro-felsitic matter. Occasionally, as in the sanidine-trachyte of Berkum near Bonn on the Rhine, the ground-mass consists almost exclusively of minute sanidine crystals mingled with microliths and granules of hornblende, and containing some grains or interstitial patches of glassy matter and magnetite. In this matrix no quartz is to be recognised, although the rock contains over 72 per 1 On the Microscopical Structure of Crystals, &c.,' by II. C. Sorby, Quart. Journ. Geol. Soc. Lond. vol. xiv. p. 484. Trachyte Proper. , 225 cent, of silica. Sometimes the matrix of sanidine-rhyolites seems almost wholly amorphous, or shows a finely-fibrous structure, often hazy in appearance, while it occasionally assumes a radiate arrangement around certain points, thus giving rise to spherulitic structure. The spherulites in a rock of this character at Tolcsva, near Tokay, attain from one to two inches diameter. Well- individualised quartz sometimes occurs in the matrix of sanidine-trachyte, some- times none is visible. The minerals which are porphyritically developed in these rocks are, for the most part, crystals of sanidine, either single, or twinned on the Carlsbad type, crystals of plagio- clastic felspar, which, as a rule, show more decomposition than the sanidine crystals, magnesian mica, magnetite, and occasionally hornblende, tridymite, and sphene. TRACHYTE PROPER (Quartzless trachyte, quartzless sanidine- porphyry, domite). (^The matrix of true trachytes consists generally of an aggregate of colourless felspar microliths^ which, by their arrangement in certain directions, frequently indicate fluxion. Spiculae and granules of greenish hornblende and specks of magnetite are also, as a rule, plentifully mixed up with the felspar microliths. By rotation of the section on the stage of the microscope, a very small quantity of interstitial glass may usually be detected, by its persistent darkness between crossed Nicols.(The general colour of the matrix of trachyte is very variable, but greyish, yellowish, and reddish-brown tints are the most common) ( The larger porphyritic crystals which occur in trachytes are sanidine and sometimes plagioclastic felspars : l the latter are not, however, always present. Hornblende is com- mon in these rocks) and magnesian mica also frequently occurs in small crystals or scales, which are visible to the 1 The true sanidine-trachytes contain but very little plagioclastic felspars, and, in some instances, none. 226 Descriptive Petrology. naked eye. The sanidine crystals are usually traversed by numerous irregular fissures, along which they are often dis- placed or faulted, as though they had been subjected to strain and pressure. They are very commonly twinned on the Carlsbad type, and the same may be said of the very minute crystals of this mineral which occur so plentifully in the matrix. The larger sanidine crystals are sometimes an inch or two in diameter, as in the well-known trachyte of the Drachenfels, in the Siebengebirge on the Rhine. The triclinic felspars are, as a rule, developed only on a small scale ; and, as observed by v. Lasaulx, their glassy and cracked appearance often renders it difficult to distinguish between them and the smaller sanidine crystals. Small crystals and spiculae of hornblende are common in many trachytes. To the naked eye they look black, or greenish-black ; while, when seen in thin sections by trans- mitted light, they appear green or brown. (Magnetite and apatite are also present, as a rule, in considerable quantity, the former mineral frequently forming black, granular envelopes around the hornblende crystal^) Tridymite sometimes occurs as a constituent of the matrix, and also in small cavities and druses in the rock. Sphene, hauyne, sodalite, nepheline, and specular iron are not of uncommon occurrence, while augite is sometimes, but rarely, met with in trachytes. In some of the ejected trachytic blocks, as in those of the Laacher See, a great number of mineral species occur, including, besides those al- ready mentioned, zircon, corundum, meionite, garnet, spinel, staurolite, nosean, olivine, leucite, and various zeolites. ''The name domite (from the Pny de Dome, in Auvergne) has been applied to trachytes which contain a high percentage of silica, in some instances over 68 per cent! considered to be due to the presence of tridymite, since(^quartz is never observed in these rocksj) while the former 'mineral occurs rather plentifully in the granular-microlitic matrix, which sometimes contains a small quantity of vitreous matter. Scheme of Deviations from Trachyte as a Type. 227 PLATE IV. FELSPAR (Mostly Sanidine) LEUCITE Qvarfo-lrachi/tc. orRhyolite proper AUCITE 228 Descriptive Petrology. ^ In mineral constitution the domites do not materially differ from ordinary trachytes^ their somewhat higher percentage of silica being their chief characteristic, as pointed out by Zirkel and Rosenbusch. (The porphyritic crystals in them, however, seldom attain any great size.) /The domites form some of the most conspicuous dome- shaped hills or ' puys ' which constitute such striking features in the scenery of Auvergne) (Vide Scrope's ' Volcanos of Central France.') fTrachytic conglomerates and tuffs are composed of fragments of trachytic rocks, together with fragments of other eruptive and sedimentary rocks); these are frequently rounded, and cemented by crumbling earthy matter, mostly derived from the fine detritus resulting from the disintegration of trachytes. VjThe tuffs are either of an earthy or granular and sandy character, mostly light-coloured pale buff, grey, or yellowish-white. They contain fewer rock-fragments than the conglomerates, but pass into the latter as the fragments become more numerous^ They generally contain crystals of sanidine, biotite, and other mineral components of trachyte. It seems difficult or impossible to draw any hard line in the classification of these rocks. Sometimes they appear to have resulted simply from the weathering of trachytes, at others they have more the character of volcanic ejectamenta, ashes, &c., which have been deposited in water ; and, in both cases, there is frequently an admixture, to a greater or less extent, of detrital matter derived from sedimentary rocks. PHONOLITE GROUP. / (The rocks termed phonolite or clinkstone are in a certain degree related to the trachytes proper.; The name phonolite, from ^ovot, sound, was first given to them by Klaproth. (JBoth this and the other name, clinkstone, bear reference to the ringing or clinking sound which slabs or thin fragments emit when struck with the hammer} (The constituent minerals of phonolite are sanidine, Phono lite. 229 nepheline, and generally more or less hornblende and magnetite.^) Nosean and hauyne are often present in con- siderable quantity, also leucite, and sometimes tridymite. The minerals which are less common, and less important as constituents of phonolite, are augite, olivine, sphene, zircon, apatite, and titaniferous iron. Oligoclase as well as sanidine occurs in some of these rocks, and these oligoclase-sanidine phonolites are so closely related to the trachytes that they have received the name of trachy-phonolite. ! \The matrix or ground-mass of phonolite is micro- crystalline, and presents either a rough and porous, or a compact, character. The colour is usually grey, of a yellowish or slightly greenish tint. It is partly soluble in hydrochloric acid) the soluble portion being represented by nepheline and zeolitic decomposition products of that mineral, while the felspathic portion of the matrix constitutes the insoluble part. ([The smaller the percentage of the insoluble matter, the higher, as a rule, is the percentage of water which the rock contains, and this is usually accompanied by an increase in its specific gravity. The larger the percentage of silica which a phonolite contains, the less, as a rule, is its percentage of soluble material.) The amount of silica in phonolites generally ranges fro'm 50 to a little over 60 per cent. (Jhe phonolites fuse easily, before the blowpipe, to a whitish or greenish glass, and yield more or less water when heated.) The phonolites may be divided, according to their dominant mineral constituents, into the following sub- groups : Nepheline-phonolite, Hauyne-phonolite, Nosean-phonolite, and Felspar-phono lite. 1 A very complete account of the different varieties of phonolite is given in Boricky's Petrographische Studien an den Phonolithgesteinen Bohmens (Archiv d. Naturw. Landesdurchforschimg v. Bbhmen. Band iii. Geql. Abth.) Prag, 1873. 230 Descriptive Petrology. The phonolites have, however, also been classified by Dr. Emanuel Boficky in the following manner : i. Nepheline-phonolite. With a compact matrix, and containing much nepheline, and porphyritic crystals of sanidine. ii. Leiicite-nepheline-phonolite. With a matrix of leucite and nepheline, much pyroxene, amphibole, and magnetite, but with sanidine poorly represented. iii. Nepheline-nosean-phonolite (Nepheline-hauyne-pho- nolite). Containing much nosean and some hauyne, with a little sanidine, pyroxene, amphi- bole, titaniferous-iron, and sphene. iv. Leu cite-nosean-phono lite (Leucite-hauyne-phonolite). Consisting mainly of leucite, together with some nosean or hauyne, and more or less nepheline and sanidine. The leucite occurs both of microscopic ,and megascopic dimensions. v. Sanidine-nosean-phonolite (Sanidine-hauyne-phono- lite). A light-coloured rock, speckled with nosean and with a variable amount of porphyritic sanidine. vi. Nepheline-sanidine-phonolite. A greenish, yellowish- grey, or dark grey, slaty or compact rock, weather- ing greyish-white, containing numerous porphyritic crystals of sanidine, and a few of augite or horn- blende. vii. Oligodase-sanidine-phonolite (Trachy-phonolite). Containing from about 5 to 30 per cent, of triclinic felspar. viii. Sanidine-phonolite. Sanidine is abundant. Nephe- line and nosean occur in variable quantity up to 30 per cent., the former mineral constituting a large proportion of the matrix. Sanidine, augite, and hornblende occur porphyritically ; also occa- sionally a little mica and sphene. The relative proportions of the different minerals which constitute the matrix of phonolite vary considerably, the soluble portion varying in its relation to the insoluble Phonolite. 231 portion from 15 to 55 per cent, of the rock, according to A. von Lasaulx. The microscopic character of the matrix is generally micro-crystalline, and, in the phonolites of some localities, consists almost exclusively of superposed layers of small, well-defined crystals of nepheline and sanidine, with little, sparsely scattered crystals of hornblende and magnetite. No amorphous or micro-aphanitic substance occurs in the matrix of phonolite ; but its micro-crystalline nature is not always clearly perceptible, owing to the transparent and colourless character which it often exhibits. The sanidine is very commonly twinned on the Carlsbad type, and is frequently more or less altered. Minute crystals of nepheline, nosean, and hornblende, and granules of magnetite, are sometimes seen lying within the crystals of sanidine, and often appear closely ranged along the margins of the sections of these crystals, which also, at times, contain inclosures of glass. Plagioclastic felspars are only of exceptional occurrence in phonolites. The nepheline crystals, although they occasionally attain moderate dimensions, are, as a rule, very minute, especially those which enter into the constitution of the matrix. They show sharply denned boundaries, and in some phonolites are very numerous ; while in others, the mineral is so poorly represented, that the rocks approximate to trachytes. The nepheline crystals frequently show signs of alteration which, in its ultimate phase, results in the development of zeolitic matter, probably natrolite. Other zeolites, such as stilbite, thomsonite, chabasite, analcime, apophyllite, &c., also occur in phonolites. The decomposition commences by the development of a yellowish fringe, which gradually passes from the exterior to the interior of the crystal, and, until these fringes unite, a nucleus of unaltered nepheline remains. Hauyne is often very plentiful, and there are few 232 Descriptive Petrology. phonolites in which it is totally absent, except in those rocks in which leucite takes the place of nepheline. Hauyne and nosean are so closely related, both chemically and mor- phologically, that some mineralogists regard them as one species, and Rosenbusch includes them both under the older name hauyne. After treating a section containing these minerals with a drop of hydrochloric acid, it is possible to distinguish the sulphate- of-lime-bearing hauyne from the sulphate-of-soda-bearing nosean, by examining the section under the microscope, since, after a little time, the decom- position of the hauyne gives rise to little needles of gypsum, frequently associated, if a gentle heat be previously applied, with little rhombic, cube-like, doubly-refracting crystals of anhydrite. 1 These minerals vary considerably in colour, appearing brown, blue, yellow, green, black, and colourless. Some observers consider that, in their normal condition, they are colourless, and that, at all events, some of the colours are due to changes engendered by an elevated temperature, since colourless hauyne may be artificially coloured by heat. The decomposition of these minerals gives rise to the development of zeolitic matter, and also [when, as in the case of hauyne, they contain a fair amount of sulphate of lime] calcspar is formed. Both hornblende and augite occur in some phonolites, and it is often very difficult to distinguish the one mineral from the other, since the augite in some cases exhibits strong dichroism, while in hornblende this character is sometimes quite absent. In such cases the angles of inter- section of the cleavage planes, when they can be observed, afford a much safer means of discrimination than the phochroic characters of these minerals. Magnetite is nearly " always, and titaniferous-iron is occasionally, present in phonolites. Biotite, leucite, and tridymite are also often present in moderate quantity ; while olivine, apatite, garnet, and zircon are among the less 1 Mik. Phys. d. Massien Gesteine. Rosenbusch, 1877, p. 218. Phonolite. 233 frequently occurring constituents. The phonolites which decompose most readily are, as a rule, those which are richest in nosean. ^Phonolite occurs occasionally in the form of lava flows but more commonly in conical masses or hills. It some- times exhibits well-marked columnar structure, and has a very general tendency to split into slabs or slates, the more finely-cleavable varieties being used for roofing purposes in certain localities) In advanced stages of weathering the rock passes into an earthy condition, known as phonolite- wacke. Phonolite-conglomerate. In some stages of disintegration phonolite-conglomerates are also formed ; these consist of fragments of phonolite, and often of other rocks, together with fine, disintegrated phonolitic matter ; the whole being frequently bound together by a calcareous cement. These conglomerates are mostly found at the bottoms of the phonolite hills, from which their materials have been de- rived. Phonolite-tiiff is an earthy rock of somewhat similar character, except that it contains but few actual rock- fragments. This earthy phonolitic matter often contains numerous crystals of the constituent minerals of phonolite, and the rock is generally cemented by more or less carbonate of lime. The eutaxites of the Canary Islands, and the piperno of Pianura, near Naples, are agglomeratic and banded lavas, which are considered to be more or less closely related to phonolite. The former have a partly vitreous character, and contain rock-fragments lying in tolerably regular layers, which impart a flecked or banded appearance to the lava, into which the fragments are partially fused. Want of space precludes any detailed account of these rocks, but descriptions will be found in the ' Geolog- ische Beschreibung der Insel Teneriffe,' Fritsch u. Reiss. Winterthur, 1868 ; and in the works of Rosenbusch and v. Lasaulx, already cited. 234 Descriptive Petrology. ANDESITE GROUP. (The name andesite was first used by L. von Buch for certain rocks occurring in the Andes. The felspar in these rocks is plagioclastic, and is referred sometimes to andesine and sometimes to oligoclase. The other principal consti- tuents are hornblende, augite and quartz, while more or less magnetite is also present as a rule. These rocks were first divided by Roth into hornblende-andesites and augite-ande- sites. The former are closely related to the trachytes, the latter to the basalts, and they thus constitute a connecting link between these highly basic and highly silicated rocksf) a post also occupied to some extent, although upon different mineralogical grounds, by the trachy-dolerites. (in the rocks of both divisions of the andesite group quartz is sometimes present, sometimes absent) and, upon the presence or absence of this mineral, the aridesites may be glassed as Hornblende-andesite f Q uartzose hornblende-andesite or dacite.' ( Quartzless hornblende-andesite. t Quartzless augite-andesite. Augite-andesite \ Quartzose augite-andesite (of doubtful ^ authenticity). Diallage- and hypersthene-andesites have also been de- scribed by Drasche. (Quartzose Datite consists of a finely-granular or compact, grey, brownish or greenish-grey matrix, containing crystals of plagioclase (oligoclase or andesine) and sanidine, spiculse of hornblende, and granules and crystals of quartz. Under the microscope, the matrix is seen to consist of microliths of plagioclase, sanidine and hornblende, together with fine grains of magnetite. Quartz seldom appears, according to A. von Lasaulx, to enter into the composition of the matrix) when definite megascopic grains of quartz are visible in the lock. f As a rule, the matrix is entirely micro-crystalline, but at ( So named from its extensive occurrence in DaciaA .Andesite. 235 times, when examined microscopically, it shows here and there a very small quantity of interstitial glass} (The triclinic felspars are the most numerous and impor- tant of the porphyritic constituents of this rock, and analyses indicate that they may sometimes be referred not only to andesine and oligoclase, but also to labradoriteA Crystals, showing interlamellation of triclinic felspar with sanidine, are sometimes to be seen under the microscope. The quartz usually contains fluid lacunae and magnetic dust. The hornblende is either in spiculae, or in well-developed little crystals, which sometimes show twinning, and seldom occurs in forms of purely microscopic dimensions. It shows strong dichroism, and often contains needles of apatite' and grains of magnetite. Epidote and chlorite represent the ultimate phase of alteration of the hornblende. Occa- sionally crystals of augite may be detected in these horn- blend e-andesites. Olivine is never met with in them. Some of the hornblende-andesites of Hungary may be regarded as rhyolites, in which plagioclastic felspars play the part of sanidine. The quartzose dacites have been divided into trachytic dacites, biotite dacites, &c.; in the latter hornblende is almost entirely absent, its place being represented by biotite. Some of these rocks are very poor in quartz, and they then pass into the quartzless hornblende-andesites. (The chemical composition of the dacites varies consider- ably in the amount of silica which is present, this fluctuation being due to the variable quantity of quartz which different dacites contain^) Von Lasaulx gives as a mean analysis : SiO 2 =66-io.Al 2 O 3 = 14-80. FeO=6'3o. CaO=5'3o. MgO = 2-40. K 2 O and Na 2 O=7'7o. H 2 O=--o'5o. } - 1 '"* 1-27 1-46 Gabbro. 251 ' A query has here been put against the rock of Penig, since, although for a long time regarded as a typical hyper- sthenite, it has since been suggested by Zirkel that the mine- ral in this rock, hitherto considered to be hypersthene, must now be reckoned as diallage, its almost total absence of dichroism precluding the supposition that it is hypersthene. 1 He also states that the supposed hypersthene in the so-called hypersthenites of Veltlin, Neurode, and the Isle of Skye, is simply diallage. PLAGIOCLASE-ENSTATITE SUB-GROUP. The rocks of this group appear to differ very little in mineralogical constitution from ordinary gabbros, except that their pyroxenic constituent is rhombic and not mono- clinic. Considerable difficulty frequently attends the dis- crimination between hypersthene, enstatite, and bronzite, and it is therefore at times very unsafe to express any strong and decided opinion as to the precise nature of the rhombic mineral which represents the pyroxenic constituent of these rocks, which appear to be generally massed by Rosenbusch under the name norite. The rock of St. Paul's Island, on the coast of Labrador, in which the most typical hypersthene occurs, is placed by this author among the diallage- and olivine-bearing hypersthene norites. The norites of Hitteroe consist of plagioclase and hypersthene, in which the inter- posed plates, &c., so characteristic of the typical hypersthene, are very generally absent. These rocks also contain a little orthoclase and diallage. Olivine and mica occur in some of the norites, and bronzite has been recorded in one or two localities as an essential constituent. Serpentine and Schillerspar (Bastite) are sometimes present in these rocks when they are more or less weathered. The norites never contain any glassy matter. Gabbro occurs in the form of intrusive masses, often of 1 Mikroskop. Beschaff. d. Min. u. Gest. Zirkel. Leipzig, 1873, p. 181. 252 Descriptive Petrology. considerable magnitude, and in dykes, veins, and intrusive sheets, which are sometimes forced along the planes of bedding in the adjacent stratified rocks. BASALT GROUP. iDolerite, anamesite, and basalt, or basaltite, are names applied to the rocks of this group, which imply different conditions of texture and crystalline development, rather than any marked difference in mineralogical constitution or chemical composition. Still some difference between them frequently exists in the relative percentages of silica which they contain, and also in their specific gravities?) (The rocks of the basalt group all contain augite, magne- tite, and titaniferous iron (of the last two minerals sometimes one, sometimes both are present), but they have in addition other mineral constituents which generally form a very considerable proportion of the rock) and indeed in some instances play quite a dominant part. Of these the felspars may claim the most prominent place. They are triclimc. Monoclinic felspar, although met with at times, is of com- paratively exceptional occurrence. ^Olivine, nepheline, and leucite are minerals which exist very plentifully in some of the basalts \ in the constitution of others they occupy quite a subordinate place ; while in some, again, they are totally absent. The occurrence of leucite seems to be restricted to certain localities, and this mineral has not as yet been de- tected in any British rocks. Hauyne and nosean (which latter may be included under the former name) are sometimes sparsely disseminated ; at other times they occur in such considerable quantity as to give a distinctive character to the rock. (Micas occur rather plentifully in some of the basalts, occasionally to such an extent as to impart a special character to them.) The basalts have been conveniently classified by Mohl Basalt. 253 according to their mineral constitution, in the following manner : i. Magma-basalts, with a colourless or brown glass matrix, ii. Plagioclase-basalts, containing notably plagioclase and occasionally nepheline in addition to the essential augite, magnetite, &c. Leucite seldom, iii. Nepheline-basalts, containing notably nepheline, and sometimes leucite, in addition to augite, magnetite, &c. Plagioclase rare or absent. iv. Leucite-basalls. v. Hauyne- and nosean-basalts. vi. Mica-basalts. The old term divine-basalt is not included in this classification, apparently for the reason that olivine may, and very commonly does, occur to a greater or less extent in all of the basalts. The rocks termed magma -basalts have already been alluded to under the name augite-tachylyte. 1 (The basalts vary considerably in structure : the coarsely crystalline varieties, and those in which the different mineral constituents are sufficiently well developed to be distinguished by the naked eye, are termed dolerite^ those in which the constituents are too small to be recognised without a magni- fying power, but in which a crystalline texture is yet clearly discernible, are styled anamesites /while the still more com- pact varieties, which, to unassisted vision, present a more or less homogeneous appearance, are called basalts (basalts proper) or basaltites. Plagioclase basalts. The constituents of these rocks are 1 Boftcky classifies the basalts as Melaphyr-basalt, Felspar-basalt, Phonolite- and andesite-basalt, Trachy-basalt, Tachylyte-basalt. Rosenbusch considers that most of the rocks included in Boficky's last three groups are more or less closely allied to the tephrites, or those rocks which are characterised by the presence of nepheline or leucite 254 Descriptive Petrology. plagioclastic felspars, augite, and magnetite. Titaniferous iron is frequently present. Apatite, olivine, nepheline, and hauyne may also be accessory: Carbonate of iron, calcspar, zeolites, chalcedony, &c., occur as secondary products, and very commonly fill the interior of vesicles. The spaces between the individual crystals are often filled with a glass-magma, usually of a brownish tint, and fre- quently containing great numbers of opaque trichitesi As a rule, the glassy matter represents only a very small pro- portion of the entire rock. The plagioclase in these rocks is sometimes oligoclase, sometimes labradorite, anorthite, or andesine. It is, however, in most cases oligoclase. Ortho- clase also occurs at times in these rocks, but its presence is quite exceptional. Olivine frequently forms an important constituent of the plagioclase basalts. In microscopic sections of basalts which have undergone partial decomposition, the olivine and augite crystals are often merely represented by pseudomorphs of green matter, which is serpentine or some other hydrous silicate. The augite in basalts is generally rich in glass inclosures. Steam pores and fluid lacunae are also of common occurrence in *them. The olivine sometimes appears in tolerably well- defined crystals ; but it is more usually in roundish grains, or in granular aggregates. The latter are sometimes of con- siderable size, and occasionally show, in external configu- ration, that they are large, rudely-developed crystals. The plagioclase basalts are of more frequent occurrence than any of the other rocks belonging to the basalt group. in conjunction with plagioclase. Rosenbusch defines basalt as a rock consisting essentially of olivine, augite, and plagioclase, and regards these rocks as the tertiary and recent equivalents of olivine-diabase and melaphyre. Sandberger has proposed a division of these rocks into those which contain titanic-iron and those which contain magnetite. The former he designates dolerites, the latter basalts. This classifica- tion, however, as suggested by Rosenbusch, is by no means satisfactory, owing to the frequent difficulty in distinguishing between these two minerals, and also from the fact that magnetite is very commonly titaniferous. Basalt. 255 Von Lasaulx gives the two following analyses as repre- senting the average composition of a coarsely crystalline and of a compact variety, the former being a doleritic, the latter a basaltic, type : Plagioclase dolerite : SiO 3 = 50-59, A1 2 O 3 = 14-10, Fe 2 O 3 = 16-02, CaO =9-20, MgO = 5-09, K 2 O = 1-05, Na 2 O = 2-19, Plagioclase basaltite : Si0 2 = 43'O, A1 2 O 3 = 14-0, Fe 2 O 3 and FeO = 15-30, CaO = 12-10, MgO = 9-10, K 2 O = i -30, Na 2 O - 3-87, H 2 O = I -30. The lavas of Etna appear for the most part to be plagioclase basalts, rich in olivine. The plagioclase crystals in these lavas contain great numbers of irregularly-shaped glass inclosures. Nepheline basalt, or nephelinite.-(This is a crystalline granular admixture of nepheline, augite, and magnetite. More or less olivine is always present.) Apatite, sphene, hauyne, melilite, and garnet are among the more common accessory minerals. The nephelinite of Katzen- buckel in the Odenwald, described by Rosenbusch, 1 may be taken as one of the most typical examples. ( Only mere"* traces of interstitial glass are ever to be seen in these rocks r some however contain interstitial nepheline, which may be easily distinguished from glassy matter by its polarisation, and by the crystalline aggregate character of the patches, although no definitely developed crystals may be visible^ The following is an analysis of the nephelinite of Katzenbuckel by Rosenbusch : SiO 2 = 42'3, A1 2 O 3 and Fe 2 O 3 = 28-o, CaO and MgO = 13-65, \The rock also contains 0-65 per cent, of phosphoric acid, and traces of the oxides of nickel, cobalt, and manganese) 1 Der NcpheKnit vom Katzenbuckel. Freiburg, 1869. 256 Descriptive Petrology. t iLeucite, sodalite, and sanidine are occasionally met with as accessories in nepheline basalts. These rocks sometimes assume a very vesicular character,\as in the millstone-lavas of the Eifel, and of Niedermentlig on the Rhine. The latter rock is, however, often so rich in hauyne that it may rather be classed in the sub-group of hauyne-basalts. (.These vesicular rocks assume an amygdaloidal charac- ter when the vesicles are filled with various minerals!) Leucite-basalt (Leucitophyr, Leucilite).-(;The rocks of this sub-group are seldom coarse-grained, and are mostly of a greyish colour, the leucite crystals often giving them a light speckled appearance. They are essentially aggregates of leucite, augite, and magnetite^ Olivine and nepheline are very generally present, sometimes in considerable quan- tity. Nosean is sometimes plentiful, and biotite and sphene also occur as accessories. ^Under the microscope, scarcely any trace of vitreous matter is ever to be detected in the leucite-basalts) unless the leucite-sanidine-lavas of Vesuvius may be included under this name. In most of these rocks felspars are totally absent, although, in some of the leuci- tophyrs of Vesuvius and the Eifel, sanidine crystals are met with of tolerably large dimensions. In all the rocks of this sub-group, leucite is, as a rule, the dominant constituent. In some of the leucitophyrs, as for example in the rock termed sperone, which occurs in the neighbourhood of Rome, the leucite constitutes almost the entire mass, and the crystals, which are mostly of minute size, are very closely packed together. These crystals, when very small, no longer exhibit their characteristic crystalline form, but appear under the microscope as round spots having rather ill-defined boun- daries. >( The leucite crystals are generally rich in interposi- tions,) such as those previously described at page no. The leucite-sanidine lavas of Vesuvius have, as a rule, such a very complex mineralogical constitution, that they cannot be regarded as the equivalents of basalts. They number among their constituents leucite, sanidine, plagio- Basalt. 257 clastic felspar [mainly anorthite], nepheline, sodalite, hauyne, augite, hornblende, olivine, biotite, apatite, &c. The majority of the Vesuvian lavas consist of seven or eight of these minerals. An account of them will be found in the Trans- actions of the Royal Irish Academy, vol. xxvi. * Hanyne-basalt (Hauynophyr). Leucite, nepheline, hauyne, augite, and magnetite are the principal constituents, with usually some olivine and apatite. Vitreous matter occurs sparingly in these rocks and generally contains nume- rous trichites. Felspars, both monoclinic and triclinic, are absent. The hauyne crystals, which for the most part are blue, but also greyish or colourless at times, although fre- quently small, are seldom of very minute dimensions. Some- times the rock assumes a porphyritic character, through the increased development of hauyne and augite. The most typical examples of hauyne-basalt occur at the Laacher See in the Eifel, and at Melfi near Naples. In the rock at the latter locality the hauyne crystals sometimes appear red, owing to the interposition of' lamellae of hematite. This red colour does not, however, always extend to the surface, so that the fractured crystals sometimes have a red nucleus surrounded by a blue border. Hauyne-basalts are rocks of very limited occurrence. Mica-basalts. These can scarcely be regarded as a distinct sub-group, since the mica which they contain does not exclude the occurrence, and cannot be considered as the representative, of any of the essential constituents of the sub-groups already described, unless, in any cases, its mode of occurrence could be reconciled with the observations of Kjerulf on the mica-pseudomorphs after augite, which he procured from the Eifel ; or those of J. D. Dana, on the alteration of olivine into mica. 2 The mica-basalts are rocks pertaining to the plagioclase, the nepheline, the leucite, or 1 ' Report on the Chemical, Mineralogical, and Microscopical Characters of the Lavas of Vesuvius from 1631 to 1868,' by Professors Haughton and Hull. Dublin, 1876. 2 System of Mineralogy, J. D. Dana, 1871, p. 258. S 258 Descriptive Petrology. the hauyne basalts, and since any or all of these rocks may at times contain mica as an accessory, the only distinction which exists between them and the mica basalts appears to be summed up in the statement that mica basalts are rich in mica, while the other basalts contain that mineral in very limited quantity, or as an accessory. The mica crystals in these rocks vary considerably in size ; sometimes they are quite large, at others, they occur as fine microscopic scales, distributed very closely and uniformly through the rock. These micas are mostly dark brown, reddish brown, or black, and may, in many cases, be referred to biotite. The basalts occur in lava streams, plugs, intrusive sheets (' Whin Sill ' of the north of England), and dykes. They are often traversed by structural planes which are, in some cases, so disposed that the rock assumes a columnar charac- ter, as at the Giant's Causeway, Fingal's Cave, and at many foreign localities. The columns are occasionally curved. They sometimes stand in vertical, at others in horizontal or inclined positions, which, in all cases, are directed at right angles to the surfaces upon which the rock cooled. This columnar structure is caused by the contraction of the basalt on cooling, but it is not exclusively in basalts that it occurs ; it is occasionally to be met with in trachytes, phono- lites, pitchstones, felstones, also in argillaceous rocks at their contact with eruptive masses. 1 Sometimes a platy or tabular structure is developed in basalt, especially near the margins of intrusive plugs or dykes. Spheroidal structure also occurs in these rocks, and the spheroids or balls may be seen often closely packed between the divisional planes which constitute the boundaries of the columns. The 1 Some interesting experiments were made by Mr. W. Chandler Roberts in connection with the artificial production of columnar struc- ture, and he has kindly supplied the following note. ' A mixture of clay and sand, in the form of Windsor-brick, was heated to about 1020 c. and slowly cooled. The mass was found to have contracted by about 6 per cent, (cubical), and columnars tructure was well deve- loped in it.' Basalt. 259 columns are sometimes divided by cup-like joints, so that one portion of the column is convex and fits into a concave surface on the adjacent part of the column. The number FIG. 8 7 . of sides, which basalt columns present, varies. Occasionally they have only three sides, at other times five, six, or eight, as shown in the accompanying figure 87.* The subjoined papers on these struc- tures may be consulted with advantage. 2 Basalt occurs in the form of wide-spread lava flows, and coulees or streams, in dykes, in irregular bosses, and in plugs or pipes, which represent the filled-up flues or feeders, from which lava streams were once poured out. 1 From illustrations in the late G< V. Du Noyer's 'Notes on the Giant's Causeway,' The Geologist, vol. iii. 1860. 'It appears now to be pretty certainly established that the peculiar structure of columnar basalt is due to contraction and splitting, consequent upon cooling. The idea entertained by some of the older geologists, that the hexagonal form, so frequently found, was caused by the squeezing together of masses originally spherical, is geometrically incorrect. This process would give rise to rhombic dodecahedra, more or less regular, and could under no circumstances lead to six-sided columns. The cup-shaped joints, so frequently found, have also been shown to be a natural consequence of the contraction on cooling, to which the columnar structure is ascribed. In this view, the analogy of columnar basalt is rather to the splitting, often seen in the mud bottom of a dried-up pool, than to ordinary crystallisation. The direction of the columnar axis with reference to the apparent planes of cooling the confusion of structure towards the middle of the dykes or beds the cup joints the irregularity of the prisms, whose cross sections are seldom regular hexagons the way in which a hexagon passes into a pentagon through a heptagon, and not directly all point to the contractile origin of the structure, at the same time that the result suggests a curious mimicry of imperfect crystallisa- tion.' C. W. M. 2 Gregory Watt, 'Observations on Basalt,' Phil. . Trans. 1804, pp. 279-313. Scrope ' On Volcanoes.' James Thomson, 'On the Jointed Prismatic Structure of the Giant's s 2 260 Descriptive Petrology. Melaphyre. The precise grounds upon which the rocks termed melaphyre have been raised to the dignity of a distinct petrological group are by no means apparent. Rosenbusch seems to regard them as closely related to, if not identical with, olivine-diabase. It is evidently a some- what doubtful question whether they should be classed with diabase or basalt. Melaphyre may be denned as a fine grained or compact, black, greenish-black, or brownish-black aggregate of plagioclase, augite, olivine, magnetite, or titani- ferous iron, and delessite or chlorophoeite. These two last constituents are considered to distinguish melaphyre from basalt, [but melaphyres possess a vitreous, or a devitrified, magma which allies them more to basalt than to diabase]. Now, delessite is a ferruginous chlorite, and chlorophceite is a hydrous silicate of protoxide of iron, also allied to chlorite, or embraced by that very comprehensive term. Both of these minerals are decomposition products, and it therefore appears that their presence should serve to render the true nature of the rock a matter of doubt, rather than to constitute one of its distinctive characters. Allport's suggestion that melaphyre should be included in the term dolerite, of which he regards it simply as a partially- altered condition, seems at least plausible. 1 The definition given by Boficky in the introduction to his 1 Petrographische Studien an den Melaphyr-Gesteinen Bohmens,' 2 appears to a great extent to confirm the foregoing statement. The melaphyres are of palaeozoic age, and this Causeway,' read at the British Association Meeting at Belfast in 1874, but only the title given in the report. R. Mallett, ' On the Origin and Mechanism of Production ... of Basalt,' R. S. Proceedings, 1874-5, vo1 - xxiii - PP- 180-84. T. G. Bonney, ' On Columnar, Fissile, and Spheroidal Structure,' Q. y. G. S., vol. xxxii. p. 140, 1876. 1 'On the Microscopic Structure and Composition of British Car- boniferous Dolerites,' by S. Allport, Quart. Journ. Geol. Soc. vol. xxx. P- 530. 2 Archiv d. Nat. Wiss. Landesdurchforsch. v. Bohmen, Geol. Abth. bd. iii. Scheme of Deviations from Basalt as a Type. 261 PLATE VI. NEPHELINE FELSPAR (Triclinic) LEUCfTE 4 Neplieline Basalb* Q MAGNETITE & TITANIFEROUS IRON 262 Descriptive Petrology. fact seems to have been one of the very insufficient reasons for separating them from similar rocks of later date. The vitreous conditions of basalt have been already de- scribed under the head of tachylyte. Basalts often assume a vesicular character, which is generally most prevalent at and near the upper and the lower parts of the lava streams. The vesicles, when subse- quently filled with calcite, zeolites, and other minerals of secondary origin, render the rock amygdaloidal. Some of the basalts (toadstones) of Derbyshire show this character very well. Basalts occur of various ages, ranging upwards into Ter- tiary, and Post-tertiary times. Basalts of Dimetian age have been identified by Pro- fessors Judd and Bonney, and also by Mr. Tawney. 1 ROCKS OF EXCEPTIONAL MINERAL CONSTITUTION. ( Characterised by the absence of felspars) Garnet-rock. A crystalline-granular aggregate of garnet and hornblende, usually with more or less magnetite. The garnets, as a rule, constitute a far larger proportion of the rock than the other minerals. They may very commonly be referred to the iron-lime varieties, and are mostly of a brownish or yellowish colour. Other minerals are also fre- quently present, such as epidote and calcite. The garnet- rocks are of very limited occurrence, and are chiefly met with in Saxony, Bohemia, the Urals, and Canada, forming irregular veins in mica-schist. Kinzigite. A crystalline aggregate of spessartine (man- ganese garnet), magnesian-mica and oligoclase, often con- taining some iolite and fibrolite, the latter a monoclinic mine- ral, having a chemical composition identical with that of andalusite. It occurs at Wittichen, at the Kinzig, Schwarzwald. 1 ' On the Older Rocks of St. David's,' by E. B. Tawney, Ptoc. Bristol Nat. Soc. vol. ii. p. 113. Eklog ite. L herzolite. 263 Eulysite. An aggregate of reddish-brown garnet, green augite, and a mineral which, in chemical composition, is allied to the iron-olivine, fayalite. The last-named mineral is the dominant constituent of the rock. Eulysite occurs in a very thick bed in the gneiss of Tunaberg in Sweden. Eklogite (Disthene-rock). A granular aggregate of red or reddish- brown garnet, smaragdite (a green variety of diallage which, according to Descloizeaux, has the cleavage and optical properties of amphibole), hornblende, or ompha- cite (a grass-green variety of pyroxene with two sets of cleavage, one more perfect than the other, intersecting at an angle of 115). Kyanite (disthene), silvery white mica, quartz, olivine, zircon, apatite, sphene, oligoclase, and py- rites, also occur at times as accessories. The eklogite from Eppenreuth contains about 70 per cent, of omphacite and 25 of garnet. Other varieties, such as those from the Fichtel- gebirge and Baden, are, on the other hand, particularly rich in hornblende. Others again contain a large proportion of disthene, mica, and quartz, and on this account may prefer- ably be termed disthene-rock. The garnets in eklogite are often surrounded by an envelope of bright-green hornblende, while brown and feebly - dichroic hornblende also occurs in the same rock. The freshly broken surfaces of the rock present a very beautiful appearance from the juxtaposition of red garnets with bright green omphacite. Lherzolite. A granular or crystalline-granular aggregate of olivine, enstatite, diopside, and picotite (a black spinel, containing over 7 per cent, of sesquioxide of chromium). The olivine is the dominant constituent. The rock varies considerably in texture; in some instances it is coarsely granular and feebly coherent, crumbling when handled ; in others it is of a medium crystalline-granular character, and quite tough. The enstatite is of a greenish-brown or yellow colour. In thin sections it appears almost colourless by ordinary transmitted light. It has a more or less fibrous 264 Descriptive Petrology. aspect. The cleavages parallel to oo P intersect at an angle of about 87; less distinct cleavages parallel to the pinakoids are also visible, and are generally rendered more apparent by rotating the section between crossed Nicols. The diopside has a rough or stepped appearance on the abraded surfaces of sections, and shows the characteristic cleavage of augite. It occurs in roundish, green grains. The picotite appears, under the microscope, in very irregular brown, or (according to Bonney) 1 deep olive-green, patches or grains, which, in aspect, somewhat resemble dots and streaks of some gummy substance. They appear dark between crossed Nicols. The olivine is very frequently altered into serpentine, the process of decomposition taking place in the first instance along the cracks in the olivine grains and crystals ; and, as it advances, they become traversed by a mesh-work of little strings of serpentinous matter, until, in the final stage, no olivine remains, the rock often being impregnated with this decomposition product to such an extent that it is virtually a serpentine rock, as pointed out by Von Lasaulx, and more fully described by Bonney, who states his belief that Iherzo- lite is an intrusive rock. Rosenbusch, in describing the extreme phases of altera- tion into serpentine, remarks that the pseudomorphs after the enstatite and olivine may be microscopically distin- guished from one another by the rectangular, grating-like disposition of the fibrous structure in the serpentine, re- sulting from the alteration of enstatite or diallage, and the irregular character of the fibrous mesh-work which is set up in the decomposed olivine. Lherzolite occurs in veins of limestone at the Etang de Lherz, in the Eastern Pyrenees, whence it takes its name. It is also met with in the Tyrol, the department of Haute - Loire, Nassau, Norway, &c. Pyrope occurs as an accessory 1 'The Lherzolite of the Adage,' GeoL Mag. decade ii. vol. iv. p. 64. D unite. Picrite. 265 in a serpentinous condition of this rock, in certain localities. The olivine bombs met with in some basalts are, according to Von Lasaulx, closely akin to Iherzolite. The chemical composition of a Norwegian Iherzolite is cited by that author as SiO 2 = 37-42, A1 2 O 3 = o-io, MgO =48-22, FeO = 8-88, MnO - 0-17, NiO=o-23, H 2 O=o7i. Dunite (so named from Dun Mountain in New Zealand, which consists in great part of this rock and serpentine), is a crystalline-granular aggregate of olivine and chromic- iron ; the former occurring in yellowish-green grains and the latter in black octahedra. Dunite passes by alteration into serpentine. The frequent association of chromic-iron with serpentine renders it probable that many serpentines may have resulted from the alteration of some rock analogous in mineral constitution to dunite. This rock also occurs in the south of Spain and in several other European loca- lities. Diallage and enstatite are present in small quantities in some varieties of dunite, which, under these circumstances, approximates to Iherzolite. Picrite is a blackish-green crystalline rock with a compact, black matrix, containing porphyritic crystals and grains of olivine. The matrix may consist of hornblende, diallage, or biotite, associated with magnetite and calcspar. The olivine constitutes nearly half the bulk of the rock. A small amount of vitreous matter containing microliths is sometimes present. Schorl-rock, although previously mentioned in the de- scription of the granitic rocks, may also be placed in this miscellaneous group. The constituents are schorl and quartz. Topaz, mica, and tinstone sometimes occur as accessories. It is intimately connected with granite, and, by the accession of orthoclase and mica, passes into the schorlaceous varieties of that rock. 266 Descriptive Petrology. VOLCANIC EJECTAMENTA. These comprise dust, ashes, sand, lapilli, and volcanic bombs. They all consist of mineral matter which has undergone a variable amount of trituration and which has been ejected, either in a solid condition, or in a state of fusion. The expulsion of this matter from craters is due to the explosions of steam and gases which occur within the volcanic vents. The lava, which is in a fused and viscid or pasty condition, naturally becomes injected more or less completely with steam and gases, the bubbles of which, when imprisoned in the molten masses and unable to escape, produce a vesicular or spongy texture; so that it is common in volcanic ejectamenta to find fragments of rock, varying in size from fine dust to large blocks, in which a cellular, or pumiceous structure exists. These vesicles are sometimes coarse, sometimes so fine that they are not discernible to the naked eye. Most of the fragments of rocks and crystals which are shot up from the crater, fall back again, unless there be a sufficiently strong wind blowing, to carry them away. The constant attrition against one another which they undergo during these repeated journeys, up into the air, and back again into the crater, tends to round off any angles which the fragments may possess; and the process, if repeated long enough, would reduce the whole to fine sand or dust. Violent explosions also affect the matter within the flue of the volcano, forming a large amount of finely-comminuted and dusty material, which is often carried by the wind for long distances, or, if projected in calm weather, falls in showers over the cone. In some of the high volcanic moun- tains in the Andes, the flow of lava streams over the snow and ice, which rests at high levels, occasionally causes inun- dations, carrying vast quantities of fine mud, termed moya, 1 composed of volcanic dust and ashes ; and similar mud-in- 1 Dr. Theodor Wolf, ' Der Cotopaxi und seine letzte Eruption am 26 Juni, 1877,' Neues Jahrbuch fur Min. u. Geol. Jahrgang 1878, Heft ii. p. 167. Volcanic Ejectamenta. 267 undations are also produced there by the bursting of subter- ranean reservoirs of water during earthquakes. * Mud derived from this source descended, in 1797, from the sides of Tun- guragua in Quito, and filled valleys a thousand feet wide to the depth of six hundred feet, damming up rivers and causing lakes. In these currents and lakes of moya, thousands of small fish are sometimes enveloped, which, according to Humboldt, have lived and multiplied in subterranean cavities.' * Volcanic ashes commonly consist of small fragments of lavas, and crystals of felspars, augite, olivine, biotite, mag- netite, &c., and, in general, there is a more or less close relation in the minerals which constitute volcanic ashes and sands, and the mineral constitution of the lavas which have been erupted from the same crater. Volcanic ashes very often contain particles, or fused drops, of vitreous matter, and the crystals which occur in ashes also frequently contain nume- rous glass inclosures. The plagioclase crystals which occur in the ashes of Etna are especially rich in glass inciosures, but the plagioclase in the Etna lavas also contains them in great quantity. 2 Volcanic sand simply differs from ash in the constituent fragments being coarser. The puzzolana of Naples and the gravier noir of the Puy Gravenoire in Auvergne are volcanic sands, used in the manufacture of hydraulic mortar. Lapilli are moderate-sized fragments of rock, usually scoriaceous lava, which have been ejected from a crater. They may either occur imbedded in deposits of ashes and sand, or they may, of themselves, constitute accumulations. The ejected lapilli are sometimes pumice fragments and, at times, form entire volcanic cones, as in some of the craters in the Lipari Islands. Volcanic bombs vary considerably in character, but, gene- rally-speaking, they may be defined as masses of molten rock- 1 Lyell's Principles of Geology, Qth edition, p. 348. 2 Etna, a History of the Mountain and its Eruptions, by G. F, Rodwell, p. 138. London, Kegan Paul & Co. 1878. 268 Descriptive Petrology. matter, which, by rotation in the air. during their upward flight and downward fall, have assumed a more or less spherical form, and have wholly or partially solidified before again reaching the earth ; in the latter case, the imperfectly- solidified mass sometimes becomes flattened, by impact on the surface upon which it falls. Such bodies are termed slag- cakes. The identification of very old deposits of volcanic ash is not always an easy task. Where numerous lapilli of scoriaceous and other unquestionably eruptive rock occur in old indurated ashes, as in those of Brent Tor in Devon- shire, it is comparatively easy to recognise the origin of the deposits ; but when these fail, it becomes a matter of con- siderable difficulty to say with any certainty whether a rock formed of broken crystals, such as might characterise any lava, in conjunction with very finely divided matter, such as might be referred either to fine volcanic dust or to ordinary detrital sediment, really represents a volcanic, ash, or is simply a sediment formed wholly, or partly, of the detritus of pre-existing eruptive rocks. Some of the rocks mapped as ash beds in the English Lake district have undergone a very great amount of alteration, so that their originally frag- mentary character is only revealed by superficial weathering or by microscopic examination, and, when the alteration becomes extreme, it is hardly possible to distinguish them from compact porphyritic felsite. Some of these rocks closely resemble halleflinta, and a determination of their precise origin is a difficult exercise for micro-petrologists. The recognition of ash deposits is sometimes rendered trouble- some by an intimate admixture of ordinary sedimentary matter. Much yet remains to be done in the determination of old volcanic ejectamenta, a field of inquiry in which none but the most sceptic are likely to demonstrate the truth. 1 1 The student may advantageously consult the recent paper by Dr. Albrecht Penck, ' Studien iiber lockere vulkanische Auswiirflinge. ' Zeitschr. d. Deutsch. Geol. Ges. 1878. Serpentine. 269 ALTERED ERUPTIVE ROCKS. The alterations which eruptive rocks undergo, subsequently to their formation, represent, in most instances, decomposition, often accompanied by pseudomorphous replacement of their constituent minerals, due to the chemical changes effected by infiltration of water, charged, either with carbonic acid, or carrying in solution various soluble mineral substances which it has taken up during its passage through other rocks. There are comparatively few, or no eruptive rocks which do not, to some extent, show traces of such alteration, and the pseudomorphs which they contain are so numerous and interesting that they may constitute quite a special branch of study. The admirable * Recherches sur les Pseudomorphoses ' by Delesse, indicate how much may be done, and yet remains to be done, in this field of inquiry. Some of the most characteristic pseudomorphs will be found mentioned in the descriptions of the various rocks in which they occur. The following are a few rocks which have resulted from the decomposition of eruptive rocks. Serpentine has, in some instances, been demonstrated as the result of the decomposition of such rocks as Iherzolite, gabbro, &c. It is also quite possible that serpentine may sometimes represent the alteration of ordinary sedimentary rocks, especially magnesian limestones, as suggested by Jukes and other geologists, but good evidence seems as yet to be wanting upon this point. Serpentine is also stated to result from the decomposition of some gneissic rocks and other crystalline-schists, also from garnet-rock and eklogite. Serpentine is a fine-grained, massive, compact, rather tough, but soft rock, of very variable colour, dark and light shades of green, greenish-grey, and deep red being the most prevalent. It is often very beautifully veined and mottled with other colours, and, where not much exposed to atmo- spheric influences, it forms a valuable stone for decorative 270 Descriptive Petrology. work in architecture. It is easily turned in the lathe into columns or small ornamental articles, and takes a high polish. It often contains crystals of diallage, which, to some extent, add to the beauty of the stone. Serpentine is also frequently traversed by white veins of steatite, in which angular fragments of serpentine are sometimes imbedded. Serpentine is essentially a hydrous silicate of magnesia. When pure it contains at least two-thirds of silicate of magnesia, but it is frequently impure, through admixture with silicate of protoxide of iron, sesquioxide of chromium, argillaceous matter, and carbonates of lime and magnesia. The variations in its colour are due to different states of oxidation of the ferruginous matter which the rock contains. Serpentine when heated yields on an average about 12 per cent, of water. The minerals which occur as accessories in serpentine are very numerous, and are for the most part the same as those which are met with as accessories in the crystalline schists, with which rocks serpentine is very commonly associated and interbedded. To them, however, may be added chromic iron, picotite, bronzite, schiller-spar, hematite, both massive and as specular iron, dolomite, calcite, brucite, magnesite, hydro- talcite, native copper, copper pyrites, and copper glance. Gold and platinum occur in the serpentines of the Ural. Serpentine seems especially to result from the decompo- sition of rocks which are rich in olivine. Professor Bonney states that the serpentine of the Lizard in Cornwall contains decomposed olivine, enstatite, and picotite, and, from the presence of these minerals, he regards the rock as altered Iherzolite, similar to the serpentine into which he had pre- viously observed the typical Iherzolite ' of the Ariege to pass. He regards the Lizard serpentine as a truly eruptive rock, and considers that the sedimentary rocks which sur- round it had been metamorphosed before its intrusion. 1 1 ' On the Serpentine and Associated Rocks of the Lizard District,' T. G. Bonney, Quart. Journ. Geol. Soc. vol. xxxiii. p. 923. Serpentine. 271 FIG. Sections of serpentine, when examined by ordinary trans- mitted light, usually appear of a pale greenish or yellowish colour. By polarised light the substance commonly exhibits a more or less fibrous structure which displays very feeble polarisation, pale bluish-grey and neutral tints predominating. The crystals of olivine, when they are only partially altered, appear in disconnected fragments, with moderately strong chromatic polarisation, the spaces between the fragments being occupied by fibrous serpentine, which represents the incipient decomposition of the olivine along those irregular cracks by which the mineral is so frequently traversed, as in fig. 88, which shows part of a section of serpentine from Coverack Cove, in Cornwall, after a drawing by Professor Bonney, The speckled por- tion of the figure indicates unaltered olivine, the remainder serpentine. Serpentine frequently contains veins of a finely fibrous mineral, chrysolite, which may simply be regarded as a fibrous condition of the serpentine itself. Occasionally an appearance of lamination, or fine bedding, is visible on the weathered surface of serpentine, the rock appearing to consist of thin alternating hard and soft bands, but the cause of this unequal weathering has not yet been satisfactorily determined. Serpentine occurs either in intrusive bosses, in veins, or in beds interstratified with gneiss, mica-schist, chlorite- schist, talc-schist, &C. 1 Potstone is a soft, sectile, greenish-grey rock, composed of chlorite, talc, and serpentine, used in Italy for the manu- facture of cooking-pots. It is associated with serpentine and chlorite-slate. 1 The views entertained by Dr. T. Sterry Hunt on the origin of serpentine will be found in his Chemical and Geological Essays. 272 Descriptive Petrology. \Laterite is a red, earthy rock, which occurs in beds lying between basalt and other lava flows, and results from their decomposition.) It is strongly impregnated with sesquioxide of iron. Hematite and beauxite sometimes occur in beds of this character. (From the varying nature of the rocks from which it is derived, laterite has naturally a very variable composition, and indeed there is, as yet, no precise defini- tion of this rock.) Palagonite-rock. This results from the action of heated water or steam upon flows of lava, which effects the decom- position of many of the constituent minerals, and especially causes the peroxidation of any protoxide of iron compounds which the rock contains. The result is an amorphous, semi- vitreous substance of extremely variable colour (yellow, red, brown, and black). The chemical composition of palago- nite corresponds more or less with that of the rock from which it is derived, except that no protoxide of iron remains, as it is all converted into sesquioxide, save in a few rare instances where magnetite occurs. The qualitative compo- sition of the rock is represented by silica, alumina, sesqui- oxide of iron, magnesia, lime, soda, potash, and water. The percentage of silica mostly ranges between 30 and 40. Under the microscope, palagonite appears as a perfectly amorphous substance, in which triclinic felspars, augite, olivine, undetermined microliths, and patches of colourless devitrified matter with a radiating fibrous structure occur. These last show dark interference-crosses in polarised light. Palagonite-tuffs differ from palagonite rock in consisting not wholly of palagonite, but of fragments of that mineral, mixed with crystals of augite, olivine, and fragments of eruptive rocks. Kaolin or China-clay is a soft, white, earthy rock, which results from the decomposition of the felspar in granites. When pure, it may be regarded as a bisilicate of alumina, plus two equivalents of water; but the composition varies. It may also result from the decomposition of leucite, Kaolin. 273 beryl, &c., but all the important deposits of China-clay are in the main derived from orthoclase. These deposits are sometimes rather impure from the presence of other consti- tuent minerals of granite. Some of them, in which quartz is plentiful, are termed China-stone. The use of these clays for the manufacture of porcelain is too well known to need more than mention. The kaolin of Cornwall was first employed for this purpose by William Cookworthy of Plymouth in 1755. It has to be carefully levigated before it' is fit for the potteries. 274 Descriptive Petrology. SEDIMENTARY ROCKS. The general character of sedimentary rocks has already been described at page 15 et seqq. In this place, merely the lithological characters and industrial applications of the most typical varieties will be dealt with. These rocks con- stitute so large a proportion of the earth's crust, and have such an important bearing upon water-supply, agriculture, and mining and enginering operations, their application for building, road-making, and other industrial purposes is so extensive, and the history of their formation, and of the past conditions of life which existed at the time of their depo- sition, as shown by their fossils, presents so many points of scientific interest, that it would be impossible, even within the limits of a very large volume, to do even moderate justice to so great a subject. It has not been with any desire to underrate the importance of the sedimentary rocks that so comparatively large a proportion of this work has been devoted to the description of their eruptive brethren, but, because an elementary knowledge of their mineral constitu- tion and structure is, when compared with that of the eruptive rocks, far more simple for the student to acquire. The sedimentary rocks, as already stated, may be divided into two series, the unaltered or normal, and the altered or metamorphic. In the latter series, extreme phases of alte- ration carry the metamorphic rocks out of the sedimentary, and into the eruptive division. These eruptive rocks, when brought to the surface, and subsequently denuded, supply the materials from which fresh sediments are partly formed, so that petrology becomes the study of an endless cycle of changes from eruptive to sedimentary, and from sedimentary to eruptive rocks. The former class of changes are the result of atmospheric and marine denudation, and are due, more to mechanical than to chemical agency. The changes of the latter class are chemical and physical in their nature. Sedimentary Rocks. 275 The altered or metamorphic rocks form, therefore, a transitional series between the unaltered, or normal-sedi- mentary, and the eruptive series. Sometimes the alteration is so slight that it is difficult to detect, and its precise nature still more difficult to demonstrate ; at others, where great alteration has taken place, it is almost, and, in extreme phases, quite, impossible to say with certainty whether a rock should be referred to the metamorphic or to the eruptive series, since there is no natural boundary between them. Before considering those which have been altered, it will be better to describe the UNALTERED OR NORMAL SEDIMENTARY SERIES. These rocks may be classed as arenaceous, argillaceous, and calcareous. They are, however, of a more or less mixed character as a rule, the arenaceous rocks being often ce- mented by calcareous matter, the argillaceous and calcareous rocks frequently containing a certain admixture of sand ; while, again, some of the argillaceous series are impregnated with a variable amount of carbonate of lime, and those of the calcareous series are sometimes more or less argillaceous. Some of the normal sedimentary rocks contain fragments of felspars, scales of mica, and other detritus, derived from the disintegration of pre-existing eruptive rocks. The sedi- mentary rocks occur in strata or beds, which rest upon one another, which have a regular order of sequence, and which generally contain fossils of characteristic types. The re- marks made in the earlier part of this work, and the nume- rous text-books of geology, which deal more or less fully with the stratigraphical and palaeontological branches of the science, render it unnecessary to say anything here upon these subjects. The materials of which sedimentary rocks consist are usually more or less rounded by attrition, the result of their transport by water, or, in the case of aeolian rocks, of their transport by wind. T 2 276 Descriptive Petrology. ARENACEOUS GROUP. (Sandstones.} These rocks consist essentially of grains of silica. They either occur as superficial accumulations of loose sand forming desert tracts, or low-lying districts on sea coasts, where the wind piles the sand up in dunes, or they may occur as beds of loose sand, interstratified with coherent beds of rock. They are also met with in a state of more or less imperfect consolidation, the grains being feebly held together by an iron-oxide or by calcareous matter or they may be excessively hard and compact, the constituent grains being cemented by either silica, carbonate of lime, iron-oxides or carbonate of iron. The rocks called grits vary considerably in lithological character. The term grit ' appears indeed to be very ill-defined. The millstone grit, which may be taken as one of the leading types, is more or less coarse- grained, while some of the Silurian rocks, such as the Coniston and Denbighshire grits, are frequently very fine- grained and compact in character. Under these circum- stances it seems that a grit may best be defined as a strongly- coherent, well-cemented, or tough sandstone, usually, but not necessarily, of coarse texture. In some few cases there even appears to be, according to Prof. Morris, no cementing matter present, as in some of the new red sandstones, the constituent grains being apparently held together merely by surface cohesion superinduced by pressure. It is not pos- sible within the limits of this work to do more than allude to some of the most important sandstones which occur in the British Isles. Those used for building-stone and paving are for the most part of old red, carboniferous, triassic, and neocomian age. Commencing with the oldest and lowest in the series, the Cambrian and Silurian grits are for the most part very tough, closely compacted sandstones, frequently containing minute fragments of felspars and sometimes scales of mica. Their constitution implies that they are Sandstones. 277 formed, at all events to some extent, from the detritus of pre-existing eruptive rocks. They are, in some instances, fusible before the blowpipe, on the edges of thin splinters, which is probably due to their admixture with felspathic matter. They are generally traversed by numerous joints, so that they are seldom used for building purposes, except locally in the construction of rough walls. They are, however, well suited for road metal, and in some places good flagstones are quarried, but these are, for the most part, rather to be regarded as sandy shales and slates, than true sandstones. The flaggy sandstones are generally micaceous, and to this circumstance their fissile character is often due. Although some beds of sandstone and grit occur in the Devonian series, they are unimportant from an economic point of view, but in the Old Red Sandstone (the chrono- logical equivalent of the Devonian series), both building- stones and flagstones are quarried. They are mainly em- ployed in the districts where the stone is procured. It is often of a deep reddish-brown or purple colour owing to the presence of peroxide of iron; at other times it is greyish, occasionally with a greenish tinge. The stone, if judiciously laid, is tolerably durable, but in some old buildings, such as Chepstow Castle and Tintern Abbey, it has suffered con- siderably from the weather. Old red sandstone is extensively used for paving in many of the large towns in England, Scotland, and Ireland, and is also largely employed as a general building- stone and as road metal in the districts where it is quarried. The red sandstone quarries of Cork and Kerry yield in some instances building- stone of a very durable character. 1 The Dundee and Arbroath sandstones, known as Caithness flagstones, quarried on the east coast of Scotland, form good and durable material for paving and building, but the former is too sombre in colour to give a pleasing effect when used for architectural purposes. 2 1 Building and Ornamental Stones, E. Hull, p. 266. London, 1872. 2 Applications of Geology to the Arts and Manufactures, Ansted, p. 153. 1865. 278 Descriptive Petrology. The carboniferous sandstones, including those of the Yoredale series, the millstone grit, and the coal-measures, are very important from an industrial point of view, since they afford good material for building and paving. The Halifax, Bradford, and Rochdale flags are extensively used for the latter purpose in the North of England, and are well known to builders under the name of Yorkshire flags. Some of them absorb water very readily, consequently, in very exposed and damp situations, they are liable to flake, espe- cially if placed in positions where they are unable to part with their moisture. The stone from Bramley Fall near Leeds belongs to the millstone grit, and is largely used for architectural purposes. The Rotherham stone is worked for building purposes and for grindstones, and that at Hart Hill for scythe-stones. The Wickersley stone (Middle Coal- measure Sandstone) makes good grindstones. The pennant grits and sandstones occurring in the coal- measures of the Bristol coal-field are important building- stones. The fine-grained pale-brown and grey sandstones from Craigleith, near Edinburgh, and the Binnie quarry in Linlithgowshire, are also extensively employed for buildings. They darken somewhat on exposure, but are amongst the most durable of building-stones. The Craigleith stone con- tains only about i per cent, of carbonate of lime, the cementing medium being mainly siliceous. A little mica and carbonaceous matter is also present to the extent of about i per cent., the remainder of the rock, 98 per cent., consisting of silica. According to Ansted, a cubic foot of Craigleith stone, weighing about 146 lbs.,will absorb 4 pints of water, and good samples will resist crushing weights to the extent of 5,800 Ibs. to the square inch. Of the carboni- ferous sandstones used in Ireland the Carlow flags are perhaps the most important: they are sometimes more or less micaceous, and are of dark bluish or grey colour. The Permian sandstones, which are the equivalents of the Continental Rothliegende, or lower division of the Per- mian system, are but little used in this country, except locally, for building-stone, as in some parts of Cumberland, 1 Staffordshire, Nottinghamshire, and Yorkshire. At Mans- field, in Nottinghamshire, reddish brown and almost white varieties of triassic sandstone are quarried, and are said to be durable. As a rule the Permian sandstones are not well suited for building, being very absorbent and liable to decay. The Permian sandstone from the neighbourhood of St. Bees was used for the construction of Furness Abbey. These rocks have mostly a deep red colour, due to the presence of peroxide of iron, which, together with dolomitic matter, constitutes their cement. Triassic sandstones. Those belonging to the upper trias or Keuper are the most important as building-stones, the sandstones of the lower trias or Bunter being, as a rule, of too loosely-cemented and friable a character for such purposes. The latter are, however, used for moulds in foundries, and, occasionally, for buildings. They are often variegated and mottled, whence the name Bunter from the German bunt, variegated or coloured and they frequently exhibit false bedding. The Keuper series affords good building-stone, especially the lower Keuper sandstones, which are extensively used in the midland and north-western counties. It is of pale red, brown, and yellow colours, sometimes almost white, and is mostly fine-grained and easy to work. This stone has been largely used in the cathedrals of Chester and Worcester. The loosely-coherent triassic sandstone of Alderley Edge, in Cheshire, is partly cemented by carbonates of copper and other mineral matter, derived from infiltrating solutions. To obtain the copper, the sandstone is crushed, the copper salts dissolved in sulphuric acid, and redeposited on scrap iron in the metallic condition. The sandstone yields but little more than i per cent, of copper. Keuper sandstone is quarried 1 The Penrith and St. Bees sandstones are much used as building- stones. 28 o Descriptive Petrology. in Antrim, and is stated by Professor Hull to be exceedingly well adapted for architectural purposes. Amongst the carboniferous and triassic rocks of some countries a sandstone occurs to which the name Arkose is given. It consists essentially of the same constituents as granite, and has been derived from the disintegration of granitic rocks. Some valuable notes upon arkose occur in the * Me'moire sur les Roches dites Plutoniennes de la Belgique et de 1'Ardenne Frangaise/ by MM. Ch. de la Vallee Poussin and A. Renard, Brussels, 1876, p. 120. Jurassic sandstones. The rocks of the Jurassic period are for the most part limestones, but good sandstone is quarried at Aislaby, near Whitby, in Yorkshire, and has been used in the construction of Whitby Abbey and several other important buildings. 1 In Lincolnshire, Northamptonshire, and Dorsetshire, sandstone, belonging to the inferior oolite, is employed for building. ' The ferruginous, or calcareous rock of the lower part of the Northampton sand is locally largely used for building purposes, but it does not usually possess much durability. The white sands in the upper part of the series are extensively dug at many points for making mortar.' 2 Cretaceous sandstones. Those of most importance, from an economic point of view, are the sandstones belonging to the Hastings sand series, and some of the hard sandstone beds intercalated with the Kentish rag, derived from the lower greensand. The sand-rock of the Hastings Sands is not a very coherent stone when first dug, but it hardens on exposure, and, although largely used for building in the neighbourhood where it is quarried, it is not of a very durable character. It is generally of a warm yellowish or brownish colour, and has a somewhat ferruginous cement. Bargate stone is quarried at Godalming for building 1 'Mineral Statistics,' Mem. Geol. Sura. part. ii. 1858. R. Hunt. 2 < The Geology of Rutland,' J. W. Judd. Mem. Geol. Swv. p. 92. 1875. Sandstones. 281 purposes. It is a calcareous sandstone, and occurs in the upper part of the Hythe beds. The calcareous sandstones in the Hythe beds in. Kent are locally termed hassock, and are also used for building. The Folkestone beds of the lower greensand also afford hard sandstone and grit, suitable for building and road-making. In the upper greensand at Godstone and Merstham, a pale calcareous sandstone called fire-stone occurs, which is well suited for the floors of fur- naces, and is also a durable building-stone. 1 Flints are procured from the upper chalk, and are extensively used as building material and road-metal. Tertiary sandstones. Although, in this country, beds of sand are of constant occurrence in the tertiary formations, they are not, as a rule, sufficiently coherent to be of value for building purposes, except for making mortar. They are, when pure, used in the manufacture of glass. The Headon Hill sands, which occur in the Bagshot series in the Isle of Wight, are largely used for this purpose. The less pure sands are applied to various other uses. There are, how- ever, a few very hard tertiary sandstones, which are used in this country for building and paving, some of which are, according to Prof. Morris, derived from the Woolwich series, and others from the Bagshot beds. In some parts of the world tertiary sandstones attain great importance. Sand- stones of miocene age constitute a considerable part of the Himalayas. As an appendix to the rocks of this group we may place Tripoli, a fine white pulverulent, chalk-like deposit, which consists almost exclusively of the siliceous skeletons of diatoms. To demonstrate this it is only need- ful to wash a little of the powder on to a slip of glass and examine it under the microscope. The mud of the river Parret (which runs into the Bristol Channel) affords the material of which 'Bath bricks' are made. This mud has been stated to consist almost wholly 1 See also 'Geology of the Weald,' by W, Topley, Mem. Geol. Surv. Eng. 6 Wales i p. 371. 282 Descriptive Petrology. of the remains of diatoms. Microscopic examination, how- ever, disproves the statement, and shows that diatoms form but a very small proportion of the mud. ARGILLACEOUS GROUP. (Clays, Shales, and Slates.} These rocks are, chemically speaking, impure hydrous silicates of alumina. Sometimes the impurity consists of sand, sometimes of carbonate of lime ; and more or less carbonaceous matter is in many cases present. Their coarseness of texture is mainly dependent upon the coarse- ness of the sand which often occurs in them. When free from sand, they are usually of fine texture. They have all originally been deposited as mud, in most instances at the bottom of the sea, in others at the bottoms of lakes or as deltas, and, exceptionally, over land, when temporarily flooded by the overflow of rivers, as in the case of the Nile. Clay deposits often have a well-laminated structure, and, in the older geological formations, have assumed a more or less indurated character, frequently accompanied by a tendency to split along the planes of bedding. Very often another and more strongly-marked fissile structure is superinduced in directions cutting across the planes of stratification at various angles. This is slaty cleavage, described at page 35. Those argillaceous rocks which split parallel with the planes of lamination or bedding are called flags, but the term flag is applied to a rock of any character which splits along its bedding into large flat slabs, and consequently it is common to find the term used to denote sandstones which are suffi- ciently fissile, when quarried, to yield slabs or flags. To the argillaceous rocks which split in directions other than that of bedding the term slate is given. Still, in this case, the term is also applied to rocks which differ widely from ordi- 1 ' Geology of East Somerset and Bristol Coal Fields,' Memoirs of Gcol. Surv. p. 161. H. B. Woodward, 1876. Clays, S hales, and Slates. 283 nary slate. The Collyweston slates, calcareous sandstones of the inferior oolite, and the green-slates of the Lake District, which have been mapped as volcanic ash by the Geological Survey, are examples of the application of the term slate as indicative of fissile structure, and not of lithological character. Cambrian slates. These are very important rocks, affording compact roofing-slates of admirable quality, mostly of a dark purple or greenish colour, and capable of being split into very thin and large slates, exceedingly free from pyrites, which is common in many slates, but, from its decomposition, is most detrimental to them as roofing material. The slates of the Penrhyn and Bangor and of the Dinorwig or Llanberis quarries in North Wales are of Cambrian age. Silurian slates and flags. The Skiddaw [lower Silurian slates of Cumberland], are black, or dark-grey rocks, which are often traversed by many sets of cleavage planes, causing them to break up into splinters or dice, so that no good roofing-slate can, as a rule, be procured from them. The best lower Silurian slates of North Wales are quarried in the Llandeilo and Bala beds. They are black, dark grey, and pale grey. Ffestiniog, Llangollen, and Aberdovey are among the principal quarries. The cleavage in these rocks is often wonderfully perfect and even, so that occasionally slates ten feet long, six inches or a foot wide, and scarcely thicker than a stout piece of cardboard, are procured. These remarkably thin slates are tolerably flexible. The upper Silurian rocks also afford good slates and flags in certain localities, while the rough material serves for local building purposes. In many parts of the English Lake District the houses are commonly constructed of rough slates and flags derived from the Bannisdale and Coniston series. The quoins of the better class of these houses are often built of a light-coloured freestone, and the general effect is good, although sombre. Silurian slates are quarried 284 Descriptive Petrology. in Scotland in Inverness-shire, Perthshire, and Aberdeen- shire; also at Killaloe and some other localities in Ire- land. Devonian slates of a grey colour are worked in Cornwall, at the Delabole and Tintagel quarries, and in Devonshire, in the neighbourhood of Tavistock, at Wiveliscombe and Treborough in Somersetshire, and in other parts of the United Kingdom. The carboniferous flags are quarried for roofing and paving purposes at several places in Yorkshire, Lancashire, and other counties, where carboniferous rocks occur, and are mainly procured from the coal measures. They are of dark-grey colour or black, and are principally used in the neighbourhoods where they are quarried. There are no true clay slates of later age in Great Britain, but in other parts of the world slates of even tertiary age occur. Of the clays, used in this country for economic purposes, may be mentioned the china clays or kaolins of Cornwall, which have been formed from the decomposition of the felspathic constituents of granite ; the Watcombe clay, which occurs in the trias, and is now used in the manufacture of terra-cotta pottery \ the calcareous liassic clays, used for brick-making and burning for lime and hydraulic cement ; the various clays of oolitic and neocomian age, some of which are used for brick-making, &c. ; the gault, the clays of the Woolwich and Reading beds, and the London clay, all of which are used for bricks ; the celebrated Poole clay, dug at Wareham, which belongs to the Bagshot series, and is extensively used for pottery. The clays of the Bovey beds, large quantities of which are annually shipped at Teignmouth, afford good pottery-clays and pipe-clays. There are also many brick-earths and clays of post- tertiary age which are extensively used for brick-making and other purposes. Fuller's Earth is a yellowish, greenish or bluish clayey rock containing about 50 per cent, of silica, 20 per Limestones. 285 cent, of alumina, 25 per cent, of water, and a little oxide of iron. It chiefly occurs between the Inferior- and Great- Oolite, and in the Lower Greensand. The river-mud in the Medway and at the mouth of the Thames is largely used in the manufacture of Portland cement, after being artificially mixed with chalk and burnt. By the careful levigation of some clays, Dr. John Percy has eliminated minute, but beautifully- developed, crystals of kaolinite. CALCAREOUS GROUP. (Limestones.) These, in some cases, consist almost exclusively of car- bonates of lime and magnesia (magnesian-limestones or dolomites), while occasionally the}' are very impure, con- taining a considerable admixture of sand, clay, and, in some instances, bituminous matter. The British Silurian lime- stones are of comparatively little value, except for lime, locally employed for agricultural purposes. The Devonian limestones are, however, extensively used for building and paving, and some of them are well adapted for ornamental purposes on account of the richly coloured mottling and veinings which they frequently exhibit. The carboniferous limestone is largely used for building, and is a very durable stone. Some of the highly fossiliferous beds especially those which contain numerous fragments of encrinite stems, locally termed screw-stonesconstitute handsome marbles, the fossils being white, and the rock itself dark grey, or almost black. Good encrinital marble is quarried at Dent in Yorkshire, and the stone is much used for chimney- pieces. The carboniferous limestone often contains bands and nodules of chert. The magnesian limestone of Per- mian age 1 is a very well known and, when judiciously selected and properly laid, a very durable building-stone. This is 1 Beds of magnesian limestone also occur in the carbonifeious lime- stone series in Derbyshire and elsewhere. 286 Descriptive Petrology. well shown in the keep of Conisborough Castle and York Minster, which have both been built of Magnesian Limestone. The Museum of Practical Geology is fronted with this stone, and has stood well. The Houses of Parlia- ment are also built of Magnesian Limestone. It works freely, and can generally be procured in large blocks. It may here be observed that many limestones contain only a very small percentage of carbonate of magnesia, and since magnesia and lime are isomorphous, the amount of mag- nesian carbonate in limestones may fluctuate from mere traces to a ratio of CaCO 3 to MgCO 3 = 1:3. No sharp line of demarcation can therefore be drawn between the dolomitic limestones and the true dolomites, in which the ratio of CaCO 3 to MgCO 3 = i : i giving the percentage com- position as CaCO 3 =54'35, MgCO 3 = 45'65 for normal dolomite. The oolitic limestones are so numerous and constitute such valuable building-stones that it is only possible to mention a few of those principally employed. These are the Doulting stone, belonging to the inferior oolite, the Bath-stones belonging to the great oolite, of which the chief kinds used are the Box Hill and Corsham Down stones. The Ketton stone, belonging to the Lincolnshire oolites, is an exceedingly valuable building-stone, possessing great tenacity, working freely, and resisting atmospheric influences, even when placed in unfavourable situations. The Ancaster stone, which also occurs in the Lincolnshire oolites, is a less expensive stone, but is very durable, and is extensively used for building. The Portland oolites afford remarkably good stone, which is used very largely, and constitutes one of the most important building-stones in this country. The Purbeck limestones, which, unlike the preceding, are of fresh-water origin, are largely used for paving ; while, in the upper part of the series, a compact limestone, crowded with fossil shells, of the genus Paludina, is known as Purbeck marble, and has been used for small L imes tones. 287 columns and other architectural decoration, for some cen- turies. The Petworth marble has also been applied to similar purposes. The oolitic limestones, as a rule, differ structurally from the limestones of older and of more recent date, inasmuch as that they are usually aggregates of little spherical deposits of carbonate of lime, which have formed in concentric crusts around nuclei. These nuclei consist sometimes of a granule of sand, sometimes of the remains of a minute organism. The little spherules are seldom much bigger than a large pin's-head, and they are also cemented together by calcareous matter. The name oolite is derived from the egg-like or fish-roe-like appearance of the stone ; but oolitic structure, although characteristic of the limestones of oolitic age, is, however, not exclusively peculiar to them, for well-developed oolitic structure occurs in certain beds of the carboniferous limestone near Bristol, while it is also developed in the coarser pisolites or pea- travertines of recent date. The older limestones, such as those of the Devonian, carboniferous, and Permian forma- tions, are either granular or crystalline-granular, and the latter character is beautifully shown in certain limestones at and near their contact with eruptive rocks. The saccharoid statuary marbles of Italy are good examples of this struc- ture, and, when examined in thin slices under the micro- scope, are seen to consist of closely aggregated crystalline grains, in each of which polarised light reveals the existence of numerous twin lamellae, the twinning taking place along planes parallel to the face JR. It seems impossible, in many cases, to say whether this structure in limestones has been due to the metamorphism engendered by the contact or proximity of eruptive rocks, or whether it is owing to other causes, since we find precisely the same structure in the amygdaloids of calcspar which have been infiltered, into the vesicles and crevices in basalts, long after their solidifi- cation ; we find it in the fossils of the chalk and of other formations, which have not had the opportunity of becoming 288 Descriptive Petrology. altered by the presence of intrusive rocks ; and we also find it in limestones, at and near their contact with eruptive masses, as already observed. The earthy variety of lime- stone, chalk, has been stated to consist exclusively of the calcareous tests of foraminifera. This, however, is not always the case. Samples of chalk may sometimes be care- fully levigated and examined, and foraminiferal remains may only be detected here and there, the greater part of the matter having merely the character of an ordinary amor- phous precipitate ; while, again, other samples may be found to consist in great part or almost entirely of the re- mains of these organisms. Some writers have even gone so far as to express an opinion that nearly all limestones have been formed out of the calcareous remains of foraminifereti corals, &c. In controversion of these statements, Credner, besides giving other good reasons, appeals to microscopic evidence, which shows, he observes, 1 'that our ordinary compact limestones are by no means always formed of broken and finely-ground organic remains, but rather of little rhombohedra of calcspar.' On the other hand, how- ever, we are bound to admit that the fossils which occur so plentifully in limestones, at all events, represent something more than an insignificant proportion of their bulk, and in some cases seem even to constitute the greater part of the rock. Cretaceous limestones. The Kentish rags are mostly very hard sandy limestones, and contain more or less dark- green glauconite, generally in fine, occasionally in coarse, roundish grains. Glauconite is stated to sometimes form the cementing medium in these rocks, but more or less car- bonate of lime is always present in this capacity. By decomposition, the protoxide of iron in the glauconite is converted into peroxide of iron, and the rock, under these circumstances, assumes a reddish-brown tint. According to 1 Elemente der Geologie, Leipzig, 1876, p. 290. Limestones. 289 :nberg, the glauconite grains often fill, invest, or replace the tests of foraminifera. These rocks form very durable building-stones. Besides their use in ashlar work they are often laid in irregularly-shaped blocks, giving rise to a honeycomb pattern on the surfaces of the walls built of them. Kentish rag is chiefly quarried at Maidstone, Hythe, and Folkestone, and is extensively used for building in the South of England. It is derived from the Hythe beds. 1 Lime- stone, either as ordinary chalk or as subordinate beds of com- pact limestone, represents a considerable part of the creta- ceous series of rocks, while most of the cretaceous sandstones are very calcareous. The chalk attains a great thickness in some parts of the kingdom ; the lower portion, termed the grey chalk or chalk marl, is generally slightly glauconitic at the base. The upper chalk contains numerous nodules, and occasionally bands of flint, which follow the stratification, although at times vertical bands of flint occur, filling up what once were open fissures. Chalk, besides being largely burnt for lime, is also locally used for building. Certain hard beds occur in the chalk which are better suited for this purpose than the softer material. Tertiary limestones. In the British Isles these are but poorly represented. The Binstead limestone, occurring in the Bembridge beds in the Isle of Wight, has, however, been extensively quarried, and has been employed in the con- struction of some of our early churches. In other parts of the world tertiary limestones often attain great thicknesses, and constitute important building stones. The pyramids, for example, are built of nummulitic limestone. There are many other interesting tertiary limestones, but want of space precludes any mention of them. ALTERED SEDIMENTARY SERIES. With regard to the rocks of this series it is difficult to 1 An account of the Wealden marbles will be found in Mr. W. Topley's 'Geology of the Weald,' Mem. Geol. Surv. Eng. 6 Wales, p. 368. U 290 Descriptive Petrology. say where alteration begins and where it ends, still more difficult in some cases to define the nature of the alteration. It is common for geologists to talk about altered slates where they show the slightest perceptible difference from slates which they regard as normal types, but it is often open to question how far the normal types are really normal, and to what extent the microscopic crystalline constituents of these rocks are to be considered normal or of secondary origin. Von Lasaulx, for example, states a belief that some of the microliths in slates may be referred to hornblende and epidote. Now the latter mineral, if present, must certainly be regarded as a secondary product, and the rock which contains it must, in a certain sense, be considered as an altered rock. The rocks of the altered sedimentary series may be divided into A. Those with no apparent crystallisation. B. sporadic crystallisation. C Crystalline \ a ' ^on-foliated. (b. foliated and schistose. Altered sedimentary Rocks with no apparent crystallisation. In the case of limestones, a crystalline or crystalline- granular condition frequently results from alteration, but sometimes the change simply appears to cause induration, without developing any crystalline structure, as in some of the Antrim chalk, altered by the proximity of basalt. Sand- stones, from the character of their constituent particles, can hardly be included under this division of the altered sedi- ments. The alteration which argillaceous rocks undergo without begetting any perceptible crystallisation consists mainly of changes which appear to be of a purely physical character, generally slight and difficult to describe intelli- gibly. Perhaps the best example of such alteration is to be found in the so-called porcelain jaspers, clays, or shales, which have been baked, either by the combustion of adjacent Chiastolite Slate. 291 coal-seams or by the contact or proximity of eruptive rocks. Porcelain jasper has a fused or fritted appearance, a slight gloss, and the different bands or laminae often assume strongly-marked differences of colour, in which dark green and brick-red sometimes predominate. Alte?'ed sedimentary Rocks with sporadic crystallisation.-*- In these rocks the development of the crystals is often very imperfect and obscure ; in some cases, however, the crystals are distinct and well-developed. The sporadic crystals which occur in altered limestones are varieties of pyroxene, usually coccolite, hornblende, garnet, sphene, tourmaline, spinel, phlogopite, chlorite, talc, &c., but the limestones themselves, in which these crystals occur, almost invariably have a crystalline structure en- gendered by metamorphism, and consequently they should rather be placed amongst those altered sedimentary rocks which have a crystalline structure. The altered slates frequently exhibit sporadic crystalli- sation. The crystals developed in them are usually silicates of alumina, such as staurolite, andalusite, chiastolite, &c. Chiastolite slate occurs in the neighbourhood of granitic masses, as in the Skiddaw district in Cumberland, where, according to J. Clifton Ward, ' on approaching the altered area the slate first becomes faintly spotty, the spots being of a somewhat oblong or oval form, and a few crystals of chias- tolite appear. Then these crystals become more numerous, so as to entitle the rock to the name of chiastolite slate. This passes into a harder, more thickly-bedded, foliated and massive rock, spotted (or andalusite) schist ; and this again into mica schist of a generally grey or brown colour, and occurring immediately around the granite/ ! The chiasto- Tite slates are mostly of dark grey or bluish-black colour, and contain pale yellowish-white crystals of chiastolite, some- times more than half an inch in length. These in trans- 1 'Geology of the Northern Part of the English Lake District.' Memoirs of the Geol. Surv. England and Wales > 1876, p. 9. U 2 292 Descriptive Petrology. verse section often show a dark central spot and, occasion- ally, the chiasmal interpositions which characterise this mineral. Staurolite slate is a dark micaceous slate containing crystals of staurolite ; in some localities passages have been observed from this rock into andalusite slates. In some altered slates the staurolite is so imperfectly developed that it merely appears in roundish or lenticular knots or lumps, which exhibit no approximation to crystal- line form. The knoten- frucht- garben- and fleckschiefer of Ger- man petrologists consist of micaceous slates containing small irregular concretions or little lenticular or ovoid bodies, which, in some cases, may be referred to andalusite, but in many instances they are shown by microscopic examination to be aggregates of small scales of mica, carbonaceous matter, quartz granules, and other constituents of the rocks in which they occur. They are often surrounded by ferru- ginous stains resulting from decomposition. The deter- mination of the precise character of these bodies is often a matter of considerable difficulty. Want of space precludes any description of other inter- esting rocks which belong to this group. ALTERED SEDIMENTARY ROCKS (CRYSTALLINE). These rocks may be divided into (A] non-foliated, and (B) foliated and schistose groups. A. Non-foliated Group. Under this head come the limestones in which a crystal- line structure has been superinduced by the proximity of eruptive rocks. This structure differs, in no essential respect, from that of the crystalline limestones described at page 284. The sporadic crystals, which sometimes occur in these lime- stones, are coccolite, tremolite, tourmaline, sphene, chondro- Quart zite. Lydian-stone. 293 dite, spinel, garnet, mica, chlorite, &c. Those crystalline limestones which are suitable for ornamental architecture are termed marbles, and many marbles are rocks of this kind, which owe their crystalline character to alteration by intrusive masses ; still there are also many in which the crystalline structure is not due to this cause. The term marble is, however, very loosely employed, and may be generally taken to signify any rock which takes a good polish and is employed for decorative or architectural pur- poses. It is impossible, from want of space, to allude even to the most important marbles. l Quartzite is a compact crystalline-granular aggregate of quartz, either in irregular crystalline grains, or in well- developed crystals. Some quartzit.es exhibit a schistose structure, which is partly, or wholly, due to the presence of small quantities of mica, the scales of mica lying in the direction of the fissile planes in the rock. These schistose quartzites may therefore be regarded as mica schists poor in mica. Lydian-stone (basanite, touch-stone, kieselschiefer) is a dark-coloured, generally velvet-black or brownish-black rock. It is an altered sandy slate. Under the microscope, it is seen, in great part, to consist of crystalline grains of quartz mixed with particles of argillaceous, carbonaceous, and ferruginous matter. The percentage of carbonaceous matter is sometimes considerable, and accounts for the extremely black colour of the rock. Lydian-stone is often traversed by small veins of crystalline quartz, and frequently contains a little pyrites. B. Foliated and Schistose Group. These rocks which are commonly designated crystal- 1 Good accounts of the Italian marbles will be found in the Official Catalogue of the Exhibition of 1862, ' Kingdom of Italy,' section v. p. 44 ; and in The Mineral Resources of Central Italy, by W. P. Jervis. London, 1868. 294 Descriptive Petrology. line schists, afford, perhaps, the best-defined instances of metamorphism, in the sense in which that term is usually applied. They are generally characterised by the presence of one of the following minerals, hornblende, mica, chlorite, talc, and occasionally schorl ; but gneiss, described under the granitic rocks at p. 2 TO, has the same constituents as granite, and is in many, if not in nearly all, instances, also an altered sedimentary rock. How far its foliation always represents bedding, is, however, a point which does not yet appear to be fully demonstrated. The term gneiss has been applied to many rocks which have not the same mineral constitution as granite, and which should rather be referred to hornblendic, and other, schists, into which, however, gneiss sometimes passes. There are many varieties of gneiss, of which the most important were alluded to under the gra- nitic rocks at page 210, more on account of their mineral constitution than of their origin, mode of occurrence, and structure, which latter would entitle them to be placed in this group. Gneiss 1 is a foliated crystalline aggregate of the same minerals which constitute the different varieties of granite, typically, of orthoclase, plagioclase, quartz, and mica. These minerals are arranged in more or less distinct layers or foliae which are approximately parallel to one another. The mica, especially, forms very distinct, although thin, bands, and it is to this arrangement of the mica that the schistose and often fissile character of the rock is due. Sometimes the mica is a potash, sometimes a magnesian mica, and, at others, both kinds are present. Gneiss varies in colour, the orthoclase in some varieties being red, while in others it is white or greyish. These different rocks are on this account desig- nated red gneiss and grey gneiss, and it has been shown, by analysis of some of the most typical examples, which occur in the neighbourhood of Freiberg, that there is a marked chemical difference between them, the red gneiss containing 1 See also page 211. Gneiss. Granulite. 295 from 75 to 76 per cent of silica, while the grey variety pos- sesses only from 65 to 66. From these analyses the per- centage of the constituent minerals in the two rocks have been deduced as follows : Red gneiss : orthoclase = 60, quartz = 30 mica = 10 per cent. Grey gneiss: =45, =25 =30 The above . represent extremes of variation, while numerous transitional conditions of gneiss exist between them. In some varieties of gneiss the mica, instead of lying in parallel foliae, wraps round lenticular aggregates of felspar and quartz, or round crystals of orthoclase. These varieties are termed pseudo-porphyritic gneiss, eye gneiss (augen gneiss], wood gneiss, &c. Oligoclase, dichroite, garnet, micaceous iron, magnetite, and chlorite sometimes occur plentifully in gneiss, so that they impart a distinct character to the rock, which is then accordingly termed dichroite gneiss, magnetite gneiss, &c. Protogine gneiss. This rock has already been described at page 212. Syenitic or hornblendic gneiss has the same mineral constitution as syenitic granite. The felspar is, in great part, represented by oligoclase. It is a rock of very exten- sive occurrence, and passages have been observed from hornblende gneiss 1 into hornblende schist. Granulite 1 or leptinite is a schistose rock composed of orthoclase and quartz, and contains, as a rule, numerous small garnets. When mica is present the rock assumes a more or less schistose structure, and passes over into gneiss- granulite, or gneiss. Schorl is of common occurrence in this rock. The margins of granitic masses sometimes approxi- mate very closely in character to, or are even identical, in mineral constitution, with, granulite. When this is the case the rock must certainly be eruptive, unless it can be 1 See also page 210. 296 Descriptive Petrology. regarded as an alteration of adjacent sedimentary rocks. At Brazen Tor, on the western margin of Dartmoor, the granite assumes quite a granulitic character, so far as its constituents are concerned, although it shows no foliation or schistose structure, while the eruptive rock which intervenes between it and the sedimentary rocks seems to preclude the idea that its origin is other than eruptive, and that it is anything more than a phase of the granite. 1 The schistose varieties of granulite must, however, be regarded as altered sedi- mentary rocks. Brian rock occurs in mica-slate at Erlhammer, near Schwarzenberg, in Saxony. It is a fine-grained or compact aggregate of garnet, albite, and quartz, and occupies a place intermediate between granulite and garnet rock. Porphyroid. Under this name are included certain altered sedimentary schistose rocks, consisting of a fine- grained matrix of felspar and quartz, and containing a large quantity of a sericite-like or micaceous mineral. Within this matrix lie numerous crystals of felspar (orthoclase or albit) and rounded grains or crystals of quartz. The por- phyroids occur interbedded with other old sedimentary rocks. Seriate- Schist. This is a schistose rock closely allied to the porphyroids, and consists of sericite, fragments of quartz, albite, and usually more or less chlorite and mica. There are two varieties, the green and the red, which differ some- what in composition, the former containing much albite and little or no mica, while the latter contains mica and very little albite. In some cases the rock is rich, in others poor, in quartz. The very fine-grained, compact varieties in which the constituent minerals cannot be distinguished by the naked eye, are termed sericite-slates or sericite-phyl- lites. Occasionally augite may be detected in some varieties of this rock, which, in such cases, are evidently allied to the diabase-schists or tuffs. 1 ' The Eruptive Rocks of Brent Tor,' Memoirs of Geol. Surv. 1878. Mica-Schist. Chlorite- Schist. 297 Mica-Schist is an aggregate of mica and quartz. The relative amounts of these two minerals varies considerably, some varieties of the rock consisting almost wholly of mica, while others are composed almost wholly of quartz, and contain only a very small proportion of mica. The latter approximate and pass into quartz-schist. All of these rocks have a schistose structure due to the parallel arrangement of the crystals and scales of mica. The mica in these schists is sometimes silvery-white potash-mica, sometimes dark magnesian-mica, but the former is by far the most common. The quartz is in small grains, often of a flattened or lenticular form, and it is very usual for the quartz and mica to constitute alternating parallel layers, so that the rock exhibits more or less distinct foliation. The percentage of silica in mica-schists varies according to the amount of quartz which is present, the extremes of fluctuation being between 40 and a little over 80 per cent. Garnets are of common occurrence in these rocks, and numerous other minerals are met with as accessories, such as tourmaline, hornblende, kyanite, staurolite, felspars, epidote, chlorite, talc, magnetite, pyrites, specular-iron, gold, and graphite. When certain of these minerals preponderate, the rocks pass over into schists of a different character, to which special names are given. In some of these rocks calcspar or dolo- mite partly or wholly replaces the quartz and films of argil- laceous matter; chlorite and sericite are also present at times. Such rocks are known as calcareous mica-schist, calcareous chlorite-schist, &c. Itacolumite is a somewhat schistose, micaceous sandstone consisting of granules of quartz and scales of mica, talc, chlorite, &c. The rock is flexible, when in thin slabs. It occurs in N. America and Brazil, and contains, amongst other minerals, gold and diamonds. Chlorite- Schist usually occurs interbedded with gneiss and with other metamorphic rocks. It is of a green or greenish-grey colour, and, as a rule, consists in great part of 298 Descriptive Petrology. scales of mica closely matted together. Quartz, however, is usually present, and felspars, mica, and talc are also of common occurrence. Sometimes it exhibits a fine-grained and evenly- cleavable slaty character (chlorite-slate), but the coarser schistose varieties usually have a somewhat imperfect, or irregular and wavy, fissile structure. The chlorite-schists are frequently rich in accessory minerals, among which may be cited mica, hornblende, actinolite, schorl, epidote, sphene, corundum, garnet, rutile, specular-iron, magnetite, pyrites, copper-pyrites, and gold. Passages occur from chlorite- schist into mica-, talc-, and hornblende-schists, and occa- sionally into rocks of a serpentinous character. Talc-Schist consists of scales of talc with a small admix- ture of quartz. It is a greenish, greyish- white, or yellowish- white rock, very soft, and has a smooth greasy feeling when rubbed with the fingers. Chlorite and mica are often present, and sometimes a little felspar. The rock contains as accessories many of the minerals which occur in chlorite - schist, in addition to which may be mentioned asbestus, hydrargillite, fahlunite, gahnite, chlorospinel, &c. Schorl-Schist is mainly composed of schorl and quartz, and often shows well-marked foliation, the greyish crystal- line-granular rock being traversed by approximately parallel thin black bands of schorl which are often more or less contorted. Cassiterite very commonly occurs in this rock. Mica, chlorite, felspars, topaz, and arsenical pyrites occur as accessory constituents. Schorl rock is a granular, non-foliated rock of similar mineral constitution. Amphibolite- or Hornblende- Schist consists of hornblende and quartz. The latter mineral is often present only in small quantities. It is a dark greenish-grey or iron-grey rock, and when quartz is absent, and it consists exclusively or almost exclusively of hornblende, it is then termed amphi- bolite. Some of these rocks are rather difficult to identify without microscopic examination. The hornblende is some- Coarse Fragment al Rocks. // ,299 times granular, at others in radiating prisms or iih/ a fibrous* condition. Magnetite is common as an-' accessory ,' drid fel- spars, garnet, biotite, epidote, and pyrites arWaJsO .met with/, , although they are always subordinate constituents.' * / Actinolite- Schist may simply be regarded as a variety of hornblende-schist. The constituents are actinolite and quartz. Felspars are occasionally present in amphibolite. These are sometimes triclinic, and the rock then approxi- mates to diorite. COARSE FRAGMENTAL ROCKS. These may be divided into Breccias and Conglomerates. They consist of materials derived from the waste of various rocks and are made up of fragments either angular, or sub- angular, or of rounded, waterworn pebbles or boulders. Similar but much finer material constitutes many of the sedimentary rocks, and it is merely in the size of the frag- ments that many sandstones, grits, &c., differ from breccias and conglomerates, the former being as much entitled to be placed among the fragmentary or clastic rocks as the latter. This view is very clearly represented by Naumann's classifi- cation of the clastic rocks, 1 which he divides into the psephitic (from ^r]0oe, a small stone) ; the psammitic (from ^o'/z^oc, sand) ; and the pelitic (from Trr/Xde, mud). The psammites and pelites of the two last groups are respectively represented by the various sandstones, arkose, &c., and by the tuffs which have already been described in conjunction with the rocks from which they have been derived ; so that it only remains to describe the breccias and conglomerates which constitute the psephitic division of the clastic rocks, although the coarse materials of the latter are often mixed with or cemented by psammitic and pelitic matter. BRECCIAS. The breccias differ from the conglomerates in consisting 1 So named from K\acrT6s, broken. 300 Descriptive Petrology. of angular or sub-angular fragments, instead of rounded water- worn pebbles or boulders ; but it often happens that these coarse clastic rocks have a mixed character ; consisting partly of angular and coniparatively unwater-worn, and partly of rounded water- worn materials. To such rocks the name breccio-conglomerate is sometimes given. The frag- ments of which breccias are composed are usually large enough to permit the recognition of their lithological cha- racter. They are often derived from various sources, frag- ments of sandstone, quartz, jasper, and various eruptive rocks being common, while occasionally they consist of lime- stone fragments, or of broken pieces of bone, as in some of the bone-breccias which occur in cave-deposits. The gene- rally angular character of breccias indicates that they have been formed at or near the spots where they occur and that their materials have never ' travelled far. They may be formed from the superficial disintegration of rocks in their immediate vicinity ; they may represent talus and rubbish heaps which have been subsequently cemented by the infiltra- tion of calcareous, siliceous, or ferruginous matter ; they may result from the breaking in or ' creep ' of rocks with which soluble deposits have been interstratified and subsequently dissolved out, as in the case of the haselgebirge, occurring in the Northern Alps, where, from the removal of underlying salt beds, a breccia has been formed, consisting of fragments of various rocks imbedded in a matrix of clay. Breccias may also result from the breaking-away of small fragments from the sides of fissures (friction breccias) through which dykes of eruptive rock have been subse- quently injected, or into which mineral matter in solution may have subsequently filtered. The latter condition is a very common one, and may be well observed in some metal- liferous lodes. In such cases it is not unusual to find the fragments enveloped in successive deposits of different mineral character. The accumulations of angular and sub- angular ice-worn and scratched stones which constitute the Conglomerates. 301 moraines of glaciers, and the angular unworn rubbish dropped by the melting of icebergs, may also give rise to deposits which may be regarded as unconsolidated breccias. Sometimes, also, they are formed at the bottom of lava-flows, the once molten rock having caught up and enveloped frag- ments of various kinds and sizes ; and, finally, coarse volcanic ejectamenta, or lapilli, may also constitute rocks which may justly be described as volcanic breccias or volcanic agglo- merates. Some breccias, when polished, are well adapted for ornamental purposes. They occur in formations of very different ages, and they have an especial geological interest as indicating either the local disintegration of older rocks, or the transport of materials from distant countries by glacial agency. CONGLOMERATES. These are composed of pebbles or boulders which have been either carried by rivers, or washed about on shores, and consequently rounded by attrition. The process of round- ing is constantly going on in the beds of our rivers and along our sea-coasts ; and the beaches which are formed along our shores, would, if cemented, become true conglo- merates. The materials composing conglomerates are, like those of breccias, of very variable character, being derived from many different sources, but they are mostly formed from rocks of considerable hardness, since softer fragments become totally disintegrated by constant trituration and abrasion. The beaches now forming on the south-eastern coasts of England are in great part composed of pebbles which have resulted from the wear and tear of flints derived from the chalk-cliffs. The puddingstone of Hertfordshire consists of flint pebbles, held together by a siliceous cement. The nagelfluhe, formed on the northern flanks of the Alps, consists, in great part, of limestone fragments, partly mixed with fragments of quartz, granite, gneiss, &c. The new-red conglomerate of the Keuper consists mainly of pebbles and 3O2 Descriptive Petrology. boulders of carboniferous limestone, usually cemented by dolomitic matter, whence it is also called dolomitic conglo- merate. The pebble beds of the Bunter are formed to a large extent of pebbles of quartz and quartzite, in a matrix of sandstone. Conglomerates occur at the base of the old- red-sandstone in Caithness and elsewhere. These contain not merely pebbles, but large boulders, and the deposit is considered by Prof. Ramsay to be of glacial origin. Con- glomerates are sometimes used as building-stones. The dolomitic conglomerate of Triassic age which occurs in the Mendips consists of pebbles and small boulders derived from old-red-sandstone and carboniferous rocks and the cementing material is often, but not invariably, dolomitic. Conglomerates have a special geological interest, inasmuch as they usually represent old sea-beaches, and consequently indi- cate the former existence of coast lines. There, are, however, instances in which they may not represent littoral deposits. CALCAREOUS TUFAS. Calcareous tufa, travertine, pisolite, osteocolla, &c., are deposits formed by the chemical precipitation of carbonate of lime from waters holding bicarbonate of lime in solu- tion. Deposits of this kind are generally formed in the valleys of limestone districts. Any foreign bodies which occur in the solution from which the precipitation takes place, become externally incrusted, just as kettles and boilers become furred internally with carbonate of lime. Successive deposits are thus formed, and the result is a light and often spongy rock, in which more or less distinct layers represent the successive deposits. Twigs, leaves, and other objects become, in this manner, incrusted with carbonate of lime ; and a small trade is carried on at Matlock, Knares- borough, and elsewhere, by submitting natural and artificial objects to the incrusting influence of the waters of these petrifying-wells, as they are termed. The variety of tufa named osteocolla consists of calcareous deposits around Tufas. Siliceous Sinters. 303 twigs and mosses, while pisolite is composed of little pea- like spherical concretions of carbonate of lime around small nuclei. The oolitic limestones have also, doubtless, been formed in a similar manner, although under very different circumstances; the latter representing marine deposits, while the ordinary calcareous tufas are usually formed in valleys, and constitute, as a rule, deposits of very limited extent. In Italy, however, some extensive deposits of travertine occur, especially at Tivoli, where the waters of the Anio have formed beds of tufa four or five hundred feet in thickness. 1 At this place spheroidal concretions from six to eight feet in diameter were observed by Sir Charles Lyell. Calcareous tufa is sometimes used as a building-stone, and appears to be very durable, even in old edifices built by the Romans. Siliceous Sinters. These are rocks formed by the deposition of silica from the waters of hot springs and geysers. Geyserite is a snow- white siliceous sinter of this kind which occurs incrusting the pipes of geysers and forming tolerably thick deposits on the adjacent ground over which the water of the geysers is ejected. The deposits of siliceous sinter at Rotomahana, near Lake Taupo, in New Zealand, are perhaps the most wonderful in the world. At this place the thermal waters charged with silica in solution flow down the hill-sides, forming snow-white terraces of siliceous sinter. The in- fluence which thermal waters, holding silica in solution, have exerted upon many of our older rocks, is a question which well deserves the attention of petrologists. MINERAL DEPOSITS CONSTITUTING ROCK-MASSES. Rock-salt, in some districts, constitutes deposits of great thickness : coal also forms beds or seams of variable thick- ness in the carboniferous series, and in rocks of oolitic and miocene age in certain countries. 1 Lyell's Principle* of Geology, pth edition, p. 244. 304 Descriptive Petrology. Gypsum sometimes forms beds of considerable thickness. Iron ores occasionally occur in large masses, as in the case of the Pilot Knob, Missouri, which consists almost wholly of hematite, of some of the large deposits of this ore at Norberg and Langbanshytta in Sweden, Gellivara in Lap- land, and Santander in Spain, and of the extensive deposits of limonite and clay-ironstone which occur in various parts of the world. Felspars, such as labradorite, sometimes con- stitute, by themselves, rock-masses of considerable thickness and extent. Mr. Bauerman states that, in Labrador, rocks consisting sometimes of labradorite, at others of oligoclase, form large and important beds. Various metallic ores at times constitute lodes and beds of considerable magnitude, but for a description of them the student must refer to larger works, treating more fully upon this part of the subject, which rather belongs to mineralogy and mining than to petrology. > The deposits of cinnabar in Spain, of zinc ores in New Jersey, U.S., and in the Harz, of copper in the Lake Superior district, and of other minerals of the heavy metals which occur more or less plentifully in all parts of the world, the deposits of rock-salt in Russia, Poland, Gallicia, Ger- many, Austria, Spain, England, Algeria, Abyssinia, and in various parts of America, the phosphatic beds met with in certain formations, and the important occurrences of coal, lignite, bituminous schists, asphaltum, and petroleum, are matters which can be studied in most geological, minera- logical, and mining books. Ice, which must be regarded as a rock, occurs in thick and extensive sheets over the Arctic and Antarctic regions, while perennial snow exists at great ele- vations, even at the equator. Ice is sometimes interstratined with sands, &c., as in some parts of Siberia. Deception Island, in New South Shetland, lat. 62 55' S., is principally composed of alternate layers of volcanic ashes and ice, and similar alternations of beds have been observed on Cotopaxi, while a large glacier has been discovered by Gemmellaro, Ice. 305 beneath a lava- stream, at the foot of the highest cone of Etna. The latter phenomenon is described in Lyell's * Prin- ciples of Geology,' and the author attributes the preservation of the glacier to a mass of snow having been covered by a badly conducting layer of volcanic ash, scoriae, &c., prior to the eruption of the lava which now caps the glacier. Ice exhibits stratification in the upper portions of glaciers, but the motion of the latter in their descent to lower levels gradually obliterates these traces of bedding. Glaciers and ice-sheets, in creeping over the subjacent rocks, scrape them into smooth, hummocky forms, termed roches moutonnees, by means of the stones, gravel, &c., which get imprisoned be- tween the ice and the rock-surface over which it moves. The phenomena of glaciation will be found fully discussed in many books and papers which have been published on the subject during the last ten or fifteen years. APPENDIX. A. SINCE this book was written, Mr. Watson, of Pall Mall, has, at my suggestion, constructed a microscope specially suited for petrological work. This instrument is in one respect decidedly preferable to the microscopes commonly made on the Continent, inasmuch as it is supported upon trunnions, like the ordinary English microscopes, and consequently allows of any inclination of the working part of the instrument. The foot, up to the trunnion-bearings, is cast in one piece, after the model of Ross, Crouch, and other well- known makers. Upon this a brass limb is supported. The limb, below the trunnions, is cylindrical, and carries an ordi- nary mirror with a jointed arm. The limb, above the axis, describes such a curve as is most convenient for lifting and carrying the instrument, without incurring any risk of strain to the working parts. The upper portion of the limb is planed to receive the rackwork, which constitutes the coarse adjust- ment, in a manner similar to that employed in the construc- tion of the Jackson-pattern microscopes, as in the instruments by Beck. The tube or body carries the rack, and, by it, is moved against these planed surfaces, for focussing. At the lower end of the tube, immediately above the thread which carries the objective, a slot is cut to receive a Klein's quartz plate. The quartz plate is in a small brass mount, which fits this slot, and can be removed from the instrument at pleasure. At such times a revolving collar can be turned over the aperture. The eye-piece, at the upper end of the tube, is made with a disc, about i| inch diameter, having its outer edge divided, and, immediately above this, a similar disc, connected with the eye- X 2 308 Appendix. piece analyser, revolves with an index, so that the analyser can be set in any desired position, or the amount of rotation imparted to it can be recorded. The eye-piece is also furnished with crossed cobwebs for centering. A space is left between the bottom of the analyser and the eye-glass, sufficient to permit the introduction of a plate of calcspar for stauroscopic examina- tions, and an eye-piece-fitting, without lenses, is also supplied, so that the instrument can, by the superposition of convergent lenses over the polariser, be used for viewing the systems of rings and interference crosses, presented by crystals when examined by convergent, polarised light. The polarising prism is mounted upon a movable arm, beneath the stage, and carries a graduated disc, so that it can be set in any desired direction, or be instantly displaced when ordinary illumination is requisite. The stage of the microscope is circular, and can be rotated and centered. It is divided on silver to half degrees, and a vernier is attached to the front of the stage for goniometric purposes. It has also two rectangular, divided lines, to serve as a finder. The centring is effected by two screws working against a spring on the opposite side. These screws enable the observer to centre the instrument for any objective. They are conveniently situated, so as not to be liable to derangement during the ordinary manipulation of the instrument. The fine adjustment carries a divided head, for the approximate measurement of the thickness of sections. The body of the microscope can be clamped in any position by a lever, attached to the right trunnion. In the instrument which I have examined the adjustments work very smoothly. The thread for the objectives is of the, now, almost universal gauge, so that any English objectives may be used. Foreign ones can also be employed, by means of an adapter. The engraving on page 306, for which I am indebted to Mr. Watson, shows the general plan upon which the instrument is constructed. The smaller figures represent the quartz plate, the calcspar plate, and a section of the polariser and its fittings. Appendix. 309 B Books consulted in the preparation of this Work. Ansted, D. T. ' Elementary Course of Geology and Mineralogy.' London, 1850. Ansted, D. T. ' Applications of Geology to the Arts and Manu- factures.' London, 1865. Blum, J. R. Handbuch der Lithologie oder Gesteinslehre.' Erlangen, 1860. Boficky, E. ' Petrographische Studien an den Basaltgesteinen Bohmens.' Prag, 1874. Boficky, E. ' Petrographische Studien an den Phonolithge- steinen Bohmens.' Prag, 1873. Boficky, E. ' Petrographische Studien an den Melaphyrge- steinen Bohmens.' Prag, 1876. Boficky, E. c Elemente einer neuen chemisch-mikroskopischen mineral- und Gesteins-Analyse.' Prag, 1877. Brewster, D. l Optics' (' Cabinet Cyclopaedia'). London, 1831. Bristow, H. W. * Glossary of Mineralogy.' London, 1 86 1. Bryce, J. ' Geology of Arran and other Clyde Islands.' Glas- gow and London, 1872. Coquand, H. ' Traitd des Roches.' Paris, 1857. Cotta, B. von. ' Rocks Classified and Described.' Translation by P. H. Lawrence. London, 1866. Credner, H. 'Elemente der Geologic' Leipzig, 1876. Dana, J. D. ' System of Mineralogy.' Fifth Edition. London and New York, 1871. De la Beche, H. ' How to Observe Geology.' London, 1836. De la Beche, H. ' Report on the Geology of Cornwall, Devon, and West Somerset.' London, 1839. De la Beche, H. ' Researches in Theoretical Geology.' London, 1834. De la Fosse. ' Nouveau Cours de Mineralogie.' Paris, 1858-62. 3io Appendix. Delesse, A. ' Recherches sur 1'Origin des Roches.' Paris, 1865. Delesse, A. l Recherches sur les Pseudomorphoses.' l Annales des Mines/ XVI. Paris, 1859. Delesse, A. and De Lapparent. ' ReVue de Gdologie.' Paris. Descloizeaux, A. ' Manuel de Mineralogie.' Paris, 1862. Fischer, H, ' Kritische mikroskopisch-mineralogische Studien.' Freiburg, 1869 and 1871. Ganot. ' Elementary Treatise on Physics.' Translation by Atkinson. London, 1863. Green, A. H. ' Geology for Students.' Parti. London, 1876. Hunt, R. ' Mineral Statistics of the United Kingdom, 1858. Part II. Embracing Clays, Bricks, &c., Building and other Stones.' (Mem. Geol. Surv.) London, 1 860. Jannetaz, E. ' Les Roches.' Paris, 1874. Judd, J. W. ' Geology of Rutland, &c.' (Mem. Geol. Surv.) London, 1875. Jukes, J. B. ' Student's Manual of Geology.' Edinburgh, 1862. Kengott, A. ' Elemente der Petrographie.' Leipzig, 1868. Kinahan, G. H. ' Handy-Book of Rock Names.' London, 1873. La Vallde Poussin and Regard, A. ' Mdmoire sur les Roches dites Plutoniennes de la Belgique et de 1'Ardenne Franchise.' (Acad. Royale des Sciences de Belgique.) Bruxelles, 1876. Lasaulx, A. von. ' Elemente der Petrographie.' Bonn, 1875. Lommel, E. ( The Nature of Light, with a General Account of Physical Optics.' London, 1875. Lyell, C. ' Principles of Geology.' 9th Edition. London, 1853. Lyell, C. Student's Elements of Geology.' London, 1871. Macculloch, J. l A Geological Classification of Rocks.' London, 1821. Mohl, H. ' Die Basalte und Phonolithe Sachsens ' (Nova Acta). Dresden, 1873. Pereira, J. P. 'Lectures on Polarised Light.' London, 1843. Phillips, J. ' Manual of Geology.' London and Glasgow, 1855. Appendix. 31 1 Pinkerton, J. < Petralogy.' London, 1811. Playfair, J. ' Illustrations of the Huttonian Theory of the Earth/ Edinburgh, 1802. Ramsay, A. C. ' Physical Geology and Geography of Great Britain/ 3rd Edition. London, 1872. Renard, A. ' Memoire sur la Structure et la Composition Mineralogique du Coticule.' (Acad. Royale des Sciences de Belgique.) Bruxelles, 1877. Rosenbusch, H. ' Die Steiger Schiefer und ihre Contactzone an den Granititen von Barr-Andlau und Hohwald.' Strass- burg, 1877. Rosenbusch, H. ' Mikroskopische Physiographic der petro- graphisch wichtigen Mineralien.' Stuttgart, 1873. Rosenbusch, H. ' Mikroskopische Physiographic der massigen Gesteine.' 1877. Roth, J. ' Die Gesteins-Analysen.' Berlin, 1861. Scrope, G. P. 'Volcanos of Central France.' London, 1858. Smith, C. H. * Lithology, or Observations on Stone used for Building.' London, 1845. Spottiswoode, W. ' Polarisation of Light ' (Nature Series). London, 1874. Tyndall, J. ' Six Lectures on Light.' Delivered in America. London, 1875. Tyndall, J. ' Notes on a Course of Nine Lectures on Light.' Delivered at the Royal Institution. London, 1872. Vogelsang, H. l Die Krystalliten.' Bonn, 1875. Ward, J. C. Geology of the Northern Part of the English Lake District.' (Mem. Geol. Surv.) London, 1876. Whitaker, W. ' Geological Record.' London. Woodward, C. ' Familiar Introduction to the Study of Polarised Light.' London, 1861. Woodward, H. B. ' Geology of England and Wales.' London, 1876. Woodward, H. B. ' Geology of East Somerset and the Bristol Coal-Fields.' (Mem. Geol. Surv.) London, 1876. 312 Appendix. Zirkel, F. ' Lehrbuch der Petrographie.' Bonn, 1866. Zirkel, F. ' Microscopic Petrography.' (U. S. Geol. Explora- tion of Fortieth Parallel.) Washington, 1876. Zirkel, F. ' Mikroskopische Beschaffenheit der Mineralien und Gesteine.' Leipzig, 1873. Zirkel, F. ' Untersuchungen iiber mikroskopisch Zusammen- setzung und Structur der Basaltgesteine.' Bonn, 1870. Also numerous papers published in the ' Quarterly Journal of the Geological Society/ the ' Geological Magazine,' ' Leonhard's Jahrbuch fiir Mineralogie/ 'Zeitschrift der deutschen geolo- gischen Gesellschaft,' 'Annales des Mines,' and numerous other publications. Among these, books and papers by the following authors have been specially consulted : Allport, S. Bonney, T. G. Boficky, E. Cotta, B. v. Dana, J. D. De la Beche, Sir H. Delesse, A. Forbes, D. Judd, J. W. Kengott, A. Lasaulx, A. v. Lyell, Sir C. Phillips, J. A. Rdnard, A. Rosenbusch, H. Sorby, H. C. Zirkel, F. Errata. Page 8, line 20, for marine read submarine. I0 3 33> ,, regarded sometimes regarded. 220, 33, ,, basalt mica-basalt. INDEX. ACI BAT BYS ACID, sub class of eruptive rocks, 34, Arenaceous limestone, 20 sedimentary rocks, 276 Bath stone, 286 Baveno type of twinning, 177 Argillaceous limestone, 92 Actinolite, 131 20 Beale, Lionel S., 59 schist, 295 rocks, 282 Beale's reflector, 49 Aden, globular silica in Arkose, 280 Belonites in pitchstone, quartz-trachyte of, 152 Age of strata, determined Arran, pitchstone of, 196 Arranging rock speci- 190 Berkum, near Bonn, by means of fossils, 30 mens, 39 sanidine-rhyolite of, Aislaby sandstone, 280 Asbestus, 131 224 Albite, 97 Ascension, obsidian of, Binstead limestone, 289 type of twinning, 98 181 Biotite, 135 Alderley Edge, sandstone Ashes, volcanic, 267 Blowpipe apparatus, 45 of, 279 Asparagus stone, 147 Bobenhausen, tachylyte Allport, S., 107, 122, 152, Aubuisson, 209 of, 201 182, 190, 193, 197, 245, Augen-gneiss, 295 Bombs, volcanic, 267 260 Augite-andesite, 236 Bonney, T. G., 182, 193, Altered conditions of py- porphyrite, 240 211, 260, 262, 270 roxene, 126 syenite, 218 Boficky, ., method of eruptive rocks, 269 Aussig, tridymite from, analysis, 100 sedimentary series, 152 120, 125, 132, 229, 289 Auvergne, domite of, 226 253, 260 Amazon stone, 97 Axes of elasticity, 78 Bouteillenstein, 187 Amianthus, 131 optical, 78 Box Hill stone, 286 Amorphous matter, 170 Axiolitic structure in vit- Bradford flags, 278 Amphibole, 127 Amphibolite, 298 reous rocks, 184 .Bramley Fall stone, 278 Brazen Tor, granulitic Amphigene, 108 rock of, 296 Analcime, 159 BAGSHOT sands, 281 Breccias, 299 Anamesite, 252 Banded structure in defined, 17 Andalusite, 143 vitreous rocks, 181 Breislackite, 125 Andesine, 99 Andesite, 234 Bangor quarries, 283 Bannisdale slates, 283 Breithaupt, 97 Brent Tor, schalstein Anisotropic substances, 77 Bargate stone, 280 near, 248 Anorthite, 98 Basalt, 252 volcanic ejecta- Ansted, 277 classification of, 253 menta of, 268 Anticlinal, term defined, columnar structure of, Brewster, 142, 150, 159 21 Antrim, chalk of, 290 258 diagram of deviations Brezina, stauroscope of, 83 keuper sandstone of, from, 261 Bristol Coal-Field, sand- 280 Basaltite, 252 stones of, 278 Apatite, 145 Basanite, 293 Buch, L. von, 171, 234 Aplite, 211 Basic rocks, 34 Bunsen, 250 Aqueous rocks defined, sub-class of rocks, 177 Bunter sandstones, 279 15 Basis, term defined, 168 Buschbad, perlite of. 183 Arbroath sandstone, 277 Bath bricks, 281 Byssolite, 131 Index. CAB DAR ERL CABINETS, 43 Caithness flagstone, Cliffs and escarpments, 27 Dathe, J. F. E., 87, 245 Daubeny, 37 277 Clinkstone, 228 Definition of the term Calabria, mica syenite of, Clinochlore, 137 rock, 6 218 Clip-lens, 44 De la Beche, Sir H.,37 Calc-aphanite, 247 Calcareous rocks, 285 Coal, preparation of sec- tion of, 71 De la Fosse, 78 Delesse, A., 130, 171, 221, sandstone, 19 Coast lines affected by 269 tufa, 1 8, 302 relative hardness of Dent marble, 285 Calcite, 148 rocks, 27 Denudation, 25 Calcspar, 148 fluid inclosures in, 149 twinning of, 149 Cambrian sandstones, 276 changes in, u Collecting rock speci- mens, 39 Columnar structure of Descloizeaux, 96, 97, 120, 122, 135, ISO, 159, l6o Devitrification, 185 of pitchstones, 198 slates, 283 basalt, 258 Devonian limestones, 285 Canada, gneissic syenite of, 218 in phonolite, 233 Collyweston slates, 283 sandstones, 277 slates, 284 moroxite of, 147 Camera lucida, 49 Condensers, 49 achromatic, 50 Devonshire, schalstein of, 248 Canary islands, eutaxite Conglomerate, 17, 19, Diabase, 244 of, 232 297 amygdaloidal, 247 Cancrinite, 108 dolomitic, 302 aphanite, 247 Cape Wrath, amphibolite schist of, 218 Carbonic acid, action of, phonolite, 233 Coniston flags, 283 Contortion of strata, 21 porphyrite, 239 schist, 247 Diabasmandelstein, 247 upon certain rocks, 29 Carboniferous limestone, Cookworthy, Wm., 273 Copper pyrites, 157 Diallage, 124 andesite, 234 285 Corals, 1 8 Diallagoid augite, 132 sandstone, 278 Cordier, 209 Dimetian basalts, 262 slates and flags, 284 Carlow flags, 278 Cordierite granite, 210 Cornubianite, 213 Dinorwig quarries, 283 Diorite, 241 Carlsbad type of twin- Corriegills, Arran, pitch- porphyrite, 238 ning, 92 stone of, 197 Dip, term defined, 21 Cassiterite, 148 Corsham Down stone, 286 Disthene rock, 263 Craigleith sandstone, 278 Coticule, 142 Disturbances of the Cast-iron plate for grind- ing sections, 61 Cotopaxi, 304 Cotta, B. von, 219, 220 earth's crust, 9 _Dleiite, 252 Centering 57 Credner, H., 206, 288 Dolomites, 285 Chabasite, 160 Cretaceous sandstones. Dolomitic conglomerate, Chalcopyrite, 157 280 302 Chalk, 289 Cross-hatching in ortho- Domite, 226 Chalk of Antrim, 290 clase, 94 Doulting stone, 286 Chert, 285 Crust of the earth, dislo- Drachenfels, tridymite Chiastolite, 143 cations of, ii from, 152 slate, 291 Crypto-crystalline matter, Dressel, 114 China clay, 272, 284 170 Dundee sandstone, 277 formation of, 30 Crystalline eruptive rocks, Dunite, 265 Chips for section-cutting, 202 Durocher's theory, 34 60 limestones, 292 Chlorite, 136 Crystallites, 160 in quartz, 151 schist, 297 in obsidian, 187 Cubic system, cleavages EARTHQUAKES, n Eifel, leucitophyrs Chrysolite, 116 in, 171 of, 2 6 pseudo, 187 Czertochin, Bohemia, Eisenglimmer, 155 Classification of rocks, tachylyte of, 202 Eklogite, 263 174 Elba, specular iron of, Clastic rocks, 299 !55 Clays, 17, 282 DACITE, quartzose, Elaeolite, 108 of Bovey Beds, 284 234 Elevation and subsidence &c., washing of, 73 Damascened structure in of land, ii Cleavage, cause of, 35 vitreous rocks, 181 El van, 204, 210 in rocks, 12 Damourite, 134 Encrinital marble, 285 Cleavages, diagrams of, Dana, J. D., 90, 117, Enstatite, 120 of minerals, 171 121, 131, 147, 257 Darwin, C, 36, 37, 212 Epidote, 127, 139 Erlan, 296 Index. 315 ERU HAP INT Eruptive rocks, general characters of, 32 Friction breccias, 300 Fritsch and Reiss, 233 Hardness, scale of, 44 Hart Hill, sandstone of, Eruptive rocks, vitreous, Fntzgartner, M. G. R., 278 177 117, 146 Hartley, W. N., 164 Escarpments and cliffs, Fruchtschiefer, 292 Haselgebirge, 300 27 Fuchsite, 134 Haughton, S., 113, 207, Estimation of amount of Fuller's earth, 284 208, 257 waste by denudation, Fusion of vitreous matter, Hauyne, 112 24 179 and nosean, 232 Etna, ashes of, 267 basalt, 257 glacier on, 304 Hauynophyr, 257 lavas of, 255, 267 /~* ABBRO, 249 Hawaii, filiform lava of, Eulysite, 263 vJT Garbenschiefer, 292 186 Eurite, 209, 214 Garnet, 140 obsidian of, 191 Eutaxite, 233 rock. 262 Headon Hill sands, 281 Eye-pieces, 52 Gemmellaro, 304 Hematite, 155, 304 Geyserite, 303 Henslow, 212 Geysers, 33, 303 Hertfordshire, pudding- FALSE bedding, 15 Faroe Isles, heu- Giant's Causeway, basalt of, 258 stone of, 301 Heulandite, 160 landite of, 160 Glass inclosures, 165 Hexagonal system, cleav- Faults, origin of, 13 Felsi-dolentes, 34 natural, 170 Globular silica, 152 ages in, 172 Hills and mountains, P'elsitic matter, 167 Globulites, 161 causes affecting their pitchstone, 197 Felsite, 168, 213 Gneiss, 212, 294 augen, 295 forms, 29 and valleys, formation Felspars, 86 _ ' foliation in, 212 of, 28 decomposition of, 30 protogine, 212, 295 Himalayas, sandstone of, Felspar-porphyries, 209 rocks, 304 syenitic, 295 Gneissic syenite, 218 281 Homogeneous vitreous Felstune, 209, 214 Goniometers for micro- rocks, 1 80 Ferrite, 167 scopes, 53 Horizontal strata, map Ferruginous sandstone, Granite, 202 and section of district 19 Filiform condition of vit- diagram of deviations from, 215 composed of, 22 denudation of, 23 reous lavas, 1 86 origin of, 206 Hornblende, 128 Finders, 51 Granitell, 211 andesite, 234 Fingal's Cave, 258 Granitic type, deviations schist, 295, 298 Fire-stone, 281 from, 213 syenite, 217 Flags, 282 Granitite, 210 Hornblendic granite, 203 Silurian, 283 Granular diabase, 247 Hydrometamorphism, 208 Fleckschiefer, 292 Granulite, 211, 295 Hypersthene, 119 Flexure of strata, 21 Green, A. H., 37 andesite, 234 Flints, 281 Greenstone, 241 Hypersthenite, 250 Fluid inclosures, 164 tuffs, 249 Hull, E., 257, 277 Fluxion structure, 163 Greisen, 211 Hungary, tridymite of, Foliated rocks, 293 Grinding of rock sections, 152 Foliation, 36 65 Hunt, R., 280 Foraminifera, 18 Grit, 17, 19 T. Sterry, 36, 271 Forbes, David, 2 Groth, 140, 150 theory of fels- Forces operating in the Groundmass, term de- pars, 95 interior of the earth, 10 fined, 1 68 Formation, term defined, Gas inclosures, 164 30 Gumbel, 155, 248 TCE, 304 Fossiliferous rocks de- Gypsum, 304 _L Idocrase, 142 fined, 15 Inclined strata, denuda- Fossils indicative of con- tion of, 23 ditions under which rocks have been depo- HALB-GRANIT, 211 Halleflinta, 209, Inclosures in quartz, 150 of fluid, 164 sited, 15 214 of glass, 165 Fossils replaced by py- Halifax flags, 278 Indicator for eye-piece, 52 rites, 157 Hammers, 40 Internal heat of the earth, Foster, C. Le Neve, 212 Hassock, 281 evidence of, 10 Fragmental rocks, 299 Hastings sands, 280 Interpositions in felspars, Freiberg, gneiss of, 294 Haplite, 211 95 Index. INT MAR Intrusive sheets distin- Laminar fission, 12 guished from lava-flows, Lapilli, 267 3 2 Lasaulx, A. von, 59, 103, Iron glance, 155 114, 119, 133, 195, 216, ores, 304 218, 231, 232, 234, 247, pyrites, 156 255, 264, 290 Isle of Wight, tertiary Latente, 272 limestone of, 289 Isotropic substances, 76 Lava-flows distinguished from intrusive sheets, Itacolumite, 297 3 2 Italian marbles, 287, 293 Lavas of Etna, 255 Vesuvius, 256 Lebour, G. A., 32 TADE, 131 Leeson's microscope, 59 J Jasper, porcelain, 290 Lepidolite, 134 Lepidomelane, 136 Jervis, W. P., 293 Leptinite, 211, 295 Jointing, 12 Jordan's section cutter, Leucilite, 256 Leucite, 108 61 basalt, 256 Judd, J. W., 37, 191, 193, Leucitophyr, 256 262, 280 Leucoxene, 155 Jukes, J. B., 13 Jurassic sandstones, 280 Levigation, 73 Levy, Michel, 152 Lherzolite, 263 Liassic clays, 284 KALKOWSKY, 171 limestones, 286 Kaolin, 203, 272, Limestones, 17, 20 284 Limestone, carboniferous, formation of, 30 285 Kaolinite, 284 cretaceous, 288 Katzenbuckel, nepheli- crystalline, 287, 293 nite of, 255 Devonian, 285 Kengott, 120, 219 Kersantite, 220, 239 magnesian, 20, 285 oolitic, 286 Kersanton, 220, 239 Silurian, 285 Ketton stone, 286 tertiary, 289 Keuper sandstone, 279 Limonite, 156 Kieselschiefer, 293 Lincolnshire oolites, 286 Kilanea, lavas of, 199 Lipari, obsidian of, 183, Kinahan, G. H., 209, 238 189 Kinzigite, 262 Liparite, 222 Kjerulf, 257 Lithia mica, 134 Klaproth, 228 Lithoidite, 222 Klein's quartz-plate, 58 Knaresborough, calcare- Llanberis quarries, 283 London clay, 284 ous tufa of, 302 Knotenschiefer, 292 Luxullianite, 210 Lydian stone, 293 Kobell, V., stauroscope Lyell, SirC., IT, 37, 303, of, 81 305 Kupferberg, bronzite of, 121 Kyanite, 144 in quartz, 151 MACHINES for grinding micro- scopic sections, 61 Made, 143 LAACHER See, hauyne-basalt 0^257 Magma, term defined, 1 68 hauyne of, 114 Magnesian limestone, 20, Labelling specimens, 42 285 Labradorite, 99 Magnetite, 153 Lake District, mica traps of, 220 Magnets, 45 Maltwood's finder, 51 - volcanic ejecta- Marble, encrinital, 285 menta of, 268 Marbles, Italian, 287, 293 NEP Marbles, microscopic cha- racter of, 287 Marcasite, 157 Margarodite, 135 Marialite, 112 Marine denudation, effects of, 25 Marl, 19 Maskelyne, N. S., 120 Matlock, calcareous tufa of, 302 Medway, mud of, 285 Meionite, 112 Melaphyre, 260 Merrifield, C.W., 259 Metamorphism, 36, 208 Mica basalt, 257 Micaceous felstone, 209 Micas, 132 Mica schist, 297 syenite, 218 Microcline, 96 Micro-crystalline matter, felsitic matter, 171 Microliths, 162, 185 in perlite, 194 Microscopes, 46 Microscopic analysis of E. Boiicky, 100 preparations, 59 Millstone lavas, 256 Mineral deposits, 303 Minerals, optical charac- ters of, 74 rock-forming, 86 Minette, 219 Miocene sandstones, 281 Missouri, hematite of, 156, 304 Mohl, H., 152, 252 Monoclinic system, cleav- ages in, 172 Moraines, 301 Moroxite, 147 Morris, J., 281 Mountains and hills, causes affecting their forms, 29 Mounting sections of rocks, &c., 69 Mudstones, 17 Murchison, Sir R. I., 212 Muscovite, 133 Museums, arrangement of rock collections in, 43 NAPLES, piperno of, 2 33 Nassau, schalstein of, 248 Natrolite, 159 Negative crystals, 165 Nepheline, 104 alteration of, into na- trolite, 159 Index. 317 NEP PYR Nepheline basalt, 255 Perlitic structure, 182 Pyrometamorphism, 208 Nephelinite, 255 Neurode, hypersthenite . in tachylyte, 194 Permian limestones, 285 Pumice, 191, 267 Purbeck limestone, 286 of, 251 sandstones, 278 marble, 286 Nevada, U.S., vitreous rocks of, 184 Perthite, 97 Petrosilex, 209, 214 Pyramids, nummulitic limestone of the, 289 New red conglomerate, Petworth marble, 287 Pyroxene, 121 301 Phillips, J., 36 sandstone, 279 J. A., 37, 70, 151, 211 Newton, E. T., 71 Niedermendig, lava of, Phlogopite, 135 Phonolite, 228 QUARTZ, 149 diabase, 247 256 classification of, 229 inclosures in, 150 Nile mud, 282 conglomerate, 233 porphyry, 210 Norite, 251 tuff, 233 rhyolite, 223 Normal sedimentary wacke, 233 trachyte, 223 rocks, 275 Phosphoric acid, detec- Quartzite, 293 Northampton sand, 280 tion of, 145 Quartzose dacite, 234 North Elmsley, Canada, moroxite of, 147 Picrite, 265 Pilot Knob, Missouri, 156 Quartzless diabase, 245 hornblende andesite, Nosean, 112 Piperno, 233 235 and hauyne, 232 Pisolite, 1 8, 303 trachyte, 223 Nose-pieces, 52 Pitchstone, 195 devitrification of, 198 felsitic, 197 RAIN, effect of, upon OBJECT-glasses, 48 Obsidian, 186 trachytic, 195 Plagioclase, 91 rocks, 29 Rammelsberg, 150 crystallites in, 187 basalt, 253 Ramsay, A. C., 36, 212, spherulites in, 188 enstatite rocks, 251 302 Old red conglomerate, 302 Plagioclastic felspars, 91 Rath, G. vom, 108, 250 sandstone, 277 Pliny, 203, 217 Reiss and Fritsch, 233 Oligoclase, 99 Plutonic and volcanic Renard, A., 59, 86, 135, diorite, 241 rocks, 33 142, 165, 280 Olivine, 116 Pocket lens, 44 Reusch, 149 alteration of, 264, 271 Polarisation of light, 75 Rhombic system, cleav- basalt, 253 Polarising apparatus, 48, ages in, 172- gabbro, 249 75 Rhyolite, 178 Oolite, 286 Poole clay, 284 Richthofen,Von, 178, 193, Oolitic limestones, 286 Porcelain jasper, 290 2 37 Opacite, 166 Porphyrite, 237 Roberts, W. Chandler, 258 Optical axes, 78 properties of minerals, Porphyritic structure in vitreous rocks, 185 Rodwell, G. F., 267 Rosenbusch, H., 108, 142, 74 Porphyroid, 296 143, 144, 148, 152, 159, determination Portland cement, 285 160, 167, 169, 170, 171, of, 79 oolites, 286 196, 199, 2ig, 221, 228, Orthoclase, 92 Potstone, 271 232, 245, 249, 254, 264 Orthoclastic felspars, 91 Poussin, Ch. de la Valee, Rosenbusch's microscope, Osteocolla, 302 280 54 and Renard, 135 Practical value of petro- Roth, J., 178, 234 Rocche Rosse, obsidian PACHUCHA, Mexico, tridymite from, 152 logical research, 3 Pre-Cambrian basalts, 262 of, 184, 189, 191 Rock salt, 304 Palagonite rock, 272 Preliminary examination in fluid inclosures, luff, 272 of rocks, 44 165 Paragonite, 135 Propylite, 237 Rochdale flags, 278 Paranthine, in Protogine gneiss, 212, Roches moutonnees, 305 Pearlite, 192 295 Rocks, condition under Pearlstone, 192 granite, 213 which formed, 6 Pele's hair, 186 Provisional names applied general characters of, 6 Pelitic rocks, 299 to minerals, 166 Rotheram stone, 278 Penck, A., 268 Pennant grit, 278 Penrhyn quarries, 283 Psammitic rocks, 299 Psephitic rocks, 299 Pseudo-chrysolite, 187 Rothliegende, 278 Rothweil, analcime from, 160 Penrith sandstone, 279 Percy, J., 285 Pseudomorphs, 30 Puddingstone, 301 Roto-mahana, 33, 303 Rounded crystals, 178 Perlite, 192 Pyrites, 156 Rounding of stones, 16 Index. RUT TAB VER Rutile, 147 Silurian slates, 283 Tachylyte, 199 in quartz, 151 Sinter, siliceous, 19, 33, Talc, 137 33 schist, 298 Skiddaw slates, 283 Taupo Lake, siliceous SAINT Bees sand- Skye, hypersthenite of, sinter of, 303 stone, 279 251 Tawney, E. B., 262 Salt, crystals of, in fluid Slabs, 282 Tertiary limestones, 289 inclosures, 165 Slates, 17, 282 sandstones, 281 Sands, 17 Cambrian, 283 Tetragonal system, cleav- Sand, volcanic, 267 Silurian, 283 ages in, 171 Sandberger, 254 Sand-rock, 19, 280 Slaty cleavage, cause of,35 Slicing rocks with dia- Thames, mud of, 285 Thermal springs, 33 Sandstones, 17, 276 mond dust, 63 Thickness of beds, mea- Cambrian, 276 carboniferous, 278 Slievenalargy, tachylyte of, 200 surement of, 24 of the earth's crust, 10 cretaceous, 280 Smith, Lawrence, micro- Tin stone, 148 Devonian, 277 scope of, 59 Tintagel Quarries, 284 Jurassic, 280 Sphene, 140 Titaniferous iron, 154 old red, 277 Sodalite, 112 Titanite, 140 oolitic, 280 of Somma, 116 Tivbli, travertine of, 303 Silurian, 276 Sorby, H. C, 2, 36, 107, Tolcsva, obsidian of, 181 tertiary, 281 Sandwich Islands, lavas 142, 151, 224 Some, Isle of Mull, ta- Topaz, 142 Topley, W., 32, 281, 289 of, 199 chylyte of, 202 Tourmaline, 137 Sanidine, 94 South Burgess, Canada, Trachy-phonolite, 279 ' rhyolite, 224 moroxite of, 147 Trachyte, 221 -- trachyte, 224 Specimens, collection of, proper, 225 Scapolite, in 39 Trachytes, classification Scenery, on what depen- dressing of, 41 of, 222 dent, ii Schalstein, 248 Specular iron, 155 Spherulites in obsidian. Trachytic pitchstone, 195 Travertine, 18, 303 Scheerer, 108 1 88 Tremolite, 131 Schistose rocks, 293 Spherulitic structure, 183 Triassic sandstones, 279 Schmidt's goniometer, 53 Spilite, 247 Trichites, 162, 185 Schorl, 138 Stache, 237 Triclinic ' system, cleav- rock, 265, 298 Statuary marble, 287 ages in, 172 schist, 298 Staurolite slate, 292 Tridymite, 152 Scotch slates, 284 StaUroscope, 81 Tripoli, 281 Scrope, G. Poulett, 37, Stauroscopic examina- Tschermak, 88, 95, 103 228 tion, 84 Tufa, calcareous, 18, 302 Sedimentary matter, sort- Stelzner, 171 Tuff", greenstone, 249 ing of, in water, 16 rocks, 274 Stockwerksporphyr, 211 Stone inclosures, 165 phonolite, 233 Twinning of calcspar, 149 classification of, 19 Strata, flexure of, 21 felspars, 87, 102 denned, 15 Stratigraphical breaks, 31 Seifersdorf, minette of, Streak of minerals, 45 219 Striations in labradorite, T TNCONFORMITY, Seismology, 9 102 \^J Stratigraphical, 31 Semi-granite, 211 Strike, term denned, 21 United States, propylite Serpentine, 269 Structures in basalt, &c., of, 237 Sericite, 134 due to contraction, 14, rhyolite of, 185 - schist, 296 259 Setton, les, 152 vitreous rocks, 180 Shales, 17, 282 Sharp, D., 36 Structural planes, 12 Subsidence and elevation T 7ARIOLITE, 248 V Vibration, principal Silica, globular condition of land, ii directions of, in crys- of, 152 Syenite, 203, 217 tals, 58 percentage of in erup- tive rocks, 177 Syenitic gneiss, 295 granite, 203 Vitreous rocks, 177 phenomena effusion Siliceous breccia, 19 limestone, 20 Synclinal, term defined, 22 Szabo, loo in, 179 Ve'lain, M., 152 sinter, 19, 33, 303 Veltlin, hypersthenite of, Silurian flags, 283 25 1 limestones, 285 HP ABLE of cleavages Verde, antique porphyry, sandstones, 276 J. in minerals, 171 240 Index. 319 VES WOL Vesuvian, 142 lavas, 257 Vesuvius, leucitophyrs of, WA D S W O R T H, U.S., rhyolite of, 185 256 Ward, J. C., 34, 144, 249, nepheline of, 107 Viridite, 166 291 Washing of fine deposits, Vitreous eruptive rocks, 73 177 Vogelsang, H., 113, 119, Watcombe clay, 284 Wealden marbles, 289 142, 163, 171, 197, 201 Weathering of rocks, 29 Volcanic phenomena, 38 Weiss-stein, 211 and plutonic rocks, de- Welsh slates, 283 fined, 33 Wernerite, in ashes, 267 Whin Sill, 258 bombs, 267 Wickersley stone, 278 ejectamenta, 266 Witham, H., sections of sand, 267 fossils first prepared Volcanoes, general cha- racters of, 37 Wolf rock, 107 Vom Rath, 152 tridymite in, 152 Vulcanicity, 9 Wolff, Th., 171, 266 ZWI Wollaston's prism, 49 Woodward, H. B., 282 Woolwich and Reading clays, 284 \7ORKSHIRE flags, 1 278 Jurassic sandstones of, 280 yEOLITES, 158 / * Zircon, 143 Zirkel, F., 106, 107, io1, 114, 121, 125, 129, 130, 146, 147, 160, 166, 167, 168, 184, 197, 201, 219, 221, 228, 237, 249, 251 Zwitter rock, 211 LONDON : PRINTED BY SPOTT1SWOODE AND CO., NEW-STREET SQUARE AND PARLIAMENT STREET *r AY USE Mm K O CN CN ^N cS" < CO CO Ul U Z UJ t/C ^. CN ^3 in X u ! -t 8 1*